From 86f578de5ba6ea11ead9284d9f036fee01ba5893 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Thu, 15 Nov 2007 19:00:06 +0800 Subject: [CRYPTO] doc: Update api-intro.txt This patch updates the list of transforms we support and clarifies that the Block Ciphers interface in fact supports all ciphers including stream ciphers. It also removes the obsolete Configuration Notes section and adds the linux-crypto mailing list as the primary bug reporting address. Finally it documents the fact that setkey should only be called from user context. Signed-off-by: Herbert Xu --- Documentation/crypto/api-intro.txt | 41 +++++++++++++++----------------------- 1 file changed, 16 insertions(+), 25 deletions(-) (limited to 'Documentation') diff --git a/Documentation/crypto/api-intro.txt b/Documentation/crypto/api-intro.txt index a2ac6d294793..8b49302712a8 100644 --- a/Documentation/crypto/api-intro.txt +++ b/Documentation/crypto/api-intro.txt @@ -33,9 +33,16 @@ The idea is to make the user interface and algorithm registration API very simple, while hiding the core logic from both. Many good ideas from existing APIs such as Cryptoapi and Nettle have been adapted for this. -The API currently supports three types of transforms: Ciphers, Digests and -Compressors. The compression algorithms especially seem to be performing -very well so far. +The API currently supports five main types of transforms: AEAD (Authenticated +Encryption with Associated Data), Block Ciphers, Ciphers, Compressors and +Hashes. + +Please note that Block Ciphers is somewhat of a misnomer. It is in fact +meant to support all ciphers including stream ciphers. The difference +between Block Ciphers and Ciphers is that the latter operates on exactly +one block while the former can operate on an arbitrary amount of data, +subject to block size requirements (i.e., non-stream ciphers can only +process multiples of blocks). Support for hardware crypto devices via an asynchronous interface is under development. @@ -69,29 +76,12 @@ Here's an example of how to use the API: Many real examples are available in the regression test module (tcrypt.c). -CONFIGURATION NOTES - -As Triple DES is part of the DES module, for those using modular builds, -add the following line to /etc/modprobe.conf: - - alias des3_ede des - -The Null algorithms reside in the crypto_null module, so these lines -should also be added: - - alias cipher_null crypto_null - alias digest_null crypto_null - alias compress_null crypto_null - -The SHA384 algorithm shares code within the SHA512 module, so you'll -also need: - alias sha384 sha512 - - DEVELOPER NOTES Transforms may only be allocated in user context, and cryptographic -methods may only be called from softirq and user contexts. +methods may only be called from softirq and user contexts. For +transforms with a setkey method it too should only be called from +user context. When using the API for ciphers, performance will be optimal if each scatterlist contains data which is a multiple of the cipher's block @@ -130,8 +120,9 @@ might already be working on. BUGS Send bug reports to: -Herbert Xu -Cc: David S. Miller +linux-crypto@vger.kernel.org +Cc: Herbert Xu , + David S. Miller FURTHER INFORMATION -- cgit v1.2.3 From f37a7238d379bc2befbcb08a03de03f4f446131c Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 24 Oct 2007 18:26:22 +0200 Subject: [SCSI] 53c7xx: fix removal fallout This patch does some additional cleanups after the 53c7xx removal. Signed-off-by: Adrian Bunk Acked-by: Geert Uytterhoeven Signed-off-by: James Bottomley --- Documentation/dontdiff | 2 -- Documentation/m68k/kernel-options.txt | 60 ----------------------------------- Documentation/scsi/00-INDEX | 2 -- Documentation/scsi/ncr53c7xx.txt | 40 ----------------------- drivers/scsi/.gitignore | 2 -- 5 files changed, 106 deletions(-) delete mode 100644 Documentation/scsi/ncr53c7xx.txt (limited to 'Documentation') diff --git a/Documentation/dontdiff b/Documentation/dontdiff index f2d658a6a942..c09a96b99354 100644 --- a/Documentation/dontdiff +++ b/Documentation/dontdiff @@ -46,8 +46,6 @@ .mailmap .mm 53c700_d.h -53c7xx_d.h -53c7xx_u.h 53c8xx_d.h* BitKeeper COPYING diff --git a/Documentation/m68k/kernel-options.txt b/Documentation/m68k/kernel-options.txt index 248589e8bcf5..c93bed66e25d 100644 --- a/Documentation/m68k/kernel-options.txt +++ b/Documentation/m68k/kernel-options.txt @@ -867,66 +867,6 @@ controller and should be autodetected by the driver. An example is the 24 bit region which is specified by a mask of 0x00fffffe. -5.5) 53c7xx= ------------- - -Syntax: 53c7xx= - -These options affect the A4000T, A4091, WarpEngine, Blizzard 603e+, -and GForce 040/060 SCSI controllers on the Amiga, as well as the -builtin MVME 16x SCSI controller. - -The is a comma-separated list of the sub-options listed -below. - -5.5.1) nosync -------------- - -Syntax: nosync:0 - - Disables sync negotiation for all devices. Any value after the - colon is acceptable (and has the same effect). - -5.5.2) noasync --------------- - -[OBSOLETE, REMOVED] - -5.5.3) nodisconnect -------------------- - -Syntax: nodisconnect:0 - - Disables SCSI disconnects. Any value after the colon is acceptable - (and has the same effect). - -5.5.4) validids ---------------- - -Syntax: validids:0xNN - - Specify which SCSI ids the driver should pay attention to. This is - a bitmask (i.e. to only pay attention to ID#4, you'd use 0x10). - Default is 0x7f (devices 0-6). - -5.5.5) opthi -5.5.6) optlo ------------- - -Syntax: opthi:M,optlo:N - - Specify options for "hostdata->options". The acceptable definitions - are listed in drivers/scsi/53c7xx.h; the 32 high bits should be in - opthi and the 32 low bits in optlo. They must be specified in the - order opthi=M,optlo=N. - -5.5.7) next ------------ - - No argument. Used to separate blocks of keywords when there's more - than one 53c7xx host adapter in the system. - - /* Local Variables: */ /* mode: text */ /* End: */ diff --git a/Documentation/scsi/00-INDEX b/Documentation/scsi/00-INDEX index aa1f7e927834..c2e18e109858 100644 --- a/Documentation/scsi/00-INDEX +++ b/Documentation/scsi/00-INDEX @@ -64,8 +64,6 @@ lpfc.txt - LPFC driver release notes megaraid.txt - Common Management Module, shared code handling ioctls for LSI drivers -ncr53c7xx.txt - - info on driver for NCR53c7xx based adapters ncr53c8xx.txt - info on driver for NCR53c8xx based adapters osst.txt diff --git a/Documentation/scsi/ncr53c7xx.txt b/Documentation/scsi/ncr53c7xx.txt deleted file mode 100644 index 91e9552d63e5..000000000000 --- a/Documentation/scsi/ncr53c7xx.txt +++ /dev/null @@ -1,40 +0,0 @@ -README for WarpEngine/A4000T/A4091 SCSI kernels. - -Use the following options to disable options in the SCSI driver. - -Using amiboot for example..... - -To disable Synchronous Negotiation.... - - amiboot -k kernel 53c7xx=nosync:0 - -To disable Disconnection.... - - amiboot -k kernel 53c7xx=nodisconnect:0 - -To disable certain SCSI devices... - - amiboot -k kernel 53c7xx=validids:0x3F - - this allows only device ID's 0,1,2,3,4 and 5 for linux to handle. - (this is a bitmasked field - i.e. each bit represents a SCSI ID) - -These commands work on a per controller basis and use the option 'next' to -move to the next controller in the system. - -e.g. - amiboot -k kernel 53c7xx=nodisconnect:0,next,nosync:0 - - this uses No Disconnection on the first controller and Asynchronous - SCSI on the second controller. - -Known Issues: - -Two devices are known not to function with the default settings of using -synchronous SCSI. These are the Archive Viper 150 Tape Drive and the -SyQuest SQ555 removeable hard drive. When using these devices on a controller -use the 'nosync:0' option. - -Please try these options and post any problems/successes to me. - -Alan Hourihane diff --git a/drivers/scsi/.gitignore b/drivers/scsi/.gitignore index b385af314356..c89ae9a04399 100644 --- a/drivers/scsi/.gitignore +++ b/drivers/scsi/.gitignore @@ -1,3 +1 @@ 53c700_d.h -53c7xx_d.h -53c7xx_u.h -- cgit v1.2.3 From eb44820c28bc9a042e1157b41c677018a8fdfc74 Mon Sep 17 00:00:00 2001 From: Rob Landley Date: Sat, 3 Nov 2007 13:30:39 -0500 Subject: [SCSI] Add Documentation and integrate into docbook build Add Documentation/DocBook/scsi_midlayer.tmpl, add to Makefile, and update lots of kerneldoc comments in drivers/scsi/*. Updated with comments from Stefan Richter, Stephen M. Cameron, James Bottomley and Randy Dunlap. Signed-off-by: Rob Landley Signed-off-by: James Bottomley --- Documentation/DocBook/Makefile | 2 +- Documentation/DocBook/scsi_midlayer.tmpl | 409 +++++++++++++++++++++++++++++++ drivers/scsi/constants.c | 1 - drivers/scsi/hosts.c | 4 +- drivers/scsi/scsi.c | 183 +++++++------- drivers/scsi/scsi_devinfo.c | 34 ++- drivers/scsi/scsi_error.c | 46 ++-- drivers/scsi/scsi_ioctl.c | 24 +- drivers/scsi/scsi_lib.c | 42 ++-- drivers/scsi/scsi_netlink.c | 19 +- drivers/scsi/scsi_proc.c | 110 ++++++++- drivers/scsi/scsi_scan.c | 32 ++- drivers/scsi/scsi_transport_fc.c | 102 ++++---- drivers/scsi/scsi_transport_iscsi.c | 17 +- drivers/scsi/scsi_transport_sas.c | 40 +-- drivers/scsi/scsi_transport_srp.c | 10 +- drivers/scsi/scsicam.c | 35 ++- 17 files changed, 821 insertions(+), 289 deletions(-) create mode 100644 Documentation/DocBook/scsi_midlayer.tmpl (limited to 'Documentation') diff --git a/Documentation/DocBook/Makefile b/Documentation/DocBook/Makefile index 4953bc258729..c6c5e59daaca 100644 --- a/Documentation/DocBook/Makefile +++ b/Documentation/DocBook/Makefile @@ -11,7 +11,7 @@ DOCBOOKS := wanbook.xml z8530book.xml mcabook.xml videobook.xml \ procfs-guide.xml writing_usb_driver.xml \ kernel-api.xml filesystems.xml lsm.xml usb.xml \ gadget.xml libata.xml mtdnand.xml librs.xml rapidio.xml \ - genericirq.xml s390-drivers.xml uio-howto.xml + genericirq.xml s390-drivers.xml uio-howto.xml scsi_midlayer.xml ### # The build process is as follows (targets): diff --git a/Documentation/DocBook/scsi_midlayer.tmpl b/Documentation/DocBook/scsi_midlayer.tmpl new file mode 100644 index 000000000000..6255930216b3 --- /dev/null +++ b/Documentation/DocBook/scsi_midlayer.tmpl @@ -0,0 +1,409 @@ + + + + + + SCSI Mid Layer Guide + + + + James + Bottomley + +
+ James.Bottomley@steeleye.com +
+ + + + + Rob + Landley + +
+ rob@landley.net +
+ + + + + + + 2007 + Linux Foundation + + + + + This documentation is free software; you can redistribute + it and/or modify it under the terms of the GNU General Public + License version 2. + + + + This program is distributed in the hope that it will be + useful, but WITHOUT ANY WARRANTY; without even the implied + warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + For more details see the file COPYING in the source + distribution of Linux. + + + + + + + + Introduction + + Protocol vs bus + + Once upon a time, the Small Computer Systems Interface defined both + a parallel I/O bus and a data protocol to connect a wide variety of + peripherals (disk drives, tape drives, modems, printers, scanners, + optical drives, test equipment, and medical devices) to a host + computer. + + + Although the old parallel (fast/wide/ultra) SCSI bus has largely + fallen out of use, the SCSI command set is more widely used than ever + to communicate with devices over a number of different busses. + + + The SCSI protocol + is a big-endian peer-to-peer packet based protocol. SCSI commands + are 6, 10, 12, or 16 bytes long, often followed by an associated data + payload. + + + SCSI commands can be transported over just about any kind of bus, and + are the default protocol for storage devices attached to USB, SATA, + SAS, Fibre Channel, FireWire, and ATAPI devices. SCSI packets are + also commonly exchanged over Infiniband, + I20, TCP/IP + (iSCSI), even + Parallel + ports. + + + + Design of the Linux SCSI subsystem + + The SCSI subsystem uses a three layer design, with upper, mid, and low + layers. Every operation involving the SCSI subsystem (such as reading + a sector from a disk) uses one driver at each of the 3 levels: one + upper layer driver, one lower layer driver, and the scsi midlayer. + + + The SCSI upper layer provides the interface between userspace and the + kernel, in the form of block and char device nodes for I/O and + ioctl(). The SCSI lower layer contains drivers for specific hardware + devices. + + + In between is the SCSI mid-layer, analogous to a network routing + layer such as the IPv4 stack. The SCSI mid-layer routes a packet + based data protocol between the upper layer's /dev nodes and the + corresponding devices in the lower layer. It manages command queues, + provides error handling and power management functions, and responds + to ioctl() requests. + + + + + + SCSI upper layer + + The upper layer supports the user-kernel interface by providing + device nodes. + + + sd (SCSI Disk) + sd (sd_mod.o) + + + + sr (SCSI CD-ROM) + sr (sr_mod.o) + + + st (SCSI Tape) + st (st.o) + + + sg (SCSI Generic) + sg (sg.o) + + + ch (SCSI Media Changer) + ch (ch.c) + + + + + SCSI mid layer + + + SCSI midlayer implementation + + include/scsi/scsi_device.h + + +!Iinclude/scsi/scsi_device.h + + + + drivers/scsi/scsi.c + Main file for the scsi midlayer. +!Edrivers/scsi/scsi.c + + + drivers/scsi/scsicam.c + + SCSI + Common Access Method support functions, for use with + HDIO_GETGEO, etc. + +!Edrivers/scsi/scsicam.c + + + drivers/scsi/scsi_error.c + Common SCSI error/timeout handling routines. +!Edrivers/scsi/scsi_error.c + + + drivers/scsi/scsi_devinfo.c + + Manage scsi_dev_info_list, which tracks blacklisted and whitelisted + devices. + +!Idrivers/scsi/scsi_devinfo.c + + + drivers/scsi/scsi_ioctl.c + + Handle ioctl() calls for scsi devices. + +!Edrivers/scsi/scsi_ioctl.c + + + drivers/scsi/scsi_lib.c + + SCSI queuing library. + +!Edrivers/scsi/scsi_lib.c + + + drivers/scsi/scsi_lib_dma.c + + SCSI library functions depending on DMA + (map and unmap scatter-gather lists). + +!Edrivers/scsi/scsi_lib_dma.c + + + drivers/scsi/scsi_module.c + + The file drivers/scsi/scsi_module.c contains legacy support for + old-style host templates. It should never be used by any new driver. + + + + drivers/scsi/scsi_proc.c + + The functions in this file provide an interface between + the PROC file system and the SCSI device drivers + It is mainly used for debugging, statistics and to pass + information directly to the lowlevel driver. + + I.E. plumbing to manage /proc/scsi/* + +!Idrivers/scsi/scsi_proc.c + + + drivers/scsi/scsi_netlink.c + + Infrastructure to provide async events from transports to userspace + via netlink, using a single NETLINK_SCSITRANSPORT protocol for all + transports. + + See the + original patch submission for more details. + +!Idrivers/scsi/scsi_netlink.c + + + drivers/scsi/scsi_scan.c + + Scan a host to determine which (if any) devices are attached. + + The general scanning/probing algorithm is as follows, exceptions are + made to it depending on device specific flags, compilation options, + and global variable (boot or module load time) settings. + + A specific LUN is scanned via an INQUIRY command; if the LUN has a + device attached, a scsi_device is allocated and setup for it. + + For every id of every channel on the given host, start by scanning + LUN 0. Skip hosts that don't respond at all to a scan of LUN 0. + Otherwise, if LUN 0 has a device attached, allocate and setup a + scsi_device for it. If target is SCSI-3 or up, issue a REPORT LUN, + and scan all of the LUNs returned by the REPORT LUN; else, + sequentially scan LUNs up until some maximum is reached, or a LUN is + seen that cannot have a device attached to it. + +!Idrivers/scsi/scsi_scan.c + + + drivers/scsi/scsi_sysctl.c + + Set up the sysctl entry: "/dev/scsi/logging_level" + (DEV_SCSI_LOGGING_LEVEL) which sets/returns scsi_logging_level. + + + + drivers/scsi/scsi_sysfs.c + + SCSI sysfs interface routines. + +!Edrivers/scsi/scsi_sysfs.c + + + drivers/scsi/hosts.c + + mid to lowlevel SCSI driver interface + +!Edrivers/scsi/hosts.c + + + drivers/scsi/constants.c + + mid to lowlevel SCSI driver interface + +!Edrivers/scsi/constants.c + + + + + Transport classes + + Transport classes are service libraries for drivers in the scsi + lower layer, which expose transport attributes in sysfs. + + + Fibre Channel transport + + The file drivers/scsi/scsi_transport_fc.c defines transport attributes + for Fibre Channel. + +!Edrivers/scsi/scsi_transport_fc.c + + + iSCSI transport class + + The file drivers/scsi/scsi_transport_iscsi.c defines transport + attributes for the iSCSI class, which sends SCSI packets over TCP/IP + connections. + +!Edrivers/scsi/scsi_transport_iscsi.c + + + Serial Attached SCSI (SAS) transport class + + The file drivers/scsi/scsi_transport_sas.c defines transport + attributes for Serial Attached SCSI, a variant of SATA aimed at + large high-end systems. + + + The SAS transport class contains common code to deal with SAS HBAs, + an aproximated representation of SAS topologies in the driver model, + and various sysfs attributes to expose these topologies and managment + interfaces to userspace. + + + In addition to the basic SCSI core objects this transport class + introduces two additional intermediate objects: The SAS PHY + as represented by struct sas_phy defines an "outgoing" PHY on + a SAS HBA or Expander, and the SAS remote PHY represented by + struct sas_rphy defines an "incoming" PHY on a SAS Expander or + end device. Note that this is purely a software concept, the + underlying hardware for a PHY and a remote PHY is the exactly + the same. + + + There is no concept of a SAS port in this code, users can see + what PHYs form a wide port based on the port_identifier attribute, + which is the same for all PHYs in a port. + +!Edrivers/scsi/scsi_transport_sas.c + + + SATA transport class + + The SATA transport is handled by libata, which has its own book of + documentation in this directory. + + + + Parallel SCSI (SPI) transport class + + The file drivers/scsi/scsi_transport_spi.c defines transport + attributes for traditional (fast/wide/ultra) SCSI busses. + +!Edrivers/scsi/scsi_transport_spi.c + + + SCSI RDMA (SRP) transport class + + The file drivers/scsi/scsi_transport_srp.c defines transport + attributes for SCSI over Remote Direct Memory Access. + +!Edrivers/scsi/scsi_transport_srp.c + + + + + + + SCSI lower layer + + Host Bus Adapter transport types + + Many modern device controllers use the SCSI command set as a protocol to + communicate with their devices through many different types of physical + connections. + + + In SCSI language a bus capable of carrying SCSI commands is + called a "transport", and a controller connecting to such a bus is + called a "host bus adapter" (HBA). + + + Debug transport + + The file drivers/scsi/scsi_debug.c simulates a host adapter with a + variable number of disks (or disk like devices) attached, sharing a + common amount of RAM. Does a lot of checking to make sure that we are + not getting blocks mixed up, and panics the kernel if anything out of + the ordinary is seen. + + + To be more realistic, the simulated devices have the transport + attributes of SAS disks. + + + For documentation see + http://www.torque.net/sg/sdebug26.html + + + + + todo + Parallel (fast/wide/ultra) SCSI, USB, SATA, + SAS, Fibre Channel, FireWire, ATAPI devices, Infiniband, + I20, iSCSI, Parallel ports, netlink... + + + + + diff --git a/drivers/scsi/constants.c b/drivers/scsi/constants.c index fac4314cd2ad..403a7f2d8f9b 100644 --- a/drivers/scsi/constants.c +++ b/drivers/scsi/constants.c @@ -362,7 +362,6 @@ void scsi_print_command(struct scsi_cmnd *cmd) EXPORT_SYMBOL(scsi_print_command); /** - * * scsi_print_status - print scsi status description * @scsi_status: scsi status value * diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c index 24271a871b8c..9a10b4335e76 100644 --- a/drivers/scsi/hosts.c +++ b/drivers/scsi/hosts.c @@ -54,8 +54,7 @@ static struct class shost_class = { }; /** - * scsi_host_set_state - Take the given host through the host - * state model. + * scsi_host_set_state - Take the given host through the host state model. * @shost: scsi host to change the state of. * @state: state to change to. * @@ -431,7 +430,6 @@ EXPORT_SYMBOL(scsi_unregister); /** * scsi_host_lookup - get a reference to a Scsi_Host by host no - * * @hostnum: host number to locate * * Return value: diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c index 0fb1709ce5e3..412a351c6cf7 100644 --- a/drivers/scsi/scsi.c +++ b/drivers/scsi/scsi.c @@ -122,6 +122,11 @@ static const char *const scsi_device_types[] = { "Automation/Drive ", }; +/** + * scsi_device_type - Return 17 char string indicating device type. + * @type: type number to look up + */ + const char * scsi_device_type(unsigned type) { if (type == 0x1e) @@ -156,6 +161,14 @@ static struct scsi_host_cmd_pool scsi_cmd_dma_pool = { static DEFINE_MUTEX(host_cmd_pool_mutex); +/** + * __scsi_get_command - Allocate a struct scsi_cmnd + * @shost: host to transmit command + * @gfp_mask: allocation mask + * + * Description: allocate a struct scsi_cmd from host's slab, recycling from the + * host's free_list if necessary. + */ struct scsi_cmnd *__scsi_get_command(struct Scsi_Host *shost, gfp_t gfp_mask) { struct scsi_cmnd *cmd; @@ -179,13 +192,10 @@ struct scsi_cmnd *__scsi_get_command(struct Scsi_Host *shost, gfp_t gfp_mask) } EXPORT_SYMBOL_GPL(__scsi_get_command); -/* - * Function: scsi_get_command() - * - * Purpose: Allocate and setup a scsi command block - * - * Arguments: dev - parent scsi device - * gfp_mask- allocator flags +/** + * scsi_get_command - Allocate and setup a scsi command block + * @dev: parent scsi device + * @gfp_mask: allocator flags * * Returns: The allocated scsi command structure. */ @@ -217,6 +227,12 @@ struct scsi_cmnd *scsi_get_command(struct scsi_device *dev, gfp_t gfp_mask) } EXPORT_SYMBOL(scsi_get_command); +/** + * __scsi_put_command - Free a struct scsi_cmnd + * @shost: dev->host + * @cmd: Command to free + * @dev: parent scsi device + */ void __scsi_put_command(struct Scsi_Host *shost, struct scsi_cmnd *cmd, struct device *dev) { @@ -237,12 +253,9 @@ void __scsi_put_command(struct Scsi_Host *shost, struct scsi_cmnd *cmd, } EXPORT_SYMBOL(__scsi_put_command); -/* - * Function: scsi_put_command() - * - * Purpose: Free a scsi command block - * - * Arguments: cmd - command block to free +/** + * scsi_put_command - Free a scsi command block + * @cmd: command block to free * * Returns: Nothing. * @@ -263,12 +276,13 @@ void scsi_put_command(struct scsi_cmnd *cmd) } EXPORT_SYMBOL(scsi_put_command); -/* - * Function: scsi_setup_command_freelist() - * - * Purpose: Setup the command freelist for a scsi host. +/** + * scsi_setup_command_freelist - Setup the command freelist for a scsi host. + * @shost: host to allocate the freelist for. * - * Arguments: shost - host to allocate the freelist for. + * Description: The command freelist protects against system-wide out of memory + * deadlock by preallocating one SCSI command structure for each host, so the + * system can always write to a swap file on a device associated with that host. * * Returns: Nothing. */ @@ -282,7 +296,7 @@ int scsi_setup_command_freelist(struct Scsi_Host *shost) /* * Select a command slab for this host and create it if not - * yet existant. + * yet existent. */ mutex_lock(&host_cmd_pool_mutex); pool = (shost->unchecked_isa_dma ? &scsi_cmd_dma_pool : &scsi_cmd_pool); @@ -318,12 +332,9 @@ int scsi_setup_command_freelist(struct Scsi_Host *shost) } -/* - * Function: scsi_destroy_command_freelist() - * - * Purpose: Release the command freelist for a scsi host. - * - * Arguments: shost - host that's freelist is going to be destroyed +/** + * scsi_destroy_command_freelist - Release the command freelist for a scsi host. + * @shost: host whose freelist is going to be destroyed */ void scsi_destroy_command_freelist(struct Scsi_Host *shost) { @@ -441,8 +452,12 @@ void scsi_log_completion(struct scsi_cmnd *cmd, int disposition) } #endif -/* - * Assign a serial number to the request for error recovery +/** + * scsi_cmd_get_serial - Assign a serial number to a command + * @host: the scsi host + * @cmd: command to assign serial number to + * + * Description: a serial number identifies a request for error recovery * and debugging purposes. Protected by the Host_Lock of host. */ static inline void scsi_cmd_get_serial(struct Scsi_Host *host, struct scsi_cmnd *cmd) @@ -452,14 +467,12 @@ static inline void scsi_cmd_get_serial(struct Scsi_Host *host, struct scsi_cmnd cmd->serial_number = host->cmd_serial_number++; } -/* - * Function: scsi_dispatch_command - * - * Purpose: Dispatch a command to the low-level driver. - * - * Arguments: cmd - command block we are dispatching. +/** + * scsi_dispatch_command - Dispatch a command to the low-level driver. + * @cmd: command block we are dispatching. * - * Notes: + * Return: nonzero return request was rejected and device's queue needs to be + * plugged. */ int scsi_dispatch_cmd(struct scsi_cmnd *cmd) { @@ -585,7 +598,7 @@ int scsi_dispatch_cmd(struct scsi_cmnd *cmd) /** * scsi_req_abort_cmd -- Request command recovery for the specified command - * cmd: pointer to the SCSI command of interest + * @cmd: pointer to the SCSI command of interest * * This function requests that SCSI Core start recovery for the * command by deleting the timer and adding the command to the eh @@ -606,9 +619,9 @@ EXPORT_SYMBOL(scsi_req_abort_cmd); * @cmd: The SCSI Command for which a low-level device driver (LLDD) gives * ownership back to SCSI Core -- i.e. the LLDD has finished with it. * - * This function is the mid-level's (SCSI Core) interrupt routine, which - * regains ownership of the SCSI command (de facto) from a LLDD, and enqueues - * the command to the done queue for further processing. + * Description: This function is the mid-level's (SCSI Core) interrupt routine, + * which regains ownership of the SCSI command (de facto) from a LLDD, and + * enqueues the command to the done queue for further processing. * * This is the producer of the done queue who enqueues at the tail. * @@ -617,7 +630,7 @@ EXPORT_SYMBOL(scsi_req_abort_cmd); static void scsi_done(struct scsi_cmnd *cmd) { /* - * We don't have to worry about this one timing out any more. + * We don't have to worry about this one timing out anymore. * If we are unable to remove the timer, then the command * has already timed out. In which case, we have no choice but to * let the timeout function run, as we have no idea where in fact @@ -660,10 +673,11 @@ static struct scsi_driver *scsi_cmd_to_driver(struct scsi_cmnd *cmd) return *(struct scsi_driver **)cmd->request->rq_disk->private_data; } -/* - * Function: scsi_finish_command +/** + * scsi_finish_command - cleanup and pass command back to upper layer + * @cmd: the command * - * Purpose: Pass command off to upper layer for finishing of I/O + * Description: Pass command off to upper layer for finishing of I/O * request, waking processes that are waiting on results, * etc. */ @@ -708,18 +722,14 @@ void scsi_finish_command(struct scsi_cmnd *cmd) } EXPORT_SYMBOL(scsi_finish_command); -/* - * Function: scsi_adjust_queue_depth() - * - * Purpose: Allow low level drivers to tell us to change the queue depth - * on a specific SCSI device - * - * Arguments: sdev - SCSI Device in question - * tagged - Do we use tagged queueing (non-0) or do we treat - * this device as an untagged device (0) - * tags - Number of tags allowed if tagged queueing enabled, - * or number of commands the low level driver can - * queue up in non-tagged mode (as per cmd_per_lun). +/** + * scsi_adjust_queue_depth - Let low level drivers change a device's queue depth + * @sdev: SCSI Device in question + * @tagged: Do we use tagged queueing (non-0) or do we treat + * this device as an untagged device (0) + * @tags: Number of tags allowed if tagged queueing enabled, + * or number of commands the low level driver can + * queue up in non-tagged mode (as per cmd_per_lun). * * Returns: Nothing * @@ -742,8 +752,8 @@ void scsi_adjust_queue_depth(struct scsi_device *sdev, int tagged, int tags) spin_lock_irqsave(sdev->request_queue->queue_lock, flags); - /* Check to see if the queue is managed by the block layer - * if it is, and we fail to adjust the depth, exit */ + /* Check to see if the queue is managed by the block layer. + * If it is, and we fail to adjust the depth, exit. */ if (blk_queue_tagged(sdev->request_queue) && blk_queue_resize_tags(sdev->request_queue, tags) != 0) goto out; @@ -772,20 +782,17 @@ void scsi_adjust_queue_depth(struct scsi_device *sdev, int tagged, int tags) } EXPORT_SYMBOL(scsi_adjust_queue_depth); -/* - * Function: scsi_track_queue_full() +/** + * scsi_track_queue_full - track QUEUE_FULL events to adjust queue depth + * @sdev: SCSI Device in question + * @depth: Current number of outstanding SCSI commands on this device, + * not counting the one returned as QUEUE_FULL. * - * Purpose: This function will track successive QUEUE_FULL events on a + * Description: This function will track successive QUEUE_FULL events on a * specific SCSI device to determine if and when there is a * need to adjust the queue depth on the device. * - * Arguments: sdev - SCSI Device in question - * depth - Current number of outstanding SCSI commands on - * this device, not counting the one returned as - * QUEUE_FULL. - * - * Returns: 0 - No change needed - * >0 - Adjust queue depth to this new depth + * Returns: 0 - No change needed, >0 - Adjust queue depth to this new depth, * -1 - Drop back to untagged operation using host->cmd_per_lun * as the untagged command depth * @@ -824,10 +831,10 @@ int scsi_track_queue_full(struct scsi_device *sdev, int depth) EXPORT_SYMBOL(scsi_track_queue_full); /** - * scsi_device_get - get an addition reference to a scsi_device + * scsi_device_get - get an additional reference to a scsi_device * @sdev: device to get a reference to * - * Gets a reference to the scsi_device and increments the use count + * Description: Gets a reference to the scsi_device and increments the use count * of the underlying LLDD module. You must hold host_lock of the * parent Scsi_Host or already have a reference when calling this. */ @@ -849,8 +856,8 @@ EXPORT_SYMBOL(scsi_device_get); * scsi_device_put - release a reference to a scsi_device * @sdev: device to release a reference on. * - * Release a reference to the scsi_device and decrements the use count - * of the underlying LLDD module. The device is freed once the last + * Description: Release a reference to the scsi_device and decrements the use + * count of the underlying LLDD module. The device is freed once the last * user vanishes. */ void scsi_device_put(struct scsi_device *sdev) @@ -867,7 +874,7 @@ void scsi_device_put(struct scsi_device *sdev) } EXPORT_SYMBOL(scsi_device_put); -/* helper for shost_for_each_device, thus not documented */ +/* helper for shost_for_each_device, see that for documentation */ struct scsi_device *__scsi_iterate_devices(struct Scsi_Host *shost, struct scsi_device *prev) { @@ -895,6 +902,8 @@ EXPORT_SYMBOL(__scsi_iterate_devices); /** * starget_for_each_device - helper to walk all devices of a target * @starget: target whose devices we want to iterate over. + * @data: Opaque passed to each function call. + * @fn: Function to call on each device * * This traverses over each device of @starget. The devices have * a reference that must be released by scsi_host_put when breaking @@ -946,13 +955,13 @@ EXPORT_SYMBOL(__starget_for_each_device); * @starget: SCSI target pointer * @lun: SCSI Logical Unit Number * - * Looks up the scsi_device with the specified @lun for a give - * @starget. The returned scsi_device does not have an additional + * Description: Looks up the scsi_device with the specified @lun for a given + * @starget. The returned scsi_device does not have an additional * reference. You must hold the host's host_lock over this call and * any access to the returned scsi_device. * * Note: The only reason why drivers would want to use this is because - * they're need to access the device list in irq context. Otherwise you + * they need to access the device list in irq context. Otherwise you * really want to use scsi_device_lookup_by_target instead. **/ struct scsi_device *__scsi_device_lookup_by_target(struct scsi_target *starget, @@ -974,9 +983,9 @@ EXPORT_SYMBOL(__scsi_device_lookup_by_target); * @starget: SCSI target pointer * @lun: SCSI Logical Unit Number * - * Looks up the scsi_device with the specified @channel, @id, @lun for a - * give host. The returned scsi_device has an additional reference that - * needs to be release with scsi_host_put once you're done with it. + * Description: Looks up the scsi_device with the specified @channel, @id, @lun + * for a given host. The returned scsi_device has an additional reference that + * needs to be released with scsi_device_put once you're done with it. **/ struct scsi_device *scsi_device_lookup_by_target(struct scsi_target *starget, uint lun) @@ -996,19 +1005,19 @@ struct scsi_device *scsi_device_lookup_by_target(struct scsi_target *starget, EXPORT_SYMBOL(scsi_device_lookup_by_target); /** - * scsi_device_lookup - find a device given the host (UNLOCKED) + * __scsi_device_lookup - find a device given the host (UNLOCKED) * @shost: SCSI host pointer * @channel: SCSI channel (zero if only one channel) - * @pun: SCSI target number (physical unit number) + * @id: SCSI target number (physical unit number) * @lun: SCSI Logical Unit Number * - * Looks up the scsi_device with the specified @channel, @id, @lun for a - * give host. The returned scsi_device does not have an additional reference. - * You must hold the host's host_lock over this call and any access to the - * returned scsi_device. + * Description: Looks up the scsi_device with the specified @channel, @id, @lun + * for a given host. The returned scsi_device does not have an additional + * reference. You must hold the host's host_lock over this call and any access + * to the returned scsi_device. * * Note: The only reason why drivers would want to use this is because - * they're need to access the device list in irq context. Otherwise you + * they need to access the device list in irq context. Otherwise you * really want to use scsi_device_lookup instead. **/ struct scsi_device *__scsi_device_lookup(struct Scsi_Host *shost, @@ -1033,9 +1042,9 @@ EXPORT_SYMBOL(__scsi_device_lookup); * @id: SCSI target number (physical unit number) * @lun: SCSI Logical Unit Number * - * Looks up the scsi_device with the specified @channel, @id, @lun for a - * give host. The returned scsi_device has an additional reference that - * needs to be release with scsi_host_put once you're done with it. + * Description: Looks up the scsi_device with the specified @channel, @id, @lun + * for a given host. The returned scsi_device has an additional reference that + * needs to be released with scsi_device_put once you're done with it. **/ struct scsi_device *scsi_device_lookup(struct Scsi_Host *shost, uint channel, uint id, uint lun) diff --git a/drivers/scsi/scsi_devinfo.c b/drivers/scsi/scsi_devinfo.c index 348cc5a6e3cd..6839c1004b29 100644 --- a/drivers/scsi/scsi_devinfo.c +++ b/drivers/scsi/scsi_devinfo.c @@ -276,11 +276,12 @@ static void scsi_strcpy_devinfo(char *name, char *to, size_t to_length, } /** - * scsi_dev_info_list_add: add one dev_info list entry. + * scsi_dev_info_list_add - add one dev_info list entry. + * @compatible: if true, null terminate short strings. Otherwise space pad. * @vendor: vendor string * @model: model (product) string * @strflags: integer string - * @flag: if strflags NULL, use this flag value + * @flags: if strflags NULL, use this flag value * * Description: * Create and add one dev_info entry for @vendor, @model, @strflags or @@ -322,8 +323,7 @@ static int scsi_dev_info_list_add(int compatible, char *vendor, char *model, } /** - * scsi_dev_info_list_add_str: parse dev_list and add to the - * scsi_dev_info_list. + * scsi_dev_info_list_add_str - parse dev_list and add to the scsi_dev_info_list. * @dev_list: string of device flags to add * * Description: @@ -374,15 +374,15 @@ static int scsi_dev_info_list_add_str(char *dev_list) } /** - * get_device_flags - get device specific flags from the dynamic device - * list. Called during scan time. + * get_device_flags - get device specific flags from the dynamic device list. + * @sdev: &scsi_device to get flags for * @vendor: vendor name * @model: model name * * Description: * Search the scsi_dev_info_list for an entry matching @vendor and * @model, if found, return the matching flags value, else return - * the host or global default settings. + * the host or global default settings. Called during scan time. **/ int scsi_get_device_flags(struct scsi_device *sdev, const unsigned char *vendor, @@ -483,13 +483,11 @@ stop_output: } /* - * proc_scsi_dev_info_write: allow additions to the scsi_dev_info_list via - * /proc. + * proc_scsi_dev_info_write - allow additions to scsi_dev_info_list via /proc. * - * Use: echo "vendor:model:flag" > /proc/scsi/device_info - * - * To add a black/white list entry for vendor and model with an integer - * value of flag to the scsi device info list. + * Description: Adds a black/white list entry for vendor and model with an + * integer value of flag to the scsi device info list. + * To use, echo "vendor:model:flag" > /proc/scsi/device_info */ static int proc_scsi_devinfo_write(struct file *file, const char __user *buf, unsigned long length, void *data) @@ -532,8 +530,7 @@ MODULE_PARM_DESC(default_dev_flags, "scsi default device flag integer value"); /** - * scsi_dev_info_list_delete: called from scsi.c:exit_scsi to remove - * the scsi_dev_info_list. + * scsi_dev_info_list_delete - called from scsi.c:exit_scsi to remove the scsi_dev_info_list. **/ void scsi_exit_devinfo(void) { @@ -552,13 +549,12 @@ void scsi_exit_devinfo(void) } /** - * scsi_dev_list_init: set up the dynamic device list. - * @dev_list: string of device flags to add + * scsi_dev_list_init - set up the dynamic device list. * * Description: - * Add command line @dev_list entries, then add + * Add command line entries from scsi_dev_flags, then add * scsi_static_device_list entries to the scsi device info list. - **/ + */ int __init scsi_init_devinfo(void) { #ifdef CONFIG_SCSI_PROC_FS diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index ebaca4ca4a13..22f8140598ba 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -779,7 +779,7 @@ static int scsi_request_sense(struct scsi_cmnd *scmd) * Notes: * We don't want to use the normal command completion while we are are * still handling errors - it may cause other commands to be queued, - * and that would disturb what we are doing. thus we really want to + * and that would disturb what we are doing. Thus we really want to * keep a list of pending commands for final completion, and once we * are ready to leave error handling we handle completion for real. **/ @@ -794,7 +794,7 @@ EXPORT_SYMBOL(scsi_eh_finish_cmd); /** * scsi_eh_get_sense - Get device sense data. * @work_q: Queue of commands to process. - * @done_q: Queue of proccessed commands.. + * @done_q: Queue of processed commands. * * Description: * See if we need to request sense information. if so, then get it @@ -802,7 +802,7 @@ EXPORT_SYMBOL(scsi_eh_finish_cmd); * * Notes: * This has the unfortunate side effect that if a shost adapter does - * not automatically request sense information, that we end up shutting + * not automatically request sense information, we end up shutting * it down before we request it. * * All drivers should request sense information internally these days, @@ -858,7 +858,7 @@ EXPORT_SYMBOL_GPL(scsi_eh_get_sense); /** * scsi_eh_tur - Send TUR to device. - * @scmd: Scsi cmd to send TUR + * @scmd: &scsi_cmnd to send TUR * * Return value: * 0 - Device is ready. 1 - Device NOT ready. @@ -887,14 +887,14 @@ retry_tur: } /** - * scsi_eh_abort_cmds - abort canceled commands. - * @shost: scsi host being recovered. - * @eh_done_q: list_head for processed commands. + * scsi_eh_abort_cmds - abort pending commands. + * @work_q: &list_head for pending commands. + * @done_q: &list_head for processed commands. * * Decription: * Try and see whether or not it makes sense to try and abort the - * running command. this only works out to be the case if we have one - * command that has timed out. if the command simply failed, it makes + * running command. This only works out to be the case if we have one + * command that has timed out. If the command simply failed, it makes * no sense to try and abort the command, since as far as the shost * adapter is concerned, it isn't running. **/ @@ -931,7 +931,7 @@ static int scsi_eh_abort_cmds(struct list_head *work_q, /** * scsi_eh_try_stu - Send START_UNIT to device. - * @scmd: Scsi cmd to send START_UNIT + * @scmd: &scsi_cmnd to send START_UNIT * * Return value: * 0 - Device is ready. 1 - Device NOT ready. @@ -956,8 +956,9 @@ static int scsi_eh_try_stu(struct scsi_cmnd *scmd) /** * scsi_eh_stu - send START_UNIT if needed - * @shost: scsi host being recovered. - * @eh_done_q: list_head for processed commands. + * @shost: &scsi host being recovered. + * @work_q: &list_head for pending commands. + * @done_q: &list_head for processed commands. * * Notes: * If commands are failing due to not ready, initializing command required, @@ -1008,10 +1009,11 @@ static int scsi_eh_stu(struct Scsi_Host *shost, /** * scsi_eh_bus_device_reset - send bdr if needed * @shost: scsi host being recovered. - * @eh_done_q: list_head for processed commands. + * @work_q: &list_head for pending commands. + * @done_q: &list_head for processed commands. * * Notes: - * Try a bus device reset. still, look to see whether we have multiple + * Try a bus device reset. Still, look to see whether we have multiple * devices that are jammed or not - if we have multiple devices, it * makes no sense to try bus_device_reset - we really would need to try * a bus_reset instead. @@ -1063,8 +1065,9 @@ static int scsi_eh_bus_device_reset(struct Scsi_Host *shost, /** * scsi_eh_bus_reset - send a bus reset - * @shost: scsi host being recovered. - * @eh_done_q: list_head for processed commands. + * @shost: &scsi host being recovered. + * @work_q: &list_head for pending commands. + * @done_q: &list_head for processed commands. **/ static int scsi_eh_bus_reset(struct Scsi_Host *shost, struct list_head *work_q, @@ -1440,7 +1443,8 @@ static void scsi_restart_operations(struct Scsi_Host *shost) /** * scsi_eh_ready_devs - check device ready state and recover if not. * @shost: host to be recovered. - * @eh_done_q: list_head for processed commands. + * @work_q: &list_head for pending commands. + * @done_q: &list_head for processed commands. * **/ void scsi_eh_ready_devs(struct Scsi_Host *shost, @@ -1824,9 +1828,7 @@ int scsi_command_normalize_sense(struct scsi_cmnd *cmd, EXPORT_SYMBOL(scsi_command_normalize_sense); /** - * scsi_sense_desc_find - search for a given descriptor type in - * descriptor sense data format. - * + * scsi_sense_desc_find - search for a given descriptor type in descriptor sense data format. * @sense_buffer: byte array of descriptor format sense data * @sb_len: number of valid bytes in sense_buffer * @desc_type: value of descriptor type to find @@ -1865,9 +1867,7 @@ const u8 * scsi_sense_desc_find(const u8 * sense_buffer, int sb_len, EXPORT_SYMBOL(scsi_sense_desc_find); /** - * scsi_get_sense_info_fld - attempts to get information field from - * sense data (either fixed or descriptor format) - * + * scsi_get_sense_info_fld - get information field from sense data (either fixed or descriptor format) * @sense_buffer: byte array of sense data * @sb_len: number of valid bytes in sense_buffer * @info_out: pointer to 64 integer where 8 or 4 byte information diff --git a/drivers/scsi/scsi_ioctl.c b/drivers/scsi/scsi_ioctl.c index 32293f451669..83e144716901 100644 --- a/drivers/scsi/scsi_ioctl.c +++ b/drivers/scsi/scsi_ioctl.c @@ -174,10 +174,15 @@ static int scsi_ioctl_get_pci(struct scsi_device *sdev, void __user *arg) } -/* - * the scsi_ioctl() function differs from most ioctls in that it does - * not take a major/minor number as the dev field. Rather, it takes - * a pointer to a scsi_devices[] element, a structure. +/** + * scsi_ioctl - Dispatch ioctl to scsi device + * @sdev: scsi device receiving ioctl + * @cmd: which ioctl is it + * @arg: data associated with ioctl + * + * Description: The scsi_ioctl() function differs from most ioctls in that it + * does not take a major/minor number as the dev field. Rather, it takes + * a pointer to a &struct scsi_device. */ int scsi_ioctl(struct scsi_device *sdev, int cmd, void __user *arg) { @@ -264,9 +269,12 @@ int scsi_ioctl(struct scsi_device *sdev, int cmd, void __user *arg) } EXPORT_SYMBOL(scsi_ioctl); -/* - * the scsi_nonblock_ioctl() function is designed for ioctls which may - * be executed even if the device is in recovery. +/** + * scsi_nonblock_ioctl() - Handle SG_SCSI_RESET + * @sdev: scsi device receiving ioctl + * @cmd: Must be SC_SCSI_RESET + * @arg: pointer to int containing SG_SCSI_RESET_{DEVICE,BUS,HOST} + * @filp: either NULL or a &struct file which must have the O_NONBLOCK flag. */ int scsi_nonblockable_ioctl(struct scsi_device *sdev, int cmd, void __user *arg, struct file *filp) @@ -276,7 +284,7 @@ int scsi_nonblockable_ioctl(struct scsi_device *sdev, int cmd, /* The first set of iocts may be executed even if we're doing * error processing, as long as the device was opened * non-blocking */ - if (filp && filp->f_flags & O_NONBLOCK) { + if (filp && (filp->f_flags & O_NONBLOCK)) { if (scsi_host_in_recovery(sdev->host)) return -ENODEV; } else if (!scsi_block_when_processing_errors(sdev)) diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index a9ac5b1b1667..199b02452259 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -175,7 +175,7 @@ int scsi_queue_insert(struct scsi_cmnd *cmd, int reason) * * returns the req->errors value which is the scsi_cmnd result * field. - **/ + */ int scsi_execute(struct scsi_device *sdev, const unsigned char *cmd, int data_direction, void *buffer, unsigned bufflen, unsigned char *sense, int timeout, int retries, int flags) @@ -274,7 +274,7 @@ static void scsi_bi_endio(struct bio *bio, int error) /** * scsi_req_map_sg - map a scatterlist into a request * @rq: request to fill - * @sg: scatterlist + * @sgl: scatterlist * @nsegs: number of elements * @bufflen: len of buffer * @gfp: memory allocation flags @@ -365,14 +365,16 @@ free_bios: * @sdev: scsi device * @cmd: scsi command * @cmd_len: length of scsi cdb - * @data_direction: data direction + * @data_direction: DMA_TO_DEVICE, DMA_FROM_DEVICE, or DMA_NONE * @buffer: data buffer (this can be a kernel buffer or scatterlist) * @bufflen: len of buffer * @use_sg: if buffer is a scatterlist this is the number of elements * @timeout: request timeout in seconds * @retries: number of times to retry request - * @flags: or into request flags - **/ + * @privdata: data passed to done() + * @done: callback function when done + * @gfp: memory allocation flags + */ int scsi_execute_async(struct scsi_device *sdev, const unsigned char *cmd, int cmd_len, int data_direction, void *buffer, unsigned bufflen, int use_sg, int timeout, int retries, void *privdata, @@ -1804,7 +1806,7 @@ void scsi_exit_queue(void) * @timeout: command timeout * @retries: number of retries before failing * @data: returns a structure abstracting the mode header data - * @sense: place to put sense data (or NULL if no sense to be collected). + * @sshdr: place to put sense data (or NULL if no sense to be collected). * must be SCSI_SENSE_BUFFERSIZE big. * * Returns zero if successful; negative error number or scsi @@ -1871,8 +1873,7 @@ scsi_mode_select(struct scsi_device *sdev, int pf, int sp, int modepage, EXPORT_SYMBOL_GPL(scsi_mode_select); /** - * scsi_mode_sense - issue a mode sense, falling back from 10 to - * six bytes if necessary. + * scsi_mode_sense - issue a mode sense, falling back from 10 to six bytes if necessary. * @sdev: SCSI device to be queried * @dbd: set if mode sense will allow block descriptors to be returned * @modepage: mode page being requested @@ -1881,13 +1882,13 @@ EXPORT_SYMBOL_GPL(scsi_mode_select); * @timeout: command timeout * @retries: number of retries before failing * @data: returns a structure abstracting the mode header data - * @sense: place to put sense data (or NULL if no sense to be collected). + * @sshdr: place to put sense data (or NULL if no sense to be collected). * must be SCSI_SENSE_BUFFERSIZE big. * * Returns zero if unsuccessful, or the header offset (either 4 * or 8 depending on whether a six or ten byte command was * issued) if successful. - **/ + */ int scsi_mode_sense(struct scsi_device *sdev, int dbd, int modepage, unsigned char *buffer, int len, int timeout, int retries, @@ -2007,14 +2008,13 @@ scsi_test_unit_ready(struct scsi_device *sdev, int timeout, int retries) EXPORT_SYMBOL(scsi_test_unit_ready); /** - * scsi_device_set_state - Take the given device through the device - * state model. + * scsi_device_set_state - Take the given device through the device state model. * @sdev: scsi device to change the state of. * @state: state to change to. * * Returns zero if unsuccessful or an error if the requested * transition is illegal. - **/ + */ int scsi_device_set_state(struct scsi_device *sdev, enum scsi_device_state state) { @@ -2264,7 +2264,7 @@ EXPORT_SYMBOL_GPL(sdev_evt_send_simple); * Must be called with user context, may sleep. * * Returns zero if unsuccessful or an error if not. - **/ + */ int scsi_device_quiesce(struct scsi_device *sdev) { @@ -2289,7 +2289,7 @@ EXPORT_SYMBOL(scsi_device_quiesce); * queues. * * Must be called with user context, may sleep. - **/ + */ void scsi_device_resume(struct scsi_device *sdev) { @@ -2326,8 +2326,7 @@ scsi_target_resume(struct scsi_target *starget) EXPORT_SYMBOL(scsi_target_resume); /** - * scsi_internal_device_block - internal function to put a device - * temporarily into the SDEV_BLOCK state + * scsi_internal_device_block - internal function to put a device temporarily into the SDEV_BLOCK state * @sdev: device to block * * Block request made by scsi lld's to temporarily stop all @@ -2342,7 +2341,7 @@ EXPORT_SYMBOL(scsi_target_resume); * state, all commands are deferred until the scsi lld reenables * the device with scsi_device_unblock or device_block_tmo fires. * This routine assumes the host_lock is held on entry. - **/ + */ int scsi_internal_device_block(struct scsi_device *sdev) { @@ -2382,7 +2381,7 @@ EXPORT_SYMBOL_GPL(scsi_internal_device_block); * (which must be a legal transition) allowing the midlayer to * goose the queue for this device. This routine assumes the * host_lock is held upon entry. - **/ + */ int scsi_internal_device_unblock(struct scsi_device *sdev) { @@ -2460,7 +2459,7 @@ EXPORT_SYMBOL_GPL(scsi_target_unblock); /** * scsi_kmap_atomic_sg - find and atomically map an sg-elemnt - * @sg: scatter-gather list + * @sgl: scatter-gather list * @sg_count: number of segments in sg * @offset: offset in bytes into sg, on return offset into the mapped area * @len: bytes to map, on return number of bytes mapped @@ -2509,8 +2508,7 @@ void *scsi_kmap_atomic_sg(struct scatterlist *sgl, int sg_count, EXPORT_SYMBOL(scsi_kmap_atomic_sg); /** - * scsi_kunmap_atomic_sg - atomically unmap a virtual address, previously - * mapped with scsi_kmap_atomic_sg + * scsi_kunmap_atomic_sg - atomically unmap a virtual address, previously mapped with scsi_kmap_atomic_sg * @virt: virtual address to be unmapped */ void scsi_kunmap_atomic_sg(void *virt) diff --git a/drivers/scsi/scsi_netlink.c b/drivers/scsi/scsi_netlink.c index 40579edca101..3e1591828171 100644 --- a/drivers/scsi/scsi_netlink.c +++ b/drivers/scsi/scsi_netlink.c @@ -32,11 +32,12 @@ EXPORT_SYMBOL_GPL(scsi_nl_sock); /** - * scsi_nl_rcv_msg - - * Receive message handler. Extracts message from a receive buffer. + * scsi_nl_rcv_msg - Receive message handler. + * @skb: socket receive buffer + * + * Description: Extracts message from a receive buffer. * Validates message header and calls appropriate transport message handler * - * @skb: socket receive buffer * **/ static void @@ -99,9 +100,7 @@ next_msg: /** - * scsi_nl_rcv_event - - * Event handler for a netlink socket. - * + * scsi_nl_rcv_event - Event handler for a netlink socket. * @this: event notifier block * @event: event type * @ptr: event payload @@ -129,9 +128,7 @@ static struct notifier_block scsi_netlink_notifier = { /** - * scsi_netlink_init - - * Called by SCSI subsystem to intialize the SCSI transport netlink - * interface + * scsi_netlink_init - Called by SCSI subsystem to intialize the SCSI transport netlink interface * **/ void @@ -160,9 +157,7 @@ scsi_netlink_init(void) /** - * scsi_netlink_exit - - * Called by SCSI subsystem to disable the SCSI transport netlink - * interface + * scsi_netlink_exit - Called by SCSI subsystem to disable the SCSI transport netlink interface * **/ void diff --git a/drivers/scsi/scsi_proc.c b/drivers/scsi/scsi_proc.c index bb6f051beda8..ed395154a5b1 100644 --- a/drivers/scsi/scsi_proc.c +++ b/drivers/scsi/scsi_proc.c @@ -45,6 +45,16 @@ static struct proc_dir_entry *proc_scsi; /* Protect sht->present and sht->proc_dir */ static DEFINE_MUTEX(global_host_template_mutex); +/** + * proc_scsi_read - handle read from /proc by calling host's proc_info() command + * @buffer: passed to proc_info + * @start: passed to proc_info + * @offset: passed to proc_info + * @length: passed to proc_info + * @eof: returns whether length read was less than requested + * @data: pointer to a &struct Scsi_Host + */ + static int proc_scsi_read(char *buffer, char **start, off_t offset, int length, int *eof, void *data) { @@ -57,6 +67,13 @@ static int proc_scsi_read(char *buffer, char **start, off_t offset, return n; } +/** + * proc_scsi_write_proc - Handle write to /proc by calling host's proc_info() + * @file: not used + * @buf: source of data to write. + * @count: number of bytes (at most PROC_BLOCK_SIZE) to write. + * @data: pointer to &struct Scsi_Host + */ static int proc_scsi_write_proc(struct file *file, const char __user *buf, unsigned long count, void *data) { @@ -80,6 +97,13 @@ out: return ret; } +/** + * scsi_proc_hostdir_add - Create directory in /proc for a scsi host + * @sht: owner of this directory + * + * Sets sht->proc_dir to the new directory. + */ + void scsi_proc_hostdir_add(struct scsi_host_template *sht) { if (!sht->proc_info) @@ -97,6 +121,10 @@ void scsi_proc_hostdir_add(struct scsi_host_template *sht) mutex_unlock(&global_host_template_mutex); } +/** + * scsi_proc_hostdir_rm - remove directory in /proc for a scsi host + * @sht: owner of directory + */ void scsi_proc_hostdir_rm(struct scsi_host_template *sht) { if (!sht->proc_info) @@ -110,6 +138,11 @@ void scsi_proc_hostdir_rm(struct scsi_host_template *sht) mutex_unlock(&global_host_template_mutex); } + +/** + * scsi_proc_host_add - Add entry for this host to appropriate /proc dir + * @shost: host to add + */ void scsi_proc_host_add(struct Scsi_Host *shost) { struct scsi_host_template *sht = shost->hostt; @@ -133,6 +166,10 @@ void scsi_proc_host_add(struct Scsi_Host *shost) p->owner = sht->module; } +/** + * scsi_proc_host_rm - remove this host's entry from /proc + * @shost: which host + */ void scsi_proc_host_rm(struct Scsi_Host *shost) { char name[10]; @@ -143,7 +180,14 @@ void scsi_proc_host_rm(struct Scsi_Host *shost) sprintf(name,"%d", shost->host_no); remove_proc_entry(name, shost->hostt->proc_dir); } - +/** + * proc_print_scsidevice - return data about this host + * @dev: A scsi device + * @data: &struct seq_file to output to. + * + * Description: prints Host, Channel, Id, Lun, Vendor, Model, Rev, Type, + * and revision. + */ static int proc_print_scsidevice(struct device *dev, void *data) { struct scsi_device *sdev = to_scsi_device(dev); @@ -189,6 +233,21 @@ static int proc_print_scsidevice(struct device *dev, void *data) return 0; } +/** + * scsi_add_single_device - Respond to user request to probe for/add device + * @host: user-supplied decimal integer + * @channel: user-supplied decimal integer + * @id: user-supplied decimal integer + * @lun: user-supplied decimal integer + * + * Description: called by writing "scsi add-single-device" to /proc/scsi/scsi. + * + * does scsi_host_lookup() and either user_scan() if that transport + * type supports it, or else scsi_scan_host_selected() + * + * Note: this seems to be aimed exclusively at SCSI parallel busses. + */ + static int scsi_add_single_device(uint host, uint channel, uint id, uint lun) { struct Scsi_Host *shost; @@ -206,6 +265,16 @@ static int scsi_add_single_device(uint host, uint channel, uint id, uint lun) return error; } +/** + * scsi_remove_single_device - Respond to user request to remove a device + * @host: user-supplied decimal integer + * @channel: user-supplied decimal integer + * @id: user-supplied decimal integer + * @lun: user-supplied decimal integer + * + * Description: called by writing "scsi remove-single-device" to + * /proc/scsi/scsi. Does a scsi_device_lookup() and scsi_remove_device() + */ static int scsi_remove_single_device(uint host, uint channel, uint id, uint lun) { struct scsi_device *sdev; @@ -226,6 +295,25 @@ static int scsi_remove_single_device(uint host, uint channel, uint id, uint lun) return error; } +/** + * proc_scsi_write - handle writes to /proc/scsi/scsi + * @file: not used + * @buf: buffer to write + * @length: length of buf, at most PAGE_SIZE + * @ppos: not used + * + * Description: this provides a legacy mechanism to add or remove devices by + * Host, Channel, ID, and Lun. To use, + * "echo 'scsi add-single-device 0 1 2 3' > /proc/scsi/scsi" or + * "echo 'scsi remove-single-device 0 1 2 3' > /proc/scsi/scsi" with + * "0 1 2 3" replaced by the Host, Channel, Id, and Lun. + * + * Note: this seems to be aimed at parallel SCSI. Most modern busses (USB, + * SATA, Firewire, Fibre Channel, etc) dynamically assign these values to + * provide a unique identifier and nothing more. + */ + + static ssize_t proc_scsi_write(struct file *file, const char __user *buf, size_t length, loff_t *ppos) { @@ -291,6 +379,11 @@ static ssize_t proc_scsi_write(struct file *file, const char __user *buf, return err; } +/** + * proc_scsi_show - show contents of /proc/scsi/scsi (attached devices) + * @s: output goes here + * @p: not used + */ static int proc_scsi_show(struct seq_file *s, void *p) { seq_printf(s, "Attached devices:\n"); @@ -298,10 +391,17 @@ static int proc_scsi_show(struct seq_file *s, void *p) return 0; } +/** + * proc_scsi_open - glue function + * @inode: not used + * @file: passed to single_open() + * + * Associates proc_scsi_show with this file + */ static int proc_scsi_open(struct inode *inode, struct file *file) { /* - * We don't really needs this for the write case but it doesn't + * We don't really need this for the write case but it doesn't * harm either. */ return single_open(file, proc_scsi_show, NULL); @@ -315,6 +415,9 @@ static const struct file_operations proc_scsi_operations = { .release = single_release, }; +/** + * scsi_init_procfs - create scsi and scsi/scsi in procfs + */ int __init scsi_init_procfs(void) { struct proc_dir_entry *pde; @@ -336,6 +439,9 @@ err1: return -ENOMEM; } +/** + * scsi_exit_procfs - Remove scsi/scsi and scsi from procfs + */ void scsi_exit_procfs(void) { remove_proc_entry("scsi/scsi", NULL); diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index 40ea71cd2ca6..7e8b2045ccfa 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -221,6 +221,9 @@ static void scsi_unlock_floptical(struct scsi_device *sdev, /** * scsi_alloc_sdev - allocate and setup a scsi_Device + * @starget: which target to allocate a &scsi_device for + * @lun: which lun + * @hostdata: usually NULL and set by ->slave_alloc instead * * Description: * Allocate, initialize for io, and return a pointer to a scsi_Device. @@ -472,7 +475,6 @@ static void scsi_target_reap_usercontext(struct work_struct *work) /** * scsi_target_reap - check to see if target is in use and destroy if not - * * @starget: target to be checked * * This is used after removing a LUN or doing a last put of the target @@ -928,8 +930,7 @@ static inline void scsi_destroy_sdev(struct scsi_device *sdev) #ifdef CONFIG_SCSI_LOGGING /** - * scsi_inq_str - print INQUIRY data from min to max index, - * strip trailing whitespace + * scsi_inq_str - print INQUIRY data from min to max index, strip trailing whitespace * @buf: Output buffer with at least end-first+1 bytes of space * @inq: Inquiry buffer (input) * @first: Offset of string into inq @@ -957,9 +958,10 @@ static unsigned char *scsi_inq_str(unsigned char *buf, unsigned char *inq, * scsi_probe_and_add_lun - probe a LUN, if a LUN is found add it * @starget: pointer to target device structure * @lun: LUN of target device - * @sdevscan: probe the LUN corresponding to this scsi_device - * @sdevnew: store the value of any new scsi_device allocated * @bflagsp: store bflags here if not NULL + * @sdevp: probe the LUN corresponding to this scsi_device + * @rescan: if nonzero skip some code only needed on first scan + * @hostdata: passed to scsi_alloc_sdev() * * Description: * Call scsi_probe_lun, if a LUN with an attached device is found, @@ -1110,6 +1112,8 @@ static int scsi_probe_and_add_lun(struct scsi_target *starget, * scsi_sequential_lun_scan - sequentially scan a SCSI target * @starget: pointer to target structure to scan * @bflags: black/white list flag for LUN 0 + * @scsi_level: Which version of the standard does this device adhere to + * @rescan: passed to scsi_probe_add_lun() * * Description: * Generally, scan from LUN 1 (LUN 0 is assumed to already have been @@ -1220,7 +1224,7 @@ EXPORT_SYMBOL(scsilun_to_int); /** * int_to_scsilun: reverts an int into a scsi_lun - * @int: integer to be reverted + * @lun: integer to be reverted * @scsilun: struct scsi_lun to be set. * * Description: @@ -1252,18 +1256,22 @@ EXPORT_SYMBOL(int_to_scsilun); /** * scsi_report_lun_scan - Scan using SCSI REPORT LUN results - * @sdevscan: scan the host, channel, and id of this scsi_device + * @starget: which target + * @bflags: Zero or a mix of BLIST_NOLUN, BLIST_REPORTLUN2, or BLIST_NOREPORTLUN + * @rescan: nonzero if we can skip code only needed on first scan * * Description: - * If @sdevscan is for a SCSI-3 or up device, send a REPORT LUN - * command, and scan the resulting list of LUNs by calling - * scsi_probe_and_add_lun. + * Fast scanning for modern (SCSI-3) devices by sending a REPORT LUN command. + * Scan the resulting list of LUNs by calling scsi_probe_and_add_lun. * - * Modifies sdevscan->lun. + * If BLINK_REPORTLUN2 is set, scan a target that supports more than 8 + * LUNs even if it's older than SCSI-3. + * If BLIST_NOREPORTLUN is set, return 1 always. + * If BLIST_NOLUN is set, return 0 always. * * Return: * 0: scan completed (or no memory, so further scanning is futile) - * 1: no report lun scan, or not configured + * 1: could not scan with REPORT LUN **/ static int scsi_report_lun_scan(struct scsi_target *starget, int bflags, int rescan) diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c index 7a7cfe583b2a..b1119da6e88c 100644 --- a/drivers/scsi/scsi_transport_fc.c +++ b/drivers/scsi/scsi_transport_fc.c @@ -481,9 +481,9 @@ MODULE_PARM_DESC(dev_loss_tmo, " exceeded, the scsi target is removed. Value should be" " between 1 and SCSI_DEVICE_BLOCK_MAX_TIMEOUT."); -/** +/* * Netlink Infrastructure - **/ + */ static atomic_t fc_event_seq; @@ -491,10 +491,10 @@ static atomic_t fc_event_seq; * fc_get_event_number - Obtain the next sequential FC event number * * Notes: - * We could have inline'd this, but it would have required fc_event_seq to + * We could have inlined this, but it would have required fc_event_seq to * be exposed. For now, live with the subroutine call. * Atomic used to avoid lock/unlock... - **/ + */ u32 fc_get_event_number(void) { @@ -505,7 +505,6 @@ EXPORT_SYMBOL(fc_get_event_number); /** * fc_host_post_event - called to post an even on an fc_host. - * * @shost: host the event occurred on * @event_number: fc event number obtained from get_fc_event_number() * @event_code: fc_host event being posted @@ -513,7 +512,7 @@ EXPORT_SYMBOL(fc_get_event_number); * * Notes: * This routine assumes no locks are held on entry. - **/ + */ void fc_host_post_event(struct Scsi_Host *shost, u32 event_number, enum fc_host_event_code event_code, u32 event_data) @@ -579,17 +578,16 @@ EXPORT_SYMBOL(fc_host_post_event); /** - * fc_host_post_vendor_event - called to post a vendor unique event on - * a fc_host - * + * fc_host_post_vendor_event - called to post a vendor unique event on an fc_host * @shost: host the event occurred on * @event_number: fc event number obtained from get_fc_event_number() * @data_len: amount, in bytes, of vendor unique data * @data_buf: pointer to vendor unique data + * @vendor_id: Vendor id * * Notes: * This routine assumes no locks are held on entry. - **/ + */ void fc_host_post_vendor_event(struct Scsi_Host *shost, u32 event_number, u32 data_len, char * data_buf, u64 vendor_id) @@ -1900,7 +1898,6 @@ static int fc_vport_match(struct attribute_container *cont, /** * fc_timed_out - FC Transport I/O timeout intercept handler - * * @scmd: The SCSI command which timed out * * This routine protects against error handlers getting invoked while a @@ -1920,7 +1917,7 @@ static int fc_vport_match(struct attribute_container *cont, * * Notes: * This routine assumes no locks are held on entry. - **/ + */ static enum scsi_eh_timer_return fc_timed_out(struct scsi_cmnd *scmd) { @@ -2133,7 +2130,7 @@ EXPORT_SYMBOL(fc_release_transport); * 1 - work queued for execution * 0 - work is already queued * -EINVAL - work queue doesn't exist - **/ + */ static int fc_queue_work(struct Scsi_Host *shost, struct work_struct *work) { @@ -2152,7 +2149,7 @@ fc_queue_work(struct Scsi_Host *shost, struct work_struct *work) /** * fc_flush_work - Flush a fc_host's workqueue. * @shost: Pointer to Scsi_Host bound to fc_host. - **/ + */ static void fc_flush_work(struct Scsi_Host *shost) { @@ -2175,7 +2172,7 @@ fc_flush_work(struct Scsi_Host *shost) * * Return value: * 1 on success / 0 already queued / < 0 for error - **/ + */ static int fc_queue_devloss_work(struct Scsi_Host *shost, struct delayed_work *work, unsigned long delay) @@ -2195,7 +2192,7 @@ fc_queue_devloss_work(struct Scsi_Host *shost, struct delayed_work *work, /** * fc_flush_devloss - Flush a fc_host's devloss workqueue. * @shost: Pointer to Scsi_Host bound to fc_host. - **/ + */ static void fc_flush_devloss(struct Scsi_Host *shost) { @@ -2212,21 +2209,20 @@ fc_flush_devloss(struct Scsi_Host *shost) /** - * fc_remove_host - called to terminate any fc_transport-related elements - * for a scsi host. - * @rport: remote port to be unblocked. + * fc_remove_host - called to terminate any fc_transport-related elements for a scsi host. + * @shost: Which &Scsi_Host * * This routine is expected to be called immediately preceeding the * a driver's call to scsi_remove_host(). * * WARNING: A driver utilizing the fc_transport, which fails to call - * this routine prior to scsi_remote_host(), will leave dangling + * this routine prior to scsi_remove_host(), will leave dangling * objects in /sys/class/fc_remote_ports. Access to any of these * objects can result in a system crash !!! * * Notes: * This routine assumes no locks are held on entry. - **/ + */ void fc_remove_host(struct Scsi_Host *shost) { @@ -2281,10 +2277,10 @@ EXPORT_SYMBOL(fc_remove_host); /** * fc_starget_delete - called to delete the scsi decendents of an rport - * (target and all sdevs) - * * @work: remote port to be operated on. - **/ + * + * Deletes target and all sdevs. + */ static void fc_starget_delete(struct work_struct *work) { @@ -2303,9 +2299,8 @@ fc_starget_delete(struct work_struct *work) /** * fc_rport_final_delete - finish rport termination and delete it. - * * @work: remote port to be deleted. - **/ + */ static void fc_rport_final_delete(struct work_struct *work) { @@ -2375,7 +2370,7 @@ fc_rport_final_delete(struct work_struct *work) * * Notes: * This routine assumes no locks are held on entry. - **/ + */ static struct fc_rport * fc_rport_create(struct Scsi_Host *shost, int channel, struct fc_rport_identifiers *ids) @@ -2462,8 +2457,7 @@ delete_rport: } /** - * fc_remote_port_add - notifies the fc transport of the existence - * of a remote FC port. + * fc_remote_port_add - notify fc transport of the existence of a remote FC port. * @shost: scsi host the remote port is connected to. * @channel: Channel on shost port connected to. * @ids: The world wide names, fc address, and FC4 port @@ -2499,7 +2493,7 @@ delete_rport: * * Notes: * This routine assumes no locks are held on entry. - **/ + */ struct fc_rport * fc_remote_port_add(struct Scsi_Host *shost, int channel, struct fc_rport_identifiers *ids) @@ -2683,19 +2677,18 @@ EXPORT_SYMBOL(fc_remote_port_add); /** - * fc_remote_port_delete - notifies the fc transport that a remote - * port is no longer in existence. + * fc_remote_port_delete - notifies the fc transport that a remote port is no longer in existence. * @rport: The remote port that no longer exists * * The LLDD calls this routine to notify the transport that a remote * port is no longer part of the topology. Note: Although a port * may no longer be part of the topology, it may persist in the remote * ports displayed by the fc_host. We do this under 2 conditions: - * - If the port was a scsi target, we delay its deletion by "blocking" it. + * 1) If the port was a scsi target, we delay its deletion by "blocking" it. * This allows the port to temporarily disappear, then reappear without * disrupting the SCSI device tree attached to it. During the "blocked" * period the port will still exist. - * - If the port was a scsi target and disappears for longer than we + * 2) If the port was a scsi target and disappears for longer than we * expect, we'll delete the port and the tear down the SCSI device tree * attached to it. However, we want to semi-persist the target id assigned * to that port if it eventually does exist. The port structure will @@ -2709,7 +2702,8 @@ EXPORT_SYMBOL(fc_remote_port_add); * temporary blocked state. From the LLDD's perspective, the rport no * longer exists. From the SCSI midlayer's perspective, the SCSI target * exists, but all sdevs on it are blocked from further I/O. The following - * is then expected: + * is then expected. + * * If the remote port does not return (signaled by a LLDD call to * fc_remote_port_add()) within the dev_loss_tmo timeout, then the * scsi target is removed - killing all outstanding i/o and removing the @@ -2731,7 +2725,7 @@ EXPORT_SYMBOL(fc_remote_port_add); * * Notes: * This routine assumes no locks are held on entry. - **/ + */ void fc_remote_port_delete(struct fc_rport *rport) { @@ -2792,12 +2786,12 @@ fc_remote_port_delete(struct fc_rport *rport) EXPORT_SYMBOL(fc_remote_port_delete); /** - * fc_remote_port_rolechg - notifies the fc transport that the roles - * on a remote may have changed. + * fc_remote_port_rolechg - notifies the fc transport that the roles on a remote may have changed. * @rport: The remote port that changed. + * @roles: New roles for this port. * - * The LLDD calls this routine to notify the transport that the roles - * on a remote port may have changed. The largest effect of this is + * Description: The LLDD calls this routine to notify the transport that the + * roles on a remote port may have changed. The largest effect of this is * if a port now becomes a FCP Target, it must be allocated a * scsi target id. If the port is no longer a FCP target, any * scsi target id value assigned to it will persist in case the @@ -2810,7 +2804,7 @@ EXPORT_SYMBOL(fc_remote_port_delete); * * Notes: * This routine assumes no locks are held on entry. - **/ + */ void fc_remote_port_rolechg(struct fc_rport *rport, u32 roles) { @@ -2875,12 +2869,12 @@ fc_remote_port_rolechg(struct fc_rport *rport, u32 roles) EXPORT_SYMBOL(fc_remote_port_rolechg); /** - * fc_timeout_deleted_rport - Timeout handler for a deleted remote port, - * which we blocked, and has now failed to return - * in the allotted time. - * + * fc_timeout_deleted_rport - Timeout handler for a deleted remote port. * @work: rport target that failed to reappear in the allotted time. - **/ + * + * Description: An attempt to delete a remote port blocks, and if it fails + * to return in the allotted time this gets called. + */ static void fc_timeout_deleted_rport(struct work_struct *work) { @@ -2984,14 +2978,12 @@ fc_timeout_deleted_rport(struct work_struct *work) } /** - * fc_timeout_fail_rport_io - Timeout handler for a fast io failing on a - * disconnected SCSI target. - * + * fc_timeout_fail_rport_io - Timeout handler for a fast io failing on a disconnected SCSI target. * @work: rport to terminate io on. * * Notes: Only requests the failure of the io, not that all are flushed * prior to returning. - **/ + */ static void fc_timeout_fail_rport_io(struct work_struct *work) { @@ -3008,9 +3000,8 @@ fc_timeout_fail_rport_io(struct work_struct *work) /** * fc_scsi_scan_rport - called to perform a scsi scan on a remote port. - * * @work: remote port to be scanned. - **/ + */ static void fc_scsi_scan_rport(struct work_struct *work) { @@ -3047,7 +3038,7 @@ fc_scsi_scan_rport(struct work_struct *work) * * Notes: * This routine assumes no locks are held on entry. - **/ + */ static int fc_vport_create(struct Scsi_Host *shost, int channel, struct device *pdev, struct fc_vport_identifiers *ids, struct fc_vport **ret_vport) @@ -3172,7 +3163,7 @@ delete_vport: * * Notes: * This routine assumes no locks are held on entry. - **/ + */ int fc_vport_terminate(struct fc_vport *vport) { @@ -3232,9 +3223,8 @@ EXPORT_SYMBOL(fc_vport_terminate); /** * fc_vport_sched_delete - workq-based delete request for a vport - * * @work: vport to be deleted. - **/ + */ static void fc_vport_sched_delete(struct work_struct *work) { diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index 5428d15f23c6..cb48b80c0865 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -328,9 +328,10 @@ EXPORT_SYMBOL_GPL(iscsi_add_session); * iscsi_create_session - create iscsi class session * @shost: scsi host * @transport: iscsi transport + * @target_id: which target * * This can be called from a LLD or iscsi_transport. - **/ + */ struct iscsi_cls_session * iscsi_create_session(struct Scsi_Host *shost, struct iscsi_transport *transport, @@ -382,7 +383,7 @@ EXPORT_SYMBOL_GPL(iscsi_free_session); * * Can be called by a LLD or iscsi_transport. There must not be * any running connections. - **/ + */ int iscsi_destroy_session(struct iscsi_cls_session *session) { iscsi_remove_session(session); @@ -418,7 +419,7 @@ static int iscsi_is_conn_dev(const struct device *dev) * for software iscsi we could be trying to preallocate a connection struct * in which case there could be two connection structs and cid would be * non-zero. - **/ + */ struct iscsi_cls_conn * iscsi_create_conn(struct iscsi_cls_session *session, uint32_t cid) { @@ -465,10 +466,10 @@ EXPORT_SYMBOL_GPL(iscsi_create_conn); /** * iscsi_destroy_conn - destroy iscsi class connection - * @session: iscsi cls session + * @conn: iscsi cls session * - * This can be called from a LLD or iscsi_transport. - **/ + * This can be called from an LLD or iscsi_transport. + */ int iscsi_destroy_conn(struct iscsi_cls_conn *conn) { transport_unregister_device(&conn->dev); @@ -690,7 +691,7 @@ iscsi_if_get_stats(struct iscsi_transport *transport, struct nlmsghdr *nlh) * * This is called by HW iscsi LLDs to notify userpsace that its HW has * removed a session. - **/ + */ int iscsi_if_destroy_session_done(struct iscsi_cls_conn *conn) { struct iscsi_internal *priv; @@ -751,7 +752,7 @@ EXPORT_SYMBOL_GPL(iscsi_if_destroy_session_done); * * This is called by HW iscsi LLDs to notify userpsace that its HW has * created a session or a existing session is back in the logged in state. - **/ + */ int iscsi_if_create_session_done(struct iscsi_cls_conn *conn) { struct iscsi_internal *priv; diff --git a/drivers/scsi/scsi_transport_sas.c b/drivers/scsi/scsi_transport_sas.c index 3120f4b3a11a..87e786daa713 100644 --- a/drivers/scsi/scsi_transport_sas.c +++ b/drivers/scsi/scsi_transport_sas.c @@ -323,7 +323,7 @@ static int do_sas_phy_delete(struct device *dev, void *data) } /** - * sas_remove_children -- tear down a devices SAS data structures + * sas_remove_children - tear down a devices SAS data structures * @dev: device belonging to the sas object * * Removes all SAS PHYs and remote PHYs for a given object @@ -336,7 +336,7 @@ void sas_remove_children(struct device *dev) EXPORT_SYMBOL(sas_remove_children); /** - * sas_remove_host -- tear down a Scsi_Host's SAS data structures + * sas_remove_host - tear down a Scsi_Host's SAS data structures * @shost: Scsi Host that is torn down * * Removes all SAS PHYs and remote PHYs for a given Scsi_Host. @@ -577,7 +577,7 @@ static void sas_phy_release(struct device *dev) } /** - * sas_phy_alloc -- allocates and initialize a SAS PHY structure + * sas_phy_alloc - allocates and initialize a SAS PHY structure * @parent: Parent device * @number: Phy index * @@ -618,7 +618,7 @@ struct sas_phy *sas_phy_alloc(struct device *parent, int number) EXPORT_SYMBOL(sas_phy_alloc); /** - * sas_phy_add -- add a SAS PHY to the device hierarchy + * sas_phy_add - add a SAS PHY to the device hierarchy * @phy: The PHY to be added * * Publishes a SAS PHY to the rest of the system. @@ -638,7 +638,7 @@ int sas_phy_add(struct sas_phy *phy) EXPORT_SYMBOL(sas_phy_add); /** - * sas_phy_free -- free a SAS PHY + * sas_phy_free - free a SAS PHY * @phy: SAS PHY to free * * Frees the specified SAS PHY. @@ -655,7 +655,7 @@ void sas_phy_free(struct sas_phy *phy) EXPORT_SYMBOL(sas_phy_free); /** - * sas_phy_delete -- remove SAS PHY + * sas_phy_delete - remove SAS PHY * @phy: SAS PHY to remove * * Removes the specified SAS PHY. If the SAS PHY has an @@ -677,7 +677,7 @@ sas_phy_delete(struct sas_phy *phy) EXPORT_SYMBOL(sas_phy_delete); /** - * scsi_is_sas_phy -- check if a struct device represents a SAS PHY + * scsi_is_sas_phy - check if a struct device represents a SAS PHY * @dev: device to check * * Returns: @@ -843,7 +843,6 @@ EXPORT_SYMBOL(sas_port_alloc_num); /** * sas_port_add - add a SAS port to the device hierarchy - * * @port: port to be added * * publishes a port to the rest of the system @@ -868,7 +867,7 @@ int sas_port_add(struct sas_port *port) EXPORT_SYMBOL(sas_port_add); /** - * sas_port_free -- free a SAS PORT + * sas_port_free - free a SAS PORT * @port: SAS PORT to free * * Frees the specified SAS PORT. @@ -885,7 +884,7 @@ void sas_port_free(struct sas_port *port) EXPORT_SYMBOL(sas_port_free); /** - * sas_port_delete -- remove SAS PORT + * sas_port_delete - remove SAS PORT * @port: SAS PORT to remove * * Removes the specified SAS PORT. If the SAS PORT has an @@ -924,7 +923,7 @@ void sas_port_delete(struct sas_port *port) EXPORT_SYMBOL(sas_port_delete); /** - * scsi_is_sas_port -- check if a struct device represents a SAS port + * scsi_is_sas_port - check if a struct device represents a SAS port * @dev: device to check * * Returns: @@ -1309,6 +1308,7 @@ static void sas_rphy_initialize(struct sas_rphy *rphy) /** * sas_end_device_alloc - allocate an rphy for an end device + * @parent: which port * * Allocates an SAS remote PHY structure, connected to @parent. * @@ -1345,6 +1345,8 @@ EXPORT_SYMBOL(sas_end_device_alloc); /** * sas_expander_alloc - allocate an rphy for an end device + * @parent: which port + * @type: SAS_EDGE_EXPANDER_DEVICE or SAS_FANOUT_EXPANDER_DEVICE * * Allocates an SAS remote PHY structure, connected to @parent. * @@ -1383,7 +1385,7 @@ struct sas_rphy *sas_expander_alloc(struct sas_port *parent, EXPORT_SYMBOL(sas_expander_alloc); /** - * sas_rphy_add -- add a SAS remote PHY to the device hierarchy + * sas_rphy_add - add a SAS remote PHY to the device hierarchy * @rphy: The remote PHY to be added * * Publishes a SAS remote PHY to the rest of the system. @@ -1430,8 +1432,8 @@ int sas_rphy_add(struct sas_rphy *rphy) EXPORT_SYMBOL(sas_rphy_add); /** - * sas_rphy_free -- free a SAS remote PHY - * @rphy SAS remote PHY to free + * sas_rphy_free - free a SAS remote PHY + * @rphy: SAS remote PHY to free * * Frees the specified SAS remote PHY. * @@ -1459,7 +1461,7 @@ void sas_rphy_free(struct sas_rphy *rphy) EXPORT_SYMBOL(sas_rphy_free); /** - * sas_rphy_delete -- remove and free SAS remote PHY + * sas_rphy_delete - remove and free SAS remote PHY * @rphy: SAS remote PHY to remove and free * * Removes the specified SAS remote PHY and frees it. @@ -1473,7 +1475,7 @@ sas_rphy_delete(struct sas_rphy *rphy) EXPORT_SYMBOL(sas_rphy_delete); /** - * sas_rphy_remove -- remove SAS remote PHY + * sas_rphy_remove - remove SAS remote PHY * @rphy: SAS remote phy to remove * * Removes the specified SAS remote PHY. @@ -1504,7 +1506,7 @@ sas_rphy_remove(struct sas_rphy *rphy) EXPORT_SYMBOL(sas_rphy_remove); /** - * scsi_is_sas_rphy -- check if a struct device represents a SAS remote PHY + * scsi_is_sas_rphy - check if a struct device represents a SAS remote PHY * @dev: device to check * * Returns: @@ -1604,7 +1606,7 @@ static int sas_user_scan(struct Scsi_Host *shost, uint channel, SETUP_TEMPLATE(expander_attrs, expander_##field, S_IRUGO, 1) /** - * sas_attach_transport -- instantiate SAS transport template + * sas_attach_transport - instantiate SAS transport template * @ft: SAS transport class function template */ struct scsi_transport_template * @@ -1715,7 +1717,7 @@ sas_attach_transport(struct sas_function_template *ft) EXPORT_SYMBOL(sas_attach_transport); /** - * sas_release_transport -- release SAS transport template instance + * sas_release_transport - release SAS transport template instance * @t: transport template instance */ void sas_release_transport(struct scsi_transport_template *t) diff --git a/drivers/scsi/scsi_transport_srp.c b/drivers/scsi/scsi_transport_srp.c index 65c584db33bd..739e327bba09 100644 --- a/drivers/scsi/scsi_transport_srp.c +++ b/drivers/scsi/scsi_transport_srp.c @@ -242,8 +242,8 @@ struct srp_rport *srp_rport_add(struct Scsi_Host *shost, EXPORT_SYMBOL_GPL(srp_rport_add); /** - * srp_rport_del -- remove a SRP remote port - * @port: SRP remote port to remove + * srp_rport_del - remove a SRP remote port + * @rport: SRP remote port to remove * * Removes the specified SRP remote port. */ @@ -271,7 +271,7 @@ static int do_srp_rport_del(struct device *dev, void *data) } /** - * srp_remove_host -- tear down a Scsi_Host's SRP data structures + * srp_remove_host - tear down a Scsi_Host's SRP data structures * @shost: Scsi Host that is torn down * * Removes all SRP remote ports for a given Scsi_Host. @@ -297,7 +297,7 @@ static int srp_it_nexus_response(struct Scsi_Host *shost, u64 nexus, int result) } /** - * srp_attach_transport -- instantiate SRP transport template + * srp_attach_transport - instantiate SRP transport template * @ft: SRP transport class function template */ struct scsi_transport_template * @@ -337,7 +337,7 @@ srp_attach_transport(struct srp_function_template *ft) EXPORT_SYMBOL_GPL(srp_attach_transport); /** - * srp_release_transport -- release SRP transport template instance + * srp_release_transport - release SRP transport template instance * @t: transport template instance */ void srp_release_transport(struct scsi_transport_template *t) diff --git a/drivers/scsi/scsicam.c b/drivers/scsi/scsicam.c index cd68a66c7bb3..3f21bc65e8c6 100644 --- a/drivers/scsi/scsicam.c +++ b/drivers/scsi/scsicam.c @@ -24,6 +24,14 @@ static int setsize(unsigned long capacity, unsigned int *cyls, unsigned int *hds, unsigned int *secs); +/** + * scsi_bios_ptable - Read PC partition table out of first sector of device. + * @dev: from this device + * + * Description: Reads the first sector from the device and returns %0x42 bytes + * starting at offset %0x1be. + * Returns: partition table in kmalloc(GFP_KERNEL) memory, or NULL on error. + */ unsigned char *scsi_bios_ptable(struct block_device *dev) { unsigned char *res = kmalloc(66, GFP_KERNEL); @@ -43,15 +51,17 @@ unsigned char *scsi_bios_ptable(struct block_device *dev) } EXPORT_SYMBOL(scsi_bios_ptable); -/* - * Function : int scsicam_bios_param (struct block_device *bdev, ector_t capacity, int *ip) +/** + * scsicam_bios_param - Determine geometry of a disk in cylinders/heads/sectors. + * @bdev: which device + * @capacity: size of the disk in sectors + * @ip: return value: ip[0]=heads, ip[1]=sectors, ip[2]=cylinders * - * Purpose : to determine the BIOS mapping used for a drive in a + * Description : determine the BIOS mapping/geometry used for a drive in a * SCSI-CAM system, storing the results in ip as required * by the HDIO_GETGEO ioctl(). * * Returns : -1 on failure, 0 on success. - * */ int scsicam_bios_param(struct block_device *bdev, sector_t capacity, int *ip) @@ -98,15 +108,18 @@ int scsicam_bios_param(struct block_device *bdev, sector_t capacity, int *ip) } EXPORT_SYMBOL(scsicam_bios_param); -/* - * Function : static int scsi_partsize(unsigned char *buf, unsigned long - * capacity,unsigned int *cyls, unsigned int *hds, unsigned int *secs); +/** + * scsi_partsize - Parse cylinders/heads/sectors from PC partition table + * @buf: partition table, see scsi_bios_ptable() + * @capacity: size of the disk in sectors + * @cyls: put cylinders here + * @hds: put heads here + * @secs: put sectors here * - * Purpose : to determine the BIOS mapping used to create the partition + * Description: determine the BIOS mapping/geometry used to create the partition * table, storing the results in *cyls, *hds, and *secs * - * Returns : -1 on failure, 0 on success. - * + * Returns: -1 on failure, 0 on success. */ int scsi_partsize(unsigned char *buf, unsigned long capacity, @@ -194,7 +207,7 @@ EXPORT_SYMBOL(scsi_partsize); * * WORKING X3T9.2 * DRAFT 792D - * + * see http://www.t10.org/ftp/t10/drafts/cam/cam-r12b.pdf * * Revision 6 * 10-MAR-94 -- cgit v1.2.3 From 74286a3c276bca449f934b2cd921f9bced996e04 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Wed, 14 Nov 2007 18:01:21 -0600 Subject: [SCSI] DocBook scsi_midlayer.tmpl->scsi.tmpl Signed-off-by: James Bottomley --- Documentation/DocBook/Makefile | 2 +- Documentation/DocBook/scsi.tmpl | 409 +++++++++++++++++++++++++++++++ Documentation/DocBook/scsi_midlayer.tmpl | 409 ------------------------------- 3 files changed, 410 insertions(+), 410 deletions(-) create mode 100644 Documentation/DocBook/scsi.tmpl delete mode 100644 Documentation/DocBook/scsi_midlayer.tmpl (limited to 'Documentation') diff --git a/Documentation/DocBook/Makefile b/Documentation/DocBook/Makefile index c6c5e59daaca..6a0ad4715e9f 100644 --- a/Documentation/DocBook/Makefile +++ b/Documentation/DocBook/Makefile @@ -11,7 +11,7 @@ DOCBOOKS := wanbook.xml z8530book.xml mcabook.xml videobook.xml \ procfs-guide.xml writing_usb_driver.xml \ kernel-api.xml filesystems.xml lsm.xml usb.xml \ gadget.xml libata.xml mtdnand.xml librs.xml rapidio.xml \ - genericirq.xml s390-drivers.xml uio-howto.xml scsi_midlayer.xml + genericirq.xml s390-drivers.xml uio-howto.xml scsi.xml ### # The build process is as follows (targets): diff --git a/Documentation/DocBook/scsi.tmpl b/Documentation/DocBook/scsi.tmpl new file mode 100644 index 000000000000..6255930216b3 --- /dev/null +++ b/Documentation/DocBook/scsi.tmpl @@ -0,0 +1,409 @@ + + + + + + SCSI Mid Layer Guide + + + + James + Bottomley + +
+ James.Bottomley@steeleye.com +
+ + + + + Rob + Landley + +
+ rob@landley.net +
+ + + + + + + 2007 + Linux Foundation + + + + + This documentation is free software; you can redistribute + it and/or modify it under the terms of the GNU General Public + License version 2. + + + + This program is distributed in the hope that it will be + useful, but WITHOUT ANY WARRANTY; without even the implied + warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + For more details see the file COPYING in the source + distribution of Linux. + + + + + + + + Introduction + + Protocol vs bus + + Once upon a time, the Small Computer Systems Interface defined both + a parallel I/O bus and a data protocol to connect a wide variety of + peripherals (disk drives, tape drives, modems, printers, scanners, + optical drives, test equipment, and medical devices) to a host + computer. + + + Although the old parallel (fast/wide/ultra) SCSI bus has largely + fallen out of use, the SCSI command set is more widely used than ever + to communicate with devices over a number of different busses. + + + The SCSI protocol + is a big-endian peer-to-peer packet based protocol. SCSI commands + are 6, 10, 12, or 16 bytes long, often followed by an associated data + payload. + + + SCSI commands can be transported over just about any kind of bus, and + are the default protocol for storage devices attached to USB, SATA, + SAS, Fibre Channel, FireWire, and ATAPI devices. SCSI packets are + also commonly exchanged over Infiniband, + I20, TCP/IP + (iSCSI), even + Parallel + ports. + + + + Design of the Linux SCSI subsystem + + The SCSI subsystem uses a three layer design, with upper, mid, and low + layers. Every operation involving the SCSI subsystem (such as reading + a sector from a disk) uses one driver at each of the 3 levels: one + upper layer driver, one lower layer driver, and the scsi midlayer. + + + The SCSI upper layer provides the interface between userspace and the + kernel, in the form of block and char device nodes for I/O and + ioctl(). The SCSI lower layer contains drivers for specific hardware + devices. + + + In between is the SCSI mid-layer, analogous to a network routing + layer such as the IPv4 stack. The SCSI mid-layer routes a packet + based data protocol between the upper layer's /dev nodes and the + corresponding devices in the lower layer. It manages command queues, + provides error handling and power management functions, and responds + to ioctl() requests. + + + + + + SCSI upper layer + + The upper layer supports the user-kernel interface by providing + device nodes. + + + sd (SCSI Disk) + sd (sd_mod.o) + + + + sr (SCSI CD-ROM) + sr (sr_mod.o) + + + st (SCSI Tape) + st (st.o) + + + sg (SCSI Generic) + sg (sg.o) + + + ch (SCSI Media Changer) + ch (ch.c) + + + + + SCSI mid layer + + + SCSI midlayer implementation + + include/scsi/scsi_device.h + + +!Iinclude/scsi/scsi_device.h + + + + drivers/scsi/scsi.c + Main file for the scsi midlayer. +!Edrivers/scsi/scsi.c + + + drivers/scsi/scsicam.c + + SCSI + Common Access Method support functions, for use with + HDIO_GETGEO, etc. + +!Edrivers/scsi/scsicam.c + + + drivers/scsi/scsi_error.c + Common SCSI error/timeout handling routines. +!Edrivers/scsi/scsi_error.c + + + drivers/scsi/scsi_devinfo.c + + Manage scsi_dev_info_list, which tracks blacklisted and whitelisted + devices. + +!Idrivers/scsi/scsi_devinfo.c + + + drivers/scsi/scsi_ioctl.c + + Handle ioctl() calls for scsi devices. + +!Edrivers/scsi/scsi_ioctl.c + + + drivers/scsi/scsi_lib.c + + SCSI queuing library. + +!Edrivers/scsi/scsi_lib.c + + + drivers/scsi/scsi_lib_dma.c + + SCSI library functions depending on DMA + (map and unmap scatter-gather lists). + +!Edrivers/scsi/scsi_lib_dma.c + + + drivers/scsi/scsi_module.c + + The file drivers/scsi/scsi_module.c contains legacy support for + old-style host templates. It should never be used by any new driver. + + + + drivers/scsi/scsi_proc.c + + The functions in this file provide an interface between + the PROC file system and the SCSI device drivers + It is mainly used for debugging, statistics and to pass + information directly to the lowlevel driver. + + I.E. plumbing to manage /proc/scsi/* + +!Idrivers/scsi/scsi_proc.c + + + drivers/scsi/scsi_netlink.c + + Infrastructure to provide async events from transports to userspace + via netlink, using a single NETLINK_SCSITRANSPORT protocol for all + transports. + + See the + original patch submission for more details. + +!Idrivers/scsi/scsi_netlink.c + + + drivers/scsi/scsi_scan.c + + Scan a host to determine which (if any) devices are attached. + + The general scanning/probing algorithm is as follows, exceptions are + made to it depending on device specific flags, compilation options, + and global variable (boot or module load time) settings. + + A specific LUN is scanned via an INQUIRY command; if the LUN has a + device attached, a scsi_device is allocated and setup for it. + + For every id of every channel on the given host, start by scanning + LUN 0. Skip hosts that don't respond at all to a scan of LUN 0. + Otherwise, if LUN 0 has a device attached, allocate and setup a + scsi_device for it. If target is SCSI-3 or up, issue a REPORT LUN, + and scan all of the LUNs returned by the REPORT LUN; else, + sequentially scan LUNs up until some maximum is reached, or a LUN is + seen that cannot have a device attached to it. + +!Idrivers/scsi/scsi_scan.c + + + drivers/scsi/scsi_sysctl.c + + Set up the sysctl entry: "/dev/scsi/logging_level" + (DEV_SCSI_LOGGING_LEVEL) which sets/returns scsi_logging_level. + + + + drivers/scsi/scsi_sysfs.c + + SCSI sysfs interface routines. + +!Edrivers/scsi/scsi_sysfs.c + + + drivers/scsi/hosts.c + + mid to lowlevel SCSI driver interface + +!Edrivers/scsi/hosts.c + + + drivers/scsi/constants.c + + mid to lowlevel SCSI driver interface + +!Edrivers/scsi/constants.c + + + + + Transport classes + + Transport classes are service libraries for drivers in the scsi + lower layer, which expose transport attributes in sysfs. + + + Fibre Channel transport + + The file drivers/scsi/scsi_transport_fc.c defines transport attributes + for Fibre Channel. + +!Edrivers/scsi/scsi_transport_fc.c + + + iSCSI transport class + + The file drivers/scsi/scsi_transport_iscsi.c defines transport + attributes for the iSCSI class, which sends SCSI packets over TCP/IP + connections. + +!Edrivers/scsi/scsi_transport_iscsi.c + + + Serial Attached SCSI (SAS) transport class + + The file drivers/scsi/scsi_transport_sas.c defines transport + attributes for Serial Attached SCSI, a variant of SATA aimed at + large high-end systems. + + + The SAS transport class contains common code to deal with SAS HBAs, + an aproximated representation of SAS topologies in the driver model, + and various sysfs attributes to expose these topologies and managment + interfaces to userspace. + + + In addition to the basic SCSI core objects this transport class + introduces two additional intermediate objects: The SAS PHY + as represented by struct sas_phy defines an "outgoing" PHY on + a SAS HBA or Expander, and the SAS remote PHY represented by + struct sas_rphy defines an "incoming" PHY on a SAS Expander or + end device. Note that this is purely a software concept, the + underlying hardware for a PHY and a remote PHY is the exactly + the same. + + + There is no concept of a SAS port in this code, users can see + what PHYs form a wide port based on the port_identifier attribute, + which is the same for all PHYs in a port. + +!Edrivers/scsi/scsi_transport_sas.c + + + SATA transport class + + The SATA transport is handled by libata, which has its own book of + documentation in this directory. + + + + Parallel SCSI (SPI) transport class + + The file drivers/scsi/scsi_transport_spi.c defines transport + attributes for traditional (fast/wide/ultra) SCSI busses. + +!Edrivers/scsi/scsi_transport_spi.c + + + SCSI RDMA (SRP) transport class + + The file drivers/scsi/scsi_transport_srp.c defines transport + attributes for SCSI over Remote Direct Memory Access. + +!Edrivers/scsi/scsi_transport_srp.c + + + + + + + SCSI lower layer + + Host Bus Adapter transport types + + Many modern device controllers use the SCSI command set as a protocol to + communicate with their devices through many different types of physical + connections. + + + In SCSI language a bus capable of carrying SCSI commands is + called a "transport", and a controller connecting to such a bus is + called a "host bus adapter" (HBA). + + + Debug transport + + The file drivers/scsi/scsi_debug.c simulates a host adapter with a + variable number of disks (or disk like devices) attached, sharing a + common amount of RAM. Does a lot of checking to make sure that we are + not getting blocks mixed up, and panics the kernel if anything out of + the ordinary is seen. + + + To be more realistic, the simulated devices have the transport + attributes of SAS disks. + + + For documentation see + http://www.torque.net/sg/sdebug26.html + + + + + todo + Parallel (fast/wide/ultra) SCSI, USB, SATA, + SAS, Fibre Channel, FireWire, ATAPI devices, Infiniband, + I20, iSCSI, Parallel ports, netlink... + + + + + diff --git a/Documentation/DocBook/scsi_midlayer.tmpl b/Documentation/DocBook/scsi_midlayer.tmpl deleted file mode 100644 index 6255930216b3..000000000000 --- a/Documentation/DocBook/scsi_midlayer.tmpl +++ /dev/null @@ -1,409 +0,0 @@ - - - - - - SCSI Mid Layer Guide - - - - James - Bottomley - -
- James.Bottomley@steeleye.com -
- - - - - Rob - Landley - -
- rob@landley.net -
- - - - - - - 2007 - Linux Foundation - - - - - This documentation is free software; you can redistribute - it and/or modify it under the terms of the GNU General Public - License version 2. - - - - This program is distributed in the hope that it will be - useful, but WITHOUT ANY WARRANTY; without even the implied - warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - For more details see the file COPYING in the source - distribution of Linux. - - - - - - - - Introduction - - Protocol vs bus - - Once upon a time, the Small Computer Systems Interface defined both - a parallel I/O bus and a data protocol to connect a wide variety of - peripherals (disk drives, tape drives, modems, printers, scanners, - optical drives, test equipment, and medical devices) to a host - computer. - - - Although the old parallel (fast/wide/ultra) SCSI bus has largely - fallen out of use, the SCSI command set is more widely used than ever - to communicate with devices over a number of different busses. - - - The SCSI protocol - is a big-endian peer-to-peer packet based protocol. SCSI commands - are 6, 10, 12, or 16 bytes long, often followed by an associated data - payload. - - - SCSI commands can be transported over just about any kind of bus, and - are the default protocol for storage devices attached to USB, SATA, - SAS, Fibre Channel, FireWire, and ATAPI devices. SCSI packets are - also commonly exchanged over Infiniband, - I20, TCP/IP - (iSCSI), even - Parallel - ports. - - - - Design of the Linux SCSI subsystem - - The SCSI subsystem uses a three layer design, with upper, mid, and low - layers. Every operation involving the SCSI subsystem (such as reading - a sector from a disk) uses one driver at each of the 3 levels: one - upper layer driver, one lower layer driver, and the scsi midlayer. - - - The SCSI upper layer provides the interface between userspace and the - kernel, in the form of block and char device nodes for I/O and - ioctl(). The SCSI lower layer contains drivers for specific hardware - devices. - - - In between is the SCSI mid-layer, analogous to a network routing - layer such as the IPv4 stack. The SCSI mid-layer routes a packet - based data protocol between the upper layer's /dev nodes and the - corresponding devices in the lower layer. It manages command queues, - provides error handling and power management functions, and responds - to ioctl() requests. - - - - - - SCSI upper layer - - The upper layer supports the user-kernel interface by providing - device nodes. - - - sd (SCSI Disk) - sd (sd_mod.o) - - - - sr (SCSI CD-ROM) - sr (sr_mod.o) - - - st (SCSI Tape) - st (st.o) - - - sg (SCSI Generic) - sg (sg.o) - - - ch (SCSI Media Changer) - ch (ch.c) - - - - - SCSI mid layer - - - SCSI midlayer implementation - - include/scsi/scsi_device.h - - -!Iinclude/scsi/scsi_device.h - - - - drivers/scsi/scsi.c - Main file for the scsi midlayer. -!Edrivers/scsi/scsi.c - - - drivers/scsi/scsicam.c - - SCSI - Common Access Method support functions, for use with - HDIO_GETGEO, etc. - -!Edrivers/scsi/scsicam.c - - - drivers/scsi/scsi_error.c - Common SCSI error/timeout handling routines. -!Edrivers/scsi/scsi_error.c - - - drivers/scsi/scsi_devinfo.c - - Manage scsi_dev_info_list, which tracks blacklisted and whitelisted - devices. - -!Idrivers/scsi/scsi_devinfo.c - - - drivers/scsi/scsi_ioctl.c - - Handle ioctl() calls for scsi devices. - -!Edrivers/scsi/scsi_ioctl.c - - - drivers/scsi/scsi_lib.c - - SCSI queuing library. - -!Edrivers/scsi/scsi_lib.c - - - drivers/scsi/scsi_lib_dma.c - - SCSI library functions depending on DMA - (map and unmap scatter-gather lists). - -!Edrivers/scsi/scsi_lib_dma.c - - - drivers/scsi/scsi_module.c - - The file drivers/scsi/scsi_module.c contains legacy support for - old-style host templates. It should never be used by any new driver. - - - - drivers/scsi/scsi_proc.c - - The functions in this file provide an interface between - the PROC file system and the SCSI device drivers - It is mainly used for debugging, statistics and to pass - information directly to the lowlevel driver. - - I.E. plumbing to manage /proc/scsi/* - -!Idrivers/scsi/scsi_proc.c - - - drivers/scsi/scsi_netlink.c - - Infrastructure to provide async events from transports to userspace - via netlink, using a single NETLINK_SCSITRANSPORT protocol for all - transports. - - See the - original patch submission for more details. - -!Idrivers/scsi/scsi_netlink.c - - - drivers/scsi/scsi_scan.c - - Scan a host to determine which (if any) devices are attached. - - The general scanning/probing algorithm is as follows, exceptions are - made to it depending on device specific flags, compilation options, - and global variable (boot or module load time) settings. - - A specific LUN is scanned via an INQUIRY command; if the LUN has a - device attached, a scsi_device is allocated and setup for it. - - For every id of every channel on the given host, start by scanning - LUN 0. Skip hosts that don't respond at all to a scan of LUN 0. - Otherwise, if LUN 0 has a device attached, allocate and setup a - scsi_device for it. If target is SCSI-3 or up, issue a REPORT LUN, - and scan all of the LUNs returned by the REPORT LUN; else, - sequentially scan LUNs up until some maximum is reached, or a LUN is - seen that cannot have a device attached to it. - -!Idrivers/scsi/scsi_scan.c - - - drivers/scsi/scsi_sysctl.c - - Set up the sysctl entry: "/dev/scsi/logging_level" - (DEV_SCSI_LOGGING_LEVEL) which sets/returns scsi_logging_level. - - - - drivers/scsi/scsi_sysfs.c - - SCSI sysfs interface routines. - -!Edrivers/scsi/scsi_sysfs.c - - - drivers/scsi/hosts.c - - mid to lowlevel SCSI driver interface - -!Edrivers/scsi/hosts.c - - - drivers/scsi/constants.c - - mid to lowlevel SCSI driver interface - -!Edrivers/scsi/constants.c - - - - - Transport classes - - Transport classes are service libraries for drivers in the scsi - lower layer, which expose transport attributes in sysfs. - - - Fibre Channel transport - - The file drivers/scsi/scsi_transport_fc.c defines transport attributes - for Fibre Channel. - -!Edrivers/scsi/scsi_transport_fc.c - - - iSCSI transport class - - The file drivers/scsi/scsi_transport_iscsi.c defines transport - attributes for the iSCSI class, which sends SCSI packets over TCP/IP - connections. - -!Edrivers/scsi/scsi_transport_iscsi.c - - - Serial Attached SCSI (SAS) transport class - - The file drivers/scsi/scsi_transport_sas.c defines transport - attributes for Serial Attached SCSI, a variant of SATA aimed at - large high-end systems. - - - The SAS transport class contains common code to deal with SAS HBAs, - an aproximated representation of SAS topologies in the driver model, - and various sysfs attributes to expose these topologies and managment - interfaces to userspace. - - - In addition to the basic SCSI core objects this transport class - introduces two additional intermediate objects: The SAS PHY - as represented by struct sas_phy defines an "outgoing" PHY on - a SAS HBA or Expander, and the SAS remote PHY represented by - struct sas_rphy defines an "incoming" PHY on a SAS Expander or - end device. Note that this is purely a software concept, the - underlying hardware for a PHY and a remote PHY is the exactly - the same. - - - There is no concept of a SAS port in this code, users can see - what PHYs form a wide port based on the port_identifier attribute, - which is the same for all PHYs in a port. - -!Edrivers/scsi/scsi_transport_sas.c - - - SATA transport class - - The SATA transport is handled by libata, which has its own book of - documentation in this directory. - - - - Parallel SCSI (SPI) transport class - - The file drivers/scsi/scsi_transport_spi.c defines transport - attributes for traditional (fast/wide/ultra) SCSI busses. - -!Edrivers/scsi/scsi_transport_spi.c - - - SCSI RDMA (SRP) transport class - - The file drivers/scsi/scsi_transport_srp.c defines transport - attributes for SCSI over Remote Direct Memory Access. - -!Edrivers/scsi/scsi_transport_srp.c - - - - - - - SCSI lower layer - - Host Bus Adapter transport types - - Many modern device controllers use the SCSI command set as a protocol to - communicate with their devices through many different types of physical - connections. - - - In SCSI language a bus capable of carrying SCSI commands is - called a "transport", and a controller connecting to such a bus is - called a "host bus adapter" (HBA). - - - Debug transport - - The file drivers/scsi/scsi_debug.c simulates a host adapter with a - variable number of disks (or disk like devices) attached, sharing a - common amount of RAM. Does a lot of checking to make sure that we are - not getting blocks mixed up, and panics the kernel if anything out of - the ordinary is seen. - - - To be more realistic, the simulated devices have the transport - attributes of SAS disks. - - - For documentation see - http://www.torque.net/sg/sdebug26.html - - - - - todo - Parallel (fast/wide/ultra) SCSI, USB, SATA, - SAS, Fibre Channel, FireWire, ATAPI devices, Infiniband, - I20, iSCSI, Parallel ports, netlink... - - - - - -- cgit v1.2.3 From f28cd7cf8f696eafe42d1632b5a306fbf784d3cd Mon Sep 17 00:00:00 2001 From: bo yang Date: Fri, 9 Nov 2007 04:44:56 -0500 Subject: [SCSI] megaraid_sas: Update version and changelog Update version and changelog Signed-off-by: Bo Yang Signed-off-by: James Bottomley --- Documentation/scsi/ChangeLog.megaraid_sas | 159 ++++++++++++++++++++++++++++++ drivers/scsi/megaraid/megaraid_sas.c | 10 +- drivers/scsi/megaraid/megaraid_sas.h | 8 +- 3 files changed, 168 insertions(+), 9 deletions(-) (limited to 'Documentation') diff --git a/Documentation/scsi/ChangeLog.megaraid_sas b/Documentation/scsi/ChangeLog.megaraid_sas index 5eb927544990..91c81db0ba71 100644 --- a/Documentation/scsi/ChangeLog.megaraid_sas +++ b/Documentation/scsi/ChangeLog.megaraid_sas @@ -1,3 +1,162 @@ +1 Release Date : Thur. Nov. 07 16:30:43 PST 2007 - + (emaild-id:megaraidlinux@lsi.com) + Sumant Patro + Bo Yang + +2 Current Version : 00.00.03.16 +3 Older Version : 00.00.03.15 + +1. Increased MFI_POLL_TIMEOUT_SECS to 60 seconds from 10. FW may take + a max of 60 seconds to respond to the INIT cmd. + +1 Release Date : Fri. Sep. 07 16:30:43 PST 2007 - + (emaild-id:megaraidlinux@lsi.com) + Sumant Patro + Bo Yang + +2 Current Version : 00.00.03.15 +3 Older Version : 00.00.03.14 + +1. Added module parameter "poll_mode_io" to support for "polling" + (reduced interrupt operation). In this mode, IO completion + interrupts are delayed. At the end of initiating IOs, the + driver schedules for cmd completion if there are pending cmds + to be completed. A timer-based interrupt has also been added + to prevent IO completion processing from being delayed + indefinitely in the case that no new IOs are initiated. + +1 Release Date : Fri. Sep. 07 16:30:43 PST 2007 - + (emaild-id:megaraidlinux@lsi.com) + Sumant Patro + Bo Yang + +2 Current Version : 00.00.03.14 +3 Older Version : 00.00.03.13 + +1. Setting the max_sectors_per_req based on max SGL supported by the + FW. Prior versions calculated this value from controller info + (max_sectors_1, max_sectors_2). For certain controllers/FW, + this was resulting in a value greater than max SGL supported + by the FW. Issue was first reported by users running LUKS+XFS + with megaraid_sas. Thanks to RB for providing the logs and + duplication steps that helped to get to the root cause of the + issue. 2. Increased MFI_POLL_TIMEOUT_SECS to 60 seconds from + 10. FW may take a max of 60 seconds to respond to the INIT + cmd. + +1 Release Date : Fri. June. 15 16:30:43 PST 2007 - + (emaild-id:megaraidlinux@lsi.com) + Sumant Patro + Bo Yang + +2 Current Version : 00.00.03.13 +3 Older Version : 00.00.03.12 + +1. Added the megasas_reset_timer routine to intercept cmd timeout and throttle io. + +On Fri, 2007-03-16 at 16:44 -0600, James Bottomley wrote: +It looks like megaraid_sas at least needs this to throttle its commands +> as they begin to time out. The code keeps the existing transport +> template use of eh_timed_out (and allows the transport to override the +> host if they both have this callback). +> +> James + +1 Release Date : Sat May. 12 16:30:43 PST 2007 - + (emaild-id:megaraidlinux@lsi.com) + Sumant Patro + Bo Yang + +2 Current Version : 00.00.03.12 +3 Older Version : 00.00.03.11 + +1. When MegaSAS driver receives reset call from OS, driver waits in reset +routine for max 3 minutes for all pending command completion. Now driver will +call completion routine every 5 seconds from the reset routine instead of +waiting for depending on cmd completion from isr path. + +1 Release Date : Mon Apr. 30 10:25:52 PST 2007 - + (emaild-id:megaraidlinux@lsi.com) + Sumant Patro + Bo Yang + +2 Current Version : 00.00.03.11 +3 Older Version : 00.00.03.09 + + 1. Memory Manager for IOCTL removed for 2.6 kernels. + pci_alloc_consistent replaced by dma_alloc_coherent. With this + change there is no need of memory manager in the driver code + + On Wed, 2007-02-07 at 13:30 -0800, Andrew Morton wrote: + > I suspect all this horror is due to stupidity in the DMA API. + > + > pci_alloc_consistent() just goes and assumes GFP_ATOMIC, whereas + > the caller (megasas_mgmt_fw_ioctl) would have been perfectly happy + > to use GFP_KERNEL. + > + > I bet this fixes it + + It does, but the DMA API was expanded to cope with this exact case, so + use dma_alloc_coherent() directly in the megaraid code instead. The dev + is just &pci_dev->dev. + + James + + 3. SYNCHRONIZE_CACHE is not supported by FW and thus blocked by driver. + 4. Hibernation support added + 5. Performing diskdump while running IO in RHEL 4 was failing. Fixed. + +1 Release Date : Fri Feb. 09 14:36:28 PST 2007 - + (emaild-id:megaraidlinux@lsi.com) + Sumant Patro + Bo Yang + +2 Current Version : 00.00.03.09 +3 Older Version : 00.00.03.08 + +i. Under heavy IO mid-layer prints "DRIVER_TIMEOUT" errors + + The driver now waits for 10 seconds to elapse instead of 5 (as in + previous release) to resume IO. + +1 Release Date : Mon Feb. 05 11:35:24 PST 2007 - + (emaild-id:megaraidlinux@lsi.com) + Sumant Patro + Bo Yang +2 Current Version : 00.00.03.08 +3 Older Version : 00.00.03.07 + +i. Under heavy IO mid-layer prints "DRIVER_TIMEOUT" errors + + Fix: The driver is now throttling IO. + Checks added in megasas_queue_command to know if FW is able to + process commands within timeout period. If number of retries + is 2 or greater,the driver stops sending cmd to FW temporarily. IO is + resumed if pending cmd count reduces to 16 or 5 seconds has elapsed + from the time cmds were last sent to FW. + +ii. FW enables WCE bit in Mode Sense cmd for drives that are configured + as WriteBack. The OS may send "SYNCHRONIZE_CACHE" cmd when Logical + Disks are exposed with WCE=1. User is advised to enable Write Back + mode only when the controller has battery backup. At this time + Synhronize cache is not supported by the FW. Driver will short-cycle + the cmd and return sucess without sending down to FW. + +1 Release Date : Sun Jan. 14 11:21:32 PDT 2007 - + Sumant Patro /Bo Yang +2 Current Version : 00.00.03.07 +3 Older Version : 00.00.03.06 + +i. bios_param entry added in scsi_host_template that returns disk geometry + information. + +1 Release Date : Fri Oct 20 11:21:32 PDT 2006 - Sumant Patro /Bo Yang +2 Current Version : 00.00.03.06 +3 Older Version : 00.00.03.05 + +1. Added new memory management module to support the IOCTL memory allocation. For IOCTL we try to allocate from the memory pool created during driver initialization. If mem pool is empty then we allocate at run time. +2. Added check in megasas_queue_command and dpc/isr routine to see if we have already declared adapter dead + (hw_crit_error=1). If hw_crit_error==1, now we donot accept any processing of pending cmds/accept any cmd from OS 1 Release Date : Mon Oct 02 11:21:32 PDT 2006 - Sumant Patro 2 Current Version : 00.00.03.05 diff --git a/drivers/scsi/megaraid/megaraid_sas.c b/drivers/scsi/megaraid/megaraid_sas.c index 499f994f503d..d7ec921865c4 100644 --- a/drivers/scsi/megaraid/megaraid_sas.c +++ b/drivers/scsi/megaraid/megaraid_sas.c @@ -2,7 +2,7 @@ * * Linux MegaRAID driver for SAS based RAID controllers * - * Copyright (c) 2003-2005 LSI Logic Corporation. + * Copyright (c) 2003-2005 LSI Corporation. * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -10,7 +10,7 @@ * 2 of the License, or (at your option) any later version. * * FILE : megaraid_sas.c - * Version : v00.00.03.10-rc5 + * Version : v00.00.03.16-rc1 * * Authors: * (email-id : megaraidlinux@lsi.com) @@ -58,7 +58,7 @@ MODULE_PARM_DESC(poll_mode_io, MODULE_LICENSE("GPL"); MODULE_VERSION(MEGASAS_VERSION); MODULE_AUTHOR("megaraidlinux@lsi.com"); -MODULE_DESCRIPTION("LSI Logic MegaRAID SAS Driver"); +MODULE_DESCRIPTION("LSI MegaRAID SAS Driver"); /* * PCI ID table for all supported controllers @@ -1182,7 +1182,7 @@ megasas_service_aen(struct megasas_instance *instance, struct megasas_cmd *cmd) static struct scsi_host_template megasas_template = { .module = THIS_MODULE, - .name = "LSI Logic SAS based MegaRAID driver", + .name = "LSI SAS based MegaRAID driver", .proc_name = "megaraid_sas", .slave_configure = megasas_slave_configure, .queuecommand = megasas_queue_command, @@ -1966,7 +1966,7 @@ static int megasas_init_mfi(struct megasas_instance *instance) */ instance->base_addr = pci_resource_start(instance->pdev, 0); - if (pci_request_regions(instance->pdev, "megasas: LSI Logic")) { + if (pci_request_regions(instance->pdev, "megasas: LSI")) { printk(KERN_DEBUG "megasas: IO memory region busy!\n"); return -EBUSY; } diff --git a/drivers/scsi/megaraid/megaraid_sas.h b/drivers/scsi/megaraid/megaraid_sas.h index dcc66fc9c2cb..6466bdf548c2 100644 --- a/drivers/scsi/megaraid/megaraid_sas.h +++ b/drivers/scsi/megaraid/megaraid_sas.h @@ -2,7 +2,7 @@ * * Linux MegaRAID driver for SAS based RAID controllers * - * Copyright (c) 2003-2005 LSI Logic Corporation. + * Copyright (c) 2003-2005 LSI Corporation. * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -18,9 +18,9 @@ /* * MegaRAID SAS Driver meta data */ -#define MEGASAS_VERSION "00.00.03.10-rc5" -#define MEGASAS_RELDATE "May 17, 2007" -#define MEGASAS_EXT_VERSION "Thu May 17 10:09:32 PDT 2007" +#define MEGASAS_VERSION "00.00.03.16-rc1" +#define MEGASAS_RELDATE "Nov. 07, 2007" +#define MEGASAS_EXT_VERSION "Thu. Nov. 07 10:09:32 PDT 2007" /* * Device IDs -- cgit v1.2.3 From 6af6632a1c1cac169fe7fbedb2f642b4ebb38323 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 14 Nov 2007 16:52:25 -0800 Subject: [SCSI] boot options: correct option name and tell where to find docs for it Minor corrections and additions to 'scsi_logging_level', as pointed out by Chuck Ebbert. Also point out the IBM S390-tools 'scsi_logging_level' script. Signed-off-by: Randy Dunlap Signed-off-by: James Bottomley --- Documentation/kernel-parameters.txt | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index c4178778e7fd..99938b8b50e1 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1593,7 +1593,13 @@ and is between 256 and 4096 characters. It is defined in the file Format: :: (flags are integer value) - scsi_logging= [SCSI] + scsi_logging_level= [SCSI] a bit mask of logging levels + See drivers/scsi/scsi_logging.h for bits. Also + settable via sysctl at dev.scsi.logging_level + (/proc/sys/dev/scsi/logging_level). + There is also a nice 'scsi_logging_level' script in the + S390-tools package, available for download at + http://www-128.ibm.com/developerworks/linux/linux390/s390-tools-1.5.4.html scsi_mod.scan= [SCSI] sync (default) scans SCSI busses as they are discovered. async scans them in kernel threads, -- cgit v1.2.3 From dc8875e1078961591af4fbf9556c44648bf28d11 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 15 Nov 2007 15:42:30 -0800 Subject: [SCSI] docbook and kernel-doc updates - Change title to remove "Mid-Layer" since the doc is about all of the SCSI layers. - Use "SCSI" instead of "scsi" in docbook text. - Use "*/" to end kernel-doc notation blocks. - A few other minor typo fixes. Signed-off-by: Randy Dunlap Signed-off-by: James Bottomley --- Documentation/DocBook/scsi.tmpl | 10 ++--- drivers/scsi/scsi.c | 2 +- drivers/scsi/scsi_error.c | 79 +++++++++++++++++++-------------------- drivers/scsi/scsi_transport_srp.c | 3 +- 4 files changed, 45 insertions(+), 49 deletions(-) (limited to 'Documentation') diff --git a/Documentation/DocBook/scsi.tmpl b/Documentation/DocBook/scsi.tmpl index 6255930216b3..f299ab182bbe 100644 --- a/Documentation/DocBook/scsi.tmpl +++ b/Documentation/DocBook/scsi.tmpl @@ -4,7 +4,7 @@ - SCSI Mid Layer Guide + SCSI Interfaces Guide @@ -92,7 +92,7 @@ The SCSI subsystem uses a three layer design, with upper, mid, and low layers. Every operation involving the SCSI subsystem (such as reading a sector from a disk) uses one driver at each of the 3 levels: one - upper layer driver, one lower layer driver, and the scsi midlayer. + upper layer driver, one lower layer driver, and the SCSI midlayer. The SCSI upper layer provides the interface between userspace and the @@ -154,7 +154,7 @@ drivers/scsi/scsi.c - Main file for the scsi midlayer. + Main file for the SCSI midlayer. !Edrivers/scsi/scsi.c @@ -182,7 +182,7 @@ drivers/scsi/scsi_ioctl.c - Handle ioctl() calls for scsi devices. + Handle ioctl() calls for SCSI devices. !Edrivers/scsi/scsi_ioctl.c @@ -287,7 +287,7 @@ Transport classes - Transport classes are service libraries for drivers in the scsi + Transport classes are service libraries for drivers in the SCSI lower layer, which expose transport attributes in sysfs. diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c index 412a351c6cf7..ebc019380835 100644 --- a/drivers/scsi/scsi.c +++ b/drivers/scsi/scsi.c @@ -960,7 +960,7 @@ EXPORT_SYMBOL(__starget_for_each_device); * reference. You must hold the host's host_lock over this call and * any access to the returned scsi_device. * - * Note: The only reason why drivers would want to use this is because + * Note: The only reason why drivers should use this is because * they need to access the device list in irq context. Otherwise you * really want to use scsi_device_lookup_by_target instead. **/ diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index 22f8140598ba..169bc595648a 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -62,7 +62,7 @@ void scsi_eh_wakeup(struct Scsi_Host *shost) * @shost: SCSI host to invoke error handling on. * * Schedule SCSI EH without scmd. - **/ + */ void scsi_schedule_eh(struct Scsi_Host *shost) { unsigned long flags; @@ -86,7 +86,7 @@ EXPORT_SYMBOL_GPL(scsi_schedule_eh); * * Return value: * 0 on failure. - **/ + */ int scsi_eh_scmd_add(struct scsi_cmnd *scmd, int eh_flag) { struct Scsi_Host *shost = scmd->device->host; @@ -121,7 +121,7 @@ int scsi_eh_scmd_add(struct scsi_cmnd *scmd, int eh_flag) * This should be turned into an inline function. Each scsi command * has its own timer, and as it is added to the queue, we set up the * timer. When the command completes, we cancel the timer. - **/ + */ void scsi_add_timer(struct scsi_cmnd *scmd, int timeout, void (*complete)(struct scsi_cmnd *)) { @@ -155,7 +155,7 @@ void scsi_add_timer(struct scsi_cmnd *scmd, int timeout, * Return value: * 1 if we were able to detach the timer. 0 if we blew it, and the * timer function has already started to run. - **/ + */ int scsi_delete_timer(struct scsi_cmnd *scmd) { int rtn; @@ -181,7 +181,7 @@ int scsi_delete_timer(struct scsi_cmnd *scmd) * only in that the normal completion handling might run, but if the * normal completion function determines that the timer has already * fired, then it mustn't do anything. - **/ + */ void scsi_times_out(struct scsi_cmnd *scmd) { enum scsi_eh_timer_return (* eh_timed_out)(struct scsi_cmnd *); @@ -224,7 +224,7 @@ void scsi_times_out(struct scsi_cmnd *scmd) * * Return value: * 0 when dev was taken offline by error recovery. 1 OK to proceed. - **/ + */ int scsi_block_when_processing_errors(struct scsi_device *sdev) { int online; @@ -245,7 +245,7 @@ EXPORT_SYMBOL(scsi_block_when_processing_errors); * scsi_eh_prt_fail_stats - Log info on failures. * @shost: scsi host being recovered. * @work_q: Queue of scsi cmds to process. - **/ + */ static inline void scsi_eh_prt_fail_stats(struct Scsi_Host *shost, struct list_head *work_q) { @@ -295,7 +295,7 @@ static inline void scsi_eh_prt_fail_stats(struct Scsi_Host *shost, * Notes: * When a deferred error is detected the current command has * not been executed and needs retrying. - **/ + */ static int scsi_check_sense(struct scsi_cmnd *scmd) { struct scsi_sense_hdr sshdr; @@ -398,7 +398,7 @@ static int scsi_check_sense(struct scsi_cmnd *scmd) * queued during error recovery. the main difference here is that we * don't allow for the possibility of retries here, and we are a lot * more restrictive about what we consider acceptable. - **/ + */ static int scsi_eh_completed_normally(struct scsi_cmnd *scmd) { /* @@ -452,7 +452,7 @@ static int scsi_eh_completed_normally(struct scsi_cmnd *scmd) /** * scsi_eh_done - Completion function for error handling. * @scmd: Cmd that is done. - **/ + */ static void scsi_eh_done(struct scsi_cmnd *scmd) { struct completion *eh_action; @@ -469,7 +469,7 @@ static void scsi_eh_done(struct scsi_cmnd *scmd) /** * scsi_try_host_reset - ask host adapter to reset itself * @scmd: SCSI cmd to send hsot reset. - **/ + */ static int scsi_try_host_reset(struct scsi_cmnd *scmd) { unsigned long flags; @@ -498,7 +498,7 @@ static int scsi_try_host_reset(struct scsi_cmnd *scmd) /** * scsi_try_bus_reset - ask host to perform a bus reset * @scmd: SCSI cmd to send bus reset. - **/ + */ static int scsi_try_bus_reset(struct scsi_cmnd *scmd) { unsigned long flags; @@ -533,7 +533,7 @@ static int scsi_try_bus_reset(struct scsi_cmnd *scmd) * unreliable for a given host, then the host itself needs to put a * timer on it, and set the host back to a consistent state prior to * returning. - **/ + */ static int scsi_try_bus_device_reset(struct scsi_cmnd *scmd) { int rtn; @@ -568,7 +568,7 @@ static int __scsi_try_to_abort_cmd(struct scsi_cmnd *scmd) * author of the low-level driver wishes this operation to be timed, * they can provide this facility themselves. helper functions in * scsi_error.c can be supplied to make this easier to do. - **/ + */ static int scsi_try_to_abort_cmd(struct scsi_cmnd *scmd) { /* @@ -601,7 +601,7 @@ static void scsi_abort_eh_cmnd(struct scsi_cmnd *scmd) * sent must be one that does not transfer any data. If @sense_bytes != 0 * @cmnd is ignored and this functions sets up a REQUEST_SENSE command * and cmnd buffers to read @sense_bytes into @scmd->sense_buffer. - **/ + */ void scsi_eh_prep_cmnd(struct scsi_cmnd *scmd, struct scsi_eh_save *ses, unsigned char *cmnd, int cmnd_size, unsigned sense_bytes) { @@ -667,7 +667,7 @@ EXPORT_SYMBOL(scsi_eh_prep_cmnd); * @ses: saved information from a coresponding call to scsi_prep_eh_cmnd * * Undo any damage done by above scsi_prep_eh_cmnd(). - **/ + */ void scsi_eh_restore_cmnd(struct scsi_cmnd* scmd, struct scsi_eh_save *ses) { /* @@ -697,7 +697,7 @@ EXPORT_SYMBOL(scsi_eh_restore_cmnd); * * Return value: * SUCCESS or FAILED or NEEDS_RETRY - **/ + */ static int scsi_send_eh_cmnd(struct scsi_cmnd *scmd, unsigned char *cmnd, int cmnd_size, int timeout, unsigned sense_bytes) { @@ -765,7 +765,7 @@ static int scsi_send_eh_cmnd(struct scsi_cmnd *scmd, unsigned char *cmnd, * Some hosts automatically obtain this information, others require * that we obtain it on our own. This function will *not* return until * the command either times out, or it completes. - **/ + */ static int scsi_request_sense(struct scsi_cmnd *scmd) { return scsi_send_eh_cmnd(scmd, NULL, 0, SENSE_TIMEOUT, ~0); @@ -782,7 +782,7 @@ static int scsi_request_sense(struct scsi_cmnd *scmd) * and that would disturb what we are doing. Thus we really want to * keep a list of pending commands for final completion, and once we * are ready to leave error handling we handle completion for real. - **/ + */ void scsi_eh_finish_cmd(struct scsi_cmnd *scmd, struct list_head *done_q) { scmd->device->host->host_failed--; @@ -810,7 +810,7 @@ EXPORT_SYMBOL(scsi_eh_finish_cmd); * * XXX: Long term this code should go away, but that needs an audit of * all LLDDs first. - **/ + */ int scsi_eh_get_sense(struct list_head *work_q, struct list_head *done_q) { @@ -862,7 +862,7 @@ EXPORT_SYMBOL_GPL(scsi_eh_get_sense); * * Return value: * 0 - Device is ready. 1 - Device NOT ready. - **/ + */ static int scsi_eh_tur(struct scsi_cmnd *scmd) { static unsigned char tur_command[6] = {TEST_UNIT_READY, 0, 0, 0, 0, 0}; @@ -897,7 +897,7 @@ retry_tur: * command that has timed out. If the command simply failed, it makes * no sense to try and abort the command, since as far as the shost * adapter is concerned, it isn't running. - **/ + */ static int scsi_eh_abort_cmds(struct list_head *work_q, struct list_head *done_q) { @@ -935,7 +935,7 @@ static int scsi_eh_abort_cmds(struct list_head *work_q, * * Return value: * 0 - Device is ready. 1 - Device NOT ready. - **/ + */ static int scsi_eh_try_stu(struct scsi_cmnd *scmd) { static unsigned char stu_command[6] = {START_STOP, 0, 0, 0, 1, 0}; @@ -963,7 +963,7 @@ static int scsi_eh_try_stu(struct scsi_cmnd *scmd) * Notes: * If commands are failing due to not ready, initializing command required, * try revalidating the device, which will end up sending a start unit. - **/ + */ static int scsi_eh_stu(struct Scsi_Host *shost, struct list_head *work_q, struct list_head *done_q) @@ -1017,7 +1017,7 @@ static int scsi_eh_stu(struct Scsi_Host *shost, * devices that are jammed or not - if we have multiple devices, it * makes no sense to try bus_device_reset - we really would need to try * a bus_reset instead. - **/ + */ static int scsi_eh_bus_device_reset(struct Scsi_Host *shost, struct list_head *work_q, struct list_head *done_q) @@ -1068,7 +1068,7 @@ static int scsi_eh_bus_device_reset(struct Scsi_Host *shost, * @shost: &scsi host being recovered. * @work_q: &list_head for pending commands. * @done_q: &list_head for processed commands. - **/ + */ static int scsi_eh_bus_reset(struct Scsi_Host *shost, struct list_head *work_q, struct list_head *done_q) @@ -1125,7 +1125,7 @@ static int scsi_eh_bus_reset(struct Scsi_Host *shost, * scsi_eh_host_reset - send a host reset * @work_q: list_head for processed commands. * @done_q: list_head for processed commands. - **/ + */ static int scsi_eh_host_reset(struct list_head *work_q, struct list_head *done_q) { @@ -1160,8 +1160,7 @@ static int scsi_eh_host_reset(struct list_head *work_q, * scsi_eh_offline_sdevs - offline scsi devices that fail to recover * @work_q: list_head for processed commands. * @done_q: list_head for processed commands. - * - **/ + */ static void scsi_eh_offline_sdevs(struct list_head *work_q, struct list_head *done_q) { @@ -1194,7 +1193,7 @@ static void scsi_eh_offline_sdevs(struct list_head *work_q, * is woken. In cases where the error code indicates an error that * doesn't require the error handler read (i.e. we don't need to * abort/reset), this function should return SUCCESS. - **/ + */ int scsi_decide_disposition(struct scsi_cmnd *scmd) { int rtn; @@ -1375,7 +1374,7 @@ int scsi_decide_disposition(struct scsi_cmnd *scmd) * * If scsi_allocate_request() fails for what ever reason, we * completely forget to lock the door. - **/ + */ static void scsi_eh_lock_door(struct scsi_device *sdev) { unsigned char cmnd[MAX_COMMAND_SIZE]; @@ -1399,7 +1398,7 @@ static void scsi_eh_lock_door(struct scsi_device *sdev) * Notes: * When we entered the error handler, we blocked all further i/o to * this device. we need to 'reverse' this process. - **/ + */ static void scsi_restart_operations(struct Scsi_Host *shost) { struct scsi_device *sdev; @@ -1445,8 +1444,7 @@ static void scsi_restart_operations(struct Scsi_Host *shost) * @shost: host to be recovered. * @work_q: &list_head for pending commands. * @done_q: &list_head for processed commands. - * - **/ + */ void scsi_eh_ready_devs(struct Scsi_Host *shost, struct list_head *work_q, struct list_head *done_q) @@ -1462,8 +1460,7 @@ EXPORT_SYMBOL_GPL(scsi_eh_ready_devs); /** * scsi_eh_flush_done_q - finish processed commands or retry them. * @done_q: list_head of processed commands. - * - **/ + */ void scsi_eh_flush_done_q(struct list_head *done_q) { struct scsi_cmnd *scmd, *next; @@ -1517,7 +1514,7 @@ EXPORT_SYMBOL(scsi_eh_flush_done_q); * scsi_finish_cmd() called for it. we do all of the retry stuff * here, so when we restart the host after we return it should have an * empty queue. - **/ + */ static void scsi_unjam_host(struct Scsi_Host *shost) { unsigned long flags; @@ -1544,7 +1541,7 @@ static void scsi_unjam_host(struct Scsi_Host *shost) * Notes: * This is the main error handling loop. This is run as a kernel thread * for every SCSI host and handles all error handling activity. - **/ + */ int scsi_error_handler(void *data) { struct Scsi_Host *shost = data; @@ -1773,7 +1770,7 @@ EXPORT_SYMBOL(scsi_reset_provider); * * Return value: * 1 if valid sense data information found, else 0; - **/ + */ int scsi_normalize_sense(const u8 *sense_buffer, int sb_len, struct scsi_sense_hdr *sshdr) { @@ -1839,7 +1836,7 @@ EXPORT_SYMBOL(scsi_command_normalize_sense); * * Return value: * pointer to start of (first) descriptor if found else NULL - **/ + */ const u8 * scsi_sense_desc_find(const u8 * sense_buffer, int sb_len, int desc_type) { @@ -1875,7 +1872,7 @@ EXPORT_SYMBOL(scsi_sense_desc_find); * * Return value: * 1 if information field found, 0 if not found. - **/ + */ int scsi_get_sense_info_fld(const u8 * sense_buffer, int sb_len, u64 * info_out) { diff --git a/drivers/scsi/scsi_transport_srp.c b/drivers/scsi/scsi_transport_srp.c index 739e327bba09..2445c98ae95e 100644 --- a/drivers/scsi/scsi_transport_srp.c +++ b/drivers/scsi/scsi_transport_srp.c @@ -185,11 +185,10 @@ static int srp_host_match(struct attribute_container *cont, struct device *dev) /** * srp_rport_add - add a SRP remote port to the device hierarchy - * * @shost: scsi host the remote port is connected to. * @ids: The port id for the remote port. * - * publishes a port to the rest of the system + * Publishes a port to the rest of the system. */ struct srp_rport *srp_rport_add(struct Scsi_Host *shost, struct srp_rport_identifiers *ids) -- cgit v1.2.3 From 00f5970193e22c48f399a2430635d6416b51befe Mon Sep 17 00:00:00 2001 From: HighPoint Linux Team Date: Thu, 13 Dec 2007 16:14:26 -0800 Subject: [SCSI] hptiop: add more adapter models and other fixes Most code changes were made to support adapters based on Marvell IOP, plus some other fixes. - add more PCI device IDs - support for adapters based on Marvell IOP - fix a result code translation error on big-endian systems - fix resource releasing bug when scsi_host_alloc() fail in hptiop_probe() - update scsi_cmnd.resid when finishing a request - correct some coding style issues [akpm@linux-foundation.org: type fixes] Signed-off-by: HighPoint Linux Team Signed-off-by: Andrew Morton Signed-off-by: James Bottomley --- Documentation/scsi/hptiop.txt | 30 ++- drivers/scsi/Kconfig | 4 +- drivers/scsi/hptiop.c | 593 +++++++++++++++++++++++++++++++++--------- drivers/scsi/hptiop.h | 124 ++++++--- 4 files changed, 576 insertions(+), 175 deletions(-) (limited to 'Documentation') diff --git a/Documentation/scsi/hptiop.txt b/Documentation/scsi/hptiop.txt index d28a31247d4c..a6eb4add1be6 100644 --- a/Documentation/scsi/hptiop.txt +++ b/Documentation/scsi/hptiop.txt @@ -1,9 +1,9 @@ -HIGHPOINT ROCKETRAID 3xxx RAID DRIVER (hptiop) +HIGHPOINT ROCKETRAID 3xxx/4xxx ADAPTER DRIVER (hptiop) Controller Register Map ------------------------- -The controller IOP is accessed via PCI BAR0. +For Intel IOP based adapters, the controller IOP is accessed via PCI BAR0: BAR0 offset Register 0x10 Inbound Message Register 0 @@ -18,6 +18,24 @@ The controller IOP is accessed via PCI BAR0. 0x40 Inbound Queue Port 0x44 Outbound Queue Port +For Marvell IOP based adapters, the IOP is accessed via PCI BAR0 and BAR1: + + BAR0 offset Register + 0x20400 Inbound Doorbell Register + 0x20404 Inbound Interrupt Mask Register + 0x20408 Outbound Doorbell Register + 0x2040C Outbound Interrupt Mask Register + + BAR1 offset Register + 0x0 Inbound Queue Head Pointer + 0x4 Inbound Queue Tail Pointer + 0x8 Outbound Queue Head Pointer + 0xC Outbound Queue Tail Pointer + 0x10 Inbound Message Register + 0x14 Outbound Message Register + 0x40-0x1040 Inbound Queue + 0x1040-0x2040 Outbound Queue + I/O Request Workflow ---------------------- @@ -73,15 +91,9 @@ The driver exposes following sysfs attributes: driver-version R driver version string firmware-version R firmware version string -The driver registers char device "hptiop" to communicate with HighPoint RAID -management software. Its ioctl routine acts as a general binary interface -between the IOP firmware and HighPoint RAID management software. New management -functions can be implemented in application/firmware without modification -in driver code. - ----------------------------------------------------------------------------- -Copyright (C) 2006 HighPoint Technologies, Inc. All Rights Reserved. +Copyright (C) 2006-2007 HighPoint Technologies, Inc. All Rights Reserved. This file is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig index becbb09d4c9b..e397599d54e3 100644 --- a/drivers/scsi/Kconfig +++ b/drivers/scsi/Kconfig @@ -573,10 +573,10 @@ config SCSI_ARCMSR_AER source "drivers/scsi/megaraid/Kconfig.megaraid" config SCSI_HPTIOP - tristate "HighPoint RocketRAID 3xxx Controller support" + tristate "HighPoint RocketRAID 3xxx/4xxx Controller support" depends on SCSI && PCI help - This option enables support for HighPoint RocketRAID 3xxx + This option enables support for HighPoint RocketRAID 3xxx/4xxx controllers. To compile this driver as a module, choose M here; the module diff --git a/drivers/scsi/hptiop.c b/drivers/scsi/hptiop.c index 0844331abb87..df1a76438e29 100644 --- a/drivers/scsi/hptiop.c +++ b/drivers/scsi/hptiop.c @@ -1,5 +1,5 @@ /* - * HighPoint RR3xxx controller driver for Linux + * HighPoint RR3xxx/4xxx controller driver for Linux * Copyright (C) 2006-2007 HighPoint Technologies, Inc. All Rights Reserved. * * This program is free software; you can redistribute it and/or modify @@ -38,80 +38,84 @@ #include "hptiop.h" MODULE_AUTHOR("HighPoint Technologies, Inc."); -MODULE_DESCRIPTION("HighPoint RocketRAID 3xxx SATA Controller Driver"); +MODULE_DESCRIPTION("HighPoint RocketRAID 3xxx/4xxx Controller Driver"); static char driver_name[] = "hptiop"; -static const char driver_name_long[] = "RocketRAID 3xxx SATA Controller driver"; -static const char driver_ver[] = "v1.2 (070830)"; - -static void hptiop_host_request_callback(struct hptiop_hba *hba, u32 tag); -static void hptiop_iop_request_callback(struct hptiop_hba *hba, u32 tag); +static const char driver_name_long[] = "RocketRAID 3xxx/4xxx Controller driver"; +static const char driver_ver[] = "v1.3 (071203)"; + +static int iop_send_sync_msg(struct hptiop_hba *hba, u32 msg, u32 millisec); +static void hptiop_finish_scsi_req(struct hptiop_hba *hba, u32 tag, + struct hpt_iop_request_scsi_command *req); +static void hptiop_host_request_callback_itl(struct hptiop_hba *hba, u32 tag); +static void hptiop_iop_request_callback_itl(struct hptiop_hba *hba, u32 tag); static void hptiop_message_callback(struct hptiop_hba *hba, u32 msg); -static inline void hptiop_pci_posting_flush(struct hpt_iopmu __iomem *iop) -{ - readl(&iop->outbound_intstatus); -} - -static int iop_wait_ready(struct hpt_iopmu __iomem *iop, u32 millisec) +static int iop_wait_ready_itl(struct hptiop_hba *hba, u32 millisec) { u32 req = 0; int i; for (i = 0; i < millisec; i++) { - req = readl(&iop->inbound_queue); + req = readl(&hba->u.itl.iop->inbound_queue); if (req != IOPMU_QUEUE_EMPTY) break; msleep(1); } if (req != IOPMU_QUEUE_EMPTY) { - writel(req, &iop->outbound_queue); - hptiop_pci_posting_flush(iop); + writel(req, &hba->u.itl.iop->outbound_queue); + readl(&hba->u.itl.iop->outbound_intstatus); return 0; } return -1; } -static void hptiop_request_callback(struct hptiop_hba *hba, u32 tag) +static int iop_wait_ready_mv(struct hptiop_hba *hba, u32 millisec) +{ + return iop_send_sync_msg(hba, IOPMU_INBOUND_MSG0_NOP, millisec); +} + +static void hptiop_request_callback_itl(struct hptiop_hba *hba, u32 tag) { if (tag & IOPMU_QUEUE_ADDR_HOST_BIT) - return hptiop_host_request_callback(hba, + hptiop_host_request_callback_itl(hba, tag & ~IOPMU_QUEUE_ADDR_HOST_BIT); else - return hptiop_iop_request_callback(hba, tag); + hptiop_iop_request_callback_itl(hba, tag); } -static inline void hptiop_drain_outbound_queue(struct hptiop_hba *hba) +static void hptiop_drain_outbound_queue_itl(struct hptiop_hba *hba) { u32 req; - while ((req = readl(&hba->iop->outbound_queue)) != IOPMU_QUEUE_EMPTY) { + while ((req = readl(&hba->u.itl.iop->outbound_queue)) != + IOPMU_QUEUE_EMPTY) { if (req & IOPMU_QUEUE_MASK_HOST_BITS) - hptiop_request_callback(hba, req); + hptiop_request_callback_itl(hba, req); else { struct hpt_iop_request_header __iomem * p; p = (struct hpt_iop_request_header __iomem *) - ((char __iomem *)hba->iop + req); + ((char __iomem *)hba->u.itl.iop + req); if (readl(&p->flags) & IOP_REQUEST_FLAG_SYNC_REQUEST) { if (readl(&p->context)) - hptiop_request_callback(hba, req); + hptiop_request_callback_itl(hba, req); else writel(1, &p->context); } else - hptiop_request_callback(hba, req); + hptiop_request_callback_itl(hba, req); } } } -static int __iop_intr(struct hptiop_hba *hba) +static int iop_intr_itl(struct hptiop_hba *hba) { - struct hpt_iopmu __iomem *iop = hba->iop; + struct hpt_iopmu_itl __iomem *iop = hba->u.itl.iop; u32 status; int ret = 0; @@ -119,6 +123,7 @@ static int __iop_intr(struct hptiop_hba *hba) if (status & IOPMU_OUTBOUND_INT_MSG0) { u32 msg = readl(&iop->outbound_msgaddr0); + dprintk("received outbound msg %x\n", msg); writel(IOPMU_OUTBOUND_INT_MSG0, &iop->outbound_intstatus); hptiop_message_callback(hba, msg); @@ -126,31 +131,115 @@ static int __iop_intr(struct hptiop_hba *hba) } if (status & IOPMU_OUTBOUND_INT_POSTQUEUE) { - hptiop_drain_outbound_queue(hba); + hptiop_drain_outbound_queue_itl(hba); + ret = 1; + } + + return ret; +} + +static u64 mv_outbound_read(struct hpt_iopmu_mv __iomem *mu) +{ + u32 outbound_tail = readl(&mu->outbound_tail); + u32 outbound_head = readl(&mu->outbound_head); + + if (outbound_tail != outbound_head) { + u64 p; + + memcpy_fromio(&p, &mu->outbound_q[mu->outbound_tail], 8); + outbound_tail++; + + if (outbound_tail == MVIOP_QUEUE_LEN) + outbound_tail = 0; + writel(outbound_tail, &mu->outbound_tail); + return p; + } else + return 0; +} + +static void mv_inbound_write(u64 p, struct hptiop_hba *hba) +{ + u32 inbound_head = readl(&hba->u.mv.mu->inbound_head); + u32 head = inbound_head + 1; + + if (head == MVIOP_QUEUE_LEN) + head = 0; + + memcpy_toio(&hba->u.mv.mu->inbound_q[inbound_head], &p, 8); + writel(head, &hba->u.mv.mu->inbound_head); + writel(MVIOP_MU_INBOUND_INT_POSTQUEUE, + &hba->u.mv.regs->inbound_doorbell); +} + +static void hptiop_request_callback_mv(struct hptiop_hba *hba, u64 tag) +{ + u32 req_type = (tag >> 5) & 0x7; + struct hpt_iop_request_scsi_command *req; + + dprintk("hptiop_request_callback_mv: tag=%llx\n", tag); + + BUG_ON((tag & MVIOP_MU_QUEUE_REQUEST_RETURN_CONTEXT) == 0); + + switch (req_type) { + case IOP_REQUEST_TYPE_GET_CONFIG: + case IOP_REQUEST_TYPE_SET_CONFIG: + hba->msg_done = 1; + break; + + case IOP_REQUEST_TYPE_SCSI_COMMAND: + req = hba->reqs[tag >> 8].req_virt; + if (likely(tag & MVIOP_MU_QUEUE_REQUEST_RESULT_BIT)) + req->header.result = cpu_to_le32(IOP_RESULT_SUCCESS); + + hptiop_finish_scsi_req(hba, tag>>8, req); + break; + + default: + break; + } +} + +static int iop_intr_mv(struct hptiop_hba *hba) +{ + u32 status; + int ret = 0; + + status = readl(&hba->u.mv.regs->outbound_doorbell); + writel(~status, &hba->u.mv.regs->outbound_doorbell); + + if (status & MVIOP_MU_OUTBOUND_INT_MSG) { + u32 msg; + msg = readl(&hba->u.mv.mu->outbound_msg); + dprintk("received outbound msg %x\n", msg); + hptiop_message_callback(hba, msg); + ret = 1; + } + + if (status & MVIOP_MU_OUTBOUND_INT_POSTQUEUE) { + u64 tag; + + while ((tag = mv_outbound_read(hba->u.mv.mu))) + hptiop_request_callback_mv(hba, tag); ret = 1; } return ret; } -static int iop_send_sync_request(struct hptiop_hba *hba, +static int iop_send_sync_request_itl(struct hptiop_hba *hba, void __iomem *_req, u32 millisec) { struct hpt_iop_request_header __iomem *req = _req; u32 i; - writel(readl(&req->flags) | IOP_REQUEST_FLAG_SYNC_REQUEST, - &req->flags); - + writel(readl(&req->flags) | IOP_REQUEST_FLAG_SYNC_REQUEST, &req->flags); writel(0, &req->context); - - writel((unsigned long)req - (unsigned long)hba->iop, - &hba->iop->inbound_queue); - - hptiop_pci_posting_flush(hba->iop); + writel((unsigned long)req - (unsigned long)hba->u.itl.iop, + &hba->u.itl.iop->inbound_queue); + readl(&hba->u.itl.iop->outbound_intstatus); for (i = 0; i < millisec; i++) { - __iop_intr(hba); + iop_intr_itl(hba); if (readl(&req->context)) return 0; msleep(1); @@ -159,19 +248,49 @@ static int iop_send_sync_request(struct hptiop_hba *hba, return -1; } -static int iop_send_sync_msg(struct hptiop_hba *hba, u32 msg, u32 millisec) +static int iop_send_sync_request_mv(struct hptiop_hba *hba, + u32 size_bits, u32 millisec) { + struct hpt_iop_request_header *reqhdr = hba->u.mv.internal_req; u32 i; hba->msg_done = 0; + reqhdr->flags |= cpu_to_le32(IOP_REQUEST_FLAG_SYNC_REQUEST); + mv_inbound_write(hba->u.mv.internal_req_phy | + MVIOP_MU_QUEUE_ADDR_HOST_BIT | size_bits, hba); + + for (i = 0; i < millisec; i++) { + iop_intr_mv(hba); + if (hba->msg_done) + return 0; + msleep(1); + } + return -1; +} + +static void hptiop_post_msg_itl(struct hptiop_hba *hba, u32 msg) +{ + writel(msg, &hba->u.itl.iop->inbound_msgaddr0); + readl(&hba->u.itl.iop->outbound_intstatus); +} + +static void hptiop_post_msg_mv(struct hptiop_hba *hba, u32 msg) +{ + writel(msg, &hba->u.mv.mu->inbound_msg); + writel(MVIOP_MU_INBOUND_INT_MSG, &hba->u.mv.regs->inbound_doorbell); + readl(&hba->u.mv.regs->inbound_doorbell); +} - writel(msg, &hba->iop->inbound_msgaddr0); +static int iop_send_sync_msg(struct hptiop_hba *hba, u32 msg, u32 millisec) +{ + u32 i; - hptiop_pci_posting_flush(hba->iop); + hba->msg_done = 0; + hba->ops->post_msg(hba, msg); for (i = 0; i < millisec; i++) { spin_lock_irq(hba->host->host_lock); - __iop_intr(hba); + hba->ops->iop_intr(hba); spin_unlock_irq(hba->host->host_lock); if (hba->msg_done) break; @@ -181,46 +300,67 @@ static int iop_send_sync_msg(struct hptiop_hba *hba, u32 msg, u32 millisec) return hba->msg_done? 0 : -1; } -static int iop_get_config(struct hptiop_hba *hba, +static int iop_get_config_itl(struct hptiop_hba *hba, struct hpt_iop_request_get_config *config) { u32 req32; struct hpt_iop_request_get_config __iomem *req; - req32 = readl(&hba->iop->inbound_queue); + req32 = readl(&hba->u.itl.iop->inbound_queue); if (req32 == IOPMU_QUEUE_EMPTY) return -1; req = (struct hpt_iop_request_get_config __iomem *) - ((unsigned long)hba->iop + req32); + ((unsigned long)hba->u.itl.iop + req32); writel(0, &req->header.flags); writel(IOP_REQUEST_TYPE_GET_CONFIG, &req->header.type); writel(sizeof(struct hpt_iop_request_get_config), &req->header.size); writel(IOP_RESULT_PENDING, &req->header.result); - if (iop_send_sync_request(hba, req, 20000)) { + if (iop_send_sync_request_itl(hba, req, 20000)) { dprintk("Get config send cmd failed\n"); return -1; } memcpy_fromio(config, req, sizeof(*config)); - writel(req32, &hba->iop->outbound_queue); + writel(req32, &hba->u.itl.iop->outbound_queue); + return 0; +} + +static int iop_get_config_mv(struct hptiop_hba *hba, + struct hpt_iop_request_get_config *config) +{ + struct hpt_iop_request_get_config *req = hba->u.mv.internal_req; + + req->header.flags = cpu_to_le32(IOP_REQUEST_FLAG_OUTPUT_CONTEXT); + req->header.type = cpu_to_le32(IOP_REQUEST_TYPE_GET_CONFIG); + req->header.size = + cpu_to_le32(sizeof(struct hpt_iop_request_get_config)); + req->header.result = cpu_to_le32(IOP_RESULT_PENDING); + req->header.context = cpu_to_le64(IOP_REQUEST_TYPE_GET_CONFIG<<5); + + if (iop_send_sync_request_mv(hba, 0, 20000)) { + dprintk("Get config send cmd failed\n"); + return -1; + } + + memcpy(config, req, sizeof(struct hpt_iop_request_get_config)); return 0; } -static int iop_set_config(struct hptiop_hba *hba, +static int iop_set_config_itl(struct hptiop_hba *hba, struct hpt_iop_request_set_config *config) { u32 req32; struct hpt_iop_request_set_config __iomem *req; - req32 = readl(&hba->iop->inbound_queue); + req32 = readl(&hba->u.itl.iop->inbound_queue); if (req32 == IOPMU_QUEUE_EMPTY) return -1; req = (struct hpt_iop_request_set_config __iomem *) - ((unsigned long)hba->iop + req32); + ((unsigned long)hba->u.itl.iop + req32); memcpy_toio((u8 __iomem *)req + sizeof(struct hpt_iop_request_header), (u8 *)config + sizeof(struct hpt_iop_request_header), @@ -232,22 +372,52 @@ static int iop_set_config(struct hptiop_hba *hba, writel(sizeof(struct hpt_iop_request_set_config), &req->header.size); writel(IOP_RESULT_PENDING, &req->header.result); - if (iop_send_sync_request(hba, req, 20000)) { + if (iop_send_sync_request_itl(hba, req, 20000)) { dprintk("Set config send cmd failed\n"); return -1; } - writel(req32, &hba->iop->outbound_queue); + writel(req32, &hba->u.itl.iop->outbound_queue); return 0; } -static int hptiop_initialize_iop(struct hptiop_hba *hba) +static int iop_set_config_mv(struct hptiop_hba *hba, + struct hpt_iop_request_set_config *config) { - struct hpt_iopmu __iomem *iop = hba->iop; + struct hpt_iop_request_set_config *req = hba->u.mv.internal_req; - /* enable interrupts */ + memcpy(req, config, sizeof(struct hpt_iop_request_set_config)); + req->header.flags = cpu_to_le32(IOP_REQUEST_FLAG_OUTPUT_CONTEXT); + req->header.type = cpu_to_le32(IOP_REQUEST_TYPE_SET_CONFIG); + req->header.size = + cpu_to_le32(sizeof(struct hpt_iop_request_set_config)); + req->header.result = cpu_to_le32(IOP_RESULT_PENDING); + req->header.context = cpu_to_le64(IOP_REQUEST_TYPE_SET_CONFIG<<5); + + if (iop_send_sync_request_mv(hba, 0, 20000)) { + dprintk("Set config send cmd failed\n"); + return -1; + } + + return 0; +} + +static void hptiop_enable_intr_itl(struct hptiop_hba *hba) +{ writel(~(IOPMU_OUTBOUND_INT_POSTQUEUE | IOPMU_OUTBOUND_INT_MSG0), - &iop->outbound_intmask); + &hba->u.itl.iop->outbound_intmask); +} + +static void hptiop_enable_intr_mv(struct hptiop_hba *hba) +{ + writel(MVIOP_MU_OUTBOUND_INT_POSTQUEUE | MVIOP_MU_OUTBOUND_INT_MSG, + &hba->u.mv.regs->outbound_intmask); +} + +static int hptiop_initialize_iop(struct hptiop_hba *hba) +{ + /* enable interrupts */ + hba->ops->enable_intr(hba); hba->initialized = 1; @@ -261,37 +431,74 @@ static int hptiop_initialize_iop(struct hptiop_hba *hba) return 0; } -static int hptiop_map_pci_bar(struct hptiop_hba *hba) +static void __iomem *hptiop_map_pci_bar(struct hptiop_hba *hba, int index) { u32 mem_base_phy, length; void __iomem *mem_base_virt; + struct pci_dev *pcidev = hba->pcidev; - if (!(pci_resource_flags(pcidev, 0) & IORESOURCE_MEM)) { + + if (!(pci_resource_flags(pcidev, index) & IORESOURCE_MEM)) { printk(KERN_ERR "scsi%d: pci resource invalid\n", hba->host->host_no); - return -1; + return 0; } - mem_base_phy = pci_resource_start(pcidev, 0); - length = pci_resource_len(pcidev, 0); + mem_base_phy = pci_resource_start(pcidev, index); + length = pci_resource_len(pcidev, index); mem_base_virt = ioremap(mem_base_phy, length); if (!mem_base_virt) { printk(KERN_ERR "scsi%d: Fail to ioremap memory space\n", hba->host->host_no); + return 0; + } + return mem_base_virt; +} + +static int hptiop_map_pci_bar_itl(struct hptiop_hba *hba) +{ + hba->u.itl.iop = hptiop_map_pci_bar(hba, 0); + if (hba->u.itl.iop) + return 0; + else + return -1; +} + +static void hptiop_unmap_pci_bar_itl(struct hptiop_hba *hba) +{ + iounmap(hba->u.itl.iop); +} + +static int hptiop_map_pci_bar_mv(struct hptiop_hba *hba) +{ + hba->u.mv.regs = hptiop_map_pci_bar(hba, 0); + if (hba->u.mv.regs == 0) + return -1; + + hba->u.mv.mu = hptiop_map_pci_bar(hba, 2); + if (hba->u.mv.mu == 0) { + iounmap(hba->u.mv.regs); return -1; } - hba->iop = mem_base_virt; - dprintk("hptiop_map_pci_bar: iop=%p\n", hba->iop); return 0; } +static void hptiop_unmap_pci_bar_mv(struct hptiop_hba *hba) +{ + iounmap(hba->u.mv.regs); + iounmap(hba->u.mv.mu); +} + static void hptiop_message_callback(struct hptiop_hba *hba, u32 msg) { dprintk("iop message 0x%x\n", msg); + if (msg == IOPMU_INBOUND_MSG0_NOP) + hba->msg_done = 1; + if (!hba->initialized) return; @@ -303,7 +510,7 @@ static void hptiop_message_callback(struct hptiop_hba *hba, u32 msg) hba->msg_done = 1; } -static inline struct hptiop_request *get_req(struct hptiop_hba *hba) +static struct hptiop_request *get_req(struct hptiop_hba *hba) { struct hptiop_request *ret; @@ -316,30 +523,19 @@ static inline struct hptiop_request *get_req(struct hptiop_hba *hba) return ret; } -static inline void free_req(struct hptiop_hba *hba, struct hptiop_request *req) +static void free_req(struct hptiop_hba *hba, struct hptiop_request *req) { dprintk("free_req(%d, %p)\n", req->index, req); req->next = hba->req_list; hba->req_list = req; } -static void hptiop_host_request_callback(struct hptiop_hba *hba, u32 _tag) +static void hptiop_finish_scsi_req(struct hptiop_hba *hba, u32 tag, + struct hpt_iop_request_scsi_command *req) { - struct hpt_iop_request_scsi_command *req; struct scsi_cmnd *scp; - u32 tag; - - if (hba->iopintf_v2) { - tag = _tag & ~ IOPMU_QUEUE_REQUEST_RESULT_BIT; - req = hba->reqs[tag].req_virt; - if (likely(_tag & IOPMU_QUEUE_REQUEST_RESULT_BIT)) - req->header.result = IOP_RESULT_SUCCESS; - } else { - tag = _tag; - req = hba->reqs[tag].req_virt; - } - dprintk("hptiop_host_request_callback: req=%p, type=%d, " + dprintk("hptiop_finish_scsi_req: req=%p, type=%d, " "result=%d, context=0x%x tag=%d\n", req, req->header.type, req->header.result, req->header.context, tag); @@ -354,6 +550,8 @@ static void hptiop_host_request_callback(struct hptiop_hba *hba, u32 _tag) switch (le32_to_cpu(req->header.result)) { case IOP_RESULT_SUCCESS: + scsi_set_resid(scp, + scsi_bufflen(scp) - le32_to_cpu(req->dataxfer_length)); scp->result = (DID_OK<<16); break; case IOP_RESULT_BAD_TARGET: @@ -371,12 +569,12 @@ static void hptiop_host_request_callback(struct hptiop_hba *hba, u32 _tag) case IOP_RESULT_INVALID_REQUEST: scp->result = (DID_ABORT<<16); break; - case IOP_RESULT_MODE_SENSE_CHECK_CONDITION: + case IOP_RESULT_CHECK_CONDITION: + scsi_set_resid(scp, + scsi_bufflen(scp) - le32_to_cpu(req->dataxfer_length)); scp->result = SAM_STAT_CHECK_CONDITION; - memset(&scp->sense_buffer, - 0, sizeof(scp->sense_buffer)); memcpy(&scp->sense_buffer, &req->sg_list, - min(sizeof(scp->sense_buffer), + min_t(size_t, sizeof(scp->sense_buffer), le32_to_cpu(req->dataxfer_length))); break; @@ -391,15 +589,33 @@ static void hptiop_host_request_callback(struct hptiop_hba *hba, u32 _tag) free_req(hba, &hba->reqs[tag]); } -void hptiop_iop_request_callback(struct hptiop_hba *hba, u32 tag) +static void hptiop_host_request_callback_itl(struct hptiop_hba *hba, u32 _tag) +{ + struct hpt_iop_request_scsi_command *req; + u32 tag; + + if (hba->iopintf_v2) { + tag = _tag & ~IOPMU_QUEUE_REQUEST_RESULT_BIT; + req = hba->reqs[tag].req_virt; + if (likely(_tag & IOPMU_QUEUE_REQUEST_RESULT_BIT)) + req->header.result = cpu_to_le32(IOP_RESULT_SUCCESS); + } else { + tag = _tag; + req = hba->reqs[tag].req_virt; + } + + hptiop_finish_scsi_req(hba, tag, req); +} + +void hptiop_iop_request_callback_itl(struct hptiop_hba *hba, u32 tag) { struct hpt_iop_request_header __iomem *req; struct hpt_iop_request_ioctl_command __iomem *p; struct hpt_ioctl_k *arg; req = (struct hpt_iop_request_header __iomem *) - ((unsigned long)hba->iop + tag); - dprintk("hptiop_iop_request_callback: req=%p, type=%d, " + ((unsigned long)hba->u.itl.iop + tag); + dprintk("hptiop_iop_request_callback_itl: req=%p, type=%d, " "result=%d, context=0x%x tag=%d\n", req, readl(&req->type), readl(&req->result), readl(&req->context), tag); @@ -427,7 +643,7 @@ void hptiop_iop_request_callback(struct hptiop_hba *hba, u32 tag) arg->result = HPT_IOCTL_RESULT_FAILED; arg->done(arg); - writel(tag, &hba->iop->outbound_queue); + writel(tag, &hba->u.itl.iop->outbound_queue); } static irqreturn_t hptiop_intr(int irq, void *dev_id) @@ -437,7 +653,7 @@ static irqreturn_t hptiop_intr(int irq, void *dev_id) unsigned long flags; spin_lock_irqsave(hba->host->host_lock, flags); - handled = __iop_intr(hba); + handled = hba->ops->iop_intr(hba); spin_unlock_irqrestore(hba->host->host_lock, flags); return handled; @@ -469,6 +685,57 @@ static int hptiop_buildsgl(struct scsi_cmnd *scp, struct hpt_iopsg *psg) return HPT_SCP(scp)->sgcnt; } +static void hptiop_post_req_itl(struct hptiop_hba *hba, + struct hptiop_request *_req) +{ + struct hpt_iop_request_header *reqhdr = _req->req_virt; + + reqhdr->context = cpu_to_le32(IOPMU_QUEUE_ADDR_HOST_BIT | + (u32)_req->index); + reqhdr->context_hi32 = 0; + + if (hba->iopintf_v2) { + u32 size, size_bits; + + size = le32_to_cpu(reqhdr->size); + if (size < 256) + size_bits = IOPMU_QUEUE_REQUEST_SIZE_BIT; + else if (size < 512) + size_bits = IOPMU_QUEUE_ADDR_HOST_BIT; + else + size_bits = IOPMU_QUEUE_REQUEST_SIZE_BIT | + IOPMU_QUEUE_ADDR_HOST_BIT; + writel(_req->req_shifted_phy | size_bits, + &hba->u.itl.iop->inbound_queue); + } else + writel(_req->req_shifted_phy | IOPMU_QUEUE_ADDR_HOST_BIT, + &hba->u.itl.iop->inbound_queue); +} + +static void hptiop_post_req_mv(struct hptiop_hba *hba, + struct hptiop_request *_req) +{ + struct hpt_iop_request_header *reqhdr = _req->req_virt; + u32 size, size_bit; + + reqhdr->context = cpu_to_le32(_req->index<<8 | + IOP_REQUEST_TYPE_SCSI_COMMAND<<5); + reqhdr->context_hi32 = 0; + size = le32_to_cpu(reqhdr->size); + + if (size <= 256) + size_bit = 0; + else if (size <= 256*2) + size_bit = 1; + else if (size <= 256*3) + size_bit = 2; + else + size_bit = 3; + + mv_inbound_write((_req->req_shifted_phy << 5) | + MVIOP_MU_QUEUE_ADDR_HOST_BIT | size_bit, hba); +} + static int hptiop_queuecommand(struct scsi_cmnd *scp, void (*done)(struct scsi_cmnd *)) { @@ -518,9 +785,6 @@ static int hptiop_queuecommand(struct scsi_cmnd *scp, req->header.flags = cpu_to_le32(IOP_REQUEST_FLAG_OUTPUT_CONTEXT); req->header.type = cpu_to_le32(IOP_REQUEST_TYPE_SCSI_COMMAND); req->header.result = cpu_to_le32(IOP_RESULT_PENDING); - req->header.context = cpu_to_le32(IOPMU_QUEUE_ADDR_HOST_BIT | - (u32)_req->index); - req->header.context_hi32 = 0; req->dataxfer_length = cpu_to_le32(scsi_bufflen(scp)); req->channel = scp->device->channel; req->target = scp->device->id; @@ -531,21 +795,7 @@ static int hptiop_queuecommand(struct scsi_cmnd *scp, + sg_count * sizeof(struct hpt_iopsg)); memcpy(req->cdb, scp->cmnd, sizeof(req->cdb)); - - if (hba->iopintf_v2) { - u32 size_bits; - if (req->header.size < 256) - size_bits = IOPMU_QUEUE_REQUEST_SIZE_BIT; - else if (req->header.size < 512) - size_bits = IOPMU_QUEUE_ADDR_HOST_BIT; - else - size_bits = IOPMU_QUEUE_REQUEST_SIZE_BIT | - IOPMU_QUEUE_ADDR_HOST_BIT; - writel(_req->req_shifted_phy | size_bits, &hba->iop->inbound_queue); - } else - writel(_req->req_shifted_phy | IOPMU_QUEUE_ADDR_HOST_BIT, - &hba->iop->inbound_queue); - + hba->ops->post_req(hba, _req); return 0; cmd_done: @@ -563,9 +813,7 @@ static int hptiop_reset_hba(struct hptiop_hba *hba) { if (atomic_xchg(&hba->resetting, 1) == 0) { atomic_inc(&hba->reset_count); - writel(IOPMU_INBOUND_MSG0_RESET, - &hba->iop->inbound_msgaddr0); - hptiop_pci_posting_flush(hba->iop); + hba->ops->post_msg(hba, IOPMU_INBOUND_MSG0_RESET); } wait_event_timeout(hba->reset_wq, @@ -601,8 +849,10 @@ static int hptiop_reset(struct scsi_cmnd *scp) static int hptiop_adjust_disk_queue_depth(struct scsi_device *sdev, int queue_depth) { - if(queue_depth > 256) - queue_depth = 256; + struct hptiop_hba *hba = (struct hptiop_hba *)sdev->host->hostdata; + + if (queue_depth > hba->max_requests) + queue_depth = hba->max_requests; scsi_adjust_queue_depth(sdev, MSG_ORDERED_TAG, queue_depth); return queue_depth; } @@ -663,6 +913,26 @@ static struct scsi_host_template driver_template = { .change_queue_depth = hptiop_adjust_disk_queue_depth, }; +static int hptiop_internal_memalloc_mv(struct hptiop_hba *hba) +{ + hba->u.mv.internal_req = dma_alloc_coherent(&hba->pcidev->dev, + 0x800, &hba->u.mv.internal_req_phy, GFP_KERNEL); + if (hba->u.mv.internal_req) + return 0; + else + return -1; +} + +static int hptiop_internal_memfree_mv(struct hptiop_hba *hba) +{ + if (hba->u.mv.internal_req) { + dma_free_coherent(&hba->pcidev->dev, 0x800, + hba->u.mv.internal_req, hba->u.mv.internal_req_phy); + return 0; + } else + return -1; +} + static int __devinit hptiop_probe(struct pci_dev *pcidev, const struct pci_device_id *id) { @@ -708,6 +978,7 @@ static int __devinit hptiop_probe(struct pci_dev *pcidev, hba = (struct hptiop_hba *)host->hostdata; + hba->ops = (struct hptiop_adapter_ops *)id->driver_data; hba->pcidev = pcidev; hba->host = host; hba->initialized = 0; @@ -725,16 +996,24 @@ static int __devinit hptiop_probe(struct pci_dev *pcidev, host->n_io_port = 0; host->irq = pcidev->irq; - if (hptiop_map_pci_bar(hba)) + if (hba->ops->map_pci_bar(hba)) goto free_scsi_host; - if (iop_wait_ready(hba->iop, 20000)) { + if (hba->ops->iop_wait_ready(hba, 20000)) { printk(KERN_ERR "scsi%d: firmware not ready\n", hba->host->host_no); goto unmap_pci_bar; } - if (iop_get_config(hba, &iop_config)) { + if (hba->ops->internal_memalloc) { + if (hba->ops->internal_memalloc(hba)) { + printk(KERN_ERR "scsi%d: internal_memalloc failed\n", + hba->host->host_no); + goto unmap_pci_bar; + } + } + + if (hba->ops->get_config(hba, &iop_config)) { printk(KERN_ERR "scsi%d: get config failed\n", hba->host->host_no); goto unmap_pci_bar; @@ -770,7 +1049,7 @@ static int __devinit hptiop_probe(struct pci_dev *pcidev, set_config.vbus_id = cpu_to_le16(host->host_no); set_config.max_host_request_size = cpu_to_le16(req_size); - if (iop_set_config(hba, &set_config)) { + if (hba->ops->set_config(hba, &set_config)) { printk(KERN_ERR "scsi%d: set config failed\n", hba->host->host_no); goto unmap_pci_bar; @@ -839,21 +1118,24 @@ static int __devinit hptiop_probe(struct pci_dev *pcidev, free_request_mem: dma_free_coherent(&hba->pcidev->dev, - hba->req_size*hba->max_requests + 0x20, + hba->req_size * hba->max_requests + 0x20, hba->dma_coherent, hba->dma_coherent_handle); free_request_irq: free_irq(hba->pcidev->irq, hba); unmap_pci_bar: - iounmap(hba->iop); + if (hba->ops->internal_memfree) + hba->ops->internal_memfree(hba); -free_pci_regions: - pci_release_regions(pcidev) ; + hba->ops->unmap_pci_bar(hba); free_scsi_host: scsi_host_put(host); +free_pci_regions: + pci_release_regions(pcidev); + disable_pci_device: pci_disable_device(pcidev); @@ -865,8 +1147,6 @@ static void hptiop_shutdown(struct pci_dev *pcidev) { struct Scsi_Host *host = pci_get_drvdata(pcidev); struct hptiop_hba *hba = (struct hptiop_hba *)host->hostdata; - struct hpt_iopmu __iomem *iop = hba->iop; - u32 int_mask; dprintk("hptiop_shutdown(%p)\n", hba); @@ -876,11 +1156,24 @@ static void hptiop_shutdown(struct pci_dev *pcidev) hba->host->host_no); /* disable all outbound interrupts */ - int_mask = readl(&iop->outbound_intmask); + hba->ops->disable_intr(hba); +} + +static void hptiop_disable_intr_itl(struct hptiop_hba *hba) +{ + u32 int_mask; + + int_mask = readl(&hba->u.itl.iop->outbound_intmask); writel(int_mask | IOPMU_OUTBOUND_INT_MSG0 | IOPMU_OUTBOUND_INT_POSTQUEUE, - &iop->outbound_intmask); - hptiop_pci_posting_flush(iop); + &hba->u.itl.iop->outbound_intmask); + readl(&hba->u.itl.iop->outbound_intmask); +} + +static void hptiop_disable_intr_mv(struct hptiop_hba *hba) +{ + writel(0, &hba->u.mv.regs->outbound_intmask); + readl(&hba->u.mv.regs->outbound_intmask); } static void hptiop_remove(struct pci_dev *pcidev) @@ -901,7 +1194,10 @@ static void hptiop_remove(struct pci_dev *pcidev) hba->dma_coherent, hba->dma_coherent_handle); - iounmap(hba->iop); + if (hba->ops->internal_memfree) + hba->ops->internal_memfree(hba); + + hba->ops->unmap_pci_bar(hba); pci_release_regions(hba->pcidev); pci_set_drvdata(hba->pcidev, NULL); @@ -910,11 +1206,50 @@ static void hptiop_remove(struct pci_dev *pcidev) scsi_host_put(host); } +static struct hptiop_adapter_ops hptiop_itl_ops = { + .iop_wait_ready = iop_wait_ready_itl, + .internal_memalloc = 0, + .internal_memfree = 0, + .map_pci_bar = hptiop_map_pci_bar_itl, + .unmap_pci_bar = hptiop_unmap_pci_bar_itl, + .enable_intr = hptiop_enable_intr_itl, + .disable_intr = hptiop_disable_intr_itl, + .get_config = iop_get_config_itl, + .set_config = iop_set_config_itl, + .iop_intr = iop_intr_itl, + .post_msg = hptiop_post_msg_itl, + .post_req = hptiop_post_req_itl, +}; + +static struct hptiop_adapter_ops hptiop_mv_ops = { + .iop_wait_ready = iop_wait_ready_mv, + .internal_memalloc = hptiop_internal_memalloc_mv, + .internal_memfree = hptiop_internal_memfree_mv, + .map_pci_bar = hptiop_map_pci_bar_mv, + .unmap_pci_bar = hptiop_unmap_pci_bar_mv, + .enable_intr = hptiop_enable_intr_mv, + .disable_intr = hptiop_disable_intr_mv, + .get_config = iop_get_config_mv, + .set_config = iop_set_config_mv, + .iop_intr = iop_intr_mv, + .post_msg = hptiop_post_msg_mv, + .post_req = hptiop_post_req_mv, +}; + static struct pci_device_id hptiop_id_table[] = { - { PCI_VDEVICE(TTI, 0x3220) }, - { PCI_VDEVICE(TTI, 0x3320) }, - { PCI_VDEVICE(TTI, 0x3520) }, - { PCI_VDEVICE(TTI, 0x4320) }, + { PCI_VDEVICE(TTI, 0x3220), (kernel_ulong_t)&hptiop_itl_ops }, + { PCI_VDEVICE(TTI, 0x3320), (kernel_ulong_t)&hptiop_itl_ops }, + { PCI_VDEVICE(TTI, 0x3520), (kernel_ulong_t)&hptiop_itl_ops }, + { PCI_VDEVICE(TTI, 0x4320), (kernel_ulong_t)&hptiop_itl_ops }, + { PCI_VDEVICE(TTI, 0x3510), (kernel_ulong_t)&hptiop_itl_ops }, + { PCI_VDEVICE(TTI, 0x3511), (kernel_ulong_t)&hptiop_itl_ops }, + { PCI_VDEVICE(TTI, 0x3521), (kernel_ulong_t)&hptiop_itl_ops }, + { PCI_VDEVICE(TTI, 0x3522), (kernel_ulong_t)&hptiop_itl_ops }, + { PCI_VDEVICE(TTI, 0x3410), (kernel_ulong_t)&hptiop_itl_ops }, + { PCI_VDEVICE(TTI, 0x3540), (kernel_ulong_t)&hptiop_itl_ops }, + { PCI_VDEVICE(TTI, 0x3120), (kernel_ulong_t)&hptiop_mv_ops }, + { PCI_VDEVICE(TTI, 0x3122), (kernel_ulong_t)&hptiop_mv_ops }, + { PCI_VDEVICE(TTI, 0x3020), (kernel_ulong_t)&hptiop_mv_ops }, {}, }; diff --git a/drivers/scsi/hptiop.h b/drivers/scsi/hptiop.h index 2a5e46e001cb..a0289f219752 100644 --- a/drivers/scsi/hptiop.h +++ b/drivers/scsi/hptiop.h @@ -1,5 +1,5 @@ /* - * HighPoint RR3xxx controller driver for Linux + * HighPoint RR3xxx/4xxx controller driver for Linux * Copyright (C) 2006-2007 HighPoint Technologies, Inc. All Rights Reserved. * * This program is free software; you can redistribute it and/or modify @@ -18,8 +18,7 @@ #ifndef _HPTIOP_H_ #define _HPTIOP_H_ -struct hpt_iopmu -{ +struct hpt_iopmu_itl { __le32 resrved0[4]; __le32 inbound_msgaddr0; __le32 inbound_msgaddr1; @@ -54,6 +53,40 @@ struct hpt_iopmu #define IOPMU_INBOUND_INT_ERROR 8 #define IOPMU_INBOUND_INT_POSTQUEUE 0x10 +#define MVIOP_QUEUE_LEN 512 + +struct hpt_iopmu_mv { + __le32 inbound_head; + __le32 inbound_tail; + __le32 outbound_head; + __le32 outbound_tail; + __le32 inbound_msg; + __le32 outbound_msg; + __le32 reserve[10]; + __le64 inbound_q[MVIOP_QUEUE_LEN]; + __le64 outbound_q[MVIOP_QUEUE_LEN]; +}; + +struct hpt_iopmv_regs { + __le32 reserved[0x20400 / 4]; + __le32 inbound_doorbell; + __le32 inbound_intmask; + __le32 outbound_doorbell; + __le32 outbound_intmask; +}; + +#define MVIOP_MU_QUEUE_ADDR_HOST_MASK (~(0x1full)) +#define MVIOP_MU_QUEUE_ADDR_HOST_BIT 4 + +#define MVIOP_MU_QUEUE_ADDR_IOP_HIGH32 0xffffffff +#define MVIOP_MU_QUEUE_REQUEST_RESULT_BIT 1 +#define MVIOP_MU_QUEUE_REQUEST_RETURN_CONTEXT 2 + +#define MVIOP_MU_INBOUND_INT_MSG 1 +#define MVIOP_MU_INBOUND_INT_POSTQUEUE 2 +#define MVIOP_MU_OUTBOUND_INT_MSG 1 +#define MVIOP_MU_OUTBOUND_INT_POSTQUEUE 2 + enum hpt_iopmu_message { /* host-to-iop messages */ IOPMU_INBOUND_MSG0_NOP = 0, @@ -72,8 +105,7 @@ enum hpt_iopmu_message { IOPMU_OUTBOUND_MSG0_REVALIDATE_DEVICE_MAX = 0x3ff, }; -struct hpt_iop_request_header -{ +struct hpt_iop_request_header { __le32 size; __le32 type; __le32 flags; @@ -104,11 +136,10 @@ enum hpt_iop_result_type { IOP_RESULT_RESET, IOP_RESULT_INVALID_REQUEST, IOP_RESULT_BAD_TARGET, - IOP_RESULT_MODE_SENSE_CHECK_CONDITION, + IOP_RESULT_CHECK_CONDITION, }; -struct hpt_iop_request_get_config -{ +struct hpt_iop_request_get_config { struct hpt_iop_request_header header; __le32 interface_version; __le32 firmware_version; @@ -121,8 +152,7 @@ struct hpt_iop_request_get_config __le32 sdram_size; }; -struct hpt_iop_request_set_config -{ +struct hpt_iop_request_set_config { struct hpt_iop_request_header header; __le32 iop_id; __le16 vbus_id; @@ -130,15 +160,13 @@ struct hpt_iop_request_set_config __le32 reserve[6]; }; -struct hpt_iopsg -{ +struct hpt_iopsg { __le32 size; __le32 eot; /* non-zero: end of table */ __le64 pci_address; }; -struct hpt_iop_request_block_command -{ +struct hpt_iop_request_block_command { struct hpt_iop_request_header header; u8 channel; u8 target; @@ -156,8 +184,7 @@ struct hpt_iop_request_block_command #define IOP_BLOCK_COMMAND_FLUSH 4 #define IOP_BLOCK_COMMAND_SHUTDOWN 5 -struct hpt_iop_request_scsi_command -{ +struct hpt_iop_request_scsi_command { struct hpt_iop_request_header header; u8 channel; u8 target; @@ -168,8 +195,7 @@ struct hpt_iop_request_scsi_command struct hpt_iopsg sg_list[1]; }; -struct hpt_iop_request_ioctl_command -{ +struct hpt_iop_request_ioctl_command { struct hpt_iop_request_header header; __le32 ioctl_code; __le32 inbuf_size; @@ -182,11 +208,11 @@ struct hpt_iop_request_ioctl_command #define HPTIOP_MAX_REQUESTS 256u struct hptiop_request { - struct hptiop_request * next; - void * req_virt; - u32 req_shifted_phy; - struct scsi_cmnd * scp; - int index; + struct hptiop_request *next; + void *req_virt; + u32 req_shifted_phy; + struct scsi_cmnd *scp; + int index; }; struct hpt_scsi_pointer { @@ -198,9 +224,21 @@ struct hpt_scsi_pointer { #define HPT_SCP(scp) ((struct hpt_scsi_pointer *)&(scp)->SCp) struct hptiop_hba { - struct hpt_iopmu __iomem * iop; - struct Scsi_Host * host; - struct pci_dev * pcidev; + struct hptiop_adapter_ops *ops; + union { + struct { + struct hpt_iopmu_itl __iomem *iop; + } itl; + struct { + struct hpt_iopmv_regs *regs; + struct hpt_iopmu_mv __iomem *mu; + void *internal_req; + dma_addr_t internal_req_phy; + } mv; + } u; + + struct Scsi_Host *host; + struct pci_dev *pcidev; /* IOP config info */ u32 interface_version; @@ -213,15 +251,15 @@ struct hptiop_hba { u32 req_size; /* host-allocated request buffer size */ - int iopintf_v2: 1; - int initialized: 1; - int msg_done: 1; + u32 iopintf_v2: 1; + u32 initialized: 1; + u32 msg_done: 1; struct hptiop_request * req_list; struct hptiop_request reqs[HPTIOP_MAX_REQUESTS]; /* used to free allocated dma area */ - void * dma_coherent; + void *dma_coherent; dma_addr_t dma_coherent_handle; atomic_t reset_count; @@ -231,19 +269,35 @@ struct hptiop_hba { wait_queue_head_t ioctl_wq; }; -struct hpt_ioctl_k -{ +struct hpt_ioctl_k { struct hptiop_hba * hba; u32 ioctl_code; u32 inbuf_size; u32 outbuf_size; - void * inbuf; - void * outbuf; - u32 * bytes_returned; + void *inbuf; + void *outbuf; + u32 *bytes_returned; void (*done)(struct hpt_ioctl_k *); int result; /* HPT_IOCTL_RESULT_ */ }; +struct hptiop_adapter_ops { + int (*iop_wait_ready)(struct hptiop_hba *hba, u32 millisec); + int (*internal_memalloc)(struct hptiop_hba *hba); + int (*internal_memfree)(struct hptiop_hba *hba); + int (*map_pci_bar)(struct hptiop_hba *hba); + void (*unmap_pci_bar)(struct hptiop_hba *hba); + void (*enable_intr)(struct hptiop_hba *hba); + void (*disable_intr)(struct hptiop_hba *hba); + int (*get_config)(struct hptiop_hba *hba, + struct hpt_iop_request_get_config *config); + int (*set_config)(struct hptiop_hba *hba, + struct hpt_iop_request_set_config *config); + int (*iop_intr)(struct hptiop_hba *hba); + void (*post_msg)(struct hptiop_hba *hba, u32 msg); + void (*post_req)(struct hptiop_hba *hba, struct hptiop_request *_req); +}; + #define HPT_IOCTL_RESULT_OK 0 #define HPT_IOCTL_RESULT_FAILED (-1) -- cgit v1.2.3 From cb1042f285c2168bd8cf10aca0e24802e682252b Mon Sep 17 00:00:00 2001 From: "Salyzyn, Mark" Date: Thu, 17 Jan 2008 09:25:07 -0800 Subject: [SCSI] aacraid: add Voodoo Lite class of cards. The cards being added are supported in a limited sense already through family matching, but we needed to add some functionality to the driver to expose selectively the physical drives. These Physical drives are specifically marked to not be part of any array and thus are declared JBODs (Just a Bunch Of Drives) for generic SCSI access. We report that this is the second patch in a set of two, but merely depends on the stand-alone functionality of the first patch which adds in that case the ability to report a driver feature flag via sysfs. We leverage that functionality by reporting that this driver now supports this new JBOD feature for the controller so that the array management applications may react accordingly and guide the user as they manage the controller. Signed-off-by: Mark Salyzyn Signed-off-by: James Bottomley --- Documentation/scsi/aacraid.txt | 4 ++++ drivers/scsi/aacraid/aachba.c | 5 ++++- drivers/scsi/aacraid/aacraid.h | 6 +++++- drivers/scsi/aacraid/commsup.c | 24 ++++++++++++++++++++++++ drivers/scsi/aacraid/linit.c | 15 +++++++++++---- 5 files changed, 48 insertions(+), 6 deletions(-) (limited to 'Documentation') diff --git a/Documentation/scsi/aacraid.txt b/Documentation/scsi/aacraid.txt index a8257840695a..d16011a8618e 100644 --- a/Documentation/scsi/aacraid.txt +++ b/Documentation/scsi/aacraid.txt @@ -56,6 +56,10 @@ Supported Cards/Chipsets 9005:0285:9005:02d1 Adaptec 5405 (Voodoo40) 9005:0285:15d9:02d2 SMC AOC-USAS-S8i-LP 9005:0285:15d9:02d3 SMC AOC-USAS-S8iR-LP + 9005:0285:9005:02d4 Adaptec 2045 (Voodoo04 Lite) + 9005:0285:9005:02d5 Adaptec 2405 (Voodoo40 Lite) + 9005:0285:9005:02d6 Adaptec 2445 (Voodoo44 Lite) + 9005:0285:9005:02d7 Adaptec 2805 (Voodoo80 Lite) 1011:0046:9005:0364 Adaptec 5400S (Mustang) 9005:0287:9005:0800 Adaptec Themisto (Jupiter) 9005:0200:9005:0200 Adaptec Themisto (Jupiter) diff --git a/drivers/scsi/aacraid/aachba.c b/drivers/scsi/aacraid/aachba.c index 8d7b7703ee2a..d7235f42cf5f 100644 --- a/drivers/scsi/aacraid/aachba.c +++ b/drivers/scsi/aacraid/aachba.c @@ -1339,6 +1339,8 @@ int aac_get_adapter_info(struct aac_dev* dev) } dev->cache_protected = 0; + dev->jbod = ((dev->supplement_adapter_info.FeatureBits & + AAC_FEATURE_JBOD) != 0); dev->nondasd_support = 0; dev->raid_scsi_mode = 0; if(dev->adapter_info.options & AAC_OPT_NONDASD) @@ -1923,7 +1925,8 @@ int aac_scsi_cmd(struct scsi_cmnd * scsicmd) } } } else { /* check for physical non-dasd devices */ - if ((dev->nondasd_support == 1) || expose_physicals) { + if (dev->nondasd_support || expose_physicals || + dev->jbod) { if (dev->in_reset) return -1; return aac_send_srb_fib(scsicmd); diff --git a/drivers/scsi/aacraid/aacraid.h b/drivers/scsi/aacraid/aacraid.h index 8a5b9c898e5b..3195d29f2177 100644 --- a/drivers/scsi/aacraid/aacraid.h +++ b/drivers/scsi/aacraid/aacraid.h @@ -12,7 +12,7 @@ *----------------------------------------------------------------------------*/ #ifndef AAC_DRIVER_BUILD -# define AAC_DRIVER_BUILD 2454 +# define AAC_DRIVER_BUILD 2455 # define AAC_DRIVER_BRANCH "-ms" #endif #define MAXIMUM_NUM_CONTAINERS 32 @@ -866,6 +866,7 @@ struct aac_supplement_adapter_info __le32 ReservedGrowth[1]; }; #define AAC_FEATURE_FALCON cpu_to_le32(0x00000010) +#define AAC_FEATURE_JBOD cpu_to_le32(0x08000000) #define AAC_OPTION_MU_RESET cpu_to_le32(0x00000001) #define AAC_OPTION_IGNORE_RESET cpu_to_le32(0x00000002) #define AAC_SIS_VERSION_V3 3 @@ -1012,6 +1013,7 @@ struct aac_dev * lets break them out so we don't have to do an AND to check them */ u8 nondasd_support; + u8 jbod; u8 cache_protected; u8 dac_support; u8 raid_scsi_mode; @@ -1777,6 +1779,8 @@ extern struct aac_common aac_config; #define AifEnExpEvent 23 /* Firmware Event Log */ #define AifExeFirmwarePanic 3 /* Firmware Event Panic */ #define AifHighPriority 3 /* Highest Priority Event */ +#define AifEnAddJBOD 30 /* JBOD created */ +#define AifEnDeleteJBOD 31 /* JBOD deleted */ #define AifCmdJobProgress 2 /* Progress report */ #define AifJobCtrZero 101 /* Array Zero progress */ diff --git a/drivers/scsi/aacraid/commsup.c b/drivers/scsi/aacraid/commsup.c index 1dd2e57c3345..81b36923e0ef 100644 --- a/drivers/scsi/aacraid/commsup.c +++ b/drivers/scsi/aacraid/commsup.c @@ -901,7 +901,31 @@ static void aac_handle_aif(struct aac_dev * dev, struct fib * fibptr) case AifEnConfigChange: break; + case AifEnAddJBOD: + case AifEnDeleteJBOD: + container = le32_to_cpu(((__le32 *)aifcmd->data)[1]); + if ((container >> 28)) + break; + channel = (container >> 24) & 0xF; + if (channel >= dev->maximum_num_channels) + break; + id = container & 0xFFFF; + if (id >= dev->maximum_num_physicals) + break; + lun = (container >> 16) & 0xFF; + channel = aac_phys_to_logical(channel); + device_config_needed = + (((__le32 *)aifcmd->data)[0] == + cpu_to_le32(AifEnAddJBOD)) ? ADD : DELETE; + break; + case AifEnEnclosureManagement: + /* + * If in JBOD mode, automatic exposure of new + * physical target to be suppressed until configured. + */ + if (dev->jbod) + break; switch (le32_to_cpu(((__le32 *)aifcmd->data)[3])) { case EM_DRIVE_INSERTION: case EM_DRIVE_REMOVAL: diff --git a/drivers/scsi/aacraid/linit.c b/drivers/scsi/aacraid/linit.c index 5ab733d4faf4..61be22774e99 100644 --- a/drivers/scsi/aacraid/linit.c +++ b/drivers/scsi/aacraid/linit.c @@ -404,6 +404,7 @@ static int aac_slave_configure(struct scsi_device *sdev) struct aac_dev *aac = (struct aac_dev *)sdev->host->hostdata; if ((sdev->type == TYPE_DISK) && (sdev_channel(sdev) != CONTAINER_CHANNEL) && + (!aac->jbod || sdev->inq_periph_qual) && (!aac->raid_scsi_mode || (sdev_channel(sdev) != 2))) { if (expose_physicals == 0) return -ENXIO; @@ -411,7 +412,8 @@ static int aac_slave_configure(struct scsi_device *sdev) sdev->no_uld_attach = 1; } if (sdev->tagged_supported && (sdev->type == TYPE_DISK) && - (sdev_channel(sdev) == CONTAINER_CHANNEL)) { + (!aac->raid_scsi_mode || (sdev_channel(sdev) != 2)) && + !sdev->no_uld_attach) { struct scsi_device * dev; struct Scsi_Host *host = sdev->host; unsigned num_lsu = 0; @@ -430,8 +432,11 @@ static int aac_slave_configure(struct scsi_device *sdev) ++num_lsu; __shost_for_each_device(dev, host) { if (dev->tagged_supported && (dev->type == TYPE_DISK) && - (sdev_channel(dev) == CONTAINER_CHANNEL)) { - if (!aac->fsa_dev[sdev_id(dev)].valid) + (!aac->raid_scsi_mode || + (sdev_channel(sdev) != 2)) && + !dev->no_uld_attach) { + if ((sdev_channel(dev) != CONTAINER_CHANNEL) + || !aac->fsa_dev[sdev_id(dev)].valid) ++num_lsu; } else ++num_one; @@ -804,6 +809,8 @@ static ssize_t aac_show_flags(struct class_device *class_dev, char *buf) if (dev->raw_io_interface && dev->raw_io_64) len += snprintf(buf + len, PAGE_SIZE - len, "SAI_READ_CAPACITY_16\n"); + if (dev->jbod) + len += snprintf(buf + len, PAGE_SIZE - len, "SUPPORTED_JBOD\n"); return len; } @@ -1157,7 +1164,7 @@ static int __devinit aac_probe_one(struct pci_dev *pdev, * all containers are on the virtual channel 0 (CONTAINER_CHANNEL) * physical channels are address by their actual physical number+1 */ - if ((aac->nondasd_support == 1) || expose_physicals) + if (aac->nondasd_support || expose_physicals || aac->jbod) shost->max_channel = aac->maximum_num_channels; else shost->max_channel = 0; -- cgit v1.2.3 From 6b2cb91efce215c3c34b1b79b15f30e860761a3e Mon Sep 17 00:00:00 2001 From: Li Yang Date: Wed, 25 Jul 2007 02:03:06 +0800 Subject: Chinese: Add the known_regression URI to the HOWTO Update translation for commit be3884943674f8ee7656b1d8b71c087ec900c836. Signed-off-by: Li Yang Signed-off-by: Greg Kroah-Hartman --- Documentation/zh_CN/HOWTO | 2 ++ 1 file changed, 2 insertions(+) (limited to 'Documentation') diff --git a/Documentation/zh_CN/HOWTO b/Documentation/zh_CN/HOWTO index 48fc67bfbe3d..c8660da7344e 100644 --- a/Documentation/zh_CN/HOWTO +++ b/Documentation/zh_CN/HOWTO @@ -218,6 +218,8 @@ kernel.org网站的pub/linux/kernel/v2.6/目录下找到它。它的开发遵循 时，一个新的-rc版本就会被发布。计划是每周都发布新的-rc版本。 - 这个过程一直持续下去直到内核被认为达到足够稳定的状态，持续时间大概是 6个星期。 + - 以下地址跟踪了在每个-rc发布中发现的退步列表： + http://kernelnewbies.org/known_regressions 关于内核发布，值得一提的是Andrew Morton在linux-kernel邮件列表中如是说： “没有人知道新内核何时会被发布，因为发布是根据已知bug的情况来决定 -- cgit v1.2.3 From 24277db3894941026662743e400e77c68c4a9e92 Mon Sep 17 00:00:00 2001 From: Li Yang Date: Wed, 25 Jul 2007 02:43:32 +0800 Subject: Chinese: rephrase English introduction in HOWTO Rephrase the introduction as suggested by Jesper Juhl. Signed-off-by: Li Yang Signed-off-by: Greg Kroah-Hartman --- Documentation/zh_CN/HOWTO | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'Documentation') diff --git a/Documentation/zh_CN/HOWTO b/Documentation/zh_CN/HOWTO index c8660da7344e..79d1c3c7071d 100644 --- a/Documentation/zh_CN/HOWTO +++ b/Documentation/zh_CN/HOWTO @@ -1,10 +1,10 @@ Chinese translated version of Documentation/HOWTO If you have any comment or update to the content, please contact the -original document maintainer directly. However, if you have problem +original document maintainer directly. However, if you have a problem communicating in English you can also ask the Chinese maintainer for -help. Contact the Chinese maintainer, if this translation is outdated -or there is problem with translation. +help. Contact the Chinese maintainer if this translation is outdated +or if there is a problem with the translation. Maintainer: Greg Kroah-Hartman Chinese maintainer: Li Yang -- cgit v1.2.3 From 5dae82cc186da487cf33d7f5648a9ab3e02eaee4 Mon Sep 17 00:00:00 2001 From: Li Yang Date: Tue, 23 Oct 2007 21:58:51 +0800 Subject: Chinese: Change man-pages maintainer address in HOWOTO The email address of the man-pages maintainer has changed. Cc: Michael Kerrisk Signed-off-by: Li Yang Signed-off-by: Greg Kroah-Hartman --- Documentation/zh_CN/HOWTO | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/zh_CN/HOWTO b/Documentation/zh_CN/HOWTO index 79d1c3c7071d..3d80e8af36ec 100644 --- a/Documentation/zh_CN/HOWTO +++ b/Documentation/zh_CN/HOWTO @@ -85,7 +85,7 @@ Linux内核源代码都是在GPL（通用公共许可证）的保护下发布的 Linux内核代码中包含有大量的文档。这些文档对于学习如何与内核社区互动有着 不可估量的价值。当一个新的功能被加入内核，最好把解释如何使用这个功能的文 档也放进内核。当内核的改动导致面向用户空间的接口发生变化时，最好将相关信 -息或手册页(manpages)的补丁发到mtk-manpages@gmx.net，以向手册页(manpages) +息或手册页(manpages)的补丁发到mtk.manpages@gmail.com，以向手册页(manpages) 的维护者解释这些变化。 以下是内核代码中需要阅读的文档： -- cgit v1.2.3 From 12428e7626378dec7968cd4f5df9aab2ee58e735 Mon Sep 17 00:00:00 2001 From: TripleX Chung Date: Wed, 24 Oct 2007 00:46:43 +0800 Subject: Chinese: add translation of SubmittingPatches Signed-off-by: TripleX Chung Signed-off-by: Li Yang Signed-off-by: Greg Kroah-Hartman --- Documentation/zh_CN/SubmittingPatches | 416 ++++++++++++++++++++++++++++++++++ 1 file changed, 416 insertions(+) create mode 100644 Documentation/zh_CN/SubmittingPatches (limited to 'Documentation') diff --git a/Documentation/zh_CN/SubmittingPatches b/Documentation/zh_CN/SubmittingPatches new file mode 100644 index 000000000000..985c92e20b73 --- /dev/null +++ b/Documentation/zh_CN/SubmittingPatches @@ -0,0 +1,416 @@ +Chinese translated version of Documentation/SubmittingPatches + +If you have any comment or update to the content, please contact the +original document maintainer directly. However, if you have a problem +communicating in English you can also ask the Chinese maintainer for +help. Contact the Chinese maintainer if this translation is outdated +or if there is a problem with the translation. + +Chinese maintainer: TripleX Chung +--------------------------------------------------------------------- +Documentation/SubmittingPatches 的中文翻译 + +如果想评论或更新本文的内容，请直接联系原文档的维护者。如果你使用英文 +交流有困难的话，也可以向中文版维护者求助。如果本翻译更新不及时或者翻 +译存在问题，请联系中文版维护者。 + +中文版维护者： 钟宇 TripleX Chung +中文版翻译者： 钟宇 TripleX Chung +中文版校译者： 李阳 Li Yang + 王聪 Wang Cong + +以下为正文 +--------------------------------------------------------------------- + + 如何让你的改动进入内核 + 或者 + 获得亲爱的 Linus Torvalds 的关注和处理 +---------------------------------- + +对于想要将改动提交到 Linux 内核的个人或者公司来说，如果不熟悉“规矩”， +提交的流程会让人畏惧。本文档收集了一系列建议，这些建议可以大大的提高你 +的改动被接受的机会。 +阅读 Documentation/SubmitChecklist 来获得在提交代码前需要检查的项目的列 +表。如果你在提交一个驱动程序，那么同时阅读一下 +Documentation/SubmittingDrivers 。 + + +-------------------------- +第一节 - 创建并发送你的改动 +-------------------------- + +1) "diff -up" +----------- + +使用 "diff -up" 或者 "diff -uprN" 来创建补丁。 + +所有内核的改动，都是以补丁的形式呈现的，补丁由 diff(1) 生成。创建补丁的 +时候，要确认它是以 "unified diff" 格式创建的，这种格式由 diff(1) 的 '-u' +参数生成。而且，请使用 '-p' 参数，那样会显示每个改动所在的C函数，使得 +产生的补丁容易读得多。补丁应该基于内核源代码树的根目录，而不是里边的任 +何子目录。 +为一个单独的文件创建补丁，一般来说这样做就够了： + + SRCTREE= linux-2.6 + MYFILE= drivers/net/mydriver.c + + cd $SRCTREE + cp $MYFILE $MYFILE.orig + vi $MYFILE # make your change + cd .. + diff -up $SRCTREE/$MYFILE{.orig,} > /tmp/patch + +为多个文件创建补丁，你可以解开一个没有修改过的内核源代码树，然后和你自 +己的代码树之间做 diff 。例如： + + MYSRC= /devel/linux-2.6 + + tar xvfz linux-2.6.12.tar.gz + mv linux-2.6.12 linux-2.6.12-vanilla + diff -uprN -X linux-2.6.12-vanilla/Documentation/dontdiff \ + linux-2.6.12-vanilla $MYSRC > /tmp/patch + +"dontdiff" 是内核在编译的时候产生的文件的列表，列表中的文件在 diff(1) +产生的补丁里会被跳过。"dontdiff" 文件被包含在2.6.12和之后版本的内核源代 +码树中。对于更早的内核版本，你可以从 + 获取它。 +确定你的补丁里没有包含任何不属于这次补丁提交的额外文件。记得在用diff(1) +生成补丁之后，审阅一次补丁，以确保准确。 +如果你的改动很散乱，你应该研究一下如何将补丁分割成独立的部分，将改动分 +割成一系列合乎逻辑的步骤。这样更容易让其他内核开发者审核，如果你想你的 +补丁被接受，这是很重要的。下面这些脚本能够帮助你做这件事情： +Quilt: +http://savannah.nongnu.org/projects/quilt + +Andrew Morton 的补丁脚本: +http://www.zip.com.au/~akpm/linux/patches/ +作为这些脚本的替代，quilt 是值得推荐的补丁管理工具(看上面的链接)。 + +2)描述你的改动。 +描述你的改动包含的技术细节。 + +要多具体就写多具体。最糟糕的描述可能是像下面这些语句：“更新了某驱动程 +序”，“修正了某驱动程序的bug”，或者“这个补丁包含了某子系统的修改，请 +使用。” + +如果你的描述开始变长，这表示你也许需要拆分你的补丁了，请看第3小节， +继续。 + +3)拆分你的改动 + +将改动拆分，逻辑类似的放到同一个补丁文件里。 + +例如，如果你的改动里同时有bug修正和性能优化，那么把这些改动才分到两个或 +者更多的补丁文件中。如果你的改动包含对API的修改，并且修改了驱动程序来适 +应这些新的API，那么把这些修改分成两个补丁。 + +另一方面，如果你将一个单独的改动做成多个补丁文件，那么将它们合并成一个 +单独的补丁文件。这样一个逻辑上单独的改动只被包含在一个补丁文件里。 + +如果有一个补丁依赖另外一个补丁来完成它的改动，那没问题。简单的在你的补 +丁描述里指出“这个补丁依赖某补丁”就好了。 + +如果你不能将补丁浓缩成更少的文件，那么每次大约发送出15个，然后等待审查 +和整合。 + +4)选择 e-mail 的收件人 + +看一遍 MAINTAINERS 文件和源代码，看看你所的改动所在的内核子系统有没有指 +定的维护者。如果有，给他们发e-mail。 + +如果没有找到维护者，或者维护者没有反馈，将你的补丁发送到内核开发者主邮 +件列表 linux-kernel@vger.kernel.org。大部分的内核开发者都跟踪这个邮件列 +表，可以评价你的改动。 + +每次不要发送超过15个补丁到 vger 邮件列表！！！ + +Linus Torvalds 是决定改动能否进入 Linux 内核的最终裁决者。他的 e-mail +地址是 。他收到的 e-mail 很多，所以一般 +的说，最好别给他发 e-mail。 + +那些修正bug，“显而易见”的修改或者是类似的只需要很少讨论的补丁可以直接 +发送或者CC给Linus。那些需要讨论或者没有很清楚的好处的补丁，一般先发送到 +linux-kernel邮件列表。只有当补丁被讨论得差不多了，才提交给Linus。 + +5)选择CC( e-mail 抄送)列表 + +除非你有理由不这样做，否则CC linux-kernel@vger.kernel.org。 + +除了 Linus 之外，其他内核开发者也需要注意到你的改动，这样他们才能评论你 +的改动并提供代码审查和建议。linux-kernel 是 Linux 内核开发者主邮件列表 +。其它的邮件列表为特定的子系统提供服务，比如 USB，framebuffer 设备，虚 +拟文件系统，SCSI 子系统，等等。查看 MAINTAINERS 文件来获得和你的改动有 +关的邮件列表。 + +Majordomo lists of VGER.KERNEL.ORG at: + + +如果改动影响了用户空间和内核之间的接口，请给 MAN-PAGES 的维护者（列在 +MAITAINERS 文件里的）发送一个手册页（man-pages）补丁，或者至少通知一下改 +变，让一些信息有途径进入手册页。 + +即使在第四步的时候，维护者没有作出回应，也要确认在修改他们的代码的时候 +，一直将维护者拷贝到CC列表中。 + +对于小的补丁，你也许会CC到 Adrian Bunk 管理的搜集琐碎补丁的邮件列表 +(Trivial Patch Monkey)trivial@kernel.org，那里专门收集琐碎的补丁。下面这样 +的补丁会被看作“琐碎的”补丁： + 文档的拼写修正。 + 修正会影响到 grep(1) 的拼写。 + 警告信息修正(频繁的打印无用的警告是不好的。) + 编译错误修正（代码逻辑的确是对的，只是编译有问题。） + 运行时修正（只要真的修正了错误。） + 移除使用了被废弃的函数/宏的代码(例如 check_region。) + 联系方式和文档修正。 + 用可移植的代码替换不可移植的代码（即使在体系结构相关的代码中，既然有 + 人拷贝，只要它是琐碎的） + 任何文件的作者/维护者对该文件的改动（例如 patch monkey 在重传模式下） + +URL: + +(译注，关于“琐碎补丁”的一些说明：因为原文的这一部分写得比较简单，所以不得不 +违例写一下译注。"trivial"这个英文单词的本意是“琐碎的，不重要的。”但是在这里 +有稍微有一些变化，例如对一些明显的NULL指针的修正，属于运行时修正，会被归类 +到琐碎补丁里。虽然NULL指针的修正很重要，但是这样的修正往往很小而且很容易得到 +检验，所以也被归入琐碎补丁。琐碎补丁更精确的归类应该是 +“simple, localized & easy to verify”，也就是说简单的，局部的和易于检验的。 +trivial@kernel.org邮件列表的目的是针对这样的补丁，为提交者提供一个中心，来 +降低提交的门槛。) + +6)没有 MIME 编码，没有链接，没有压缩，没有附件，只有纯文本。 + +Linus 和其他的内核开发者需要阅读和评论你提交的改动。对于内核开发者来说 +，可以“引用”你的改动很重要，使用一般的 e-mail 工具，他们就可以在你的 +代码的任何位置添加评论。 + +因为这个原因，所有的提交的补丁都是 e-mail 中“内嵌”的。 +警告：如果你使用剪切-粘贴你的补丁，小心你的编辑器的自动换行功能破坏你的 +补丁。 + +不要将补丁作为 MIME 编码的附件，不管是否压缩。很多流行的 e-mail 软件不 +是任何时候都将 MIME 编码的附件当作纯文本发送的，这会使得别人无法在你的 +代码中加评论。另外，MIME 编码的附件会让 Linus 多花一点时间来处理，这就 +降低了你的改动被接受的可能性。 + +警告：一些邮件软件，比如 Mozilla 会将你的信息以如下格式发送： +---- 邮件头 ---- +Content-Type: text/plain; charset=us-ascii; format=flowed +---- 邮件头 ---- +问题在于 “format=flowed” 会让接收端的某些邮件软件将邮件中的制表符替换 +成空格以及做一些类似的替换。这样，你发送的时候看起来没问题的补丁就被破 +坏了。 + +要修正这个问题，只需要将你的 mozilla 的 defaults/pref/mailnews.js 文件 +里的 +pref("mailnews.send_plaintext_flowed", false); // RFC 2646======= +修改成 +pref("mailnews.display.disable_format_flowed_support", true); +就可以了。 + +7) e-mail 的大小 + +给 Linus 发送补丁的时候，永远按照第6小节说的做。 + +大的改动对邮件列表不合适，对某些维护者也不合适。如果你的补丁，在不压缩 +的情况下，超过了40kB，那么你最好将补丁放在一个能通过 internet 访问的服 +务器上，然后用指向你的补丁的 URL 替代。 + +8) 指出你的内核版本 + +在标题和在补丁的描述中，指出补丁对应的内核的版本，是很重要的。 + +如果补丁不能干净的在最新版本的内核上打上，Linus 是不会接受它的。 + +9) 不要气馁，继续提交。 + +当你提交了改动以后，耐心地等待。如果 Linus 喜欢你的改动并且同意它，那么 +它将在下一个内核发布版本中出现。 + +然而，如果你的改动没有出现在下一个版本的内核中，可能有若干原因。减少那 +些原因，修正错误，重新提交更新后的改动，是你自己的工作。 + +Linus不给出任何评论就“丢弃”你的补丁是常见的事情。在系统中这样的事情很 +平常。如果他没有接受你的补丁，也许是由于以下原本： +* 你的补丁不能在最新版本的内核上干净的打上。 +* 你的补丁在 linux-kernel 邮件列表中没有得到充分的讨论。 +* 风格问题（参照第2小节） +* 邮件格式问题（重读本节） +* 你的改动有技术问题。 +* 他收到了成吨的 e-mail，而你的在混乱中丢失了。 +* 你让人为难。 + +有疑问的时候，在 linux-kernel 邮件列表上请求评论。 + +10) 在标题上加上 PATCH 的字样 + +Linus 和 linux-kernel 邮件列表的 e-mail 流量都很高，一个通常的约定是标 +题行以 [PATCH] 开头。这样可以让 Linus 和其他内核开发人员可以从 e-mail +的讨论中很轻易的将补丁分辨出来。 + +11）为你的工作签名 + +为了加强对谁做了何事的追踪，尤其是对那些透过好几层的维护者的补丁，我们 +建议在发送出去的补丁上加一个 “sign-off” 的过程。 + +"sign-off" 是在补丁的注释的最后的简单的一行文字，认证你编写了它或者其他 +人有权力将它作为开放源代码的补丁传递。规则很简单：如果你能认证如下信息 +： + 开发者来源证书 1.1 + 对于本项目的贡献，我认证如下信息： + （a）这些贡献是完全或者部分的由我创建，我有权利以文件中指出 + 的开放源代码许可证提交它；或者 + （b）这些贡献基于以前的工作，据我所知，这些以前的工作受恰当的开放 + 源代码许可证保护，而且，根据许可证，我有权提交修改后的贡献， + 无论是完全还是部分由我创造，这些贡献都使用同一个开放源代码许可证 + （除非我被允许用其它的许可证），正如文件中指出的；或者 + （c）这些贡献由认证（a），（b）或者（c）的人直接提供给我，而 + 且我没有修改它。 + （d）我理解并同意这个项目和贡献是公开的，贡献的记录（包括我 + 一起提交的个人记录，包括 sign-off ）被永久维护并且可以和这个项目 + 或者开放源代码的许可证同步地再发行。 + 那么加入这样一行： + Signed-off-by: Random J Developer + +使用你的真名（抱歉，不能使用假名或者匿名。） + +有人在最后加上标签。现在这些东西会被忽略，但是你可以这样做，来标记公司 +内部的过程，或者只是指出关于 sign-off 的一些特殊细节。 + +12）标准补丁格式 + +标准的补丁，标题行是： + Subject: [PATCH 001/123] 子系统:一句话概述 + +标准补丁的信体存在如下部分： + + - 一个 "from" 行指出补丁作者。 + + - 一个空行 + + - 说明的主体，这些说明文字会被拷贝到描述该补丁的永久改动记录里。 + + - 一个由"---"构成的标记行 + + - 不合适放到改动记录里的额外的注解。 + + - 补丁本身（diff 输出） + +标题行的格式，使得对标题行按字母序排序非常的容易 - 很多 e-mail 客户端都 +可以支持 - 因为序列号是用零填充的，所以按数字排序和按字母排序是一样的。 + +e-mail 标题中的“子系统”标识哪个内核子系统将被打补丁。 + +e-mail 标题中的“一句话概述”扼要的描述 e-mail 中的补丁。“一句话概述” +不应该是一个文件名。对于一个补丁系列（“补丁系列”指一系列的多个相关补 +丁），不要对每个补丁都使用同样的“一句话概述”。 + +记住 e-mail 的“一句话概述”会成为该补丁的全局唯一标识。它会蔓延到 git +的改动记录里。然后“一句话概述”会被用在开发者的讨论里，用来指代这个补 +丁。用户将希望通过 google 来搜索"一句话概述"来找到那些讨论这个补丁的文 +章。 + +一些标题的例子： + + Subject: [patch 2/5] ext2: improve scalability of bitmap searching + Subject: [PATCHv2 001/207] x86: fix eflags tracking + +"from" 行是信体里的最上面一行，具有如下格式： + From: Original Author + +"from" 行指明在永久改动日志里，谁会被确认为作者。如果没有 "from" 行，那 +么邮件头里的 "From: " 行会被用来决定改动日志中的作者。 + +说明的主题将会被提交到永久的源代码改动日志里，因此对那些早已经不记得和 +这个补丁相关的讨论细节的有能力的读者来说，是有意义的。 + +"---" 标记行对于补丁处理工具要找到哪里是改动日志信息的结束，是不可缺少 +的。 + +对于 "---" 标记之后的额外注解，一个好的用途就是用来写 diffstat，用来显 +示修改了什么文件和每个文件都增加和删除了多少行。diffstat 对于比较大的补 +丁特别有用。其余那些只是和时刻或者开发者相关的注解，不合适放到永久的改 +动日志里的，也应该放这里。 +使用 diffstat的选项 "-p 1 -w 70" 这样文件名就会从内核源代码树的目录开始 +，不会占用太宽的空间（很容易适合80列的宽度，也许会有一些缩进。） + +在后面的参考资料中能看到适当的补丁格式的更多细节。 + +------------------------------- +第二节 提示，建议和诀窍 +------------------------------- + +本节包含很多和提交到内核的代码有关的通常的"规则"。事情永远有例外...但是 +你必须真的有好的理由这样做。你可以把本节叫做Linus的计算机科学入门课。 + +1) 读 Document/CodingStyle + +Nuff 说过，如果你的代码和这个偏离太多，那么它有可能会被拒绝，没有更多的 +审查，没有更多的评价。 + +2) #ifdef 是丑陋的 +混杂了 ifdef 的代码难以阅读和维护。别这样做。作为替代，将你的 ifdef 放 +在头文件里，有条件地定义 "static inline" 函数，或者宏，在代码里用这些东 +西。让编译器把那些"空操作"优化掉。 + +一个简单的例子，不好的代码： + + dev = alloc_etherdev (sizeof(struct funky_private)); + if (!dev) + return -ENODEV; + #ifdef CONFIG_NET_FUNKINESS + init_funky_net(dev); + #endif + +清理后的例子: + +(头文件里) + #ifndef CONFIG_NET_FUNKINESS + static inline void init_funky_net (struct net_device *d) {} + #endif + +(代码文件里) + dev = alloc_etherdev (sizeof(struct funky_private)); + if (!dev) + return -ENODEV; + init_funky_net(dev); + +3) 'static inline' 比宏好 + +Static inline 函数相比宏来说，是好得多的选择。Static inline 函数提供了 +类型安全，没有长度限制，没有格式限制，在 gcc 下开销和宏一样小。 + +宏只在 static inline 函数不是最优的时候[在 fast paths 里有很少的独立的 +案例]，或者不可能用 static inline 函数的时候[例如字符串分配]。 +应该用 'static inline' 而不是 'static __inline__', 'extern inline' 和 +'extern __inline__' 。 + +4) 不要过度设计 + +不要试图预计模糊的未来事情，这些事情也许有用也许没有用："让事情尽可能的 +简单，而不是更简单"。 + +---------------- +第三节 参考文献 +---------------- + +Andrew Morton, "The perfect patch" (tpp). + + +Jeff Garzik, "Linux kernel patch submission format". + + +Greg Kroah-Hartman, "How to piss off a kernel subsystem maintainer". + + + + + +NO!!!! No more huge patch bombs to linux-kernel@vger.kernel.org people! + + +Kernel Documentation/CodingStyle: + + +Linus Torvalds's mail on the canonical patch format: + +-- -- cgit v1.2.3 From 09792200e465db0861dee25bdecfc55278907ed3 Mon Sep 17 00:00:00 2001 From: Li Yang Date: Wed, 24 Oct 2007 00:51:11 +0800 Subject: Chinese: add translation of SubmittingDrivers Signed-off-by: Li Yang Signed-off-by: Greg Kroah-Hartman --- Documentation/zh_CN/SubmittingDrivers | 168 ++++++++++++++++++++++++++++++++++ 1 file changed, 168 insertions(+) create mode 100644 Documentation/zh_CN/SubmittingDrivers (limited to 'Documentation') diff --git a/Documentation/zh_CN/SubmittingDrivers b/Documentation/zh_CN/SubmittingDrivers new file mode 100644 index 000000000000..5f4815c63ec7 --- /dev/null +++ b/Documentation/zh_CN/SubmittingDrivers @@ -0,0 +1,168 @@ +Chinese translated version of Documentation/SubmittingDrivers + +If you have any comment or update to the content, please contact the +original document maintainer directly. However, if you have a problem +communicating in English you can also ask the Chinese maintainer for +help. Contact the Chinese maintainer if this translation is outdated +or if there is a problem with the translation. + +Chinese maintainer: Li Yang +--------------------------------------------------------------------- +Documentation/SubmittingDrivers 的中文翻译 + +如果想评论或更新本文的内容，请直接联系原文档的维护者。如果你使用英文 +交流有困难的话，也可以向中文版维护者求助。如果本翻译更新不及时或者翻 +译存在问题，请联系中文版维护者。 + +中文版维护者： 李阳 Li Yang +中文版翻译者： 李阳 Li Yang +中文版校译者： 陈琦 Maggie Chen + 王聪 Wang Cong + 张巍 Zhang Wei + +以下为正文 +--------------------------------------------------------------------- + +如何向 Linux 内核提交驱动程序 +----------------------------- + +这篇文档将会解释如何向不同的内核源码树提交设备驱动程序。请注意，如果你感 +兴趣的是显卡驱动程序，你也许应该访问 XFree86 项目(http://www.xfree86.org/) +和／或 X.org 项目 (http://x.org)。 + +另请参阅 Documentation/SubmittingPatches 文档。 + + +分配设备号 +---------- + +块设备和字符设备的主设备号与从设备号是由 Linux 命名编号分配权威 LANANA（ +现在是 Torben Mathiasen）负责分配。申请的网址是 http://www.lanana.org/。 +即使不准备提交到主流内核的设备驱动也需要在这里分配设备号。有关详细信息， +请参阅 Documentation/devices.txt。 + +如果你使用的不是已经分配的设备号，那么当你提交设备驱动的时候，它将会被强 +制分配一个新的设备号，即便这个设备号和你之前发给客户的截然不同。 + +设备驱动的提交对象 +------------------ + +Linux 2.0: + 此内核源码树不接受新的驱动程序。 + +Linux 2.2: + 此内核源码树不接受新的驱动程序。 + +Linux 2.4: + 如果所属的代码领域在内核的 MAINTAINERS 文件中列有一个总维护者， + 那么请将驱动程序提交给他。如果此维护者没有回应或者你找不到恰当的 + 维护者，那么请联系 Willy Tarreau 。 + +Linux 2.6: + 除了遵循和 2.4 版内核同样的规则外，你还需要在 linux-kernel 邮件 + 列表上跟踪最新的 API 变化。向 Linux 2.6 内核提交驱动的顶级联系人 + 是 Andrew Morton 。 + +决定设备驱动能否被接受的条件 +---------------------------- + +许可： 代码必须使用 GNU 通用公开许可证 (GPL) 提交给 Linux，但是 + 我们并不要求 GPL 是唯一的许可。你或许会希望同时使用多种 + 许可证发布，如果希望驱动程序可以被其他开源社区（比如BSD） + 使用。请参考 include/linux/module.h 文件中所列出的可被 + 接受共存的许可。 + +版权： 版权所有者必须同意使用 GPL 许可。最好提交者和版权所有者 + 是相同个人或实体。否则，必需列出授权使用 GPL 的版权所有 + 人或实体，以备验证之需。 + +接口： 如果你的驱动程序使用现成的接口并且和其他同类的驱动程序行 + 为相似，而不是去发明无谓的新接口，那么它将会更容易被接受。 + 如果你需要一个 Linux 和 NT 的通用驱动接口，那么请在用 + 户空间实现它。 + +代码： 请使用 Documentation/CodingStyle 中所描述的 Linux 代码风 + 格。如果你的某些代码段（例如那些与 Windows 驱动程序包共 + 享的代码段）需要使用其他格式，而你却只希望维护一份代码， + 那么请将它们很好地区分出来，并且注明原因。 + +可移植性： 请注意，指针并不永远是 32 位的，不是所有的计算机都使用小 + 尾模式 (little endian) 存储数据，不是所有的人都拥有浮点 + 单元，不要随便在你的驱动程序里嵌入 x86 汇编指令。只能在 + x86 上运行的驱动程序一般是不受欢迎的。虽然你可能只有 x86 + 硬件，很难测试驱动程序在其他平台上是否可用，但是确保代码 + 可以被轻松地移植却是很简单的。 + +清晰度： 做到所有人都能修补这个驱动程序将会很有好处，因为这样你将 + 会直接收到修复的补丁而不是 bug 报告。如果你提交一个试图 + 隐藏硬件工作机理的驱动程序，那么它将会被扔进废纸篓。 + +电源管理： 因为 Linux 正在被很多移动设备和桌面系统使用，所以你的驱 + 动程序也很有可能被使用在这些设备上。它应该支持最基本的电 + 源管理，即在需要的情况下实现系统级休眠和唤醒要用到的 + .suspend 和 .resume 函数。你应该检查你的驱动程序是否能正 + 确地处理休眠与唤醒，如果实在无法确认，请至少把 .suspend + 函数定义成返回 -ENOSYS（功能未实现）错误。你还应该尝试确 + 保你的驱动在什么都不干的情况下将耗电降到最低。要获得驱动 + 程序测试的指导，请参阅 + Documentation/power/drivers-testing.txt。有关驱动程序电 + 源管理问题相对全面的概述，请参阅 + Documentation/power/devices.txt。 + +管理： 如果一个驱动程序的作者还在进行有效的维护，那么通常除了那 + 些明显正确且不需要任何检查的补丁以外，其他所有的补丁都会 + 被转发给作者。如果你希望成为驱动程序的联系人和更新者，最 + 好在代码注释中写明并且在 MAINTAINERS 文件中加入这个驱动 + 程序的条目。 + +不影响设备驱动能否被接受的条件 +------------------------------ + +供应商： 由硬件供应商来维护驱动程序通常是一件好事。不过，如果源码 + 树里已经有其他人提供了可稳定工作的驱动程序，那么请不要期 + 望“我是供应商”会成为内核改用你的驱动程序的理由。理想的情 + 况是：供应商与现有驱动程序的作者合作，构建一个统一完美的 + 驱动程序。 + +作者： 驱动程序是由大的 Linux 公司研发还是由你个人编写，并不影 + 响其是否能被内核接受。没有人对内核源码树享有特权。只要你 + 充分了解内核社区，你就会发现这一点。 + + +资源列表 +-------- + +Linux 内核主源码树： + ftp.??.kernel.org:/pub/linux/kernel/... + ?? == 你的国家代码，例如 "cn"、"us"、"uk"、"fr" 等等 + +Linux 内核邮件列表： + linux-kernel@vger.kernel.org + [可通过向majordomo@vger.kernel.org发邮件来订阅] + +Linux 设备驱动程序，第三版（探讨 2.6.10 版内核）： + http://lwn.net/Kernel/LDD3/ （免费版） + +LWN.net: + 每周内核开发活动摘要 - http://lwn.net/ + 2.6 版中 API 的变更： + http://lwn.net/Articles/2.6-kernel-api/ + 将旧版内核的驱动程序移植到 2.6 版： + http://lwn.net/Articles/driver-porting/ + +KernelTrap： + Linux 内核的最新动态以及开发者访谈 + http://kerneltrap.org/ + +内核新手(KernelNewbies): + 为新的内核开发者提供文档和帮助 + http://kernelnewbies.org/ + +Linux USB项目： + http://www.linux-usb.org/ + +写内核驱动的“不要”（Arjan van de Ven著）: + http://www.fenrus.org/how-to-not-write-a-device-driver-paper.pdf + +内核清洁工 (Kernel Janitor): + http://janitor.kernelnewbies.org/ -- cgit v1.2.3 From d1a1d45142ed3969b3cc3964f81e4249f9e49fbf Mon Sep 17 00:00:00 2001 From: Dave Young Date: Wed, 24 Oct 2007 01:14:29 +0800 Subject: Chinese: add translation of oops-tracing.txt Signed-off-by: Dave Young Signed-off-by: Li Yang Signed-off-by: Greg Kroah-Hartman --- Documentation/zh_CN/oops-tracing.txt | 212 +++++++++++++++++++++++++++++++++++ 1 file changed, 212 insertions(+) create mode 100644 Documentation/zh_CN/oops-tracing.txt (limited to 'Documentation') diff --git a/Documentation/zh_CN/oops-tracing.txt b/Documentation/zh_CN/oops-tracing.txt new file mode 100644 index 000000000000..9312608ffb8d --- /dev/null +++ b/Documentation/zh_CN/oops-tracing.txt @@ -0,0 +1,212 @@ +Chinese translated version of Documentation/oops-tracing.txt + +If you have any comment or update to the content, please contact the +original document maintainer directly. However, if you have a problem +communicating in English you can also ask the Chinese maintainer for +help. Contact the Chinese maintainer if this translation is outdated +or if there is a problem with the translation. + +Chinese maintainer: Dave Young +--------------------------------------------------------------------- +Documentation/oops-tracing.txt 的中文翻译 + +如果想评论或更新本文的内容，请直接联系原文档的维护者。如果你使用英文 +交流有困难的话，也可以向中文版维护者求助。如果本翻译更新不及时或者翻 +译存在问题，请联系中文版维护者。 + +中文版维护者： 杨瑞 Dave Young +中文版翻译者： 杨瑞 Dave Young +中文版校译者： 李阳 Li Yang + 王聪 Wang Cong + +以下为正文 +--------------------------------------------------------------------- + +注意： ksymoops 在2.6中是没有用的。 请以原有格式使用Oops(来自dmesg，等等)。 +忽略任何这样那样关于“解码Oops”或者“通过ksymoops运行”的文档。 如果你贴出运行过 +ksymoops的来自2.6的Oops，人们只会让你重贴一次。 + +快速总结 +------------- + +发现Oops并发送给看似相关的内核领域的维护者。别太担心对不上号。如果你不确定就发给 +和你所做的事情相关的代码的负责人。 如果可重现试着描述怎样重构。 那甚至比oops更有 +价值。 + +如果你对于发送给谁一无所知， 发给linux-kernel@vger.kernel.org。感谢你帮助Linux +尽可能地稳定。 + +Oops在哪里? +---------------------- + +通常Oops文本由klogd从内核缓冲区里读取并传给syslogd，由syslogd写到syslog文件中， +典型地是/var/log/messages(依赖于/etc/syslog.conf)。有时klogd崩溃了,这种情况下你 +能够运行dmesg > file来从内核缓冲区中读取数据并保存下来。 否则你可以 +cat /proc/kmsg > file， 然而你必须介入中止传输， kmsg是一个“永不结束的文件”。如 +果机器崩溃坏到你不能输入命令或者磁盘不可用那么你有三种选择:- + +（1） 手抄屏幕上的文本待机器重启后再输入计算机。 麻烦但如果没有针对崩溃的准备， +这是仅有的选择。 另外，你可以用数码相机把屏幕拍下来-不太好，但比没有强。 如果信 +息滚动到了终端的上面，你会发现以高分辩率启动（比如，vga=791）会让你读到更多的文 +本。（注意：这需要vesafb，所以对‘早期’的oops没有帮助） + +（2）用串口终端启动（请参看Documentation/serial-console.txt），运行一个null +modem到另一台机器并用你喜欢的通讯工具获取输出。Minicom工作地很好。 + +（3）使用Kdump（请参看Documentation/kdump/kdump.txt）， +使用在Documentation/kdump/gdbmacros.txt中定义的dmesg gdb宏，从旧的内存中提取内核 +环形缓冲区。 + +完整信息 +---------------- + +注意：以下来自于Linus的邮件适用于2.4内核。 我因为历史原因保留了它，并且因为其中 +一些信息仍然适用。 特别注意的是，请忽略任何ksymoops的引用。 + +From: Linus Torvalds + +怎样跟踪Oops.. [原发到linux-kernel的一封邮件] + +主要的窍门是有五年和这些烦人的oops消息打交道的经验;-) + +实际上，你有办法使它更简单。我有两个不同的方法： + + gdb /usr/src/linux/vmlinux + gdb> disassemble + +那是发现问题的简单办法，至少如果bug报告做的好的情况下（象这个一样-运行ksymoops +得到oops发生的函数及函数内的偏移）。 + +哦，如果报告发生的内核以相同的编译器和相似的配置编译它会有帮助的。 + +另一件要做的事是反汇编bug报告的“Code”部分：ksymoops也会用正确的工具来做这件事， +但如果没有那些工具你可以写一个傻程序： + + char str[] = "\xXX\xXX\xXX..."; + main(){} + +并用gcc -g编译它然后执行“disassemble str”（XX部分是由Oops报告的值-你可以仅剪切 +粘贴并用“\x”替换空格-我就是这么做的，因为我懒得写程序自动做这一切）。 + +另外，你可以用scripts/decodecode这个shell脚本。它的使用方法是： +decodecode < oops.txt + +“Code”之后的十六进制字节可能（在某些架构上）有一些当前指令之前的指令字节以及 +当前和之后的指令字节 + +Code: f9 0f 8d f9 00 00 00 8d 42 0c e8 dd 26 11 c7 a1 60 ea 2b f9 8b 50 08 a1 +64 ea 2b f9 8d 34 82 8b 1e 85 db 74 6d 8b 15 60 ea 2b f9 <8b> 43 04 39 42 54 +7e 04 40 89 42 54 8b 43 04 3b 05 00 f6 52 c0 + +最后，如果你想知道代码来自哪里，你可以： + + cd /usr/src/linux + make fs/buffer.s # 或任何产生BUG的文件 + +然后你会比gdb反汇编更清楚的知道发生了什么。 + +现在，问题是把你所拥有的所有数据结合起来：C源码（关于它应该怎样的一般知识）， +汇编代码及其反汇编得到的代码（另外还有从“oops”消息得到的寄存器状态-对了解毁坏的 +指针有用，而且当你有了汇编代码你也能拿其它的寄存器和任何它们对应的C表达式做匹配 +）。 + +实际上，你仅需看看哪里不匹配（这个例子是“Code”反汇编和编译器生成的代码不匹配）。 +然后你须要找出为什么不匹配。通常很简单-你看到代码使用了空指针然后你看代码想知道 +空指针是怎么出现的，还有检查它是否合法.. + +现在，如果明白这是一项耗时的工作而且需要一丁点儿的专心，没错。这就是我为什么大多 +只是忽略那些没有符号表信息的崩溃报告的原因：简单的说太难查找了（我有一些 +程序用于在内核代码段中搜索特定的模式，而且有时我也已经能找出那些崩溃的地方，但是 +仅仅是找出正确的序列也确实需要相当扎实的内核知识） + +_有时_会发生这种情况，我仅看到崩溃中的反汇编代码序列， 然后我马上就明白问题出在 +哪里。这时我才意识到自己干这个工作已经太长时间了;-) + + Linus + + +--------------------------------------------------------------------------- +关于Oops跟踪的注解： + +为了帮助Linus和其它内核开发者，klogd纳入了大量的支持来处理保护错误。为了拥有对 +地址解析的完整支持至少应该使用1.3-pl3的sysklogd包。 + +当保护错误发生时，klogd守护进程自动把内核日志信息中的重要地址翻译成它们相应的符 +号。 + +klogd执行两种类型的地址解析。首先是静态翻译其次是动态翻译。静态翻译和ksymoops +一样使用System.map文件。为了做静态翻译klogd守护进程必须在初始化时能找到system +map文件。关于klogd怎样搜索map文件请参看klogd手册页。 + +动态地址翻译在使用内核可装载模块时很重要。 因为内核模块的内存是从内核动态内存池 +里分配的，所以不管是模块开始位置还是模块中函数和符号的位置都不是固定的。 + +内核支持允许程序决定装载哪些模块和它们在内存中位置的系统调用。使用这些系统调用 +klogd守护进程生成一张符号表用于调试发生在可装载模块中的保护错误。 + +至少klogd会提供产生保护错误的模块名。还可有额外的符号信息供可装载模块开发者选择 +以从模块中输出符号信息。 + +因为内核模块环境可能是动态的，所以必须有一种机制当模块环境发生改变时来通知klogd +守护进程。 有一些可用的命令行选项允许klogd向当前执行中的守护进程发送信号，告知符 +号信息应该被刷新了。 更多信息请参看klogd手册页。 + +sysklogd发布时包含一个补丁修改了modules-2.0.0包，无论何时一个模块装载或者卸载都 +会自动向klogd发送信号。打上这个补丁提供了必要的对调试发生于内核可装载模块的保护 +错误的无缝支持。 + +以下是被klogd处理过的发生在可装载模块中的一个保护错误例子： +--------------------------------------------------------------------------- +Aug 29 09:51:01 blizard kernel: Unable to handle kernel paging request at virtual address f15e97cc +Aug 29 09:51:01 blizard kernel: current->tss.cr3 = 0062d000, %cr3 = 0062d000 +Aug 29 09:51:01 blizard kernel: *pde = 00000000 +Aug 29 09:51:01 blizard kernel: Oops: 0002 +Aug 29 09:51:01 blizard kernel: CPU: 0 +Aug 29 09:51:01 blizard kernel: EIP: 0010:[oops:_oops+16/3868] +Aug 29 09:51:01 blizard kernel: EFLAGS: 00010212 +Aug 29 09:51:01 blizard kernel: eax: 315e97cc ebx: 003a6f80 ecx: 001be77b edx: 00237c0c +Aug 29 09:51:01 blizard kernel: esi: 00000000 edi: bffffdb3 ebp: 00589f90 esp: 00589f8c +Aug 29 09:51:01 blizard kernel: ds: 0018 es: 0018 fs: 002b gs: 002b ss: 0018 +Aug 29 09:51:01 blizard kernel: Process oops_test (pid: 3374, process nr: 21, stackpage=00589000) +Aug 29 09:51:01 blizard kernel: Stack: 315e97cc 00589f98 0100b0b4 bffffed4 0012e38e 00240c64 003a6f80 00000001 +Aug 29 09:51:01 blizard kernel: 00000000 00237810 bfffff00 0010a7fa 00000003 00000001 00000000 bfffff00 +Aug 29 09:51:01 blizard kernel: bffffdb3 bffffed4 ffffffda 0000002b 0007002b 0000002b 0000002b 00000036 +Aug 29 09:51:01 blizard kernel: Call Trace: [oops:_oops_ioctl+48/80] [_sys_ioctl+254/272] [_system_call+82/128] +Aug 29 09:51:01 blizard kernel: Code: c7 00 05 00 00 00 eb 08 90 90 90 90 90 90 90 90 89 ec 5d c3 +--------------------------------------------------------------------------- + +Dr. G.W. Wettstein Oncology Research Div. Computing Facility +Roger Maris Cancer Center INTERNET: greg@wind.rmcc.com +820 4th St. N. +Fargo, ND 58122 +Phone: 701-234-7556 + + +--------------------------------------------------------------------------- +受污染的内核 + +一些oops报告在程序记数器之后包含字符串'Tainted: '。这表明内核已经被一些东西给污 +染了。 该字符串之后紧跟着一系列的位置敏感的字符，每个代表一个特定的污染值。 + + 1：'G'如果所有装载的模块都有GPL或相容的许可证，'P'如果装载了任何的专有模块。 +没有模块MODULE_LICENSE或者带有insmod认为是与GPL不相容的的MODULE_LICENSE的模块被 +认定是专有的。 + + 2：'F'如果有任何通过“insmod -f”被强制装载的模块，' '如果所有模块都被正常装载。 + + 3：'S'如果oops发生在SMP内核中，运行于没有证明安全运行多处理器的硬件。 当前这种 +情况仅限于几种不支持SMP的速龙处理器。 + + 4：'R'如果模块通过“insmod -f”被强制装载，' '如果所有模块都被正常装载。 + + 5：'M'如果任何处理器报告了机器检查异常，' '如果没有发生机器检查异常。 + + 6：'B'如果页释放函数发现了一个错误的页引用或者一些非预期的页标志。 + + 7：'U'如果用户或者用户应用程序特别请求设置污染标志，否则' '。 + + 8：'D'如果内核刚刚死掉，比如有OOPS或者BUG。 + +使用'Tainted: '字符串的主要原因是要告诉内核调试者，这是否是一个干净的内核亦或发 +生了任何的不正常的事。污染是永久的：即使出错的模块已经被卸载了，污染值仍然存在， +以表明内核不再值得信任。 -- cgit v1.2.3 From 6941ee8896bfc462c3e3ad113c769a57ecbf3b2a Mon Sep 17 00:00:00 2001 From: TripleX Chung Date: Wed, 24 Oct 2007 00:57:24 +0800 Subject: Chinese: add translation of stable_kernel_rules.txt Signed-off-by: TripleX Chung Signed-off-by: Li Yang Signed-off-by: Greg Kroah-Hartman --- Documentation/zh_CN/stable_kernel_rules.txt | 66 +++++++++++++++++++++++++++++ 1 file changed, 66 insertions(+) create mode 100644 Documentation/zh_CN/stable_kernel_rules.txt (limited to 'Documentation') diff --git a/Documentation/zh_CN/stable_kernel_rules.txt b/Documentation/zh_CN/stable_kernel_rules.txt new file mode 100644 index 000000000000..b5b9b0ab02fd --- /dev/null +++ b/Documentation/zh_CN/stable_kernel_rules.txt @@ -0,0 +1,66 @@ +Chinese translated version of Documentation/stable_kernel_rules.txt + +If you have any comment or update to the content, please contact the +original document maintainer directly. However, if you have a problem +communicating in English you can also ask the Chinese maintainer for +help. Contact the Chinese maintainer if this translation is outdated +or if there is a problem with the translation. + +Chinese maintainer: TripleX Chung +--------------------------------------------------------------------- +Documentation/stable_kernel_rules.txt 的中文翻译 + +如果想评论或更新本文的内容，请直接联系原文档的维护者。如果你使用英文 +交流有困难的话，也可以向中文版维护者求助。如果本翻译更新不及时或者翻 +译存在问题，请联系中文版维护者。 + + +中文版维护者： 钟宇 TripleX Chung +中文版翻译者： 钟宇 TripleX Chung +中文版校译者： 李阳 Li Yang + Kangkai Yin + +以下为正文 +--------------------------------------------------------------------- + +关于Linux 2.6稳定版发布，所有你想知道的事情。 + +关于哪些类型的补丁可以被接收进入稳定版代码树，哪些不可以的规则： + + - 必须是显而易见的正确，并且经过测试的。 + - 连同上下文，不能大于100行。 + - 必须只修正一件事情。 + - 必须修正了一个给大家带来麻烦的真正的bug（不是“这也许是一个问题...” + 那样的东西）。 + - 必须修正带来如下后果的问题：编译错误（对被标记为CONFIG_BROKEN的例外）， + 内核崩溃，挂起，数据损坏，真正的安全问题，或者一些类似“哦，这不 + 好”的问题。简短的说，就是一些致命的问题。 + - 没有“理论上的竞争条件”，除非能给出竞争条件如何被利用的解释。 + - 不能存在任何的“琐碎的”修正（拼写修正，去掉多余空格之类的）。 + - 必须被相关子系统的维护者接受。 + - 必须遵循Documentation/SubmittingPatches里的规则。 + +向稳定版代码树提交补丁的过程： + + - 在确认了补丁符合以上的规则后，将补丁发送到stable@kernel.org。 + - 如果补丁被接受到队列里，发送者会收到一个ACK回复，如果没有被接受，收 + 到的是NAK回复。回复需要几天的时间，这取决于开发者的时间安排。 + - 被接受的补丁会被加到稳定版本队列里，等待其他开发者的审查。 + - 安全方面的补丁不要发到这个列表，应该发送到security@kernel.org。 + +审查周期： + + - 当稳定版的维护者决定开始一个审查周期，补丁将被发送到审查委员会，以 + 及被补丁影响的领域的维护者（除非提交者就是该领域的维护者）并且抄送 + 到linux-kernel邮件列表。 + - 审查委员会有48小时的时间，用来决定给该补丁回复ACK还是NAK。 + - 如果委员会中有成员拒绝这个补丁，或者linux-kernel列表上有人反对这个 + 补丁，并提出维护者和审查委员会之前没有意识到的问题，补丁会从队列中 + 丢弃。 + - 在审查周期结束的时候，那些得到ACK回应的补丁将会被加入到最新的稳定版 + 发布中，一个新的稳定版发布就此产生。 + - 安全性补丁将从内核安全小组那里直接接收到稳定版代码树中，而不是通过 + 通常的审查周期。请联系内核安全小组以获得关于这个过程的更多细节。 + +审查委员会： + - 由一些自愿承担这项任务的内核开发者，和几个非志愿的组成。 -- cgit v1.2.3 From c642ecf874028c9f41d18d59a9d663c2a954cc45 Mon Sep 17 00:00:00 2001 From: Bryan Wu Date: Wed, 24 Oct 2007 01:00:23 +0800 Subject: Chinese: add translation of volatile-considered-harmful.txt Signed-off-by: Bryan Wu Signed-off-by: Li Yang Signed-off-by: Greg Kroah-Hartman --- .../zh_CN/volatile-considered-harmful.txt | 113 +++++++++++++++++++++ 1 file changed, 113 insertions(+) create mode 100644 Documentation/zh_CN/volatile-considered-harmful.txt (limited to 'Documentation') diff --git a/Documentation/zh_CN/volatile-considered-harmful.txt b/Documentation/zh_CN/volatile-considered-harmful.txt new file mode 100644 index 000000000000..ba8149d2233a --- /dev/null +++ b/Documentation/zh_CN/volatile-considered-harmful.txt @@ -0,0 +1,113 @@ +Chinese translated version of Documentation/volatile-considered-harmful.txt + +If you have any comment or update to the content, please contact the +original document maintainer directly. However, if you have a problem +communicating in English you can also ask the Chinese maintainer for +help. Contact the Chinese maintainer if this translation is outdated +or if there is a problem with the translation. + +Maintainer: Jonathan Corbet +Chinese maintainer: Bryan Wu +--------------------------------------------------------------------- +Documentation/volatile-considered-harmful.txt 的中文翻译 + +如果想评论或更新本文的内容，请直接联系原文档的维护者。如果你使用英文 +交流有困难的话，也可以向中文版维护者求助。如果本翻译更新不及时或者翻 +译存在问题，请联系中文版维护者。 + +英文版维护者： Jonathan Corbet +中文版维护者： 伍鹏 Bryan Wu +中文版翻译者： 伍鹏 Bryan Wu +中文版校译者： 张汉辉 Eugene Teo + 杨瑞 Dave Young +以下为正文 +--------------------------------------------------------------------- + +为什么不应该使用“volatile”类型 +------------------------------ + +C程序员通常认为volatile表示某个变量可以在当前执行的线程之外被改变；因此，在内核 +中用到共享数据结构时，常常会有C程序员喜欢使用volatile这类变量。换句话说，他们经 +常会把volatile类型看成某种简易的原子变量，当然它们不是。在内核中使用volatile几 +乎总是错误的；本文档将解释为什么这样。 + +理解volatile的关键是知道它的目的是用来消除优化，实际上很少有人真正需要这样的应 +用。在内核中，程序员必须防止意外的并发访问破坏共享的数据结构，这其实是一个完全 +不同的任务。用来防止意外并发访问的保护措施，可以更加高效的避免大多数优化相关的 +问题。 + +像volatile一样，内核提供了很多原语来保证并发访问时的数据安全（自旋锁, 互斥量,内 +存屏障等等），同样可以防止意外的优化。如果可以正确使用这些内核原语，那么就没有 +必要再使用volatile。如果仍然必须使用volatile，那么几乎可以肯定在代码的某处有一 +个bug。在正确设计的内核代码中，volatile能带来的仅仅是使事情变慢。 + +思考一下这段典型的内核代码： + + spin_lock(&the_lock); + do_something_on(&shared_data); + do_something_else_with(&shared_data); + spin_unlock(&the_lock); + +如果所有的代码都遵循加锁规则，当持有the_lock的时候，不可能意外的改变shared_data的 +值。任何可能访问该数据的其他代码都会在这个锁上等待。自旋锁原语跟内存屏障一样—— 它 +们显式的用来书写成这样 —— 意味着数据访问不会跨越它们而被优化。所以本来编译器认为 +它知道在shared_data里面将有什么，但是因为spin_lock()调用跟内存屏障一样，会强制编 +译器忘记它所知道的一切。那么在访问这些数据时不会有优化的问题。 + +如果shared_data被声名为volatile，锁操作将仍然是必须的。就算我们知道没有其他人正在 +使用它，编译器也将被阻止优化对临界区内shared_data的访问。在锁有效的同时， +shared_data不是volatile的。在处理共享数据的时候，适当的锁操作可以不再需要 +volatile —— 并且是有潜在危害的。 + +volatile的存储类型最初是为那些内存映射的I/O寄存器而定义。在内核里，寄存器访问也应 +该被锁保护，但是人们也不希望编译器“优化”临界区内的寄存器访问。内核里I/O的内存访问 +是通过访问函数完成的；不赞成通过指针对I/O内存的直接访问，并且不是在所有体系架构上 +都能工作。那些访问函数正是为了防止意外优化而写的，因此，再说一次，volatile类型不 +是必需的。 + +另一种引起用户可能使用volatile的情况是当处理器正忙着等待一个变量的值。正确执行一 +个忙等待的方法是： + + while (my_variable != what_i_want) + cpu_relax(); + +cpu_relax()调用会降低CPU的能量消耗或者让位于超线程双处理器；它也作为内存屏障一样出 +现，所以，再一次，volatile不是必需的。当然，忙等待一开始就是一种反常规的做法。 + +在内核中，一些稀少的情况下volatile仍然是有意义的： + + - 在一些体系架构的系统上，允许直接的I/0内存访问，那么前面提到的访问函数可以使用 + volatile。基本上，每一个访问函数调用它自己都是一个小的临界区域并且保证了按照 + 程序员期望的那样发生访问操作。 + + - 某些会改变内存的内联汇编代码虽然没有什么其他明显的附作用，但是有被GCC删除的可 + 能性。在汇编声明中加上volatile关键字可以防止这种删除操作。 + + - Jiffies变量是一种特殊情况，虽然每次引用它的时候都可以有不同的值，但读jiffies + 变量时不需要任何特殊的加锁保护。所以jiffies变量可以使用volatile，但是不赞成 + 其他跟jiffies相同类型变量使用volatile。Jiffies被认为是一种“愚蠢的遗留物" + （Linus的话）因为解决这个问题比保持现状要麻烦的多。 + + - 由于某些I/0设备可能会修改连续一致的内存,所以有时,指向连续一致内存的数据结构 + 的指针需要正确的使用volatile。网络适配器使用的环状缓存区正是这类情形的一个例 + 子，其中适配器用改变指针来表示哪些描述符已经处理过了。 + +对于大多代码，上述几种可以使用volatile的情况都不适用。所以，使用volatile是一种 +bug并且需要对这样的代码额外仔细检查。那些试图使用volatile的开发人员需要退一步想想 +他们真正想实现的是什么。 + +非常欢迎删除volatile变量的补丁 － 只要证明这些补丁完整的考虑了并发问题。 + +注释 +---- + +[1] http://lwn.net/Articles/233481/ +[2] http://lwn.net/Articles/233482/ + +致谢 +---- + +最初由Randy Dunlap推动并作初步研究 +由Jonathan Corbet撰写 +参考Satyam Sharma，Johannes Stezenbach，Jesper Juhl，Heikki Orsila， +H. Peter Anvin，Philipp Hahn和Stefan Richter的意见改善了本档。 -- cgit v1.2.3 From 22033d38bca82a4a511450562086c69f5dc457ee Mon Sep 17 00:00:00 2001 From: Li Yang Date: Wed, 24 Oct 2007 01:04:21 +0800 Subject: Chinese: add translation of sparse.txt Signed-off-by: Li Yang Signed-off-by: Greg Kroah-Hartman --- Documentation/zh_CN/sparse.txt | 100 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 100 insertions(+) create mode 100644 Documentation/zh_CN/sparse.txt (limited to 'Documentation') diff --git a/Documentation/zh_CN/sparse.txt b/Documentation/zh_CN/sparse.txt new file mode 100644 index 000000000000..75992a603ae3 --- /dev/null +++ b/Documentation/zh_CN/sparse.txt @@ -0,0 +1,100 @@ +Chinese translated version of Documentation/sparse.txt + +If you have any comment or update to the content, please contact the +original document maintainer directly. However, if you have a problem +communicating in English you can also ask the Chinese maintainer for +help. Contact the Chinese maintainer if this translation is outdated +or if there is a problem with the translation. + +Chinese maintainer: Li Yang +--------------------------------------------------------------------- +Documentation/sparse.txt 的中文翻译 + +如果想评论或更新本文的内容，请直接联系原文档的维护者。如果你使用英文 +交流有困难的话，也可以向中文版维护者求助。如果本翻译更新不及时或者翻 +译存在问题，请联系中文版维护者。 + +中文版维护者： 李阳 Li Yang +中文版翻译者： 李阳 Li Yang + + +以下为正文 +--------------------------------------------------------------------- + +Copyright 2004 Linus Torvalds +Copyright 2004 Pavel Machek +Copyright 2006 Bob Copeland + +使用 sparse 工具做类型检查 +~~~~~~~~~~~~~~~~~~~~~~~~~~ + +"__bitwise" 是一种类型属性，所以你应该这样使用它： + + typedef int __bitwise pm_request_t; + + enum pm_request { + PM_SUSPEND = (__force pm_request_t) 1, + PM_RESUME = (__force pm_request_t) 2 + }; + +这样会使 PM_SUSPEND 和 PM_RESUME 成为位方式(bitwise)整数（使用"__force" +是因为 sparse 会抱怨改变位方式的类型转换，但是这里我们确实需要强制进行转 +换）。而且因为所有枚举值都使用了相同的类型，这里的"enum pm_request"也将 +会使用那个类型做为底层实现。 + +而且使用 gcc 编译的时候，所有的 __bitwise/__force 都会消失，最后在 gcc +看来它们只不过是普通的整数。 + +坦白来说，你并不需要使用枚举类型。上面那些实际都可以浓缩成一个特殊的"int +__bitwise"类型。 + +所以更简单的办法只要这样做： + + typedef int __bitwise pm_request_t; + + #define PM_SUSPEND ((__force pm_request_t) 1) + #define PM_RESUME ((__force pm_request_t) 2) + +现在你就有了严格的类型检查所需要的所有基础架构。 + +一个小提醒：常数整数"0"是特殊的。你可以直接把常数零当作位方式整数使用而 +不用担心 sparse 会抱怨。这是因为"bitwise"（恰如其名）是用来确保不同位方 +式类型不会被弄混（小尾模式，大尾模式，cpu尾模式，或者其他），对他们来说 +常数"0"确实是特殊的。 + +获取 sparse 工具 +~~~~~~~~~~~~~~~~ + +你可以从 Sparse 的主页获取最新的发布版本： + + http://www.kernel.org/pub/linux/kernel/people/josh/sparse/ + +或者，你也可以使用 git 克隆最新的 sparse 开发版本： + + git://git.kernel.org/pub/scm/linux/kernel/git/josh/sparse.git + +DaveJ 把每小时自动生成的 git 源码树 tar 包放在以下地址： + + http://www.codemonkey.org.uk/projects/git-snapshots/sparse/ + +一旦你下载了源码，只要以普通用户身份运行： + + make + make install + +它将会被自动安装到你的 ~/bin 目录下。 + +使用 sparse 工具 +~~~~~~~~~~~~~~~~ + +用"make C=1"命令来编译内核，会对所有重新编译的 C 文件使用 sparse 工具。 +或者使用"make C=2"命令，无论文件是否被重新编译都会对其使用 sparse 工具。 +如果你已经编译了内核，用后一种方式可以很快地检查整个源码树。 + +make 的可选变量 CHECKFLAGS 可以用来向 sparse 工具传递参数。编译系统会自 +动向 sparse 工具传递 -Wbitwise 参数。你可以定义 __CHECK_ENDIAN__ 来进行 +大小尾检查。 + + make C=2 CHECKFLAGS="-D__CHECK_ENDIAN__" + +这些检查默认都是被关闭的，因为他们通常会产生大量的警告。 -- cgit v1.2.3 From 463e526083fdaa284eaea45b53bb917ed3c72900 Mon Sep 17 00:00:00 2001 From: Zhang Le Date: Tue, 8 Jan 2008 00:59:16 +0800 Subject: Chinese: add translation of Codingstyle Signed-off-by: Zhang Le Signed-off-by: Li Yang Signed-off-by: Greg Kroah-Hartman --- Documentation/zh_CN/CodingStyle | 701 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 701 insertions(+) create mode 100644 Documentation/zh_CN/CodingStyle (limited to 'Documentation') diff --git a/Documentation/zh_CN/CodingStyle b/Documentation/zh_CN/CodingStyle new file mode 100644 index 000000000000..ecd9307a641f --- /dev/null +++ b/Documentation/zh_CN/CodingStyle @@ -0,0 +1,701 @@ +Chinese translated version of Documentation/CodingStyle + +If you have any comment or update to the content, please post to LKML directly. +However, if you have problem communicating in English you can also ask the +Chinese maintainer for help. Contact the Chinese maintainer, if this +translation is outdated or there is problem with translation. + +Chinese maintainer: Zhang Le +--------------------------------------------------------------------- +Documentation/CodingStyle的中文翻译 + +如果想评论或更新本文的内容，请直接发信到LKML。如果你使用英文交流有困难的话，也可 +以向中文版维护者求助。如果本翻译更新不及时或者翻译存在问题，请联系中文版维护者。 + +中文版维护者： 张乐 Zhang Le +中文版翻译者： 张乐 Zhang Le +中文版校译者： 王聪 Wang Cong + wheelz + 管旭东 Xudong Guan + Li Zefan + Wang Chen +以下为正文 +--------------------------------------------------------------------- + + Linux内核代码风格 + +这是一个简短的文档，描述了linux内核的首选代码风格。代码风格是因人而异的，而且我 +不愿意把我的观点强加给任何人，不过这里所讲述的是我必须要维护的代码所遵守的风格， +并且我也希望绝大多数其他代码也能遵守这个风格。请在写代码时至少考虑一下本文所述的 +风格。 + +首先，我建议你打印一份GNU代码规范，然后不要读它。烧了它，这是一个具有重大象征性 +意义的动作。 + +不管怎样，现在我们开始： + + + 第一章：缩进 + +制表符是8个字符，所以缩进也是8个字符。有些异端运动试图将缩进变为4（乃至2）个字符 +深，这几乎相当于尝试将圆周率的值定义为3。 + +理由：缩进的全部意义就在于清楚的定义一个控制块起止于何处。尤其是当你盯着你的屏幕 +连续看了20小时之后，你将会发现大一点的缩进会使你更容易分辨缩进。 + +现在，有些人会抱怨8个字符的缩进会使代码向右边移动的太远，在80个字符的终端屏幕上 +就很难读这样的代码。这个问题的答案是，如果你需要3级以上的缩进，不管用何种方式你 +的代码已经有问题了，应该修正你的程序。 + +简而言之，8个字符的缩进可以让代码更容易阅读，还有一个好处是当你的函数嵌套太深的 +时候可以给你警告。留心这个警告。 + +在switch语句中消除多级缩进的首选的方式是让“switch”和从属于它的“case”标签对齐于同 +一列，而不要“两次缩进”“case”标签。比如： + + switch (suffix) { + case 'G': + case 'g': + mem <<= 30; + break; + case 'M': + case 'm': + mem <<= 20; + break; + case 'K': + case 'k': + mem <<= 10; + /* fall through */ + default: + break; + } + + +不要把多个语句放在一行里，除非你有什么东西要隐藏： + + if (condition) do_this; + do_something_everytime; + +也不要在一行里放多个赋值语句。内核代码风格超级简单。就是避免可能导致别人误读的表 +达式。 + +除了注释、文档和Kconfig之外，不要使用空格来缩进，前面的例子是例外，是有意为之。 + +选用一个好的编辑器，不要在行尾留空格。 + + + 第二章：把长的行和字符串打散 + +代码风格的意义就在于使用平常使用的工具来维持代码的可读性和可维护性。 + +每一行的长度的限制是80列，我们强烈建议您遵守这个惯例。 + +长于80列的语句要打散成有意义的片段。每个片段要明显短于原来的语句，而且放置的位置 +也明显的靠右。同样的规则也适用于有很长参数列表的函数头。长字符串也要打散成较短的 +字符串。唯一的例外是超过80列可以大幅度提高可读性并且不会隐藏信息的情况。 + +void fun(int a, int b, int c) +{ + if (condition) + printk(KERN_WARNING "Warning this is a long printk with " + "3 parameters a: %u b: %u " + "c: %u \n", a, b, c); + else + next_statement; +} + + 第三章：大括号和空格的放置 + +C语言风格中另外一个常见问题是大括号的放置。和缩进大小不同，选择或弃用某种放置策 +略并没有多少技术上的原因，不过首选的方式，就像Kernighan和Ritchie展示给我们的，是 +把起始大括号放在行尾，而把结束大括号放在行首，所以： + + if (x is true) { + we do y + } + +这适用于所有的非函数语句块（if、switch、for、while、do）。比如： + + switch (action) { + case KOBJ_ADD: + return "add"; + case KOBJ_REMOVE: + return "remove"; + case KOBJ_CHANGE: + return "change"; + default: + return NULL; + } + +不过，有一个例外，那就是函数：函数的起始大括号放置于下一行的开头，所以： + + int function(int x) + { + body of function + } + +全世界的异端可能会抱怨这个不一致性是……呃……不一致的，不过所有思维健全的人都知道（ +a）K&R是_正确的_，并且（b）K&R是正确的。此外，不管怎样函数都是特殊的（在C语言中 +，函数是不能嵌套的）。 + +注意结束大括号独自占据一行，除非它后面跟着同一个语句的剩余部分，也就是do语句中的 +“while”或者if语句中的“else”，像这样： + + do { + body of do-loop + } while (condition); + +和 + + if (x == y) { + .. + } else if (x > y) { + ... + } else { + .... + } + +理由：K&R。 + +也请注意这种大括号的放置方式也能使空（或者差不多空的）行的数量最小化，同时不失可 +读性。因此，由于你的屏幕上的新行是不可再生资源（想想25行的终端屏幕），你将会有更 +多的空行来放置注释。 + +当只有一个单独的语句的时候，不用加不必要的大括号。 + +if (condition) + action(); + +这点不适用于本身为某个条件语句的一个分支的单独语句。这时需要在两个分支里都使用大 +括号。 + +if (condition) { + do_this(); + do_that(); +} else { + otherwise(); +} + + 3.1：空格 + +Linux内核的空格使用方式（主要）取决于它是用于函数还是关键字。（大多数）关键字后 +要加一个空格。值得注意的例外是sizeof、typeof、alignof和__attribute__，这些关键字 +某些程度上看起来更像函数（它们在Linux里也常常伴随小括号而使用，尽管在C语言里这样 +的小括号不是必需的，就像“struct fileinfo info”声明过后的“sizeof info”）。 + +所以在这些关键字之后放一个空格： + if, switch, case, for, do, while +但是不要在sizeof、typeof、alignof或者__attribute__这些关键字之后放空格。例如， + s = sizeof(struct file); + +不要在小括号里的表达式两侧加空格。这是一个反例： + + s = sizeof( struct file ); + +当声明指针类型或者返回指针类型的函数时，“*”的首选使用方式是使之靠近变量名或者函 +数名，而不是靠近类型名。例子： + + char *linux_banner; + unsigned long long memparse(char *ptr, char **retptr); + char *match_strdup(substring_t *s); + +在大多数二元和三元操作符两侧使用一个空格，例如下面所有这些操作符： + + = + - < > * / % | & ^ <= >= == != ? : + +但是一元操作符后不要加空格： + & * + - ~ ! sizeof typeof alignof __attribute__ defined + +后缀自加和自减一元操作符前不加空格： + ++ -- + +前缀自加和自减一元操作符后不加空格： + ++ -- + +“.”和“->”结构体成员操作符前后不加空格。 + +不要在行尾留空白。有些可以自动缩进的编辑器会在新行的行首加入适量的空白，然后你 +就可以直接在那一行输入代码。不过假如你最后没有在那一行输入代码，有些编辑器就不 +会移除已经加入的空白，就像你故意留下一个只有空白的行。包含行尾空白的行就这样产 +生了。 + +当git发现补丁包含了行尾空白的时候会警告你，并且可以应你的要求去掉行尾空白；不过 +如果你是正在打一系列补丁，这样做会导致后面的补丁失败，因为你改变了补丁的上下文。 + + + 第四章：命名 + +C是一个简朴的语言，你的命名也应该这样。和Modula-2和Pascal程序员不同，C程序员不使 +用类似ThisVariableIsATemporaryCounter这样华丽的名字。C程序员会称那个变量为“tmp” +，这样写起来会更容易，而且至少不会令其难于理解。 + +不过，虽然混用大小写的名字是不提倡使用的，但是全局变量还是需要一个具描述性的名字 +。称一个全局函数为“foo”是一个难以饶恕的错误。 + +全局变量（只有当你真正需要它们的时候再用它）需要有一个具描述性的名字，就像全局函 +数。如果你有一个可以计算活动用户数量的函数，你应该叫它“count_active_users()”或者 +类似的名字，你不应该叫它“cntuser()”。 + +在函数名中包含函数类型（所谓的匈牙利命名法）是脑子出了问题——编译器知道那些类型而 +且能够检查那些类型，这样做只能把程序员弄糊涂了。难怪微软总是制造出有问题的程序。 + +本地变量名应该简短，而且能够表达相关的含义。如果你有一些随机的整数型的循环计数器 +，它应该被称为“i”。叫它“loop_counter”并无益处，如果它没有被误解的可能的话。类似 +的，“tmp”可以用来称呼任意类型的临时变量。 + +如果你怕混淆了你的本地变量名，你就遇到另一个问题了，叫做函数增长荷尔蒙失衡综合症 +。请看第六章（函数）。 + + + 第五章：Typedef + +不要使用类似“vps_t”之类的东西。 + +对结构体和指针使用typedef是一个错误。当你在代码里看到： + + vps_t a; + +这代表什么意思呢？ + +相反，如果是这样 + + struct virtual_container *a; + +你就知道“a”是什么了。 + +很多人认为typedef“能提高可读性”。实际不是这样的。它们只在下列情况下有用： + + (a) 完全不透明的对象（这种情况下要主动使用typedef来隐藏这个对象实际上是什么）。 + + 例如：“pte_t”等不透明对象，你只能用合适的访问函数来访问它们。 + + 注意！不透明性和“访问函数”本身是不好的。我们使用pte_t等类型的原因在于真的是 + 完全没有任何共用的可访问信息。 + + (b) 清楚的整数类型，如此，这层抽象就可以帮助消除到底是“int”还是“long”的混淆。 + + u8/u16/u32是完全没有问题的typedef，不过它们更符合类别(d)而不是这里。 + + 再次注意！要这样做，必须事出有因。如果某个变量是“unsigned long“，那么没有必要 + + typedef unsigned long myflags_t; + + 不过如果有一个明确的原因，比如它在某种情况下可能会是一个“unsigned int”而在 + 其他情况下可能为“unsigned long”，那么就不要犹豫，请务必使用typedef。 + + (c) 当你使用sparse按字面的创建一个新类型来做类型检查的时候。 + + (d) 和标准C99类型相同的类型，在某些例外的情况下。 + + 虽然让眼睛和脑筋来适应新的标准类型比如“uint32_t”不需要花很多时间，可是有些 + 人仍然拒绝使用它们。 + + 因此，Linux特有的等同于标准类型的“u8/u16/u32/u64”类型和它们的有符号类型是被 + 允许的——尽管在你自己的新代码中，它们不是强制要求要使用的。 + + 当编辑已经使用了某个类型集的已有代码时，你应该遵循那些代码中已经做出的选择。 + + (e) 可以在用户空间安全使用的类型。 + + 在某些用户空间可见的结构体里，我们不能要求C99类型而且不能用上面提到的“u32” + 类型。因此，我们在与用户空间共享的所有结构体中使用__u32和类似的类型。 + +可能还有其他的情况，不过基本的规则是永远不要使用typedef，除非你可以明确的应用上 +述某个规则中的一个。 + +总的来说，如果一个指针或者一个结构体里的元素可以合理的被直接访问到，那么它们就不 +应该是一个typedef。 + + + 第六章：函数 + +函数应该简短而漂亮，并且只完成一件事情。函数应该可以一屏或者两屏显示完（我们都知 +道ISO/ANSI屏幕大小是80x24），只做一件事情，而且把它做好。 + +一个函数的最大长度是和该函数的复杂度和缩进级数成反比的。所以，如果你有一个理论上 +很简单的只有一个很长（但是简单）的case语句的函数，而且你需要在每个case里做很多很 +小的事情，这样的函数尽管很长，但也是可以的。 + +不过，如果你有一个复杂的函数，而且你怀疑一个天分不是很高的高中一年级学生可能甚至 +搞不清楚这个函数的目的，你应该严格的遵守前面提到的长度限制。使用辅助函数，并为之 +取个具描述性的名字（如果你觉得它们的性能很重要的话，可以让编译器内联它们，这样的 +效果往往会比你写一个复杂函数的效果要好。） + +函数的另外一个衡量标准是本地变量的数量。此数量不应超过5－10个，否则你的函数就有 +问题了。重新考虑一下你的函数，把它分拆成更小的函数。人的大脑一般可以轻松的同时跟 +踪7个不同的事物，如果再增多的话，就会糊涂了。即便你聪颖过人，你也可能会记不清你2 +个星期前做过的事情。 + +在源文件里，使用空行隔开不同的函数。如果该函数需要被导出，它的EXPORT*宏应该紧贴 +在它的结束大括号之下。比如： + +int system_is_up(void) +{ + return system_state == SYSTEM_RUNNING; +} +EXPORT_SYMBOL(system_is_up); + +在函数原型中，包含函数名和它们的数据类型。虽然C语言里没有这样的要求，在Linux里这 +是提倡的做法，因为这样可以很简单的给读者提供更多的有价值的信息。 + + + 第七章：集中的函数退出途径 + +虽然被某些人声称已经过时，但是goto语句的等价物还是经常被编译器所使用，具体形式是 +无条件跳转指令。 + +当一个函数从多个位置退出并且需要做一些通用的清理工作的时候，goto的好处就显现出来 +了。 + +理由是： + +- 无条件语句容易理解和跟踪 +- 嵌套程度减小 +- 可以避免由于修改时忘记更新某个单独的退出点而导致的错误 +- 减轻了编译器的工作，无需删除冗余代码;) + +int fun(int a) +{ + int result = 0; + char *buffer = kmalloc(SIZE); + + if (buffer == NULL) + return -ENOMEM; + + if (condition1) { + while (loop1) { + ... + } + result = 1; + goto out; + } + ... +out: + kfree(buffer); + return result; +} + + 第八章：注释 + +注释是好的，不过有过度注释的危险。永远不要在注释里解释你的代码是如何运作的：更好 +的做法是让别人一看你的代码就可以明白，解释写的很差的代码是浪费时间。 + +一般的，你想要你的注释告诉别人你的代码做了什么，而不是怎么做的。也请你不要把注释 +放在一个函数体内部：如果函数复杂到你需要独立的注释其中的一部分，你很可能需要回到 +第六章看一看。你可以做一些小注释来注明或警告某些很聪明（或者槽糕）的做法，但不要 +加太多。你应该做的，是把注释放在函数的头部，告诉人们它做了什么，也可以加上它做这 +些事情的原因。 + +当注释内核API函数时，请使用kernel-doc格式。请看 +Documentation/kernel-doc-nano-HOWTO.txt和scripts/kernel-doc以获得详细信息。 + +Linux的注释风格是C89“/* ... */”风格。不要使用C99风格“// ...”注释。 + +长（多行）的首选注释风格是： + + /* + * This is the preferred style for multi-line + * comments in the Linux kernel source code. + * Please use it consistently. + * + * Description: A column of asterisks on the left side, + * with beginning and ending almost-blank lines. + */ + +注释数据也是很重要的，不管是基本类型还是衍生类型。为了方便实现这一点，每一行应只 +声明一个数据（不要使用逗号来一次声明多个数据）。这样你就有空间来为每个数据写一段 +小注释来解释它们的用途了。 + + + 第九章：你已经把事情弄糟了 + +这没什么，我们都是这样。可能你的使用了很长时间Unix的朋友已经告诉你“GNU emacs”能 +自动帮你格式化C源代码，而且你也注意到了，确实是这样，不过它所使用的默认值和我们 +想要的相去甚远（实际上，甚至比随机打的还要差——无数个猴子在GNU emacs里打字永远不 +会创造出一个好程序）（译注：请参考Infinite Monkey Theorem） + +所以你要么放弃GNU emacs，要么改变它让它使用更合理的设定。要采用后一个方案，你可 +以把下面这段粘贴到你的.emacs文件里。 + +(defun linux-c-mode () + "C mode with adjusted defaults for use with the Linux kernel." + (interactive) + (c-mode) + (c-set-style "K&R") + (setq tab-width 8) + (setq indent-tabs-mode t) + (setq c-basic-offset 8)) + +这样就定义了M-x linux-c-mode命令。当你hack一个模块的时候，如果你把字符串 +-*- linux-c -*-放在头两行的某个位置，这个模式将会被自动调用。如果你希望在你修改 +/usr/src/linux里的文件时魔术般自动打开linux-c-mode的话，你也可能需要添加 + +(setq auto-mode-alist (cons '("/usr/src/linux.*/.*\\.[ch]$" . linux-c-mode) + auto-mode-alist)) + +到你的.emacs文件里。 + +不过就算你尝试让emacs正确的格式化代码失败了，也并不意味着你失去了一切：还可以用“ +indent”。 + +不过，GNU indent也有和GNU emacs一样有问题的设定，所以你需要给它一些命令选项。不 +过，这还不算太糟糕，因为就算是GNU indent的作者也认同K&R的权威性（GNU的人并不是坏 +人，他们只是在这个问题上被严重的误导了），所以你只要给indent指定选项“-kr -i8” +（代表“K&R，8个字符缩进”），或者使用“scripts/Lindent”，这样就可以以最时髦的方式 +缩进源代码。 + +“indent”有很多选项，特别是重新格式化注释的时候，你可能需要看一下它的手册页。不过 +记住：“indent”不能修正坏的编程习惯。 + + + 第十章：Kconfig配置文件 + +对于遍布源码树的所有Kconfig*配置文件来说，它们缩进方式与C代码相比有所不同。紧挨 +在“config”定义下面的行缩进一个制表符，帮助信息则再多缩进2个空格。比如： + +config AUDIT + bool "Auditing support" + depends on NET + help + Enable auditing infrastructure that can be used with another + kernel subsystem, such as SELinux (which requires this for + logging of avc messages output). Does not do system-call + auditing without CONFIG_AUDITSYSCALL. + +仍然被认为不够稳定的功能应该被定义为依赖于“EXPERIMENTAL”： + +config SLUB + depends on EXPERIMENTAL && !ARCH_USES_SLAB_PAGE_STRUCT + bool "SLUB (Unqueued Allocator)" + ... + +而那些危险的功能（比如某些文件系统的写支持）应该在它们的提示字符串里显著的声明这 +一点： + +config ADFS_FS_RW + bool "ADFS write support (DANGEROUS)" + depends on ADFS_FS + ... + +要查看配置文件的完整文档，请看Documentation/kbuild/kconfig-language.txt。 + + + 第十一章：数据结构 + +如果一个数据结构，在创建和销毁它的单线执行环境之外可见，那么它必须要有一个引用计 +数器。内核里没有垃圾收集（并且内核之外的垃圾收集慢且效率低下），这意味着你绝对需 +要记录你对这种数据结构的使用情况。 + +引用计数意味着你能够避免上锁，并且允许多个用户并行访问这个数据结构——而不需要担心 +这个数据结构仅仅因为暂时不被使用就消失了，那些用户可能不过是沉睡了一阵或者做了一 +些其他事情而已。 + +注意上锁不能取代引用计数。上锁是为了保持数据结构的一致性，而引用计数是一个内存管 +理技巧。通常二者都需要，不要把两个搞混了。 + +很多数据结构实际上有2级引用计数，它们通常有不同“类”的用户。子类计数器统计子类用 +户的数量，每当子类计数器减至零时，全局计数器减一。 + +这种“多级引用计数”的例子可以在内存管理（“struct mm_struct”：mm_users和mm_count） +和文件系统（“struct super_block”：s_count和s_active）中找到。 + +记住：如果另一个执行线索可以找到你的数据结构，但是这个数据结构没有引用计数器，这 +里几乎肯定是一个bug。 + + + 第十二章：宏，枚举和RTL + +用于定义常量的宏的名字及枚举里的标签需要大写。 + +#define CONSTANT 0x12345 + +在定义几个相关的常量时，最好用枚举。 + +宏的名字请用大写字母，不过形如函数的宏的名字可以用小写字母。 + +一般的，如果能写成内联函数就不要写成像函数的宏。 + +含有多个语句的宏应该被包含在一个do-while代码块里： + +#define macrofun(a, b, c) \ + do { \ + if (a == 5) \ + do_this(b, c); \ + } while (0) + +使用宏的时候应避免的事情： + +1) 影响控制流程的宏： + +#define FOO(x) \ + do { \ + if (blah(x) < 0) \ + return -EBUGGERED; \ + } while(0) + +非常不好。它看起来像一个函数，不过却能导致“调用”它的函数退出；不要打乱读者大脑里 +的语法分析器。 + +2) 依赖于一个固定名字的本地变量的宏： + +#define FOO(val) bar(index, val) + +可能看起来像是个不错的东西，不过它非常容易把读代码的人搞糊涂，而且容易导致看起来 +不相关的改动带来错误。 + +3) 作为左值的带参数的宏： FOO(x) = y；如果有人把FOO变成一个内联函数的话，这种用 +法就会出错了。 + +4) 忘记了优先级：使用表达式定义常量的宏必须将表达式置于一对小括号之内。带参数的 +宏也要注意此类问题。 + +#define CONSTANT 0x4000 +#define CONSTEXP (CONSTANT | 3) + +cpp手册对宏的讲解很详细。Gcc internals手册也详细讲解了RTL（译注：register +transfer language），内核里的汇编语言经常用到它。 + + + 第十三章：打印内核消息 + +内核开发者应该是受过良好教育的。请一定注意内核信息的拼写，以给人以好的印象。不要 +用不规范的单词比如“dont”，而要用“do not”或者“don't”。保证这些信息简单、明了、无 +歧义。 + +内核信息不必以句号（译注：英文句号，即点）结束。 + +在小括号里打印数字(%d)没有任何价值，应该避免这样做。 + +里有一些驱动模型诊断宏，你应该使用它们，以确保信息对应于正确的 +设备和驱动，并且被标记了正确的消息级别。这些宏有：dev_err(), dev_warn(), +dev_info()等等。对于那些不和某个特定设备相关连的信息，定义了 +pr_debug()和pr_info()。 + +写出好的调试信息可以是一个很大的挑战；当你写出来之后，这些信息在远程除错的时候 +就会成为极大的帮助。当DEBUG符号没有被定义的时候，这些信息不应该被编译进内核里 +（也就是说，默认地，它们不应该被包含在内）。如果你使用dev_dbg()或者pr_debug()， +就能自动达到这个效果。很多子系统拥有Kconfig选项来启用-DDEBUG。还有一个相关的惯例 +是使用VERBOSE_DEBUG来添加dev_vdbg()消息到那些已经由DEBUG启用的消息之上。 + + + 第十四章：分配内存 + +内核提供了下面的一般用途的内存分配函数：kmalloc()，kzalloc()，kcalloc()和 +vmalloc()。请参考API文档以获取有关它们的详细信息。 + +传递结构体大小的首选形式是这样的： + + p = kmalloc(sizeof(*p), ...); + +另外一种传递方式中，sizeof的操作数是结构体的名字，这样会降低可读性，并且可能会引 +入bug。有可能指针变量类型被改变时，而对应的传递给内存分配函数的sizeof的结果不变。 + +强制转换一个void指针返回值是多余的。C语言本身保证了从void指针到其他任何指针类型 +的转换是没有问题的。 + + + 第十五章：内联弊病 + +有一个常见的误解是内联函数是gcc提供的可以让代码运行更快的一个选项。虽然使用内联 +函数有时候是恰当的（比如作为一种替代宏的方式，请看第十二章），不过很多情况下不是 +这样。inline关键字的过度使用会使内核变大，从而使整个系统运行速度变慢。因为大内核 +会占用更多的指令高速缓存（译注：一级缓存通常是指令缓存和数据缓存分开的）而且会导 +致pagecache的可用内存减少。想象一下，一次pagecache未命中就会导致一次磁盘寻址，将 +耗时5毫秒。5毫秒的时间内CPU能执行很多很多指令。 + +一个基本的原则是如果一个函数有3行以上，就不要把它变成内联函数。这个原则的一个例 +外是，如果你知道某个参数是一个编译时常量，而且因为这个常量你确定编译器在编译时能 +优化掉你的函数的大部分代码，那仍然可以给它加上inline关键字。kmalloc()内联函数就 +是一个很好的例子。 + +人们经常主张给static的而且只用了一次的函数加上inline，如此不会有任何损失，因为没 +有什么好权衡的。虽然从技术上说这是正确的，但是实际上这种情况下即使不加inline gcc +也可以自动使其内联。而且其他用户可能会要求移除inline，由此而来的争论会抵消inline +自身的潜在价值，得不偿失。 + + + 第十六章：函数返回值及命名 + +函数可以返回很多种不同类型的值，最常见的一种是表明函数执行成功或者失败的值。这样 +的一个值可以表示为一个错误代码整数（-Exxx＝失败，0＝成功）或者一个“成功”布尔值（ +0＝失败，非0＝成功）。 + +混合使用这两种表达方式是难于发现的bug的来源。如果C语言本身严格区分整形和布尔型变 +量，那么编译器就能够帮我们发现这些错误……不过C语言不区分。为了避免产生这种bug，请 +遵循下面的惯例： + + 如果函数的名字是一个动作或者强制性的命令，那么这个函数应该返回错误代码整 + 数。如果是一个判断，那么函数应该返回一个“成功”布尔值。 + +比如，“add work”是一个命令，所以add_work()函数在成功时返回0，在失败时返回-EBUSY。 +类似的，因为“PCI device present”是一个判断，所以pci_dev_present()函数在成功找到 +一个匹配的设备时应该返回1，如果找不到时应该返回0。 + +所有导出（译注：EXPORT）的函数都必须遵守这个惯例，所有的公共函数也都应该如此。私 +有（static）函数不需要如此，但是我们也推荐这样做。 + +返回值是实际计算结果而不是计算是否成功的标志的函数不受此惯例的限制。一般的，他们 +通过返回一些正常值范围之外的结果来表示出错。典型的例子是返回指针的函数，他们使用 +NULL或者ERR_PTR机制来报告错误。 + + + 第十七章：不要重新发明内核宏 + +头文件include/linux/kernel.h包含了一些宏，你应该使用它们，而不要自己写一些它们的 +变种。比如，如果你需要计算一个数组的长度，使用这个宏 + + #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0])) + +类似的，如果你要计算某结构体成员的大小，使用 + + #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f)) + +还有可以做严格的类型检查的min()和max()宏，如果你需要可以使用它们。你可以自己看看 +那个头文件里还定义了什么你可以拿来用的东西，如果有定义的话，你就不应在你的代码里 +自己重新定义。 + + + 第十八章：编辑器模式行和其他需要罗嗦的事情 + +有一些编辑器可以解释嵌入在源文件里的由一些特殊标记标明的配置信息。比如，emacs +能够解释被标记成这样的行： + +-*- mode: c -*- + +或者这样的： + +/* +Local Variables: +compile-command: "gcc -DMAGIC_DEBUG_FLAG foo.c" +End: +*/ + +Vim能够解释这样的标记： + +/* vim:set sw=8 noet */ + +不要在源代码中包含任何这样的内容。每个人都有他自己的编辑器配置，你的源文件不应 +该覆盖别人的配置。这包括有关缩进和模式配置的标记。人们可以使用他们自己定制的模 +式，或者使用其他可以产生正确的缩进的巧妙方法。 + + + + 附录 I：参考 + +The C Programming Language, 第二版, 作者Brian W. Kernighan和Denni +M. Ritchie. Prentice Hall, Inc., 1988. ISBN 0-13-110362-8 (软皮), +0-13-110370-9 (硬皮). URL: http://cm.bell-labs.com/cm/cs/cbook/ + +The Practice of Programming 作者Brian W. Kernighan和Rob Pike. Addison-Wesley, +Inc., 1999. ISBN 0-201-61586-X. URL: http://cm.bell-labs.com/cm/cs/tpop/ + +cpp，gcc，gcc internals和indent的GNU手册——和K&R及本文相符合的部分，全部可以在 +http://www.gnu.org/manual/找到 + +WG14是C语言的国际标准化工作组，URL: http://www.open-std.org/JTC1/SC22/WG14/ + +Kernel CodingStyle，作者greg@kroah.com发表于OLS 2002： +http://www.kroah.com/linux/talks/ols_2002_kernel_codingstyle_talk/html/ + +-- +最后更新于2007年7月13日。 -- cgit v1.2.3 From b1c7192df1b089ed654c76fbf14d7f7dbf4b9616 Mon Sep 17 00:00:00 2001 From: "Robert P. J. Day" Date: Wed, 7 Nov 2007 04:09:46 -0500 Subject: Documentation: Replace obsolete "driverfs" with "sysfs". Signed-off-by: Robert P. J. Day --- Documentation/pnp.txt | 4 ++-- Documentation/s390/cds.txt | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'Documentation') diff --git a/Documentation/pnp.txt b/Documentation/pnp.txt index 481faf515d53..a327db67782a 100644 --- a/Documentation/pnp.txt +++ b/Documentation/pnp.txt @@ -17,9 +17,9 @@ The User Interface ------------------ The Linux Plug and Play user interface provides a means to activate PnP devices for legacy and user level drivers that do not support Linux Plug and Play. The -user interface is integrated into driverfs. +user interface is integrated into sysfs. -In addition to the standard driverfs file the following are created in each +In addition to the standard sysfs file the following are created in each device's directory: id - displays a list of support EISA IDs options - displays possible resource configurations diff --git a/Documentation/s390/cds.txt b/Documentation/s390/cds.txt index 3081927cc2d6..c4b7b2bd369a 100644 --- a/Documentation/s390/cds.txt +++ b/Documentation/s390/cds.txt @@ -133,7 +133,7 @@ During its startup the Linux/390 system checks for peripheral devices. Each of those devices is uniquely defined by a so called subchannel by the ESA/390 channel subsystem. While the subchannel numbers are system generated, each subchannel also takes a user defined attribute, the so called device number. -Both subchannel number and device number cannot exceed 65535. During driverfs +Both subchannel number and device number cannot exceed 65535. During sysfs initialisation, the information about control unit type and device types that imply specific I/O commands (channel command words - CCWs) in order to operate the device are gathered. Device drivers can retrieve this set of hardware -- cgit v1.2.3 From 081248de0a0288a0ce4e1447a07ccf56aa4fae01 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 1 Nov 2007 09:29:06 -0600 Subject: kset: move /sys/slab to /sys/kernel/slab /sys/kernel is where these things should go. Also updated the documentation and tool that used this directory. Cc: Kay Sievers Acked-by: Christoph Lameter Signed-off-by: Greg Kroah-Hartman --- Documentation/vm/slabinfo.c | 2 +- Documentation/vm/slub.txt | 2 +- mm/slub.c | 3 ++- 3 files changed, 4 insertions(+), 3 deletions(-) (limited to 'Documentation') diff --git a/Documentation/vm/slabinfo.c b/Documentation/vm/slabinfo.c index 7047696c47a1..488c1f31b992 100644 --- a/Documentation/vm/slabinfo.c +++ b/Documentation/vm/slabinfo.c @@ -1021,7 +1021,7 @@ void read_slab_dir(void) char *t; int count; - if (chdir("/sys/slab")) + if (chdir("/sys/kernel/slab")) fatal("SYSFS support for SLUB not active\n"); dir = opendir("."); diff --git a/Documentation/vm/slub.txt b/Documentation/vm/slub.txt index d17f324db9f5..dcf8bcf846d6 100644 --- a/Documentation/vm/slub.txt +++ b/Documentation/vm/slub.txt @@ -63,7 +63,7 @@ In case you forgot to enable debugging on the kernel command line: It is possible to enable debugging manually when the kernel is up. Look at the contents of: -/sys/slab// +/sys/kernel/slab// Look at the writable files. Writing 1 to them will enable the corresponding debug option. All options can be set on a slab that does diff --git a/mm/slub.c b/mm/slub.c index 886131c5b5ca..b6c79462157e 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -4091,7 +4091,8 @@ static int __init slab_sysfs_init(void) struct kmem_cache *s; int err; - slab_kset = kset_create_and_add("slab", &slab_uevent_ops, NULL); + slab_kset = kset_create_and_add("slab", &slab_uevent_ops, + &kernel_kset->kobj); if (!slab_kset) { printk(KERN_ERR "Cannot register slab subsystem.\n"); return -ENOSYS; -- cgit v1.2.3 From 5008743dc7f98dd1ad4f20f4d7ff0b479e78895d Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Tue, 27 Nov 2007 11:28:26 -0800 Subject: kobject: remove old, outdated documentation. As we are replacing the documentation, it's easier to do this in a two stage pass, delete the old file and add the new one. Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- Documentation/kobject.txt | 289 ---------------------------------------------- 1 file changed, 289 deletions(-) delete mode 100644 Documentation/kobject.txt (limited to 'Documentation') diff --git a/Documentation/kobject.txt b/Documentation/kobject.txt deleted file mode 100644 index ca86a885ad8f..000000000000 --- a/Documentation/kobject.txt +++ /dev/null @@ -1,289 +0,0 @@ -The kobject Infrastructure - -Patrick Mochel - -Updated: 3 June 2003 - - -Copyright (c) 2003 Patrick Mochel -Copyright (c) 2003 Open Source Development Labs - - -0. Introduction - -The kobject infrastructure performs basic object management that larger -data structures and subsystems can leverage, rather than reimplement -similar functionality. This functionality primarily concerns: - -- Object reference counting. -- Maintaining lists (sets) of objects. -- Object set locking. -- Userspace representation. - -The infrastructure consists of a number of object types to support -this functionality. Their programming interfaces are described below -in detail, and briefly here: - -- kobjects a simple object. -- kset a set of objects of a certain type. -- ktype a set of helpers for objects of a common type. - - -The kobject infrastructure maintains a close relationship with the -sysfs filesystem. Each kobject that is registered with the kobject -core receives a directory in sysfs. Attributes about the kobject can -then be exported. Please see Documentation/filesystems/sysfs.txt for -more information. - -The kobject infrastructure provides a flexible programming interface, -and allows kobjects and ksets to be used without being registered -(i.e. with no sysfs representation). This is also described later. - - -1. kobjects - -1.1 Description - - -struct kobject is a simple data type that provides a foundation for -more complex object types. It provides a set of basic fields that -almost all complex data types share. kobjects are intended to be -embedded in larger data structures and replace fields they duplicate. - -1.2 Definition - -struct kobject { - const char * k_name; - struct kref kref; - struct list_head entry; - struct kobject * parent; - struct kset * kset; - struct kobj_type * ktype; - struct sysfs_dirent * sd; - wait_queue_head_t poll; -}; - -void kobject_init(struct kobject *); -int kobject_add(struct kobject *); -int kobject_register(struct kobject *); - -void kobject_del(struct kobject *); -void kobject_unregister(struct kobject *); - -struct kobject * kobject_get(struct kobject *); -void kobject_put(struct kobject *); - - -1.3 kobject Programming Interface - -kobjects may be dynamically added and removed from the kobject core -using kobject_register() and kobject_unregister(). Registration -includes inserting the kobject in the list of its dominant kset and -creating a directory for it in sysfs. - -Alternatively, one may use a kobject without adding it to its kset's list -or exporting it via sysfs, by simply calling kobject_init(). An -initialized kobject may later be added to the object hierarchy by -calling kobject_add(). An initialized kobject may be used for -reference counting. - -Note: calling kobject_init() then kobject_add() is functionally -equivalent to calling kobject_register(). - -When a kobject is unregistered, it is removed from its kset's list, -removed from the sysfs filesystem, and its reference count is decremented. -List and sysfs removal happen in kobject_del(), and may be called -manually. kobject_put() decrements the reference count, and may also -be called manually. - -A kobject's reference count may be incremented with kobject_get(), -which returns a valid reference to a kobject; and decremented with -kobject_put(). An object's reference count may only be incremented if -it is already positive. - -When a kobject's reference count reaches 0, the method struct -kobj_type::release() (which the kobject's kset points to) is called. -This allows any memory allocated for the object to be freed. - - -NOTE!!! - -It is _imperative_ that you supply a destructor for dynamically -allocated kobjects to free them if you are using kobject reference -counts. The reference count controls the lifetime of the object. -If it goes to 0, then it is assumed that the object will -be freed and cannot be used. - -More importantly, you must free the object there, and not immediately -after an unregister call. If someone else is referencing the object -(e.g. through a sysfs file), they will obtain a reference to the -object, assume it's valid and operate on it. If the object is -unregistered and freed in the meantime, the operation will then -reference freed memory and go boom. - -This can be prevented, in the simplest case, by defining a release -method and freeing the object from there only. Note that this will not -secure reference count/object management models that use a dual -reference count or do other wacky things with the reference count -(like the networking layer). - - -1.4 sysfs - -Each kobject receives a directory in sysfs. This directory is created -under the kobject's parent directory. - -If a kobject does not have a parent when it is registered, its parent -becomes its dominant kset. - -If a kobject does not have a parent nor a dominant kset, its directory -is created at the top-level of the sysfs partition. - - - -2. ksets - -2.1 Description - -A kset is a set of kobjects that are embedded in the same type. - - -struct kset { - struct kobj_type * ktype; - struct list_head list; - struct kobject kobj; - struct kset_uevent_ops * uevent_ops; -}; - - -void kset_init(struct kset * k); -int kset_add(struct kset * k); -int kset_register(struct kset * k); -void kset_unregister(struct kset * k); - -struct kset * kset_get(struct kset * k); -void kset_put(struct kset * k); - -struct kobject * kset_find_obj(struct kset *, char *); - - -The type that the kobjects are embedded in is described by the ktype -pointer. - -A kset contains a kobject itself, meaning that it may be registered in -the kobject hierarchy and exported via sysfs. More importantly, the -kset may be embedded in a larger data type, and may be part of another -kset (of that object type). - -For example, a block device is an object (struct gendisk) that is -contained in a set of block devices. It may also contain a set of -partitions (struct hd_struct) that have been found on the device. The -following code snippet illustrates how to express this properly. - - struct gendisk * disk; - ... - disk->kset.kobj.kset = &block_kset; - disk->kset.ktype = &partition_ktype; - kset_register(&disk->kset); - -- The kset that the disk's embedded object belongs to is the - block_kset, and is pointed to by disk->kset.kobj.kset. - -- The type of objects on the disk's _subordinate_ list are partitions, - and is set in disk->kset.ktype. - -- The kset is then registered, which handles initializing and adding - the embedded kobject to the hierarchy. - - -2.2 kset Programming Interface - -All kset functions, except kset_find_obj(), eventually forward the -calls to their embedded kobjects after performing kset-specific -operations. ksets offer a similar programming model to kobjects: they -may be used after they are initialized, without registering them in -the hierarchy. - -kset_find_obj() may be used to locate a kobject with a particular -name. The kobject, if found, is returned. - -There are also some helper functions which names point to the formerly -existing "struct subsystem", whose functions have been taken over by -ksets. - - -decl_subsys(name,type,uevent_ops) - -Declares a kset named '_subsys' of type with -uevent_ops . For example, - -decl_subsys(devices, &ktype_device, &device_uevent_ops); - -is equivalent to doing: - -struct kset devices_subsys = { - .ktype = &ktype_devices, - .uevent_ops = &device_uevent_ops, -}; -kobject_set_name(&devices_subsys, name); - -The objects that are registered with a subsystem that use the -subsystem's default list must have their kset ptr set properly. These -objects may have embedded kobjects or ksets. The -following helper makes setting the kset easier: - - -kobj_set_kset_s(obj,subsys) - -- Assumes that obj->kobj exists, and is a struct kobject. -- Sets the kset of that kobject to the kset . - -int subsystem_register(struct kset *s); -void subsystem_unregister(struct kset *s); - -These are just wrappers around the respective kset_* functions. - -2.3 sysfs - -ksets are represented in sysfs when their embedded kobjects are -registered. They follow the same rules of parenting, with one -exception. If a kset does not have a parent, nor is its embedded -kobject part of another kset, the kset's parent becomes its dominant -subsystem. - -If the kset does not have a parent, its directory is created at the -sysfs root. This should only happen when the kset registered is -embedded in a subsystem itself. - - -3. struct ktype - -3.1. Description - -struct kobj_type { - void (*release)(struct kobject *); - struct sysfs_ops * sysfs_ops; - struct attribute ** default_attrs; -}; - - -Object types require specific functions for converting between the -generic object and the more complex type. struct kobj_type provides -the object-specific fields, which include: - -- release: Called when the kobject's reference count reaches 0. This - should convert the object to the more complex type and free it. - -- sysfs_ops: Provides conversion functions for sysfs access. Please - see the sysfs documentation for more information. - -- default_attrs: Default attributes to be exported via sysfs when the - object is registered.Note that the last attribute has to be - initialized to NULL ! You can find a complete implementation - in block/genhd.c - - -Instances of struct kobj_type are not registered; only referenced by -the kset. A kobj_type may be referenced by an arbitrary number of -ksets, as there may be disparate sets of identical objects. - -- cgit v1.2.3 From 36d78d6c5b50ba945bbdee9bf1d8daac00154e02 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Tue, 27 Nov 2007 11:28:26 -0800 Subject: kobject: update the kobject/kset documentation This provides a much-needed kobject and kset documentation update. Thanks to Kay Sievers, Alan Stern, Jonathan Corbet, Randy Dunlap, Jan Engelhardt, and others for reviewing and providing help with this document. Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- Documentation/kobject.txt | 386 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 386 insertions(+) create mode 100644 Documentation/kobject.txt (limited to 'Documentation') diff --git a/Documentation/kobject.txt b/Documentation/kobject.txt new file mode 100644 index 000000000000..bf3256e04027 --- /dev/null +++ b/Documentation/kobject.txt @@ -0,0 +1,386 @@ +Everything you never wanted to know about kobjects, ksets, and ktypes + +Greg Kroah-Hartman + +Based on an original article by Jon Corbet for lwn.net written October 1, +2003 and located at http://lwn.net/Articles/51437/ + +Last updated December 19, 2007 + + +Part of the difficulty in understanding the driver model - and the kobject +abstraction upon which it is built - is that there is no obvious starting +place. Dealing with kobjects requires understanding a few different types, +all of which make reference to each other. In an attempt to make things +easier, we'll take a multi-pass approach, starting with vague terms and +adding detail as we go. To that end, here are some quick definitions of +some terms we will be working with. + + - A kobject is an object of type struct kobject. Kobjects have a name + and a reference count. A kobject also has a parent pointer (allowing + objects to be arranged into hierarchies), a specific type, and, + usually, a representation in the sysfs virtual filesystem. + + Kobjects are generally not interesting on their own; instead, they are + usually embedded within some other structure which contains the stuff + the code is really interested in. + + No structure should EVER have more than one kobject embedded within it. + If it does, the reference counting for the object is sure to be messed + up and incorrect, and your code will be buggy. So do not do this. + + - A ktype is the type of object that embeds a kobject. Every structure + that embeds a kobject needs a corresponding ktype. The ktype controls + what happens to the kobject when it is created and destroyed. + + - A kset is a group of kobjects. These kobjects can be of the same ktype + or belong to different ktypes. The kset is the basic container type for + collections of kobjects. Ksets contain their own kobjects, but you can + safely ignore that implementation detail as the kset core code handles + this kobject automatically. + + When you see a sysfs directory full of other directories, generally each + of those directories corresponds to a kobject in the same kset. + +We'll look at how to create and manipulate all of these types. A bottom-up +approach will be taken, so we'll go back to kobjects. + + +Embedding kobjects + +It is rare for kernel code to create a standalone kobject, with one major +exception explained below. Instead, kobjects are used to control access to +a larger, domain-specific object. To this end, kobjects will be found +embedded in other structures. If you are used to thinking of things in +object-oriented terms, kobjects can be seen as a top-level, abstract class +from which other classes are derived. A kobject implements a set of +capabilities which are not particularly useful by themselves, but which are +nice to have in other objects. The C language does not allow for the +direct expression of inheritance, so other techniques - such as structure +embedding - must be used. + +So, for example, the UIO code has a structure that defines the memory +region associated with a uio device: + +struct uio_mem { + struct kobject kobj; + unsigned long addr; + unsigned long size; + int memtype; + void __iomem *internal_addr; +}; + +If you have a struct uio_mem structure, finding its embedded kobject is +just a matter of using the kobj member. Code that works with kobjects will +often have the opposite problem, however: given a struct kobject pointer, +what is the pointer to the containing structure? You must avoid tricks +(such as assuming that the kobject is at the beginning of the structure) +and, instead, use the container_of() macro, found in : + + container_of(pointer, type, member) + +where pointer is the pointer to the embedded kobject, type is the type of +the containing structure, and member is the name of the structure field to +which pointer points. The return value from container_of() is a pointer to +the given type. So, for example, a pointer "kp" to a struct kobject +embedded within a struct uio_mem could be converted to a pointer to the +containing uio_mem structure with: + + struct uio_mem *u_mem = container_of(kp, struct uio_mem, kobj); + +Programmers often define a simple macro for "back-casting" kobject pointers +to the containing type. + + +Initialization of kobjects + +Code which creates a kobject must, of course, initialize that object. Some +of the internal fields are setup with a (mandatory) call to kobject_init(): + + void kobject_init(struct kobject *kobj, struct kobj_type *ktype); + +The ktype is required for a kobject to be created properly, as every kobject +must have an associated kobj_type. After calling kobject_init(), to +register the kobject with sysfs, the function kobject_add() must be called: + + int kobject_add(struct kobject *kobj, struct kobject *parent, const char *fmt, ...); + +This sets up the parent of the kobject and the name for the kobject +properly. If the kobject is to be associated with a specific kset, +kobj->kset must be assigned before calling kobject_add(). If a kset is +associated with a kobject, then the parent for the kobject can be set to +NULL in the call to kobject_add() and then the kobject's parent will be the +kset itself. + +As the name of the kobject is set when it is added to the kernel, the name +of the kobject should never be manipulated directly. If you must change +the name of the kobject, call kobject_rename(): + + int kobject_rename(struct kobject *kobj, const char *new_name); + +There is a function called kobject_set_name() but that is legacy cruft and +is being removed. If your code needs to call this function, it is +incorrect and needs to be fixed. + +To properly access the name of the kobject, use the function +kobject_name(): + + const char *kobject_name(const struct kobject * kobj); + +There is a helper function to both initialize and add the kobject to the +kernel at the same time, called supprisingly enough kobject_init_and_add(): + + int kobject_init_and_add(struct kobject *kobj, struct kobj_type *ktype, + struct kobject *parent, const char *fmt, ...); + +The arguments are the same as the individual kobject_init() and +kobject_add() functions described above. + + +Uevents + +After a kobject has been registered with the kobject core, you need to +announce to the world that it has been created. This can be done with a +call to kobject_uevent(): + + int kobject_uevent(struct kobject *kobj, enum kobject_action action); + +Use the KOBJ_ADD action for when the kobject is first added to the kernel. +This should be done only after any attributes or children of the kobject +have been initialized properly, as userspace will instantly start to look +for them when this call happens. + +When the kobject is removed from the kernel (details on how to do that is +below), the uevent for KOBJ_REMOVE will be automatically created by the +kobject core, so the caller does not have to worry about doing that by +hand. + + +Reference counts + +One of the key functions of a kobject is to serve as a reference counter +for the object in which it is embedded. As long as references to the object +exist, the object (and the code which supports it) must continue to exist. +The low-level functions for manipulating a kobject's reference counts are: + + struct kobject *kobject_get(struct kobject *kobj); + void kobject_put(struct kobject *kobj); + +A successful call to kobject_get() will increment the kobject's reference +counter and return the pointer to the kobject. + +When a reference is released, the call to kobject_put() will decrement the +reference count and, possibly, free the object. Note that kobject_init() +sets the reference count to one, so the code which sets up the kobject will +need to do a kobject_put() eventually to release that reference. + +Because kobjects are dynamic, they must not be declared statically or on +the stack, but instead, always allocated dynamically. Future versions of +the kernel will contain a run-time check for kobjects that are created +statically and will warn the developer of this improper usage. + +If all that you want to use a kobject for is to provide a reference counter +for your structure, please use the struct kref instead; a kobject would be +overkill. For more information on how to use struct kref, please see the +file Documentation/kref.txt in the Linux kernel source tree. + + +Creating "simple" kobjects + +Sometimes all that a developer wants is a way to create a simple directory +in the sysfs hierarchy, and not have to mess with the whole complication of +ksets, show and store functions, and other details. This is the one +exception where a single kobject should be created. To create such an +entry, use the function: + + struct kobject *kobject_create_and_add(char *name, struct kobject *parent); + +This function will create a kobject and place it in sysfs in the location +underneath the specified parent kobject. To create simple attributes +associated with this kobject, use: + + int sysfs_create_file(struct kobject *kobj, struct attribute *attr); +or + int sysfs_create_group(struct kobject *kobj, struct attribute_group *grp); + +Both types of attributes used here, with a kobject that has been created +with the kobject_create_and_add(), can be of type kobj_attribute, so no +special custom attribute is needed to be created. + +See the example module, samples/kobject/kobject-example.c for an +implementation of a simple kobject and attributes. + + + +ktypes and release methods + +One important thing still missing from the discussion is what happens to a +kobject when its reference count reaches zero. The code which created the +kobject generally does not know when that will happen; if it did, there +would be little point in using a kobject in the first place. Even +predictable object lifecycles become more complicated when sysfs is brought +in as other portions of the kernel can get a reference on any kobject that +is registered in the system. + +The end result is that a structure protected by a kobject cannot be freed +before its reference count goes to zero. The reference count is not under +the direct control of the code which created the kobject. So that code must +be notified asynchronously whenever the last reference to one of its +kobjects goes away. + +Once you registered your kobject via kobject_add(), you must never use +kfree() to free it directly. The only safe way is to use kobject_put(). It +is good practice to always use kobject_put() after kobject_init() to avoid +errors creeping in. + +This notification is done through a kobject's release() method. Usually +such a method has a form like: + + void my_object_release(struct kobject *kobj) + { + struct my_object *mine = container_of(kobj, struct my_object, kobj); + + /* Perform any additional cleanup on this object, then... */ + kfree(mine); + } + +One important point cannot be overstated: every kobject must have a +release() method, and the kobject must persist (in a consistent state) +until that method is called. If these constraints are not met, the code is +flawed. Note that the kernel will warn you if you forget to provide a +release() method. Do not try to get rid of this warning by providing an +"empty" release function; you will be mocked mercilessly by the kobject +maintainer if you attempt this. + +Note, the name of the kobject is available in the release function, but it +must NOT be changed within this callback. Otherwise there will be a memory +leak in the kobject core, which makes people unhappy. + +Interestingly, the release() method is not stored in the kobject itself; +instead, it is associated with the ktype. So let us introduce struct +kobj_type: + + struct kobj_type { + void (*release)(struct kobject *); + struct sysfs_ops *sysfs_ops; + struct attribute **default_attrs; + }; + +This structure is used to describe a particular type of kobject (or, more +correctly, of containing object). Every kobject needs to have an associated +kobj_type structure; a pointer to that structure must be specified when you +call kobject_init() or kobject_init_and_add(). + +The release field in struct kobj_type is, of course, a pointer to the +release() method for this type of kobject. The other two fields (sysfs_ops +and default_attrs) control how objects of this type are represented in +sysfs; they are beyond the scope of this document. + +The default_attrs pointer is a list of default attributes that will be +automatically created for any kobject that is registered with this ktype. + + +ksets + +A kset is merely a collection of kobjects that want to be associated with +each other. There is no restriction that they be of the same ktype, but be +very careful if they are not. + +A kset serves these functions: + + - It serves as a bag containing a group of objects. A kset can be used by + the kernel to track "all block devices" or "all PCI device drivers." + + - A kset is also a subdirectory in sysfs, where the associated kobjects + with the kset can show up. Every kset contains a kobject which can be + set up to be the parent of other kobjects; the top-level directories of + the sysfs hierarchy are constructed in this way. + + - Ksets can support the "hotplugging" of kobjects and influence how + uevent events are reported to user space. + +In object-oriented terms, "kset" is the top-level container class; ksets +contain their own kobject, but that kobject is managed by the kset code and +should not be manipulated by any other user. + +A kset keeps its children in a standard kernel linked list. Kobjects point +back to their containing kset via their kset field. In almost all cases, +the kobjects belonging to a ket have that kset (or, strictly, its embedded +kobject) in their parent. + +As a kset contains a kobject within it, it should always be dynamically +created and never declared statically or on the stack. To create a new +kset use: + struct kset *kset_create_and_add(const char *name, + struct kset_uevent_ops *u, + struct kobject *parent); + +When you are finished with the kset, call: + void kset_unregister(struct kset *kset); +to destroy it. + +An example of using a kset can be seen in the +samples/kobject/kset-example.c file in the kernel tree. + +If a kset wishes to control the uevent operations of the kobjects +associated with it, it can use the struct kset_uevent_ops to handle it: + +struct kset_uevent_ops { + int (*filter)(struct kset *kset, struct kobject *kobj); + const char *(*name)(struct kset *kset, struct kobject *kobj); + int (*uevent)(struct kset *kset, struct kobject *kobj, + struct kobj_uevent_env *env); +}; + + +The filter function allows a kset to prevent a uevent from being emitted to +userspace for a specific kobject. If the function returns 0, the uevent +will not be emitted. + +The name function will be called to override the default name of the kset +that the uevent sends to userspace. By default, the name will be the same +as the kset itself, but this function, if present, can override that name. + +The uevent function will be called when the uevent is about to be sent to +userspace to allow more environment variables to be added to the uevent. + +One might ask how, exactly, a kobject is added to a kset, given that no +functions which perform that function have been presented. The answer is +that this task is handled by kobject_add(). When a kobject is passed to +kobject_add(), its kset member should point to the kset to which the +kobject will belong. kobject_add() will handle the rest. + +If the kobject belonging to a kset has no parent kobject set, it will be +added to the kset's directory. Not all members of a kset do necessarily +live in the kset directory. If an explicit parent kobject is assigned +before the kobject is added, the kobject is registered with the kset, but +added below the parent kobject. + + +Kobject removal + +After a kobject has been registered with the kobject core successfully, it +must be cleaned up when the code is finished with it. To do that, call +kobject_put(). By doing this, the kobject core will automatically clean up +all of the memory allocated by this kobject. If a KOBJ_ADD uevent has been +sent for the object, a corresponding KOBJ_REMOVE uevent will be sent, and +any other sysfs housekeeping will be handled for the caller properly. + +If you need to do a two-stage delete of the kobject (say you are not +allowed to sleep when you need to destroy the object), then call +kobject_del() which will unregister the kobject from sysfs. This makes the +kobject "invisible", but it is not cleaned up, and the reference count of +the object is still the same. At a later time call kobject_put() to finish +the cleanup of the memory associated with the kobject. + +kobject_del() can be used to drop the reference to the parent object, if +circular references are constructed. It is valid in some cases, that a +parent objects references a child. Circular references _must_ be broken +with an explicit call to kobject_del(), so that a release functions will be +called, and the objects in the former circle release each other. + + +Example code to copy from + +For a more complete example of using ksets and kobjects properly, see the +sample/kobject/kset-example.c code. -- cgit v1.2.3 From e7ba176b47db2ed53f258a6b4fe9d9fc6fa437a9 Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Wed, 10 Oct 2007 14:58:29 +0200 Subject: [AVR32] NMI debugging Change the NMI handler to use the die notifier chain to signal anyone who cares. Add a simple "nmi debugger" which hooks into this chain and that may dump registers, task state, etc. when it happens. Signed-off-by: Haavard Skinnemoen --- Documentation/kernel-parameters.txt | 5 +++ arch/avr32/Kconfig | 10 +++++ arch/avr32/kernel/Makefile | 1 + arch/avr32/kernel/irq.c | 11 +++++ arch/avr32/kernel/nmi_debug.c | 82 +++++++++++++++++++++++++++++++++++++ arch/avr32/kernel/traps.c | 21 ++++++++-- arch/avr32/mach-at32ap/at32ap700x.c | 1 + arch/avr32/mach-at32ap/extint.c | 39 ++++++++++++++---- include/asm-avr32/irq.h | 5 +++ include/asm-avr32/kdebug.h | 1 + 10 files changed, 166 insertions(+), 10 deletions(-) create mode 100644 arch/avr32/kernel/nmi_debug.c (limited to 'Documentation') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index c4178778e7fd..17fc60e32443 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -34,6 +34,7 @@ parameter is applicable: ALSA ALSA sound support is enabled. APIC APIC support is enabled. APM Advanced Power Management support is enabled. + AVR32 AVR32 architecture is enabled. AX25 Appropriate AX.25 support is enabled. BLACKFIN Blackfin architecture is enabled. DRM Direct Rendering Management support is enabled. @@ -1123,6 +1124,10 @@ and is between 256 and 4096 characters. It is defined in the file of returning the full 64-bit number. The default is to return 64-bit inode numbers. + nmi_debug= [KNL,AVR32] Specify one or more actions to take + when a NMI is triggered. + Format: [state][,regs][,debounce][,die] + nmi_watchdog= [KNL,BUGS=X86-32] Debugging features for SMP kernels no387 [BUGS=X86-32] Tells the kernel to use the 387 maths diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig index 516015b3293b..e34e2c9c94cb 100644 --- a/arch/avr32/Kconfig +++ b/arch/avr32/Kconfig @@ -170,6 +170,16 @@ config OWNERSHIP_TRACE enabling Nexus-compliant debuggers to keep track of the PID of the currently executing task. +config NMI_DEBUGGING + bool "NMI Debugging" + default n + help + Say Y here and pass the nmi_debug command-line parameter to + the kernel to turn on NMI debugging. Depending on the value + of the nmi_debug option, various pieces of information will + be dumped to the console when a Non-Maskable Interrupt + happens. + # FPU emulation goes here source "kernel/Kconfig.hz" diff --git a/arch/avr32/kernel/Makefile b/arch/avr32/kernel/Makefile index bc224a4e39fe..e4b6d122b033 100644 --- a/arch/avr32/kernel/Makefile +++ b/arch/avr32/kernel/Makefile @@ -12,3 +12,4 @@ obj-y += init_task.o switch_to.o cpu.o obj-$(CONFIG_MODULES) += module.o avr32_ksyms.o obj-$(CONFIG_KPROBES) += kprobes.o obj-$(CONFIG_STACKTRACE) += stacktrace.o +obj-$(CONFIG_NMI_DEBUGGING) += nmi_debug.o diff --git a/arch/avr32/kernel/irq.c b/arch/avr32/kernel/irq.c index 61f2de266f62..a8e767d836aa 100644 --- a/arch/avr32/kernel/irq.c +++ b/arch/avr32/kernel/irq.c @@ -25,6 +25,17 @@ void ack_bad_irq(unsigned int irq) printk("unexpected IRQ %u\n", irq); } +/* May be overridden by platform code */ +int __weak nmi_enable(void) +{ + return -ENOSYS; +} + +void __weak nmi_disable(void) +{ + +} + #ifdef CONFIG_PROC_FS int show_interrupts(struct seq_file *p, void *v) { diff --git a/arch/avr32/kernel/nmi_debug.c b/arch/avr32/kernel/nmi_debug.c new file mode 100644 index 000000000000..3414b8566c29 --- /dev/null +++ b/arch/avr32/kernel/nmi_debug.c @@ -0,0 +1,82 @@ +/* + * Copyright (C) 2007 Atmel Corporation + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include +#include +#include + +#include + +enum nmi_action { + NMI_SHOW_STATE = 1 << 0, + NMI_SHOW_REGS = 1 << 1, + NMI_DIE = 1 << 2, + NMI_DEBOUNCE = 1 << 3, +}; + +static unsigned long nmi_actions; + +static int nmi_debug_notify(struct notifier_block *self, + unsigned long val, void *data) +{ + struct die_args *args = data; + + if (likely(val != DIE_NMI)) + return NOTIFY_DONE; + + if (nmi_actions & NMI_SHOW_STATE) + show_state(); + if (nmi_actions & NMI_SHOW_REGS) + show_regs(args->regs); + if (nmi_actions & NMI_DEBOUNCE) + mdelay(10); + if (nmi_actions & NMI_DIE) + return NOTIFY_BAD; + + return NOTIFY_OK; +} + +static struct notifier_block nmi_debug_nb = { + .notifier_call = nmi_debug_notify, +}; + +static int __init nmi_debug_setup(char *str) +{ + char *p, *sep; + + register_die_notifier(&nmi_debug_nb); + if (nmi_enable()) { + printk(KERN_WARNING "Unable to enable NMI.\n"); + return 0; + } + + if (*str != '=') + return 0; + + for (p = str + 1; *p; p = sep + 1) { + sep = strchr(p, ','); + if (sep) + *sep = 0; + if (strcmp(p, "state") == 0) + nmi_actions |= NMI_SHOW_STATE; + else if (strcmp(p, "regs") == 0) + nmi_actions |= NMI_SHOW_REGS; + else if (strcmp(p, "debounce") == 0) + nmi_actions |= NMI_DEBOUNCE; + else if (strcmp(p, "die") == 0) + nmi_actions |= NMI_DIE; + else + printk(KERN_WARNING "NMI: Unrecognized action `%s'\n", + p); + if (!sep) + break; + } + + return 0; +} +__setup("nmi_debug", nmi_debug_setup); diff --git a/arch/avr32/kernel/traps.c b/arch/avr32/kernel/traps.c index 870c075e6314..cf6f686d9b0b 100644 --- a/arch/avr32/kernel/traps.c +++ b/arch/avr32/kernel/traps.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include @@ -107,9 +108,23 @@ void _exception(long signr, struct pt_regs *regs, int code, asmlinkage void do_nmi(unsigned long ecr, struct pt_regs *regs) { - printk(KERN_ALERT "Got Non-Maskable Interrupt, dumping regs\n"); - show_regs_log_lvl(regs, KERN_ALERT); - show_stack_log_lvl(current, regs->sp, regs, KERN_ALERT); + int ret; + + nmi_enter(); + + ret = notify_die(DIE_NMI, "NMI", regs, 0, ecr, SIGINT); + switch (ret) { + case NOTIFY_OK: + case NOTIFY_STOP: + return; + case NOTIFY_BAD: + die("Fatal Non-Maskable Interrupt", regs, SIGINT); + default: + break; + } + + printk(KERN_ALERT "Got NMI, but nobody cared. Disabling...\n"); + nmi_disable(); } asmlinkage void do_critical_exception(unsigned long ecr, struct pt_regs *regs) diff --git a/arch/avr32/mach-at32ap/at32ap700x.c b/arch/avr32/mach-at32ap/at32ap700x.c index 9386e1f82fb8..14e61f05e1f6 100644 --- a/arch/avr32/mach-at32ap/at32ap700x.c +++ b/arch/avr32/mach-at32ap/at32ap700x.c @@ -13,6 +13,7 @@ #include #include +#include #include #include diff --git a/arch/avr32/mach-at32ap/extint.c b/arch/avr32/mach-at32ap/extint.c index f5bfd4c81fe7..e108e7bba8c0 100644 --- a/arch/avr32/mach-at32ap/extint.c +++ b/arch/avr32/mach-at32ap/extint.c @@ -26,16 +26,10 @@ #define EIC_MODE 0x0014 #define EIC_EDGE 0x0018 #define EIC_LEVEL 0x001c -#define EIC_TEST 0x0020 #define EIC_NMIC 0x0024 -/* Bitfields in TEST */ -#define EIC_TESTEN_OFFSET 31 -#define EIC_TESTEN_SIZE 1 - /* Bitfields in NMIC */ -#define EIC_EN_OFFSET 0 -#define EIC_EN_SIZE 1 +#define EIC_NMIC_ENABLE (1 << 0) /* Bit manipulation macros */ #define EIC_BIT(name) \ @@ -63,6 +57,9 @@ struct eic { unsigned int first_irq; }; +static struct eic *nmi_eic; +static bool nmi_enabled; + static void eic_ack_irq(unsigned int irq) { struct eic *eic = get_irq_chip_data(irq); @@ -174,6 +171,24 @@ static void demux_eic_irq(unsigned int irq, struct irq_desc *desc) } } +int nmi_enable(void) +{ + nmi_enabled = true; + + if (nmi_eic) + eic_writel(nmi_eic, NMIC, EIC_NMIC_ENABLE); + + return 0; +} + +void nmi_disable(void) +{ + if (nmi_eic) + eic_writel(nmi_eic, NMIC, 0); + + nmi_enabled = false; +} + static int __init eic_probe(struct platform_device *pdev) { struct eic *eic; @@ -230,6 +245,16 @@ static int __init eic_probe(struct platform_device *pdev) set_irq_chained_handler(int_irq, demux_eic_irq); set_irq_data(int_irq, eic); + if (pdev->id == 0) { + nmi_eic = eic; + if (nmi_enabled) + /* + * Someone tried to enable NMI before we were + * ready. Do it now. + */ + nmi_enable(); + } + dev_info(&pdev->dev, "External Interrupt Controller at 0x%p, IRQ %u\n", eic->regs, int_irq); diff --git a/include/asm-avr32/irq.h b/include/asm-avr32/irq.h index 83e6549d7783..9315724c0596 100644 --- a/include/asm-avr32/irq.h +++ b/include/asm-avr32/irq.h @@ -11,4 +11,9 @@ #define irq_canonicalize(i) (i) +#ifndef __ASSEMBLER__ +int nmi_enable(void); +void nmi_disable(void); +#endif + #endif /* __ASM_AVR32_IOCTLS_H */ diff --git a/include/asm-avr32/kdebug.h b/include/asm-avr32/kdebug.h index fd7e99046b2f..ca4f9542365a 100644 --- a/include/asm-avr32/kdebug.h +++ b/include/asm-avr32/kdebug.h @@ -5,6 +5,7 @@ enum die_val { DIE_BREAKPOINT, DIE_SSTEP, + DIE_NMI, }; #endif /* __ASM_AVR32_KDEBUG_H */ -- cgit v1.2.3 From 86ef5c9a8edd78e6bf92879f32329d89b2d55b5a Mon Sep 17 00:00:00 2001 From: Gautham R Shenoy Date: Fri, 25 Jan 2008 21:08:02 +0100 Subject: cpu-hotplug: replace lock_cpu_hotplug() with get_online_cpus() Replace all lock_cpu_hotplug/unlock_cpu_hotplug from the kernel and use get_online_cpus and put_online_cpus instead as it highlights the refcount semantics in these operations. The new API guarantees protection against the cpu-hotplug operation, but it doesn't guarantee serialized access to any of the local data structures. Hence the changes needs to be reviewed. In case of pseries_add_processor/pseries_remove_processor, use cpu_maps_update_begin()/cpu_maps_update_done() as we're modifying the cpu_present_map there. Signed-off-by: Gautham R Shenoy Signed-off-by: Ingo Molnar --- Documentation/cpu-hotplug.txt | 11 ++++++----- arch/mips/kernel/mips-mt-fpaff.c | 10 +++++----- arch/powerpc/platforms/pseries/hotplug-cpu.c | 8 ++++---- arch/powerpc/platforms/pseries/rtasd.c | 8 ++++---- arch/x86/kernel/cpu/mtrr/main.c | 8 ++++---- arch/x86/kernel/microcode.c | 16 ++++++++-------- drivers/lguest/x86/core.c | 8 ++++---- drivers/s390/char/sclp_config.c | 4 ++-- include/linux/cpu.h | 8 ++++---- kernel/cpu.c | 10 +++++----- kernel/cpuset.c | 14 +++++++------- kernel/rcutorture.c | 6 +++--- kernel/sched.c | 4 ++-- kernel/stop_machine.c | 4 ++-- net/core/flow.c | 4 ++-- 15 files changed, 62 insertions(+), 61 deletions(-) (limited to 'Documentation') diff --git a/Documentation/cpu-hotplug.txt b/Documentation/cpu-hotplug.txt index a741f658a3c9..fb94f5a71b68 100644 --- a/Documentation/cpu-hotplug.txt +++ b/Documentation/cpu-hotplug.txt @@ -109,12 +109,13 @@ Never use anything other than cpumask_t to represent bitmap of CPUs. for_each_cpu_mask(x,mask) - Iterate over some random collection of cpu mask. #include - lock_cpu_hotplug() and unlock_cpu_hotplug(): + get_online_cpus() and put_online_cpus(): -The above calls are used to inhibit cpu hotplug operations. While holding the -cpucontrol mutex, cpu_online_map will not change. If you merely need to avoid -cpus going away, you could also use preempt_disable() and preempt_enable() -for those sections. Just remember the critical section cannot call any +The above calls are used to inhibit cpu hotplug operations. While the +cpu_hotplug.refcount is non zero, the cpu_online_map will not change. +If you merely need to avoid cpus going away, you could also use +preempt_disable() and preempt_enable() for those sections. +Just remember the critical section cannot call any function that can sleep or schedule this process away. The preempt_disable() will work as long as stop_machine_run() is used to take a cpu down. diff --git a/arch/mips/kernel/mips-mt-fpaff.c b/arch/mips/kernel/mips-mt-fpaff.c index 892665bb12b1..bb4f00c0cbe9 100644 --- a/arch/mips/kernel/mips-mt-fpaff.c +++ b/arch/mips/kernel/mips-mt-fpaff.c @@ -58,13 +58,13 @@ asmlinkage long mipsmt_sys_sched_setaffinity(pid_t pid, unsigned int len, if (copy_from_user(&new_mask, user_mask_ptr, sizeof(new_mask))) return -EFAULT; - lock_cpu_hotplug(); + get_online_cpus(); read_lock(&tasklist_lock); p = find_process_by_pid(pid); if (!p) { read_unlock(&tasklist_lock); - unlock_cpu_hotplug(); + put_online_cpus(); return -ESRCH; } @@ -106,7 +106,7 @@ asmlinkage long mipsmt_sys_sched_setaffinity(pid_t pid, unsigned int len, out_unlock: put_task_struct(p); - unlock_cpu_hotplug(); + put_online_cpus(); return retval; } @@ -125,7 +125,7 @@ asmlinkage long mipsmt_sys_sched_getaffinity(pid_t pid, unsigned int len, if (len < real_len) return -EINVAL; - lock_cpu_hotplug(); + get_online_cpus(); read_lock(&tasklist_lock); retval = -ESRCH; @@ -140,7 +140,7 @@ asmlinkage long mipsmt_sys_sched_getaffinity(pid_t pid, unsigned int len, out_unlock: read_unlock(&tasklist_lock); - unlock_cpu_hotplug(); + put_online_cpus(); if (retval) return retval; if (copy_to_user(user_mask_ptr, &mask, real_len)) diff --git a/arch/powerpc/platforms/pseries/hotplug-cpu.c b/arch/powerpc/platforms/pseries/hotplug-cpu.c index 412e6b42986f..c4ad54e0f288 100644 --- a/arch/powerpc/platforms/pseries/hotplug-cpu.c +++ b/arch/powerpc/platforms/pseries/hotplug-cpu.c @@ -153,7 +153,7 @@ static int pseries_add_processor(struct device_node *np) for (i = 0; i < nthreads; i++) cpu_set(i, tmp); - lock_cpu_hotplug(); + cpu_maps_update_begin(); BUG_ON(!cpus_subset(cpu_present_map, cpu_possible_map)); @@ -190,7 +190,7 @@ static int pseries_add_processor(struct device_node *np) } err = 0; out_unlock: - unlock_cpu_hotplug(); + cpu_maps_update_done(); return err; } @@ -211,7 +211,7 @@ static void pseries_remove_processor(struct device_node *np) nthreads = len / sizeof(u32); - lock_cpu_hotplug(); + cpu_maps_update_begin(); for (i = 0; i < nthreads; i++) { for_each_present_cpu(cpu) { if (get_hard_smp_processor_id(cpu) != intserv[i]) @@ -225,7 +225,7 @@ static void pseries_remove_processor(struct device_node *np) printk(KERN_WARNING "Could not find cpu to remove " "with physical id 0x%x\n", intserv[i]); } - unlock_cpu_hotplug(); + cpu_maps_update_done(); } static int pseries_smp_notifier(struct notifier_block *nb, diff --git a/arch/powerpc/platforms/pseries/rtasd.c b/arch/powerpc/platforms/pseries/rtasd.c index 73401c820110..e3078ce41518 100644 --- a/arch/powerpc/platforms/pseries/rtasd.c +++ b/arch/powerpc/platforms/pseries/rtasd.c @@ -382,7 +382,7 @@ static void do_event_scan_all_cpus(long delay) { int cpu; - lock_cpu_hotplug(); + get_online_cpus(); cpu = first_cpu(cpu_online_map); for (;;) { set_cpus_allowed(current, cpumask_of_cpu(cpu)); @@ -390,15 +390,15 @@ static void do_event_scan_all_cpus(long delay) set_cpus_allowed(current, CPU_MASK_ALL); /* Drop hotplug lock, and sleep for the specified delay */ - unlock_cpu_hotplug(); + put_online_cpus(); msleep_interruptible(delay); - lock_cpu_hotplug(); + get_online_cpus(); cpu = next_cpu(cpu, cpu_online_map); if (cpu == NR_CPUS) break; } - unlock_cpu_hotplug(); + put_online_cpus(); } static int rtasd(void *unused) diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c index 3b20613325dc..beb45c9c0835 100644 --- a/arch/x86/kernel/cpu/mtrr/main.c +++ b/arch/x86/kernel/cpu/mtrr/main.c @@ -349,7 +349,7 @@ int mtrr_add_page(unsigned long base, unsigned long size, replace = -1; /* No CPU hotplug when we change MTRR entries */ - lock_cpu_hotplug(); + get_online_cpus(); /* Search for existing MTRR */ mutex_lock(&mtrr_mutex); for (i = 0; i < num_var_ranges; ++i) { @@ -405,7 +405,7 @@ int mtrr_add_page(unsigned long base, unsigned long size, error = i; out: mutex_unlock(&mtrr_mutex); - unlock_cpu_hotplug(); + put_online_cpus(); return error; } @@ -495,7 +495,7 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size) max = num_var_ranges; /* No CPU hotplug when we change MTRR entries */ - lock_cpu_hotplug(); + get_online_cpus(); mutex_lock(&mtrr_mutex); if (reg < 0) { /* Search for existing MTRR */ @@ -536,7 +536,7 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size) error = reg; out: mutex_unlock(&mtrr_mutex); - unlock_cpu_hotplug(); + put_online_cpus(); return error; } /** diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c index 09c315214a5e..40cfd5488719 100644 --- a/arch/x86/kernel/microcode.c +++ b/arch/x86/kernel/microcode.c @@ -436,7 +436,7 @@ static ssize_t microcode_write (struct file *file, const char __user *buf, size_ return -EINVAL; } - lock_cpu_hotplug(); + get_online_cpus(); mutex_lock(µcode_mutex); user_buffer = (void __user *) buf; @@ -447,7 +447,7 @@ static ssize_t microcode_write (struct file *file, const char __user *buf, size_ ret = (ssize_t)len; mutex_unlock(µcode_mutex); - unlock_cpu_hotplug(); + put_online_cpus(); return ret; } @@ -658,14 +658,14 @@ static ssize_t reload_store(struct sys_device *dev, const char *buf, size_t sz) old = current->cpus_allowed; - lock_cpu_hotplug(); + get_online_cpus(); set_cpus_allowed(current, cpumask_of_cpu(cpu)); mutex_lock(µcode_mutex); if (uci->valid) err = cpu_request_microcode(cpu); mutex_unlock(µcode_mutex); - unlock_cpu_hotplug(); + put_online_cpus(); set_cpus_allowed(current, old); } if (err) @@ -817,9 +817,9 @@ static int __init microcode_init (void) return PTR_ERR(microcode_pdev); } - lock_cpu_hotplug(); + get_online_cpus(); error = sysdev_driver_register(&cpu_sysdev_class, &mc_sysdev_driver); - unlock_cpu_hotplug(); + put_online_cpus(); if (error) { microcode_dev_exit(); platform_device_unregister(microcode_pdev); @@ -839,9 +839,9 @@ static void __exit microcode_exit (void) unregister_hotcpu_notifier(&mc_cpu_notifier); - lock_cpu_hotplug(); + get_online_cpus(); sysdev_driver_unregister(&cpu_sysdev_class, &mc_sysdev_driver); - unlock_cpu_hotplug(); + put_online_cpus(); platform_device_unregister(microcode_pdev); } diff --git a/drivers/lguest/x86/core.c b/drivers/lguest/x86/core.c index 482aec2a9631..96d0fd07c57d 100644 --- a/drivers/lguest/x86/core.c +++ b/drivers/lguest/x86/core.c @@ -459,7 +459,7 @@ void __init lguest_arch_host_init(void) /* We don't need the complexity of CPUs coming and going while we're * doing this. */ - lock_cpu_hotplug(); + get_online_cpus(); if (cpu_has_pge) { /* We have a broader idea of "global". */ /* Remember that this was originally set (for cleanup). */ cpu_had_pge = 1; @@ -469,20 +469,20 @@ void __init lguest_arch_host_init(void) /* Turn off the feature in the global feature set. */ clear_bit(X86_FEATURE_PGE, boot_cpu_data.x86_capability); } - unlock_cpu_hotplug(); + put_online_cpus(); }; /*:*/ void __exit lguest_arch_host_fini(void) { /* If we had PGE before we started, turn it back on now. */ - lock_cpu_hotplug(); + get_online_cpus(); if (cpu_had_pge) { set_bit(X86_FEATURE_PGE, boot_cpu_data.x86_capability); /* adjust_pge's argument "1" means set PGE. */ on_each_cpu(adjust_pge, (void *)1, 0, 1); } - unlock_cpu_hotplug(); + put_online_cpus(); } diff --git a/drivers/s390/char/sclp_config.c b/drivers/s390/char/sclp_config.c index 5322e5e54a98..9dc77f14fa52 100644 --- a/drivers/s390/char/sclp_config.c +++ b/drivers/s390/char/sclp_config.c @@ -29,12 +29,12 @@ static void sclp_cpu_capability_notify(struct work_struct *work) struct sys_device *sysdev; printk(KERN_WARNING TAG "cpu capability changed.\n"); - lock_cpu_hotplug(); + get_online_cpus(); for_each_online_cpu(cpu) { sysdev = get_cpu_sysdev(cpu); kobject_uevent(&sysdev->kobj, KOBJ_CHANGE); } - unlock_cpu_hotplug(); + put_online_cpus(); } static void sclp_conf_receiver_fn(struct evbuf_header *evbuf) diff --git a/include/linux/cpu.h b/include/linux/cpu.h index a40247e4d462..3a3ff1c5cbef 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -100,8 +100,8 @@ static inline void cpuhotplug_mutex_unlock(struct mutex *cpu_hp_mutex) mutex_unlock(cpu_hp_mutex); } -extern void lock_cpu_hotplug(void); -extern void unlock_cpu_hotplug(void); +extern void get_online_cpus(void); +extern void put_online_cpus(void); #define hotcpu_notifier(fn, pri) { \ static struct notifier_block fn##_nb = \ { .notifier_call = fn, .priority = pri }; \ @@ -118,8 +118,8 @@ static inline void cpuhotplug_mutex_lock(struct mutex *cpu_hp_mutex) static inline void cpuhotplug_mutex_unlock(struct mutex *cpu_hp_mutex) { } -#define lock_cpu_hotplug() do { } while (0) -#define unlock_cpu_hotplug() do { } while (0) +#define get_online_cpus() do { } while (0) +#define put_online_cpus() do { } while (0) #define hotcpu_notifier(fn, pri) do { (void)(fn); } while (0) /* These aren't inline functions due to a GCC bug. */ #define register_hotcpu_notifier(nb) ({ (void)(nb); 0; }) diff --git a/kernel/cpu.c b/kernel/cpu.c index 656dc3fcbbae..b0c4152995f8 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -48,7 +48,7 @@ void __init cpu_hotplug_init(void) #ifdef CONFIG_HOTPLUG_CPU -void lock_cpu_hotplug(void) +void get_online_cpus(void) { might_sleep(); if (cpu_hotplug.active_writer == current) @@ -58,9 +58,9 @@ void lock_cpu_hotplug(void) mutex_unlock(&cpu_hotplug.lock); } -EXPORT_SYMBOL_GPL(lock_cpu_hotplug); +EXPORT_SYMBOL_GPL(get_online_cpus); -void unlock_cpu_hotplug(void) +void put_online_cpus(void) { if (cpu_hotplug.active_writer == current) return; @@ -73,7 +73,7 @@ void unlock_cpu_hotplug(void) mutex_unlock(&cpu_hotplug.lock); } -EXPORT_SYMBOL_GPL(unlock_cpu_hotplug); +EXPORT_SYMBOL_GPL(put_online_cpus); #endif /* CONFIG_HOTPLUG_CPU */ @@ -110,7 +110,7 @@ void cpu_maps_update_done(void) * non zero and goes to sleep again. * * However, this is very difficult to achieve in practice since - * lock_cpu_hotplug() not an api which is called all that often. + * get_online_cpus() not an api which is called all that often. * */ static void cpu_hotplug_begin(void) diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 50f5dc463688..cfaf6419d817 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -537,10 +537,10 @@ static int cpusets_overlap(struct cpuset *a, struct cpuset *b) * * Call with cgroup_mutex held. May take callback_mutex during * call due to the kfifo_alloc() and kmalloc() calls. May nest - * a call to the lock_cpu_hotplug()/unlock_cpu_hotplug() pair. + * a call to the get_online_cpus()/put_online_cpus() pair. * Must not be called holding callback_mutex, because we must not - * call lock_cpu_hotplug() while holding callback_mutex. Elsewhere - * the kernel nests callback_mutex inside lock_cpu_hotplug() calls. + * call get_online_cpus() while holding callback_mutex. Elsewhere + * the kernel nests callback_mutex inside get_online_cpus() calls. * So the reverse nesting would risk an ABBA deadlock. * * The three key local variables below are: @@ -691,9 +691,9 @@ restart: rebuild: /* Have scheduler rebuild sched domains */ - lock_cpu_hotplug(); + get_online_cpus(); partition_sched_domains(ndoms, doms); - unlock_cpu_hotplug(); + put_online_cpus(); done: if (q && !IS_ERR(q)) @@ -1617,10 +1617,10 @@ static struct cgroup_subsys_state *cpuset_create( * * If the cpuset being removed has its flag 'sched_load_balance' * enabled, then simulate turning sched_load_balance off, which - * will call rebuild_sched_domains(). The lock_cpu_hotplug() + * will call rebuild_sched_domains(). The get_online_cpus() * call in rebuild_sched_domains() must not be made while holding * callback_mutex. Elsewhere the kernel nests callback_mutex inside - * lock_cpu_hotplug() calls. So the reverse nesting would risk an + * get_online_cpus() calls. So the reverse nesting would risk an * ABBA deadlock. */ diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index c3e165c2318f..fd599829e72a 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -726,11 +726,11 @@ static void rcu_torture_shuffle_tasks(void) cpumask_t tmp_mask = CPU_MASK_ALL; int i; - lock_cpu_hotplug(); + get_online_cpus(); /* No point in shuffling if there is only one online CPU (ex: UP) */ if (num_online_cpus() == 1) { - unlock_cpu_hotplug(); + put_online_cpus(); return; } @@ -762,7 +762,7 @@ static void rcu_torture_shuffle_tasks(void) else rcu_idle_cpu--; - unlock_cpu_hotplug(); + put_online_cpus(); } /* Shuffle tasks across CPUs, with the intent of allowing each CPU in the diff --git a/kernel/sched.c b/kernel/sched.c index 86e55a9c2de6..672aa68bfeac 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -7152,7 +7152,7 @@ static int load_balance_monitor(void *unused) int i, cpu, balanced = 1; /* Prevent cpus going down or coming up */ - lock_cpu_hotplug(); + get_online_cpus(); /* lockout changes to doms_cur[] array */ lock_doms_cur(); /* @@ -7186,7 +7186,7 @@ static int load_balance_monitor(void *unused) rcu_read_unlock(); unlock_doms_cur(); - unlock_cpu_hotplug(); + put_online_cpus(); if (!balanced) timeout = sysctl_sched_min_bal_int_shares; diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index 319821ef78af..51b5ee53571a 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -203,13 +203,13 @@ int stop_machine_run(int (*fn)(void *), void *data, unsigned int cpu) int ret; /* No CPUs can come up or down during this. */ - lock_cpu_hotplug(); + get_online_cpus(); p = __stop_machine_run(fn, data, cpu); if (!IS_ERR(p)) ret = kthread_stop(p); else ret = PTR_ERR(p); - unlock_cpu_hotplug(); + put_online_cpus(); return ret; } diff --git a/net/core/flow.c b/net/core/flow.c index 3ed2b4b1d6d4..6489f4e24ecf 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -293,7 +293,7 @@ void flow_cache_flush(void) static DEFINE_MUTEX(flow_flush_sem); /* Don't want cpus going down or up during this. */ - lock_cpu_hotplug(); + get_online_cpus(); mutex_lock(&flow_flush_sem); atomic_set(&info.cpuleft, num_online_cpus()); init_completion(&info.completion); @@ -305,7 +305,7 @@ void flow_cache_flush(void) wait_for_completion(&info.completion); mutex_unlock(&flow_flush_sem); - unlock_cpu_hotplug(); + put_online_cpus(); } static void __devinit flow_cache_cpu_prepare(int cpu) -- cgit v1.2.3 From f85d6c7168887e6659f4d00fa5f34fa23dbde1bb Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 25 Jan 2008 21:08:25 +0100 Subject: Preempt-RCU: update RCU Documentation. This patch updates the RCU documentation to reflect preemptible RCU as well as recent publications. Signed-off-by: Paul E. McKenney Signed-off-by: Gautham R Shenoy Reviewed-by: Steven Rostedt Signed-off-by: Ingo Molnar --- Documentation/RCU/RTFP.txt | 210 +++++++++++++++++++++++++++++++++++++++--- Documentation/RCU/rcu.txt | 19 +++- Documentation/RCU/torture.txt | 11 +-- 3 files changed, 221 insertions(+), 19 deletions(-) (limited to 'Documentation') diff --git a/Documentation/RCU/RTFP.txt b/Documentation/RCU/RTFP.txt index 6221464d1a7e..39ad8f56783a 100644 --- a/Documentation/RCU/RTFP.txt +++ b/Documentation/RCU/RTFP.txt @@ -9,8 +9,8 @@ The first thing resembling RCU was published in 1980, when Kung and Lehman [Kung80] recommended use of a garbage collector to defer destruction of nodes in a parallel binary search tree in order to simplify its implementation. This works well in environments that have garbage -collectors, but current production garbage collectors incur significant -read-side overhead. +collectors, but most production garbage collectors incur significant +overhead. In 1982, Manber and Ladner [Manber82,Manber84] recommended deferring destruction until all threads running at that time have terminated, again @@ -99,16 +99,25 @@ locking, reduces contention, reduces memory latency for readers, and parallelizes pipeline stalls and memory latency for writers. However, these techniques still impose significant read-side overhead in the form of memory barriers. Researchers at Sun worked along similar lines -in the same timeframe [HerlihyLM02,HerlihyLMS03]. These techniques -can be thought of as inside-out reference counts, where the count is -represented by the number of hazard pointers referencing a given data -structure (rather than the more conventional counter field within the -data structure itself). +in the same timeframe [HerlihyLM02]. These techniques can be thought +of as inside-out reference counts, where the count is represented by the +number of hazard pointers referencing a given data structure (rather than +the more conventional counter field within the data structure itself). + +By the same token, RCU can be thought of as a "bulk reference count", +where some form of reference counter covers all reference by a given CPU +or thread during a set timeframe. This timeframe is related to, but +not necessarily exactly the same as, an RCU grace period. In classic +RCU, the reference counter is the per-CPU bit in the "bitmask" field, +and each such bit covers all references that might have been made by +the corresponding CPU during the prior grace period. Of course, RCU +can be thought of in other terms as well. In 2003, the K42 group described how RCU could be used to create -hot-pluggable implementations of operating-system functions. Later that -year saw a paper describing an RCU implementation of System V IPC -[Arcangeli03], and an introduction to RCU in Linux Journal [McKenney03a]. +hot-pluggable implementations of operating-system functions [Appavoo03a]. +Later that year saw a paper describing an RCU implementation of System +V IPC [Arcangeli03], and an introduction to RCU in Linux Journal +[McKenney03a]. 2004 has seen a Linux-Journal article on use of RCU in dcache [McKenney04a], a performance comparison of locking to RCU on several @@ -117,10 +126,19 @@ number of operating-system kernels [PaulEdwardMcKenneyPhD], a paper describing how to make RCU safe for soft-realtime applications [Sarma04c], and a paper describing SELinux performance with RCU [JamesMorris04b]. -2005 has seen further adaptation of RCU to realtime use, permitting +2005 brought further adaptation of RCU to realtime use, permitting preemption of RCU realtime critical sections [PaulMcKenney05a, PaulMcKenney05b]. +2006 saw the first best-paper award for an RCU paper [ThomasEHart2006a], +as well as further work on efficient implementations of preemptible +RCU [PaulEMcKenney2006b], but priority-boosting of RCU read-side critical +sections proved elusive. An RCU implementation permitting general +blocking in read-side critical sections appeared [PaulEMcKenney2006c], +Robert Olsson described an RCU-protected trie-hash combination +[RobertOlsson2006a]. + + Bibtex Entries @article{Kung80 @@ -203,6 +221,41 @@ Bibtex Entries ,Address="New Orleans, LA" } +@conference{Pu95a, +Author = "Calton Pu and Tito Autrey and Andrew Black and Charles Consel and +Crispin Cowan and Jon Inouye and Lakshmi Kethana and Jonathan Walpole and +Ke Zhang", +Title = "Optimistic Incremental Specialization: Streamlining a Commercial +Operating System", +Booktitle = "15\textsuperscript{th} ACM Symposium on +Operating Systems Principles (SOSP'95)", +address = "Copper Mountain, CO", +month="December", +year="1995", +pages="314-321", +annotation=" + Uses a replugger, but with a flag to signal when people are + using the resource at hand. Only one reader at a time. +" +} + +@conference{Cowan96a, +Author = "Crispin Cowan and Tito Autrey and Charles Krasic and +Calton Pu and Jonathan Walpole", +Title = "Fast Concurrent Dynamic Linking for an Adaptive Operating System", +Booktitle = "International Conference on Configurable Distributed Systems +(ICCDS'96)", +address = "Annapolis, MD", +month="May", +year="1996", +pages="108", +isbn="0-8186-7395-8", +annotation=" + Uses a replugger, but with a counter to signal when people are + using the resource at hand. Allows multiple readers. +" +} + @techreport{Slingwine95 ,author="John D. Slingwine and Paul E. McKenney" ,title="Apparatus and Method for Achieving Reduced Overhead Mutual @@ -312,6 +365,49 @@ Andrea Arcangeli and Andi Kleen and Orran Krieger and Rusty Russell" [Viewed June 23, 2004]" } +@conference{Michael02a +,author="Maged M. Michael" +,title="Safe Memory Reclamation for Dynamic Lock-Free Objects Using Atomic +Reads and Writes" +,Year="2002" +,Month="August" +,booktitle="{Proceedings of the 21\textsuperscript{st} Annual ACM +Symposium on Principles of Distributed Computing}" +,pages="21-30" +,annotation=" + Each thread keeps an array of pointers to items that it is + currently referencing. Sort of an inside-out garbage collection + mechanism, but one that requires the accessing code to explicitly + state its needs. Also requires read-side memory barriers on + most architectures. +" +} + +@conference{Michael02b +,author="Maged M. Michael" +,title="High Performance Dynamic Lock-Free Hash Tables and List-Based Sets" +,Year="2002" +,Month="August" +,booktitle="{Proceedings of the 14\textsuperscript{th} Annual ACM +Symposium on Parallel +Algorithms and Architecture}" +,pages="73-82" +,annotation=" + Like the title says... +" +} + +@InProceedings{HerlihyLM02 +,author={Maurice Herlihy and Victor Luchangco and Mark Moir} +,title="The Repeat Offender Problem: A Mechanism for Supporting Dynamic-Sized, +Lock-Free Data Structures" +,booktitle={Proceedings of 16\textsuperscript{th} International +Symposium on Distributed Computing} +,year=2002 +,month="October" +,pages="339-353" +} + @article{Appavoo03a ,author="J. Appavoo and K. Hui and C. A. N. Soules and R. W. Wisniewski and D. M. {Da Silva} and O. Krieger and M. A. Auslander and D. J. Edelsohn and @@ -447,3 +543,95 @@ Oregon Health and Sciences University" Realtime turns into making RCU yet more realtime friendly. " } + +@conference{ThomasEHart2006a +,Author="Thomas E. Hart and Paul E. McKenney and Angela Demke Brown" +,Title="Making Lockless Synchronization Fast: Performance Implications +of Memory Reclamation" +,Booktitle="20\textsuperscript{th} {IEEE} International Parallel and +Distributed Processing Symposium" +,month="April" +,year="2006" +,day="25-29" +,address="Rhodes, Greece" +,annotation=" + Compares QSBR (AKA "classic RCU"), HPBR, EBR, and lock-free + reference counting. +" +} + +@Conference{PaulEMcKenney2006b +,Author="Paul E. McKenney and Dipankar Sarma and Ingo Molnar and +Suparna Bhattacharya" +,Title="Extending RCU for Realtime and Embedded Workloads" +,Booktitle="{Ottawa Linux Symposium}" +,Month="July" +,Year="2006" +,pages="v2 123-138" +,note="Available: +\url{http://www.linuxsymposium.org/2006/view_abstract.php?content_key=184} +\url{http://www.rdrop.com/users/paulmck/RCU/OLSrtRCU.2006.08.11a.pdf} +[Viewed January 1, 2007]" +,annotation=" + Described how to improve the -rt implementation of realtime RCU. +" +} + +@unpublished{PaulEMcKenney2006c +,Author="Paul E. McKenney" +,Title="Sleepable {RCU}" +,month="October" +,day="9" +,year="2006" +,note="Available: +\url{http://lwn.net/Articles/202847/} +Revised: +\url{http://www.rdrop.com/users/paulmck/RCU/srcu.2007.01.14a.pdf} +[Viewed August 21, 2006]" +,annotation=" + LWN article introducing SRCU. +" +} + +@unpublished{RobertOlsson2006a +,Author="Robert Olsson and Stefan Nilsson" +,Title="{TRASH}: A dynamic {LC}-trie and hash data structure" +,month="August" +,day="18" +,year="2006" +,note="Available: +\url{http://www.nada.kth.se/~snilsson/public/papers/trash/trash.pdf} +[Viewed February 24, 2007]" +,annotation=" + RCU-protected dynamic trie-hash combination. +" +} + +@unpublished{ThomasEHart2007a +,Author="Thomas E. Hart and Paul E. McKenney and Angela Demke Brown and Jonathan Walpole" +,Title="Performance of memory reclamation for lockless synchronization" +,journal="J. Parallel Distrib. Comput." +,year="2007" +,note="To appear in J. Parallel Distrib. Comput. + \url{doi=10.1016/j.jpdc.2007.04.010}" +,annotation={ + Compares QSBR (AKA "classic RCU"), HPBR, EBR, and lock-free + reference counting. Journal version of ThomasEHart2006a. +} +} + +@unpublished{PaulEMcKenney2007QRCUspin +,Author="Paul E. McKenney" +,Title="Using Promela and Spin to verify parallel algorithms" +,month="August" +,day="1" +,year="2007" +,note="Available: +\url{http://lwn.net/Articles/243851/} +[Viewed September 8, 2007]" +,annotation=" + LWN article describing Promela and spin, and also using Oleg + Nesterov's QRCU as an example (with Paul McKenney's fastpath). +" +} + diff --git a/Documentation/RCU/rcu.txt b/Documentation/RCU/rcu.txt index f84407cba816..95821a29ae41 100644 --- a/Documentation/RCU/rcu.txt +++ b/Documentation/RCU/rcu.txt @@ -36,6 +36,14 @@ o How can the updater tell when a grace period has completed executed in user mode, or executed in the idle loop, we can safely free up that item. + Preemptible variants of RCU (CONFIG_PREEMPT_RCU) get the + same effect, but require that the readers manipulate CPU-local + counters. These counters allow limited types of blocking + within RCU read-side critical sections. SRCU also uses + CPU-local counters, and permits general blocking within + RCU read-side critical sections. These two variants of + RCU detect grace periods by sampling these counters. + o If I am running on a uniprocessor kernel, which can only do one thing at a time, why should I wait for a grace period? @@ -46,7 +54,10 @@ o How can I see where RCU is currently used in the Linux kernel? Search for "rcu_read_lock", "rcu_read_unlock", "call_rcu", "rcu_read_lock_bh", "rcu_read_unlock_bh", "call_rcu_bh", "srcu_read_lock", "srcu_read_unlock", "synchronize_rcu", - "synchronize_net", and "synchronize_srcu". + "synchronize_net", "synchronize_srcu", and the other RCU + primitives. Or grab one of the cscope databases from: + + http://www.rdrop.com/users/paulmck/RCU/linuxusage/rculocktab.html o What guidelines should I follow when writing code that uses RCU? @@ -67,7 +78,11 @@ o I hear that RCU is patented? What is with that? o I hear that RCU needs work in order to support realtime kernels? - Yes, work in progress. + This work is largely completed. Realtime-friendly RCU can be + enabled via the CONFIG_PREEMPT_RCU kernel configuration parameter. + However, work is in progress for enabling priority boosting of + preempted RCU read-side critical sections.This is needed if you + have CPU-bound realtime threads. o Where can I find more information on RCU? diff --git a/Documentation/RCU/torture.txt b/Documentation/RCU/torture.txt index 25a3c3f7d378..2967a65269d8 100644 --- a/Documentation/RCU/torture.txt +++ b/Documentation/RCU/torture.txt @@ -46,12 +46,13 @@ stat_interval The number of seconds between output of torture shuffle_interval The number of seconds to keep the test threads affinitied - to a particular subset of the CPUs. Used in conjunction - with test_no_idle_hz. + to a particular subset of the CPUs, defaults to 5 seconds. + Used in conjunction with test_no_idle_hz. test_no_idle_hz Whether or not to test the ability of RCU to operate in a kernel that disables the scheduling-clock interrupt to idle CPUs. Boolean parameter, "1" to test, "0" otherwise. + Defaults to omitting this test. torture_type The type of RCU to test: "rcu" for the rcu_read_lock() API, "rcu_sync" for rcu_read_lock() with synchronous reclamation, @@ -82,8 +83,6 @@ be evident. ;-) The entries are as follows: -o "ggp": The number of counter flips (or batches) since boot. - o "rtc": The hexadecimal address of the structure currently visible to readers. @@ -117,8 +116,8 @@ o "Reader Pipe": Histogram of "ages" of structures seen by readers. o "Reader Batch": Another histogram of "ages" of structures seen by readers, but in terms of counter flips (or batches) rather than in terms of grace periods. The legal number of non-zero - entries is again two. The reason for this separate view is - that it is easier to get the third entry to show up in the + entries is again two. The reason for this separate view is that + it is sometimes easier to get the third entry to show up in the "Reader Batch" list than in the "Reader Pipe" list. o "Free-Block Circulation": Shows the number of torture structures -- cgit v1.2.3 From 12466577853b0d057f4416f4c7020e544f3a4209 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Wed, 24 Oct 2007 11:08:20 -0300 Subject: V4L/DVB (6422): Add the pending entries for xc2028/3028 based boards Also replaces all occurrences of TUNER_XCEIVE_XC3028 to TUNER_XC2028. Some work is still may be required to make sure that non-tm6000 drivers will be capable of using tuner-xc2028. Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.tuner | 2 +- drivers/media/video/em28xx/em28xx-cards.c | 71 +++++++++++++++++++++++++++++++ drivers/media/video/ivtv/ivtv-cards.c | 6 +-- drivers/media/video/ivtv/ivtv-driver.c | 2 +- drivers/media/video/tuner-types.c | 6 +-- include/media/tuner.h | 2 +- 6 files changed, 80 insertions(+), 9 deletions(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.tuner b/Documentation/video4linux/CARDLIST.tuner index a88c02d23805..84c3ac7c33d9 100644 --- a/Documentation/video4linux/CARDLIST.tuner +++ b/Documentation/video4linux/CARDLIST.tuner @@ -69,7 +69,7 @@ tuner=67 - Philips TD1316 Hybrid Tuner tuner=68 - Philips TUV1236D ATSC/NTSC dual in tuner=69 - Tena TNF 5335 and similar models tuner=70 - Samsung TCPN 2121P30A -tuner=71 - Xceive xc3028 +tuner=71 - Xceive xc2028/xc3028 tuner tuner=72 - Thomson FE6600 tuner=73 - Samsung TCPG 6121P30A tuner=75 - Philips TEA5761 FM Radio diff --git a/drivers/media/video/em28xx/em28xx-cards.c b/drivers/media/video/em28xx/em28xx-cards.c index 418ea8b7f85a..2d72de0e7564 100644 --- a/drivers/media/video/em28xx/em28xx-cards.c +++ b/drivers/media/video/em28xx/em28xx-cards.c @@ -156,6 +156,74 @@ struct em28xx_board em28xx_boards[] = { MSP_DSP_IN_SCART, MSP_DSP_IN_SCART), }}, }, + [EM2880_BOARD_HAUPPAUGE_WINTV_HVR_900] = { + .name = "Hauppauge WinTV HVR 900", + .vchannels = 3, + .norm = VIDEO_MODE_PAL, + .tda9887_conf = TDA9887_PRESENT, + .tuner_type = TUNER_XC2028, + .has_tuner = 1, + .decoder = EM28XX_TVP5150, + .input = {{ + .type = EM28XX_VMUX_COMPOSITE1, + .vmux = TVP5150_COMPOSITE1, + .amux = 1, + },{ + .type = EM28XX_VMUX_TELEVISION, + .vmux = TVP5150_COMPOSITE0, + .amux = 0, + },{ + .type = EM28XX_VMUX_SVIDEO, + .vmux = TVP5150_SVIDEO, + .amux = 1, + }}, + }, + [EM2880_BOARD_TERRATEC_HYBRID_XS] = { + .name = "Terratec Hybrid XS", + .vchannels = 3, + .norm = VIDEO_MODE_PAL, + .tda9887_conf = TDA9887_PRESENT, + .has_tuner = 1, + .tuner_type = TUNER_XC2028, + .decoder = EM28XX_TVP5150, + .input = {{ + .type = EM28XX_VMUX_TELEVISION, + .vmux = TVP5150_COMPOSITE0, + .amux = 0, + },{ + .type = EM28XX_VMUX_COMPOSITE1, + .vmux = TVP5150_COMPOSITE1, + .amux = 1, + },{ + .type = EM28XX_VMUX_SVIDEO, + .vmux = TVP5150_SVIDEO, + .amux = 1, + }}, + }, + /* maybe there's a reason behind it why Terratec sells the Hybrid XS as Prodigy XS with a + * different PID, let's keep it separated for now maybe we'll need it lateron */ + [EM2880_BOARD_TERRATEC_PRODIGY_XS] = { + .name = "Terratec Prodigy XS", + .vchannels = 3, + .norm = VIDEO_MODE_PAL, + .tda9887_conf = TDA9887_PRESENT, + .has_tuner = 1, + .tuner_type = TUNER_XC2028, + .decoder = EM28XX_TVP5150, + .input = {{ + .type = EM28XX_VMUX_TELEVISION, + .vmux = TVP5150_COMPOSITE0, + .amux = 0, + },{ + .type = EM28XX_VMUX_COMPOSITE1, + .vmux = TVP5150_COMPOSITE1, + .amux = 1, + },{ + .type = EM28XX_VMUX_SVIDEO, + .vmux = TVP5150_SVIDEO, + .amux = 1, + }}, + }, [EM2820_BOARD_MSI_VOX_USB_2] = { .name = "MSI VOX USB 2.0", .vchannels = 3, @@ -274,6 +342,9 @@ struct usb_device_id em28xx_id_table [] = { { USB_DEVICE(0x2304, 0x0208), .driver_info = EM2820_BOARD_PINNACLE_USB_2 }, { USB_DEVICE(0x2040, 0x4200), .driver_info = EM2820_BOARD_HAUPPAUGE_WINTV_USB_2 }, { USB_DEVICE(0x2304, 0x0207), .driver_info = EM2820_BOARD_PINNACLE_DVC_90 }, + { USB_DEVICE(0x2040, 0x6500), .driver_info = EM2880_BOARD_HAUPPAUGE_WINTV_HVR_900 }, + { USB_DEVICE(0x0ccd, 0x0042), .driver_info = EM2880_BOARD_TERRATEC_HYBRID_XS }, + { USB_DEVICE(0x0ccd, 0x0047), .driver_info = EM2880_BOARD_TERRATEC_PRODIGY_XS }, { }, }; diff --git a/drivers/media/video/ivtv/ivtv-cards.c b/drivers/media/video/ivtv/ivtv-cards.c index b6a8be622d3c..aaa114b5c268 100644 --- a/drivers/media/video/ivtv/ivtv-cards.c +++ b/drivers/media/video/ivtv/ivtv-cards.c @@ -848,7 +848,7 @@ static const struct ivtv_card ivtv_card_pg600v2 = { { IVTV_CARD_INPUT_LINE_IN1, CX25840_AUDIO_SERIAL }, }, .tuners = { - { .std = V4L2_STD_ALL, .tuner = TUNER_XCEIVE_XC3028 }, + { .std = V4L2_STD_ALL, .tuner = TUNER_XC2028 }, }, .pci_list = ivtv_pci_pg600v2, }; @@ -879,7 +879,7 @@ static const struct ivtv_card ivtv_card_club3d = { { IVTV_CARD_INPUT_LINE_IN1, CX25840_AUDIO_SERIAL }, }, .tuners = { - { .std = V4L2_STD_ALL, .tuner = TUNER_XCEIVE_XC3028 }, + { .std = V4L2_STD_ALL, .tuner = TUNER_XC2028 }, }, .pci_list = ivtv_pci_club3d, }; @@ -910,7 +910,7 @@ static const struct ivtv_card ivtv_card_avertv_mce116 = { }, .gpio_init = { .direction = 0xe000, .initial_value = 0x4000 }, /* enable line-in */ .tuners = { - { .std = V4L2_STD_ALL, .tuner = TUNER_XCEIVE_XC3028 }, + { .std = V4L2_STD_ALL, .tuner = TUNER_XC2028 }, }, .pci_list = ivtv_pci_avertv_mce116, }; diff --git a/drivers/media/video/ivtv/ivtv-driver.c b/drivers/media/video/ivtv/ivtv-driver.c index 6d2dd8764f81..ec40453e46e4 100644 --- a/drivers/media/video/ivtv/ivtv-driver.c +++ b/drivers/media/video/ivtv/ivtv-driver.c @@ -829,7 +829,7 @@ static void ivtv_load_and_init_modules(struct ivtv *itv) /* load modules */ #ifndef CONFIG_VIDEO_TUNER if (hw & IVTV_HW_TUNER) { - if (itv->options.tuner == TUNER_XCEIVE_XC3028) { + if (itv->options.tuner == TUNER_XC2028) { IVTV_INFO("Xceive tuner not yet supported, only composite and S-Video inputs will be available\n"); itv->tunerid = 1; } diff --git a/drivers/media/video/tuner-types.c b/drivers/media/video/tuner-types.c index c6a7934bd5a6..ac363f019227 100644 --- a/drivers/media/video/tuner-types.c +++ b/drivers/media/video/tuner-types.c @@ -1452,9 +1452,9 @@ struct tunertype tuners[] = { .params = tuner_samsung_tcpn_2121p30a_params, .count = ARRAY_SIZE(tuner_samsung_tcpn_2121p30a_params), }, - [TUNER_XCEIVE_XC3028] = { /* Xceive 3028 */ - .name = "Xceive xc3028", - /* see xc3028.c for details */ + [TUNER_XC2028] = { /* Xceive 2028 */ + .name = "Xceive xc2028/xc3028 tuner", + /* see tuner-xc2028.c for details */ }, [TUNER_THOMSON_FE6600] = { /* Thomson PAL / DVB-T */ .name = "Thomson FE6600", diff --git a/include/media/tuner.h b/include/media/tuner.h index c03dceb92605..d49392d90e56 100644 --- a/include/media/tuner.h +++ b/include/media/tuner.h @@ -117,7 +117,7 @@ extern int tuner_debug; #define TUNER_PHILIPS_TUV1236D 68 /* ATI HDTV Wonder */ #define TUNER_TNF_5335MF 69 /* Sabrent Bt848 */ #define TUNER_SAMSUNG_TCPN_2121P30A 70 /* Hauppauge PVR-500MCE NTSC */ -#define TUNER_XCEIVE_XC3028 71 +#define TUNER_XC2028 71 #define TUNER_THOMSON_FE6600 72 /* DViCO FusionHDTV DVB-T Hybrid */ #define TUNER_SAMSUNG_TCPG_6121P30A 73 /* Hauppauge PVR-500 PAL */ -- cgit v1.2.3 From 5bea1cd3871351d70cc7624af138f8aa68b7be77 Mon Sep 17 00:00:00 2001 From: Michael Krufky Date: Mon, 22 Oct 2007 09:56:38 -0300 Subject: V4L/DVB (6435): tda8290: add support for NXP TDA18271 tuner and TDA8295 analog demod Add basic support for NXP TDA8295 analog demod and TDA18271 tuner silicon. TDA8295 + TDA8275a not yet tested. TDA8290 + TDA18271 not yet supported. Digital mode of TDA18271 not yet tested & needs more work. Signed-off-by: Michael Krufky Reviewed-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.tuner | 1 + drivers/media/Kconfig | 3 +- drivers/media/dvb/frontends/Kconfig | 7 + drivers/media/dvb/frontends/Makefile | 1 + drivers/media/dvb/frontends/tda18271.c | 1054 ++++++++++++++++++++++++++++++ drivers/media/dvb/frontends/tda18271.h | 40 ++ drivers/media/video/tda8290.c | 319 ++++++++- drivers/media/video/tda8290.h | 8 + drivers/media/video/tuner-core.c | 5 + drivers/media/video/tuner-types.c | 3 + drivers/media/video/tveeprom.c | 2 +- include/media/tuner.h | 2 + 12 files changed, 1437 insertions(+), 8 deletions(-) create mode 100644 drivers/media/dvb/frontends/tda18271.c create mode 100644 drivers/media/dvb/frontends/tda18271.h (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.tuner b/Documentation/video4linux/CARDLIST.tuner index 84c3ac7c33d9..ac47b48715dd 100644 --- a/Documentation/video4linux/CARDLIST.tuner +++ b/Documentation/video4linux/CARDLIST.tuner @@ -73,3 +73,4 @@ tuner=71 - Xceive xc2028/xc3028 tuner tuner=72 - Thomson FE6600 tuner=73 - Samsung TCPG 6121P30A tuner=75 - Philips TEA5761 FM Radio +tuner=76 - tda8295+18271 diff --git a/drivers/media/Kconfig b/drivers/media/Kconfig index d363d0cae38b..e52229380984 100644 --- a/drivers/media/Kconfig +++ b/drivers/media/Kconfig @@ -105,9 +105,10 @@ config TUNER_MT20XX Say Y here to include support for the MT2032 / MT2050 tuner. config TUNER_TDA8290 - tristate "TDA 8290+8275(a) tuner combo" + tristate "TDA 8290/8295 + 8275(a)/18271 tuner combo" depends on I2C select DVB_TDA827X + select DVB_TDA18271 default m if VIDEO_TUNER_CUSTOMIZE help Say Y here to include support for Philips TDA8290+8275(a) tuner. diff --git a/drivers/media/dvb/frontends/Kconfig b/drivers/media/dvb/frontends/Kconfig index 59b9ed1f1aec..57178d6e1cf8 100644 --- a/drivers/media/dvb/frontends/Kconfig +++ b/drivers/media/dvb/frontends/Kconfig @@ -316,6 +316,13 @@ config DVB_TDA827X help A DVB-T silicon tuner module. Say Y when you want to support this tuner. +config DVB_TDA18271 + tristate "NXP TDA18271 silicon tuner" + depends on I2C + default m if DVB_FE_CUSTOMISE + help + A silicon tuner module. Say Y when you want to support this tuner. + config DVB_TUNER_QT1010 tristate "Quantek QT1010 silicon tuner" depends on DVB_CORE && I2C diff --git a/drivers/media/dvb/frontends/Makefile b/drivers/media/dvb/frontends/Makefile index 4b8ad1f132aa..457effcda5da 100644 --- a/drivers/media/dvb/frontends/Makefile +++ b/drivers/media/dvb/frontends/Makefile @@ -39,6 +39,7 @@ obj-$(CONFIG_DVB_ISL6421) += isl6421.o obj-$(CONFIG_DVB_TDA10086) += tda10086.o obj-$(CONFIG_DVB_TDA826X) += tda826x.o obj-$(CONFIG_DVB_TDA827X) += tda827x.o +obj-$(CONFIG_DVB_TDA18271) += tda18271.o obj-$(CONFIG_DVB_TUNER_MT2060) += mt2060.o obj-$(CONFIG_DVB_TUNER_MT2266) += mt2266.o obj-$(CONFIG_DVB_TUNER_DIB0070) += dib0070.o diff --git a/drivers/media/dvb/frontends/tda18271.c b/drivers/media/dvb/frontends/tda18271.c new file mode 100644 index 000000000000..3395f2bda498 --- /dev/null +++ b/drivers/media/dvb/frontends/tda18271.c @@ -0,0 +1,1054 @@ +/* + tda18271.c - driver for the Philips / NXP TDA18271 silicon tuner + + Copyright (C) 2007 Michael Krufky (mkrufky@linuxtv.org) + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. +*/ + +#include +#include +#include + +#include "tda18271.h" + +static int debug; +module_param(debug, int, 0644); +MODULE_PARM_DESC(debug, "Turn on/off frontend debugging (default:off)."); + +#define tuner_dbg(fmt, arg...) do {\ + if (debug > 0) \ + printk(KERN_DEBUG fmt, ##arg); } while (0) + +#define tuner_extra_dbg(fmt, arg...) do {\ + if (debug > 1) \ + printk(KERN_DEBUG fmt, ##arg); } while (0) + +#define R_ID 0x00 /* ID byte */ +#define R_TM 0x01 /* Thermo byte */ +#define R_PL 0x02 /* Power level byte */ +#define R_EP1 0x03 /* Easy Prog byte 1 */ +#define R_EP2 0x04 /* Easy Prog byte 2 */ +#define R_EP3 0x05 /* Easy Prog byte 3 */ +#define R_EP4 0x06 /* Easy Prog byte 4 */ +#define R_EP5 0x07 /* Easy Prog byte 5 */ +#define R_CPD 0x08 /* Cal Post-Divider byte */ +#define R_CD1 0x09 /* Cal Divider byte 1 */ +#define R_CD2 0x0a /* Cal Divider byte 2 */ +#define R_CD3 0x0b /* Cal Divider byte 3 */ +#define R_MPD 0x0c /* Main Post-Divider byte */ +#define R_MD1 0x0d /* Main Divider byte 1 */ +#define R_MD2 0x0e /* Main Divider byte 2 */ +#define R_MD3 0x0f /* Main Divider byte 3 */ +#define R_EB1 0x10 /* Extended byte 1 */ +#define R_EB2 0x11 /* Extended byte 2 */ +#define R_EB3 0x12 /* Extended byte 3 */ +#define R_EB4 0x13 /* Extended byte 4 */ +#define R_EB5 0x14 /* Extended byte 5 */ +#define R_EB6 0x15 /* Extended byte 6 */ +#define R_EB7 0x16 /* Extended byte 7 */ +#define R_EB8 0x17 /* Extended byte 8 */ +#define R_EB9 0x18 /* Extended byte 9 */ +#define R_EB10 0x19 /* Extended byte 10 */ +#define R_EB11 0x1a /* Extended byte 11 */ +#define R_EB12 0x1b /* Extended byte 12 */ +#define R_EB13 0x1c /* Extended byte 13 */ +#define R_EB14 0x1d /* Extended byte 14 */ +#define R_EB15 0x1e /* Extended byte 15 */ +#define R_EB16 0x1f /* Extended byte 16 */ +#define R_EB17 0x20 /* Extended byte 17 */ +#define R_EB18 0x21 /* Extended byte 18 */ +#define R_EB19 0x22 /* Extended byte 19 */ +#define R_EB20 0x23 /* Extended byte 20 */ +#define R_EB21 0x24 /* Extended byte 21 */ +#define R_EB22 0x25 /* Extended byte 22 */ +#define R_EB23 0x26 /* Extended byte 23 */ + +struct tda18271_pll_map { + u32 lomax; + u8 pd; /* post div */ + u8 d; /* div */ +}; + +static struct tda18271_pll_map tda18271_main_pll[] = { + { .lomax = 32000, .pd = 0x5f, .d = 0xf0 }, + { .lomax = 35000, .pd = 0x5e, .d = 0xe0 }, + { .lomax = 37000, .pd = 0x5d, .d = 0xd0 }, + { .lomax = 41000, .pd = 0x5c, .d = 0xc0 }, + { .lomax = 44000, .pd = 0x5b, .d = 0xb0 }, + { .lomax = 49000, .pd = 0x5a, .d = 0xa0 }, + { .lomax = 54000, .pd = 0x59, .d = 0x90 }, + { .lomax = 61000, .pd = 0x58, .d = 0x80 }, + { .lomax = 65000, .pd = 0x4f, .d = 0x78 }, + { .lomax = 70000, .pd = 0x4e, .d = 0x70 }, + { .lomax = 75000, .pd = 0x4d, .d = 0x68 }, + { .lomax = 82000, .pd = 0x4c, .d = 0x60 }, + { .lomax = 89000, .pd = 0x4b, .d = 0x58 }, + { .lomax = 98000, .pd = 0x4a, .d = 0x50 }, + { .lomax = 109000, .pd = 0x49, .d = 0x48 }, + { .lomax = 123000, .pd = 0x48, .d = 0x40 }, + { .lomax = 131000, .pd = 0x3f, .d = 0x3c }, + { .lomax = 141000, .pd = 0x3e, .d = 0x38 }, + { .lomax = 151000, .pd = 0x3d, .d = 0x34 }, + { .lomax = 164000, .pd = 0x3c, .d = 0x30 }, + { .lomax = 179000, .pd = 0x3b, .d = 0x2c }, + { .lomax = 197000, .pd = 0x3a, .d = 0x28 }, + { .lomax = 219000, .pd = 0x39, .d = 0x24 }, + { .lomax = 246000, .pd = 0x38, .d = 0x20 }, + { .lomax = 263000, .pd = 0x2f, .d = 0x1e }, + { .lomax = 282000, .pd = 0x2e, .d = 0x1c }, + { .lomax = 303000, .pd = 0x2d, .d = 0x1a }, + { .lomax = 329000, .pd = 0x2c, .d = 0x18 }, + { .lomax = 359000, .pd = 0x2b, .d = 0x16 }, + { .lomax = 395000, .pd = 0x2a, .d = 0x14 }, + { .lomax = 438000, .pd = 0x29, .d = 0x12 }, + { .lomax = 493000, .pd = 0x28, .d = 0x10 }, + { .lomax = 526000, .pd = 0x1f, .d = 0x0f }, + { .lomax = 564000, .pd = 0x1e, .d = 0x0e }, + { .lomax = 607000, .pd = 0x1d, .d = 0x0d }, + { .lomax = 658000, .pd = 0x1c, .d = 0x0c }, + { .lomax = 718000, .pd = 0x1b, .d = 0x0b }, + { .lomax = 790000, .pd = 0x1a, .d = 0x0a }, + { .lomax = 877000, .pd = 0x19, .d = 0x09 }, + { .lomax = 987000, .pd = 0x18, .d = 0x08 }, + { .lomax = 0, .pd = 0x00, .d = 0x00 }, /* end */ +}; + +static struct tda18271_pll_map tda18271_cal_pll[] = { + { .lomax = 33000, .pd = 0xdd, .d = 0xd0 }, + { .lomax = 36000, .pd = 0xdc, .d = 0xc0 }, + { .lomax = 40000, .pd = 0xdb, .d = 0xb0 }, + { .lomax = 44000, .pd = 0xda, .d = 0xa0 }, + { .lomax = 49000, .pd = 0xd9, .d = 0x90 }, + { .lomax = 55000, .pd = 0xd8, .d = 0x80 }, + { .lomax = 63000, .pd = 0xd3, .d = 0x70 }, + { .lomax = 67000, .pd = 0xcd, .d = 0x68 }, + { .lomax = 73000, .pd = 0xcc, .d = 0x60 }, + { .lomax = 80000, .pd = 0xcb, .d = 0x58 }, + { .lomax = 88000, .pd = 0xca, .d = 0x50 }, + { .lomax = 98000, .pd = 0xc9, .d = 0x48 }, + { .lomax = 110000, .pd = 0xc8, .d = 0x40 }, + { .lomax = 126000, .pd = 0xc3, .d = 0x38 }, + { .lomax = 135000, .pd = 0xbd, .d = 0x34 }, + { .lomax = 147000, .pd = 0xbc, .d = 0x30 }, + { .lomax = 160000, .pd = 0xbb, .d = 0x2c }, + { .lomax = 176000, .pd = 0xba, .d = 0x28 }, + { .lomax = 196000, .pd = 0xb9, .d = 0x24 }, + { .lomax = 220000, .pd = 0xb8, .d = 0x20 }, + { .lomax = 252000, .pd = 0xb3, .d = 0x1c }, + { .lomax = 271000, .pd = 0xad, .d = 0x1a }, + { .lomax = 294000, .pd = 0xac, .d = 0x18 }, + { .lomax = 321000, .pd = 0xab, .d = 0x16 }, + { .lomax = 353000, .pd = 0xaa, .d = 0x14 }, + { .lomax = 392000, .pd = 0xa9, .d = 0x12 }, + { .lomax = 441000, .pd = 0xa8, .d = 0x10 }, + { .lomax = 505000, .pd = 0xa3, .d = 0x0e }, + { .lomax = 543000, .pd = 0x9d, .d = 0x0d }, + { .lomax = 589000, .pd = 0x9c, .d = 0x0c }, + { .lomax = 642000, .pd = 0x9b, .d = 0x0b }, + { .lomax = 707000, .pd = 0x9a, .d = 0x0a }, + { .lomax = 785000, .pd = 0x99, .d = 0x09 }, + { .lomax = 883000, .pd = 0x98, .d = 0x08 }, + { .lomax = 1010000, .pd = 0x93, .d = 0x07 }, + { .lomax = 0, .pd = 0x00, .d = 0x00 }, /* end */ +}; + +struct tda18271_map { + u32 rfmax; + u8 val; +}; + +static struct tda18271_map tda18271_bp_filter[] = { + { .rfmax = 62000, .val = 0x00 }, + { .rfmax = 84000, .val = 0x01 }, + { .rfmax = 100000, .val = 0x02 }, + { .rfmax = 140000, .val = 0x03 }, + { .rfmax = 170000, .val = 0x04 }, + { .rfmax = 180000, .val = 0x05 }, + { .rfmax = 865000, .val = 0x06 }, + { .rfmax = 0, .val = 0x00 }, /* end */ +}; + +static struct tda18271_map tda18271_km[] = { + { .rfmax = 61100, .val = 0x74 }, + { .rfmax = 350000, .val = 0x40 }, + { .rfmax = 720000, .val = 0x30 }, + { .rfmax = 865000, .val = 0x40 }, + { .rfmax = 0, .val = 0x00 }, /* end */ +}; + +static struct tda18271_map tda18271_rf_band[] = { + { .rfmax = 47900, .val = 0x00 }, + { .rfmax = 61100, .val = 0x01 }, +/* { .rfmax = 152600, .val = 0x02 }, */ + { .rfmax = 121200, .val = 0x02 }, + { .rfmax = 164700, .val = 0x03 }, + { .rfmax = 203500, .val = 0x04 }, + { .rfmax = 457800, .val = 0x05 }, + { .rfmax = 865000, .val = 0x06 }, + { .rfmax = 0, .val = 0x00 }, /* end */ +}; + +static struct tda18271_map tda18271_gain_taper[] = { + { .rfmax = 45400, .val = 0x1f }, + { .rfmax = 45800, .val = 0x1e }, + { .rfmax = 46200, .val = 0x1d }, + { .rfmax = 46700, .val = 0x1c }, + { .rfmax = 47100, .val = 0x1b }, + { .rfmax = 47500, .val = 0x1a }, + { .rfmax = 47900, .val = 0x19 }, + { .rfmax = 49600, .val = 0x17 }, + { .rfmax = 51200, .val = 0x16 }, + { .rfmax = 52900, .val = 0x15 }, + { .rfmax = 54500, .val = 0x14 }, + { .rfmax = 56200, .val = 0x13 }, + { .rfmax = 57800, .val = 0x12 }, + { .rfmax = 59500, .val = 0x11 }, + { .rfmax = 61100, .val = 0x10 }, + { .rfmax = 67600, .val = 0x0d }, + { .rfmax = 74200, .val = 0x0c }, + { .rfmax = 80700, .val = 0x0b }, + { .rfmax = 87200, .val = 0x0a }, + { .rfmax = 93800, .val = 0x09 }, + { .rfmax = 100300, .val = 0x08 }, + { .rfmax = 106900, .val = 0x07 }, + { .rfmax = 113400, .val = 0x06 }, + { .rfmax = 119900, .val = 0x05 }, + { .rfmax = 126500, .val = 0x04 }, + { .rfmax = 133000, .val = 0x03 }, + { .rfmax = 139500, .val = 0x02 }, + { .rfmax = 146100, .val = 0x01 }, + { .rfmax = 152600, .val = 0x00 }, + { .rfmax = 154300, .val = 0x1f }, + { .rfmax = 156100, .val = 0x1e }, + { .rfmax = 157800, .val = 0x1d }, + { .rfmax = 159500, .val = 0x1c }, + { .rfmax = 161200, .val = 0x1b }, + { .rfmax = 163000, .val = 0x1a }, + { .rfmax = 164700, .val = 0x19 }, + { .rfmax = 170200, .val = 0x17 }, + { .rfmax = 175800, .val = 0x16 }, + { .rfmax = 181300, .val = 0x15 }, + { .rfmax = 186900, .val = 0x14 }, + { .rfmax = 192400, .val = 0x13 }, + { .rfmax = 198000, .val = 0x12 }, + { .rfmax = 203500, .val = 0x11 }, + { .rfmax = 216200, .val = 0x14 }, + { .rfmax = 228900, .val = 0x13 }, + { .rfmax = 241600, .val = 0x12 }, + { .rfmax = 254400, .val = 0x11 }, + { .rfmax = 267100, .val = 0x10 }, + { .rfmax = 279800, .val = 0x0f }, + { .rfmax = 292500, .val = 0x0e }, + { .rfmax = 305200, .val = 0x0d }, + { .rfmax = 317900, .val = 0x0c }, + { .rfmax = 330700, .val = 0x0b }, + { .rfmax = 343400, .val = 0x0a }, + { .rfmax = 356100, .val = 0x09 }, + { .rfmax = 368800, .val = 0x08 }, + { .rfmax = 381500, .val = 0x07 }, + { .rfmax = 394200, .val = 0x06 }, + { .rfmax = 406900, .val = 0x05 }, + { .rfmax = 419700, .val = 0x04 }, + { .rfmax = 432400, .val = 0x03 }, + { .rfmax = 445100, .val = 0x02 }, + { .rfmax = 457800, .val = 0x01 }, + { .rfmax = 476300, .val = 0x19 }, + { .rfmax = 494800, .val = 0x18 }, + { .rfmax = 513300, .val = 0x17 }, + { .rfmax = 531800, .val = 0x16 }, + { .rfmax = 550300, .val = 0x15 }, + { .rfmax = 568900, .val = 0x14 }, + { .rfmax = 587400, .val = 0x13 }, + { .rfmax = 605900, .val = 0x12 }, + { .rfmax = 624400, .val = 0x11 }, + { .rfmax = 642900, .val = 0x10 }, + { .rfmax = 661400, .val = 0x0f }, + { .rfmax = 679900, .val = 0x0e }, + { .rfmax = 698400, .val = 0x0d }, + { .rfmax = 716900, .val = 0x0c }, + { .rfmax = 735400, .val = 0x0b }, + { .rfmax = 753900, .val = 0x0a }, + { .rfmax = 772500, .val = 0x09 }, + { .rfmax = 791000, .val = 0x08 }, + { .rfmax = 809500, .val = 0x07 }, + { .rfmax = 828000, .val = 0x06 }, + { .rfmax = 846500, .val = 0x05 }, + { .rfmax = 865000, .val = 0x04 }, + { .rfmax = 0, .val = 0x00 }, /* end */ +}; + +static struct tda18271_map tda18271_rf_cal[] = { + { .rfmax = 41000, .val = 0x1e }, + { .rfmax = 43000, .val = 0x30 }, + { .rfmax = 45000, .val = 0x43 }, + { .rfmax = 46000, .val = 0x4d }, + { .rfmax = 47000, .val = 0x54 }, + { .rfmax = 47900, .val = 0x64 }, + { .rfmax = 49100, .val = 0x20 }, + { .rfmax = 50000, .val = 0x22 }, + { .rfmax = 51000, .val = 0x2a }, + { .rfmax = 53000, .val = 0x32 }, + { .rfmax = 55000, .val = 0x35 }, + { .rfmax = 56000, .val = 0x3c }, + { .rfmax = 57000, .val = 0x3f }, + { .rfmax = 58000, .val = 0x48 }, + { .rfmax = 59000, .val = 0x4d }, + { .rfmax = 60000, .val = 0x58 }, + { .rfmax = 61100, .val = 0x5f }, + { .rfmax = 0, .val = 0x00 }, /* end */ +}; + +/*---------------------------------------------------------------------*/ + +#define TDA18271_NUM_REGS 39 + +#define TDA18271_ANALOG 0 +#define TDA18271_DIGITAL 1 + +struct tda18271_priv { + u8 i2c_addr; + struct i2c_adapter *i2c_adap; + unsigned char tda18271_regs[TDA18271_NUM_REGS]; + int mode; + + u32 frequency; + u32 bandwidth; +}; + +/*---------------------------------------------------------------------*/ + +static void tda18271_dump_regs(struct dvb_frontend *fe) +{ + struct tda18271_priv *priv = fe->tuner_priv; + unsigned char *regs = priv->tda18271_regs; + + tuner_dbg("=== TDA18271 REG DUMP ===\n"); + tuner_dbg("ID_BYTE = 0x%x\n", 0xff & regs[R_ID]); + tuner_dbg("THERMO_BYTE = 0x%x\n", 0xff & regs[R_TM]); + tuner_dbg("POWER_LEVEL_BYTE = 0x%x\n", 0xff & regs[R_PL]); + tuner_dbg("EASY_PROG_BYTE_1 = 0x%x\n", 0xff & regs[R_EP1]); + tuner_dbg("EASY_PROG_BYTE_2 = 0x%x\n", 0xff & regs[R_EP2]); + tuner_dbg("EASY_PROG_BYTE_3 = 0x%x\n", 0xff & regs[R_EP3]); + tuner_dbg("EASY_PROG_BYTE_4 = 0x%x\n", 0xff & regs[R_EP4]); + tuner_dbg("EASY_PROG_BYTE_5 = 0x%x\n", 0xff & regs[R_EP5]); + tuner_dbg("CAL_POST_DIV_BYTE = 0x%x\n", 0xff & regs[R_CPD]); + tuner_dbg("CAL_DIV_BYTE_1 = 0x%x\n", 0xff & regs[R_CD1]); + tuner_dbg("CAL_DIV_BYTE_2 = 0x%x\n", 0xff & regs[R_CD2]); + tuner_dbg("CAL_DIV_BYTE_3 = 0x%x\n", 0xff & regs[R_CD3]); + tuner_dbg("MAIN_POST_DIV_BYTE = 0x%x\n", 0xff & regs[R_MPD]); + tuner_dbg("MAIN_DIV_BYTE_1 = 0x%x\n", 0xff & regs[R_MD1]); + tuner_dbg("MAIN_DIV_BYTE_2 = 0x%x\n", 0xff & regs[R_MD2]); + tuner_dbg("MAIN_DIV_BYTE_3 = 0x%x\n", 0xff & regs[R_MD3]); +} + +static void tda18271_read_regs(struct dvb_frontend *fe) +{ + struct tda18271_priv *priv = fe->tuner_priv; + unsigned char *regs = priv->tda18271_regs; + unsigned char buf = 0x00; + int ret; + struct i2c_msg msg[] = { + { .addr = priv->i2c_addr, .flags = 0, + .buf = &buf, .len = 1 }, + { .addr = priv->i2c_addr, .flags = I2C_M_RD, + .buf = regs, .len = 16 } + }; + + if (fe->ops.i2c_gate_ctrl) + fe->ops.i2c_gate_ctrl(fe, 1); + + /* read all registers */ + ret = i2c_transfer(priv->i2c_adap, msg, 2); + + if (fe->ops.i2c_gate_ctrl) + fe->ops.i2c_gate_ctrl(fe, 0); + + if (ret != 2) + printk("ERROR: %s: i2c_transfer returned: %d\n", + __FUNCTION__, ret); + + if (debug > 1) + tda18271_dump_regs(fe); +} + +static void tda18271_write_regs(struct dvb_frontend *fe, int idx, int len) +{ + struct tda18271_priv *priv = fe->tuner_priv; + unsigned char *regs = priv->tda18271_regs; + unsigned char buf[TDA18271_NUM_REGS+1]; + struct i2c_msg msg = { .addr = priv->i2c_addr, .flags = 0, + .buf = buf, .len = len+1 }; + int i, ret; + + BUG_ON((len == 0) || (idx+len > sizeof(buf))); + + buf[0] = idx; + for (i = 1; i <= len; i++) { + buf[i] = regs[idx-1+i]; + } + + if (fe->ops.i2c_gate_ctrl) + fe->ops.i2c_gate_ctrl(fe, 1); + + /* write registers */ + ret = i2c_transfer(priv->i2c_adap, &msg, 1); + + if (fe->ops.i2c_gate_ctrl) + fe->ops.i2c_gate_ctrl(fe, 0); + + if (ret != 1) + printk(KERN_WARNING "ERROR: %s: i2c_transfer returned: %d\n", + __FUNCTION__, ret); +} + +/*---------------------------------------------------------------------*/ + +static void tda18271_init_regs(struct dvb_frontend *fe) +{ + struct tda18271_priv *priv = fe->tuner_priv; + unsigned char *regs = priv->tda18271_regs; + + tda18271_read_regs(fe); + + /* test IR_CAL_OK to see if we need init */ + if ((regs[R_EP1] & 0x08) != 0) + return; + + printk(KERN_INFO "tda18271: initializing registers\n"); + + /* initialize registers */ + regs[R_ID] = 0x83; + regs[R_TM] = 0x08; + regs[R_PL] = 0x80; + regs[R_EP1] = 0xc6; + regs[R_EP2] = 0xdf; + regs[R_EP3] = 0x16; + regs[R_EP4] = 0x60; + regs[R_EP5] = 0x80; + regs[R_CPD] = 0x80; + regs[R_CD1] = 0x00; + regs[R_CD2] = 0x00; + regs[R_CD3] = 0x00; + regs[R_MPD] = 0x00; + regs[R_MD1] = 0x00; + regs[R_MD2] = 0x00; + regs[R_MD3] = 0x00; + regs[R_EB1] = 0xff; + regs[R_EB2] = 0x01; + regs[R_EB3] = 0x84; + regs[R_EB4] = 0x41; + regs[R_EB5] = 0x01; + regs[R_EB6] = 0x84; + regs[R_EB7] = 0x40; + regs[R_EB8] = 0x07; + regs[R_EB9] = 0x00; + regs[R_EB10] = 0x00; + regs[R_EB11] = 0x96; + regs[R_EB12] = 0x0f; + regs[R_EB13] = 0xc1; + regs[R_EB14] = 0x00; + regs[R_EB15] = 0x8f; + regs[R_EB16] = 0x00; + regs[R_EB17] = 0x00; + regs[R_EB18] = 0x00; + regs[R_EB19] = 0x00; + regs[R_EB20] = 0x20; + regs[R_EB21] = 0x33; + regs[R_EB22] = 0x48; + regs[R_EB23] = 0xb0; + + tda18271_write_regs(fe, 0x00, TDA18271_NUM_REGS); + /* setup AGC1 & AGC2 */ + regs[R_EB17] = 0x00; + tda18271_write_regs(fe, R_EB17, 1); + regs[R_EB17] = 0x03; + tda18271_write_regs(fe, R_EB17, 1); + regs[R_EB17] = 0x43; + tda18271_write_regs(fe, R_EB17, 1); + regs[R_EB17] = 0x4c; + tda18271_write_regs(fe, R_EB17, 1); + + regs[R_EB20] = 0xa0; + tda18271_write_regs(fe, R_EB20, 1); + regs[R_EB20] = 0xa7; + tda18271_write_regs(fe, R_EB20, 1); + regs[R_EB20] = 0xe7; + tda18271_write_regs(fe, R_EB20, 1); + regs[R_EB20] = 0xec; + tda18271_write_regs(fe, R_EB20, 1); + + /* image rejection calibration */ + + /* low-band */ + regs[R_EP3] = 0x1f; + regs[R_EP4] = 0x66; + regs[R_EP5] = 0x81; + regs[R_CPD] = 0xcc; + regs[R_CD1] = 0x6c; + regs[R_CD2] = 0x00; + regs[R_CD3] = 0x00; + regs[R_MPD] = 0xcd; + regs[R_MD1] = 0x77; + regs[R_MD2] = 0x08; + regs[R_MD3] = 0x00; + + tda18271_write_regs(fe, R_EP3, 11); + msleep(5); /* pll locking */ + + regs[R_EP1] = 0xc6; + tda18271_write_regs(fe, R_EP1, 1); + msleep(5); /* wanted low measurement */ + + regs[R_EP3] = 0x1f; + regs[R_EP4] = 0x66; + regs[R_EP5] = 0x85; + regs[R_CPD] = 0xcb; + regs[R_CD1] = 0x66; + regs[R_CD2] = 0x70; + regs[R_CD3] = 0x00; + + tda18271_write_regs(fe, R_EP3, 7); + msleep(5); /* pll locking */ + + regs[R_EP2] = 0xdf; + tda18271_write_regs(fe, R_EP2, 1); + msleep(30); /* image low optimization completion */ + + /* mid-band */ + regs[R_EP3] = 0x1f; + regs[R_EP4] = 0x66; + regs[R_EP5] = 0x82; + regs[R_CPD] = 0xa8; + regs[R_CD1] = 0x66; + regs[R_CD2] = 0x00; + regs[R_CD3] = 0x00; + regs[R_MPD] = 0xa9; + regs[R_MD1] = 0x73; + regs[R_MD2] = 0x1a; + regs[R_MD3] = 0x00; + + tda18271_write_regs(fe, R_EP3, 11); + msleep(5); /* pll locking */ + + regs[R_EP1] = 0xc6; + tda18271_write_regs(fe, R_EP1, 1); + msleep(5); /* wanted mid measurement */ + + regs[R_EP3] = 0x1f; + regs[R_EP4] = 0x66; + regs[R_EP5] = 0x86; + regs[R_CPD] = 0xa8; + regs[R_CD1] = 0x66; + regs[R_CD2] = 0xa0; + regs[R_CD3] = 0x00; + + tda18271_write_regs(fe, R_EP3, 7); + msleep(5); /* pll locking */ + + regs[R_EP2] = 0xdf; + tda18271_write_regs(fe, R_EP2, 1); + msleep(30); /* image mid optimization completion */ + + /* high-band */ + regs[R_EP3] = 0x1f; + regs[R_EP4] = 0x66; + regs[R_EP5] = 0x83; + regs[R_CPD] = 0x98; + regs[R_CD1] = 0x65; + regs[R_CD2] = 0x00; + regs[R_CD3] = 0x00; + regs[R_MPD] = 0x99; + regs[R_MD1] = 0x71; + regs[R_MD2] = 0xcd; + regs[R_MD3] = 0x00; + + tda18271_write_regs(fe, R_EP3, 11); + msleep(5); /* pll locking */ + + regs[R_EP1] = 0xc6; + tda18271_write_regs(fe, R_EP1, 1); + msleep(5); /* wanted high measurement */ + + regs[R_EP3] = 0x1f; + regs[R_EP4] = 0x66; + regs[R_EP5] = 0x87; + regs[R_CPD] = 0x98; + regs[R_CD1] = 0x65; + regs[R_CD2] = 0x50; + regs[R_CD3] = 0x00; + + tda18271_write_regs(fe, R_EP3, 7); + msleep(5); /* pll locking */ + + regs[R_EP2] = 0xdf; + + tda18271_write_regs(fe, R_EP2, 1); + msleep(30); /* image high optimization completion */ + + regs[R_EP4] = 0x64; + tda18271_write_regs(fe, R_EP4, 1); + + regs[R_EP1] = 0xc6; + tda18271_write_regs(fe, R_EP1, 1); +} + +static int tda18271_tune(struct dvb_frontend *fe, + u32 ifc, u32 freq, u32 bw, u8 std) +{ + struct tda18271_priv *priv = fe->tuner_priv; + unsigned char *regs = priv->tda18271_regs; + u32 div, N = 0; + int i; + + + tuner_dbg("%s: freq = %d, ifc = %d\n", __FUNCTION__, freq, ifc); + + tda18271_init_regs(fe); + /* RF tracking filter calibration */ + + /* calculate BP_Filter */ + i = 0; + while ((tda18271_bp_filter[i].rfmax * 1000) < freq) { + if (tda18271_bp_filter[i + 1].rfmax == 0) + break; + i++; + } + tuner_extra_dbg("bp filter = 0x%x, i = %d\n", + tda18271_bp_filter[i].val, i); + + regs[R_EP1] &= ~0x07; /* clear bp filter bits */ + regs[R_EP1] |= tda18271_bp_filter[i].val; + tda18271_write_regs(fe, R_EP1, 1); + + regs[R_EB4] &= 0x07; + regs[R_EB4] |= 0x60; + tda18271_write_regs(fe, R_EB4, 1); + + regs[R_EB7] = 0x60; + tda18271_write_regs(fe, R_EB7, 1); + + regs[R_EB14] = 0x00; + tda18271_write_regs(fe, R_EB14, 1); + + regs[R_EB20] = 0xcc; + tda18271_write_regs(fe, R_EB20, 1); + + /* set CAL mode to RF tracking filter calibration */ + regs[R_EB4] |= 0x03; + + /* calculate CAL PLL */ + + switch (priv->mode) { + case TDA18271_ANALOG: + N = freq - 1250000; + break; + case TDA18271_DIGITAL: + N = freq + bw / 2; + break; + } + + i = 0; + while ((tda18271_cal_pll[i].lomax * 1000) < N) { + if (tda18271_cal_pll[i + 1].lomax == 0) + break; + i++; + } + tuner_extra_dbg("cal pll, pd = 0x%x, d = 0x%x, i = %d\n", + tda18271_cal_pll[i].pd, tda18271_cal_pll[i].d, i); + + regs[R_CPD] = tda18271_cal_pll[i].pd; + + div = ((tda18271_cal_pll[i].d * (N / 1000)) << 7) / 125; + regs[R_CD1] = 0xff & (div >> 16); + regs[R_CD2] = 0xff & (div >> 8); + regs[R_CD3] = 0xff & div; + + /* calculate MAIN PLL */ + + switch (priv->mode) { + case TDA18271_ANALOG: + N = freq - 250000; + break; + case TDA18271_DIGITAL: + N = freq + bw / 2 + 1000000; + break; + } + + i = 0; + while ((tda18271_main_pll[i].lomax * 1000) < N) { + if (tda18271_main_pll[i + 1].lomax == 0) + break; + i++; + } + tuner_extra_dbg("main pll, pd = 0x%x, d = 0x%x, i = %d\n", + tda18271_main_pll[i].pd, tda18271_main_pll[i].d, i); + + regs[R_MPD] = (0x7f & tda18271_main_pll[i].pd); + + switch (priv->mode) { + case TDA18271_ANALOG: + regs[R_MPD] &= ~0x08; + break; + case TDA18271_DIGITAL: + regs[R_MPD] |= 0x08; + break; + } + + div = ((tda18271_main_pll[i].d * (N / 1000)) << 7) / 125; + regs[R_MD1] = 0xff & (div >> 16); + regs[R_MD2] = 0xff & (div >> 8); + regs[R_MD3] = 0xff & div; + + tda18271_write_regs(fe, R_EP3, 11); + msleep(5); /* RF tracking filter calibration initialization */ + + /* search for K,M,CO for RF Calibration */ + i = 0; + while ((tda18271_km[i].rfmax * 1000) < freq) { + if (tda18271_km[i + 1].rfmax == 0) + break; + i++; + } + tuner_extra_dbg("km = 0x%x, i = %d\n", tda18271_km[i].val, i); + + regs[R_EB13] &= 0x83; + regs[R_EB13] |= tda18271_km[i].val; + tda18271_write_regs(fe, R_EB13, 1); + + /* search for RF_BAND */ + i = 0; + while ((tda18271_rf_band[i].rfmax * 1000) < freq) { + if (tda18271_rf_band[i + 1].rfmax == 0) + break; + i++; + } + tuner_extra_dbg("rf band = 0x%x, i = %d\n", + tda18271_rf_band[i].val, i); + + regs[R_EP2] &= ~0xe0; /* clear rf band bits */ + regs[R_EP2] |= (tda18271_rf_band[i].val << 5); + + /* search for Gain_Taper */ + i = 0; + while ((tda18271_gain_taper[i].rfmax * 1000) < freq) { + if (tda18271_gain_taper[i + 1].rfmax == 0) + break; + i++; + } + tuner_extra_dbg("gain taper = 0x%x, i = %d\n", + tda18271_gain_taper[i].val, i); + + regs[R_EP2] &= ~0x1f; /* clear gain taper bits */ + regs[R_EP2] |= tda18271_gain_taper[i].val; + + tda18271_write_regs(fe, R_EP2, 1); + tda18271_write_regs(fe, R_EP1, 1); + tda18271_write_regs(fe, R_EP2, 1); + tda18271_write_regs(fe, R_EP1, 1); + + regs[R_EB4] &= 0x07; + regs[R_EB4] |= 0x40; + tda18271_write_regs(fe, R_EB4, 1); + + regs[R_EB7] = 0x40; + tda18271_write_regs(fe, R_EB7, 1); + msleep(10); + + regs[R_EB20] = 0xec; + tda18271_write_regs(fe, R_EB20, 1); + msleep(60); /* RF tracking filter calibration completion */ + + regs[R_EP4] &= ~0x03; /* set cal mode to normal */ + tda18271_write_regs(fe, R_EP4, 1); + + tda18271_write_regs(fe, R_EP1, 1); + + /* RF tracking filer correction for VHF_Low band */ + i = 0; + while ((tda18271_rf_cal[i].rfmax * 1000) < freq) { + if (tda18271_rf_cal[i].rfmax == 0) + break; + i++; + } + tuner_extra_dbg("rf cal = 0x%x, i = %d\n", tda18271_rf_cal[i].val, i); + + /* VHF_Low band only */ + if (tda18271_rf_cal[i].rfmax != 0) { + regs[R_EB14] = tda18271_rf_cal[i].val; + tda18271_write_regs(fe, R_EB14, 1); + } + + /* Channel Configuration */ + + switch (priv->mode) { + case TDA18271_ANALOG: + regs[R_EB22] = 0x2c; + break; + case TDA18271_DIGITAL: + regs[R_EB22] = 0x37; + break; + } + tda18271_write_regs(fe, R_EB22, 1); + + regs[R_EP1] |= 0x40; /* set dis power level on */ + + /* set standard */ + regs[R_EP3] &= ~0x1f; /* clear std bits */ + + /* see table 22 */ + regs[R_EP3] |= std; + + /* TO DO: * + * ================ * + * FM radio, 0x18 * + * ATSC 6MHz, 0x1c * + * DVB-T 6MHz, 0x1c * + * DVB-T 7MHz, 0x1d * + * DVB-T 8MHz, 0x1e * + * QAM 6MHz, 0x1d * + * QAM 8MHz, 0x1f */ + + regs[R_EP4] &= ~0x03; /* set cal mode to normal */ + + regs[R_EP4] &= ~0x1c; /* clear if level bits */ + switch (priv->mode) { + case TDA18271_ANALOG: + regs[R_MPD] &= ~0x80; /* IF notch = 0 */ + break; + case TDA18271_DIGITAL: + regs[R_EP4] |= 0x04; + regs[R_MPD] |= 0x80; + break; + } + + regs[R_EP4] &= ~0x80; /* turn this bit on only for fm */ + + /* FIXME: image rejection validity EP5[2:0] */ + + /* calculate MAIN PLL */ + N = freq + ifc; + + i = 0; + while ((tda18271_main_pll[i].lomax * 1000) < N) { + if (tda18271_main_pll[i + 1].lomax == 0) + break; + i++; + } + tuner_extra_dbg("main pll, pd = 0x%x, d = 0x%x, i = %d\n", + tda18271_main_pll[i].pd, tda18271_main_pll[i].d, i); + + regs[R_MPD] = (0x7f & tda18271_main_pll[i].pd); + switch (priv->mode) { + case TDA18271_ANALOG: + regs[R_MPD] &= ~0x08; + break; + case TDA18271_DIGITAL: + regs[R_MPD] |= 0x08; + break; + } + + div = ((tda18271_main_pll[i].d * (N / 1000)) << 7) / 125; + regs[R_MD1] = 0xff & (div >> 16); + regs[R_MD2] = 0xff & (div >> 8); + regs[R_MD3] = 0xff & div; + + tda18271_write_regs(fe, R_TM, 15); + msleep(5); + return 0; +} + +/* ------------------------------------------------------------------ */ + +static int tda18271_set_params(struct dvb_frontend *fe, + struct dvb_frontend_parameters *params) +{ + struct tda18271_priv *priv = fe->tuner_priv; + u8 std; + u32 bw, sgIF = 0; + + u32 freq = params->frequency; + + priv->mode = TDA18271_DIGITAL; + + /* see table 22 */ + if (fe->ops.info.type == FE_ATSC) { + switch (params->u.vsb.modulation) { + case VSB_8: + case VSB_16: + std = 0x1b; /* device-specific (spec says 0x1c) */ + sgIF = 5380000; + break; + case QAM_64: + case QAM_256: + std = 0x18; /* device-specific (spec says 0x1d) */ + sgIF = 4000000; + break; + default: + printk(KERN_WARNING "%s: modulation not set!\n", + __FUNCTION__); + return -EINVAL; + } + freq += 1750000; /* Adjust to center (+1.75MHZ) */ + bw = 6000000; + } else if (fe->ops.info.type == FE_OFDM) { + switch (params->u.ofdm.bandwidth) { + case BANDWIDTH_6_MHZ: + std = 0x1c; + bw = 6000000; + break; + case BANDWIDTH_7_MHZ: + std = 0x1d; + bw = 7000000; + break; + case BANDWIDTH_8_MHZ: + std = 0x1e; + bw = 8000000; + break; + default: + printk(KERN_WARNING "%s: bandwidth not set!\n", + __FUNCTION__); + return -EINVAL; + } + } else { + printk(KERN_WARNING "%s: modulation type not supported!\n", + __FUNCTION__); + return -EINVAL; + } + + return tda18271_tune(fe, sgIF, freq, bw, std); +} + +static int tda18271_set_analog_params(struct dvb_frontend *fe, + struct analog_parameters *params) +{ + struct tda18271_priv *priv = fe->tuner_priv; + u8 std; + unsigned int sgIF; + char *mode; + + priv->mode = TDA18271_ANALOG; + + /* see table 22 */ + if (params->std & V4L2_STD_MN) { + std = 0x0d; + sgIF = 92; + mode = "MN"; + } else if (params->std & V4L2_STD_B) { + std = 0x0e; + sgIF = 108; + mode = "B"; + } else if (params->std & V4L2_STD_GH) { + std = 0x0f; + sgIF = 124; + mode = "GH"; + } else if (params->std & V4L2_STD_PAL_I) { + std = 0x0f; + sgIF = 124; + mode = "I"; + } else if (params->std & V4L2_STD_DK) { + std = 0x0f; + sgIF = 124; + mode = "DK"; + } else if (params->std & V4L2_STD_SECAM_L) { + std = 0x0f; + sgIF = 124; + mode = "L"; + } else if (params->std & V4L2_STD_SECAM_LC) { + std = 0x0f; + sgIF = 20; + mode = "LC"; + } else { + std = 0x0f; + sgIF = 124; + mode = "xx"; + } + + if (params->mode == V4L2_TUNER_RADIO) + sgIF = 88; /* if frequency is 5.5 MHz */ + + tuner_dbg("setting tda18271 to system %s\n", mode); + + return tda18271_tune(fe, sgIF * 62500, params->frequency * 62500, + 0, std); +} + +static int tda18271_release(struct dvb_frontend *fe) +{ + kfree(fe->tuner_priv); + fe->tuner_priv = NULL; + return 0; +} + +static int tda18271_get_frequency(struct dvb_frontend *fe, u32 *frequency) +{ + struct tda18271_priv *priv = fe->tuner_priv; + *frequency = priv->frequency; + return 0; +} + +static int tda18271_get_bandwidth(struct dvb_frontend *fe, u32 *bandwidth) +{ + struct tda18271_priv *priv = fe->tuner_priv; + *bandwidth = priv->bandwidth; + return 0; +} + +static struct dvb_tuner_ops tda18271_tuner_ops = { + .info = { + .name = "NXP TDA18271HD", + .frequency_min = 45000000, + .frequency_max = 864000000, + .frequency_step = 62500 + }, + .set_params = tda18271_set_params, + .set_analog_params = tda18271_set_analog_params, + .release = tda18271_release, + .get_frequency = tda18271_get_frequency, + .get_bandwidth = tda18271_get_bandwidth, +}; + +struct dvb_frontend *tda18271_attach(struct dvb_frontend *fe, u8 addr, + struct i2c_adapter *i2c) +{ + struct tda18271_priv *priv = NULL; + + tuner_dbg("%s:\n", __FUNCTION__); + priv = kzalloc(sizeof(struct tda18271_priv), GFP_KERNEL); + if (priv == NULL) + return NULL; + + priv->i2c_addr = addr; + priv->i2c_adap = i2c; + + memcpy(&fe->ops.tuner_ops, &tda18271_tuner_ops, + sizeof(struct dvb_tuner_ops)); + + fe->tuner_priv = priv; + + return fe; +} +EXPORT_SYMBOL_GPL(tda18271_attach); +MODULE_DESCRIPTION("NXP TDA18271HD analog / digital tuner driver"); +MODULE_AUTHOR("Michael Krufky "); +MODULE_LICENSE("GPL"); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * --------------------------------------------------------------------------- + * Local variables: + * c-basic-offset: 8 + * End: + */ diff --git a/drivers/media/dvb/frontends/tda18271.h b/drivers/media/dvb/frontends/tda18271.h new file mode 100644 index 000000000000..a8a19a7197f0 --- /dev/null +++ b/drivers/media/dvb/frontends/tda18271.h @@ -0,0 +1,40 @@ +/* + tda18271.h - header for the Philips / NXP TDA18271 silicon tuner + + Copyright (C) 2007 Michael Krufky (mkrufky@linuxtv.org) + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. +*/ + +#ifndef __TDA18271_H__ +#define __TDA18271_H__ + +#include +#include "dvb_frontend.h" + +#if defined(CONFIG_DVB_TDA18271) || (defined(CONFIG_DVB_TDA18271_MODULE) && defined(MODULE)) +extern struct dvb_frontend *tda18271_attach(struct dvb_frontend *fe, u8 addr, + struct i2c_adapter *i2c); +#else +static inline struct dvb_frontend *tda18271_attach(struct dvb_frontend *fe, + u8 addr, + struct i2c_adapter *i2c) +{ + printk(KERN_WARNING "%s: driver disabled by Kconfig\n", __FUNCTION__); + return NULL; +} +#endif + +#endif /* __TDA18271_H__ */ diff --git a/drivers/media/video/tda8290.c b/drivers/media/video/tda8290.c index 5d30cbcd7361..a38ed9db6403 100644 --- a/drivers/media/video/tda8290.c +++ b/drivers/media/video/tda8290.c @@ -25,8 +25,9 @@ #include #include "tda8290.h" #include "tda827x.h" +#include "tda18271.h" -static int tuner_debug = 0; +static int tuner_debug; module_param_named(debug, tuner_debug, int, 0644); MODULE_PARM_DESC(debug, "enable verbose debug messages"); @@ -65,6 +66,34 @@ static void tda8290_i2c_bridge(struct tuner *t, int close) } } +static void tda8295_i2c_bridge(struct tuner *t, int close) +{ + struct tda8290_priv *priv = t->priv; + + unsigned char enable[2] = { 0x45, 0xc1 }; + unsigned char disable[2] = { 0x46, 0x00 }; + unsigned char buf[3] = { 0x45, 0x01, 0x00 }; + unsigned char *msg; + if (close) { + msg = enable; + tuner_i2c_xfer_send(&priv->i2c_props, msg, 2); + /* let the bridge stabilize */ + msleep(20); + } else { + msg = disable; + tuner_i2c_xfer_send(&priv->i2c_props, msg, 1); + tuner_i2c_xfer_recv(&priv->i2c_props, &msg[1], 1); + + buf[2] = msg[1]; + buf[2] &= ~0x04; + tuner_i2c_xfer_send(&priv->i2c_props, buf, 3); + msleep(5); + + msg[1] |= 0x04; + tuner_i2c_xfer_send(&priv->i2c_props, msg, 2); + } +} + /*---------------------------------------------------------------------*/ static void set_audio(struct tuner *t) @@ -233,6 +262,153 @@ static void tda8290_set_freq(struct tuner *t, unsigned int freq) /*---------------------------------------------------------------------*/ +static void tda8295_power(struct tuner *t, int enable) +{ + struct tda8290_priv *priv = t->priv; + unsigned char buf[] = { 0x30, 0x00 }; /* clb_stdbt */ + + tuner_i2c_xfer_send(&priv->i2c_props, &buf[0], 1); + tuner_i2c_xfer_recv(&priv->i2c_props, &buf[1], 1); + + if (enable) + buf[1] = 0x01; + else + buf[1] = 0x03; + + tuner_i2c_xfer_send(&priv->i2c_props, buf, 2); +} + +static void tda8295_set_easy_mode(struct tuner *t, int enable) +{ + struct tda8290_priv *priv = t->priv; + unsigned char buf[] = { 0x01, 0x00 }; + + tuner_i2c_xfer_send(&priv->i2c_props, &buf[0], 1); + tuner_i2c_xfer_recv(&priv->i2c_props, &buf[1], 1); + + if (enable) + buf[1] = 0x01; /* rising edge sets regs 0x02 - 0x23 */ + else + buf[1] = 0x00; /* reset active bit */ + + tuner_i2c_xfer_send(&priv->i2c_props, buf, 2); +} + +static void tda8295_set_video_std(struct tuner *t) +{ + struct tda8290_priv *priv = t->priv; + unsigned char buf[] = { 0x00, priv->tda8290_easy_mode }; + + tuner_i2c_xfer_send(&priv->i2c_props, buf, 2); + + tda8295_set_easy_mode(t, 1); + msleep(20); + tda8295_set_easy_mode(t, 0); +} + +/*---------------------------------------------------------------------*/ + +static void tda8295_agc1_out(struct tuner *t, int enable) +{ + struct tda8290_priv *priv = t->priv; + unsigned char buf[] = { 0x02, 0x00 }; /* DIV_FUNC */ + + tuner_i2c_xfer_send(&priv->i2c_props, &buf[0], 1); + tuner_i2c_xfer_recv(&priv->i2c_props, &buf[1], 1); + + if (enable) + buf[1] &= ~0x40; + else + buf[1] |= 0x40; + + tuner_i2c_xfer_send(&priv->i2c_props, buf, 2); +} + +static void tda8295_agc2_out(struct tuner *t, int enable) +{ + struct tda8290_priv *priv = t->priv; + unsigned char set_gpio_cf[] = { 0x44, 0x00 }; + unsigned char set_gpio_val[] = { 0x46, 0x00 }; + + tuner_i2c_xfer_send(&priv->i2c_props, &set_gpio_cf[0], 1); + tuner_i2c_xfer_recv(&priv->i2c_props, &set_gpio_cf[1], 1); + tuner_i2c_xfer_send(&priv->i2c_props, &set_gpio_val[0], 1); + tuner_i2c_xfer_recv(&priv->i2c_props, &set_gpio_val[1], 1); + + set_gpio_cf[1] &= 0xf0; /* clear GPIO_0 bits 3-0 */ + + if (enable) { + set_gpio_cf[1] |= 0x01; /* config GPIO_0 as Open Drain Out */ + set_gpio_val[1] &= 0xfe; /* set GPIO_0 pin low */ + } + tuner_i2c_xfer_send(&priv->i2c_props, set_gpio_cf, 2); + tuner_i2c_xfer_send(&priv->i2c_props, set_gpio_val, 2); +} + +static int tda8295_has_signal(struct tuner *t) +{ + struct tda8290_priv *priv = t->priv; + + unsigned char hvpll_stat = 0x26; + unsigned char ret; + + tuner_i2c_xfer_send(&priv->i2c_props, &hvpll_stat, 1); + tuner_i2c_xfer_recv(&priv->i2c_props, &ret, 1); + return (ret & 0x01) ? 65535 : 0; +} + +/*---------------------------------------------------------------------*/ + +static void tda8295_set_freq(struct tuner *t, unsigned int freq) +{ + struct tda8290_priv *priv = t->priv; + u16 ifc; + + unsigned char blanking_mode[] = { 0x1d, 0x00 }; + + struct analog_parameters params = { + .frequency = freq, + .mode = t->mode, + .audmode = t->audmode, + .std = t->std + }; + + set_audio(t); + + ifc = priv->cfg.sgIF; /* FIXME */ + + tuner_dbg("%s: ifc = %u, freq = %d\n", __FUNCTION__, ifc, freq); + + tda8295_power(t, 1); + tda8295_agc1_out(t, 1); + + tuner_i2c_xfer_send(&priv->i2c_props, &blanking_mode[0], 1); + tuner_i2c_xfer_recv(&priv->i2c_props, &blanking_mode[1], 1); + + tda8295_set_video_std(t); + + blanking_mode[1] = 0x03; + tuner_i2c_xfer_send(&priv->i2c_props, blanking_mode, 2); + msleep(20); + + tda8295_i2c_bridge(t, 1); + + if (t->fe.ops.tuner_ops.set_analog_params) + t->fe.ops.tuner_ops.set_analog_params(&t->fe, ¶ms); + + if (priv->cfg.agcf) + priv->cfg.agcf(&t->fe); + + if (tda8295_has_signal(t)) + tuner_dbg("tda8295 is locked\n"); + else + tuner_dbg("tda8295 not locked, no signal?\n"); + + tda8295_i2c_bridge(t, 0); +} + +/*---------------------------------------------------------------------*/ + static int tda8290_has_signal(struct tuner *t) { struct tda8290_priv *priv = t->priv; @@ -264,6 +440,13 @@ static void tda8290_standby(struct tuner *t) tuner_i2c_xfer_send(&priv->i2c_props, tda8290_standby, 2); } +static void tda8295_standby(struct tuner *t) +{ + tda8295_agc1_out(t, 0); /* Put AGC in tri-state */ + + tda8295_power(t, 0); +} + static void tda8290_init_if(struct tuner *t) { struct tda8290_priv *priv = t->priv; @@ -279,6 +462,35 @@ static void tda8290_init_if(struct tuner *t) tuner_i2c_xfer_send(&priv->i2c_props, set_VS, 2); } +static void tda8295_init_if(struct tuner *t) +{ + struct tda8290_priv *priv = t->priv; + + static unsigned char set_adc_ctl[] = { 0x33, 0x14 }; + static unsigned char set_adc_ctl2[] = { 0x34, 0x00 }; + static unsigned char set_pll_reg6[] = { 0x3e, 0x63 }; + static unsigned char set_pll_reg0[] = { 0x38, 0x23 }; + static unsigned char set_pll_reg7[] = { 0x3f, 0x01 }; + static unsigned char set_pll_reg10[] = { 0x42, 0x61 }; + static unsigned char set_gpio_reg0[] = { 0x44, 0x0b }; + + tda8295_power(t, 1); + + tda8295_set_easy_mode(t, 0); + tda8295_set_video_std(t); + + tuner_i2c_xfer_send(&priv->i2c_props, set_adc_ctl, 2); + tuner_i2c_xfer_send(&priv->i2c_props, set_adc_ctl2, 2); + tuner_i2c_xfer_send(&priv->i2c_props, set_pll_reg6, 2); + tuner_i2c_xfer_send(&priv->i2c_props, set_pll_reg0, 2); + tuner_i2c_xfer_send(&priv->i2c_props, set_pll_reg7, 2); + tuner_i2c_xfer_send(&priv->i2c_props, set_pll_reg10, 2); + tuner_i2c_xfer_send(&priv->i2c_props, set_gpio_reg0, 2); + + tda8295_agc1_out(t, 0); + tda8295_agc2_out(t, 0); +} + static void tda8290_init_tuner(struct tuner *t) { struct tda8290_priv *priv = t->priv; @@ -298,7 +510,7 @@ static void tda8290_init_tuner(struct tuner *t) /*---------------------------------------------------------------------*/ -static void tda8290_release(struct tuner *t) +static void tda829x_release(struct tuner *t) { if (t->fe.ops.tuner_ops.release) t->fe.ops.tuner_ops.release(&t->fe); @@ -312,7 +524,15 @@ static struct tuner_operations tda8290_tuner_ops = { .set_radio_freq = tda8290_set_freq, .has_signal = tda8290_has_signal, .standby = tda8290_standby, - .release = tda8290_release, + .release = tda829x_release, +}; + +static struct tuner_operations tda8295_tuner_ops = { + .set_tv_freq = tda8295_set_freq, + .set_radio_freq = tda8295_set_freq, + .has_signal = tda8295_has_signal, + .standby = tda8295_standby, + .release = tda829x_release, }; int tda8290_attach(struct tuner *t) @@ -403,6 +623,95 @@ int tda8290_attach(struct tuner *t) tda8290_init_if(t); return 0; } +EXPORT_SYMBOL_GPL(tda8290_attach); + +int tda8295_attach(struct tuner *t) +{ + struct tda8290_priv *priv = NULL; + u8 data; + int i, ret, tuners_found; + u32 tuner_addrs; + struct i2c_msg msg = { .flags = I2C_M_RD, .buf = &data, .len = 1 }; + + priv = kzalloc(sizeof(struct tda8290_priv), GFP_KERNEL); + if (priv == NULL) + return -ENOMEM; + t->priv = priv; + + priv->i2c_props.addr = t->i2c.addr; + priv->i2c_props.adap = t->i2c.adapter; + + tda8295_i2c_bridge(t, 1); + /* probe for tuner chip */ + tuners_found = 0; + tuner_addrs = 0; + for (i = 0x60; i <= 0x63; i++) { + msg.addr = i; + ret = i2c_transfer(priv->i2c_props.adap, &msg, 1); + if (ret == 1) { + tuners_found++; + tuner_addrs = (tuner_addrs << 8) + i; + } + } + /* if there is more than one tuner, we expect the right one is + behind the bridge and we choose the highest address that doesn't + give a response now + */ + tda8295_i2c_bridge(t, 0); + if (tuners_found > 1) + for (i = 0; i < tuners_found; i++) { + msg.addr = tuner_addrs & 0xff; + ret = i2c_transfer(priv->i2c_props.adap, &msg, 1); + if (ret == 1) + tuner_addrs = tuner_addrs >> 8; + else + break; + } + if (tuner_addrs == 0) { + tuner_addrs = 0x60; + tuner_info("could not clearly identify tuner address, " + "defaulting to %x\n", tuner_addrs); + } else { + tuner_addrs = tuner_addrs & 0xff; + tuner_info("setting tuner address to %x\n", tuner_addrs); + } + priv->tda827x_addr = tuner_addrs; + msg.addr = tuner_addrs; + + tda8295_i2c_bridge(t, 1); + ret = i2c_transfer(priv->i2c_props.adap, &msg, 1); + tda8295_i2c_bridge(t, 0); + if (ret != 1) + tuner_warn("TDA827x access failed!\n"); + if ((data & 0x3c) == 0) { + strlcpy(t->i2c.name, "tda8295+18271", sizeof(t->i2c.name)); + tda18271_attach(&t->fe, priv->tda827x_addr, + priv->i2c_props.adap); + priv->tda827x_ver = 4; + } else { + strlcpy(t->i2c.name, "tda8295+75a", sizeof(t->i2c.name)); + tda827x_attach(&t->fe, priv->tda827x_addr, + priv->i2c_props.adap, &priv->cfg); + + /* FIXME: tda827x module doesn't probe the tuner until + * tda827x_initial_sleep is called + */ + if (t->fe.ops.tuner_ops.sleep) + t->fe.ops.tuner_ops.sleep(&t->fe); + priv->tda827x_ver = 2; + } + priv->tda827x_ver |= 1; /* signifies 8295 vs 8290 */ + tuner_info("type set to %s\n", t->i2c.name); + + memcpy(&t->ops, &tda8295_tuner_ops, sizeof(struct tuner_operations)); + + priv->cfg.tda827x_lpsel = 0; + t->mode = V4L2_TUNER_ANALOG_TV; + + tda8295_init_if(t); + return 0; +} +EXPORT_SYMBOL_GPL(tda8295_attach); int tda8290_probe(struct tuner *t) { @@ -434,12 +743,10 @@ int tda8290_probe(struct tuner *t) tuner_i2c_xfer_send(&i2c_props, restore_9886, 3); return -1; } - EXPORT_SYMBOL_GPL(tda8290_probe); -EXPORT_SYMBOL_GPL(tda8290_attach); MODULE_DESCRIPTION("Philips TDA8290 + TDA8275 / TDA8275a tuner driver"); -MODULE_AUTHOR("Gerd Knorr, Hartmut Hackmann"); +MODULE_AUTHOR("Gerd Knorr, Hartmut Hackmann, Michael Krufky"); MODULE_LICENSE("GPL"); /* diff --git a/drivers/media/video/tda8290.h b/drivers/media/video/tda8290.h index 9b63e62b3a0f..dbbcb0f001e0 100644 --- a/drivers/media/video/tda8290.h +++ b/drivers/media/video/tda8290.h @@ -24,6 +24,7 @@ extern int tda8290_probe(struct tuner *t); extern int tda8290_attach(struct tuner *t); +extern int tda8295_attach(struct tuner *t); #else static inline int tda8290_probe(struct tuner *t) { @@ -37,6 +38,13 @@ static inline int tda8290_attach(struct tuner *t) __FUNCTION__); return -EINVAL; } + +static inline int tda8295_attach(struct tuner *t) +{ + printk(KERN_INFO "%s: not probed - driver disabled by Kconfig\n", + __FUNCTION__); + return -EINVAL; +} #endif #endif /* __TDA8290_H__ */ diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c index 99558c708ae7..6ae8cb205d8e 100644 --- a/drivers/media/video/tuner-core.c +++ b/drivers/media/video/tuner-core.c @@ -278,6 +278,11 @@ static void set_type(struct i2c_client *c, unsigned int type, tda8290_attach(t); break; } + case TUNER_PHILIPS_TDA8295: + { + tda8295_attach(t); + break; + } case TUNER_TEA5767: if (tea5767_attach(&t->fe, t->i2c.adapter, t->i2c.addr) == NULL) { t->type = TUNER_ABSENT; diff --git a/drivers/media/video/tuner-types.c b/drivers/media/video/tuner-types.c index ac363f019227..4a674c436cf4 100644 --- a/drivers/media/video/tuner-types.c +++ b/drivers/media/video/tuner-types.c @@ -1475,6 +1475,9 @@ struct tunertype tuners[] = { .name = "Philips TEA5761 FM Radio", /* see tea5767.c for details */ }, + [TUNER_PHILIPS_TDA8295] = { /* Philips PAL|NTSC */ + .name = "tda8295+18271", + /* see tda8290.c for details */ }, }; unsigned const int tuner_count = ARRAY_SIZE(tuners); diff --git a/drivers/media/video/tveeprom.c b/drivers/media/video/tveeprom.c index 4b2c4034f5b3..27bfe6d3a33c 100644 --- a/drivers/media/video/tveeprom.c +++ b/drivers/media/video/tveeprom.c @@ -257,7 +257,7 @@ hauppauge_tuner[] = { TUNER_ABSENT, "LG TAPQ_H702F"}, { TUNER_ABSENT, "TCL M09WPP_4N_E"}, { TUNER_ABSENT, "MaxLinear MXL5005_v2"}, - { TUNER_ABSENT, "Philips 18271_8295"}, + { TUNER_PHILIPS_TDA8295, "Philips 18271_8295"}, }; static struct HAUPPAUGE_AUDIOIC diff --git a/include/media/tuner.h b/include/media/tuner.h index d49392d90e56..faacd2d50b72 100644 --- a/include/media/tuner.h +++ b/include/media/tuner.h @@ -124,6 +124,8 @@ extern int tuner_debug; #define TUNER_TDA9887 74 /* This tuner should be used only internally */ #define TUNER_TEA5761 75 /* Only FM Radio Tuner */ +#define TUNER_PHILIPS_TDA8295 76 + /* tv card specific */ #define TDA9887_PRESENT (1<<0) #define TDA9887_PORT1_INACTIVE (1<<1) -- cgit v1.2.3 From 0b5f4a12e628893624b478523fb4e2708c605710 Mon Sep 17 00:00:00 2001 From: Michael Krufky Date: Sat, 27 Oct 2007 13:09:16 -0300 Subject: V4L/DVB (6469): tuner: remove TUNER_PHILIPS_TDA8295 TUNER_PHILIPS_TDA8290 will autodetect a TDA8290 or a TDA8295, so we don't need this separate entry anymore. Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.tuner | 3 +-- drivers/media/video/tuner-core.c | 1 - drivers/media/video/tuner-types.c | 5 +---- drivers/media/video/tveeprom.c | 2 +- include/media/tuner.h | 2 -- 5 files changed, 3 insertions(+), 10 deletions(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.tuner b/Documentation/video4linux/CARDLIST.tuner index ac47b48715dd..2211c8eac643 100644 --- a/Documentation/video4linux/CARDLIST.tuner +++ b/Documentation/video4linux/CARDLIST.tuner @@ -52,7 +52,7 @@ tuner=50 - TCL 2002N tuner=51 - Philips PAL/SECAM_D (FM 1256 I-H3) tuner=52 - Thomson DTT 7610 (ATSC/NTSC) tuner=53 - Philips FQ1286 -tuner=54 - tda8290+75 +tuner=54 - Philips/NXP TDA 8290/8295 + 8275/8275A/18271 tuner=55 - TCL 2002MB tuner=56 - Philips PAL/SECAM multi (FQ1216AME MK4) tuner=57 - Philips FQ1236A MK4 @@ -73,4 +73,3 @@ tuner=71 - Xceive xc2028/xc3028 tuner tuner=72 - Thomson FE6600 tuner=73 - Samsung TCPG 6121P30A tuner=75 - Philips TEA5761 FM Radio -tuner=76 - tda8295+18271 diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c index 6ab57ec3f5a3..600174239d09 100644 --- a/drivers/media/video/tuner-core.c +++ b/drivers/media/video/tuner-core.c @@ -295,7 +295,6 @@ static void set_type(struct i2c_client *c, unsigned int type, microtune_attach(&t->fe, t->i2c.adapter, t->i2c.addr); break; case TUNER_PHILIPS_TDA8290: - case TUNER_PHILIPS_TDA8295: { tda829x_attach(t); break; diff --git a/drivers/media/video/tuner-types.c b/drivers/media/video/tuner-types.c index 4a674c436cf4..e2cd05a05802 100644 --- a/drivers/media/video/tuner-types.c +++ b/drivers/media/video/tuner-types.c @@ -1366,7 +1366,7 @@ struct tunertype tuners[] = { .count = ARRAY_SIZE(tuner_philips_fq1286_params), }, [TUNER_PHILIPS_TDA8290] = { /* Philips PAL|NTSC */ - .name = "tda8290+75", + .name = "Philips/NXP TDA 8290/8295 + 8275/8275A/18271", /* see tda8290.c for details */ }, [TUNER_TCL_2002MB] = { /* TCL PAL */ .name = "TCL 2002MB", @@ -1475,9 +1475,6 @@ struct tunertype tuners[] = { .name = "Philips TEA5761 FM Radio", /* see tea5767.c for details */ }, - [TUNER_PHILIPS_TDA8295] = { /* Philips PAL|NTSC */ - .name = "tda8295+18271", - /* see tda8290.c for details */ }, }; unsigned const int tuner_count = ARRAY_SIZE(tuners); diff --git a/drivers/media/video/tveeprom.c b/drivers/media/video/tveeprom.c index 27bfe6d3a33c..0faa14992353 100644 --- a/drivers/media/video/tveeprom.c +++ b/drivers/media/video/tveeprom.c @@ -257,7 +257,7 @@ hauppauge_tuner[] = { TUNER_ABSENT, "LG TAPQ_H702F"}, { TUNER_ABSENT, "TCL M09WPP_4N_E"}, { TUNER_ABSENT, "MaxLinear MXL5005_v2"}, - { TUNER_PHILIPS_TDA8295, "Philips 18271_8295"}, + { TUNER_PHILIPS_TDA8290, "Philips 18271_8295"}, }; static struct HAUPPAUGE_AUDIOIC diff --git a/include/media/tuner.h b/include/media/tuner.h index faacd2d50b72..d49392d90e56 100644 --- a/include/media/tuner.h +++ b/include/media/tuner.h @@ -124,8 +124,6 @@ extern int tuner_debug; #define TUNER_TDA9887 74 /* This tuner should be used only internally */ #define TUNER_TEA5761 75 /* Only FM Radio Tuner */ -#define TUNER_PHILIPS_TDA8295 76 - /* tv card specific */ #define TDA9887_PRESENT (1<<0) #define TDA9887_PORT1_INACTIVE (1<<1) -- cgit v1.2.3 From 7d070e26a868bd72f0d32189e0f3ceba47d2db82 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Thu, 1 Nov 2007 21:52:58 -0300 Subject: V4L/DVB (6519): Fix HVR900/HVR950 entry - Television is now default; - Add HVR950 name at the entry. Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.em28xx | 2 +- drivers/media/video/em28xx/em28xx-cards.c | 10 +++++----- 2 files changed, 6 insertions(+), 6 deletions(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.em28xx b/Documentation/video4linux/CARDLIST.em28xx index 37f0e3cedf43..abdcc68d47f9 100644 --- a/Documentation/video4linux/CARDLIST.em28xx +++ b/Documentation/video4linux/CARDLIST.em28xx @@ -8,7 +8,7 @@ 7 -> Leadtek Winfast USB II (em2800) 8 -> Kworld USB2800 (em2800) 9 -> Pinnacle Dazzle DVC 90 (em2820/em2840) [2304:0207] - 10 -> Hauppauge WinTV HVR 900 (em2880) + 10 -> Hauppauge WinTV HVR 900/950 (em2880) 11 -> Terratec Hybrid XS (em2880) 12 -> Kworld PVR TV 2800 RF (em2820/em2840) 13 -> Terratec Prodigy XS (em2880) diff --git a/drivers/media/video/em28xx/em28xx-cards.c b/drivers/media/video/em28xx/em28xx-cards.c index b56e0a70d8f8..9a6a4541e3c7 100644 --- a/drivers/media/video/em28xx/em28xx-cards.c +++ b/drivers/media/video/em28xx/em28xx-cards.c @@ -158,7 +158,7 @@ struct em28xx_board em28xx_boards[] = { }}, }, [EM2880_BOARD_HAUPPAUGE_WINTV_HVR_900] = { - .name = "Hauppauge WinTV HVR 900", + .name = "Hauppauge WinTV HVR 900/950", .vchannels = 3, .norm = VIDEO_MODE_PAL, .tda9887_conf = TDA9887_PRESENT, @@ -166,13 +166,13 @@ struct em28xx_board em28xx_boards[] = { .has_tuner = 1, .decoder = EM28XX_TVP5150, .input = {{ - .type = EM28XX_VMUX_COMPOSITE1, - .vmux = TVP5150_COMPOSITE1, - .amux = 1, - },{ .type = EM28XX_VMUX_TELEVISION, .vmux = TVP5150_COMPOSITE0, .amux = 0, + },{ + .type = EM28XX_VMUX_COMPOSITE1, + .vmux = TVP5150_COMPOSITE1, + .amux = 1, },{ .type = EM28XX_VMUX_SVIDEO, .vmux = TVP5150_SVIDEO, -- cgit v1.2.3 From 3dbd85ba36ff74af87550e76f5765a768b108409 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Sat, 3 Nov 2007 08:07:07 -0300 Subject: V4L/DVB (6535): Fix: Adds the generic PCI IDs for em28xx Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.em28xx | 4 ++-- drivers/media/video/em28xx/em28xx-cards.c | 25 +++++++++---------------- drivers/media/video/em28xx/em28xx-video.c | 14 +++++--------- 3 files changed, 16 insertions(+), 27 deletions(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.em28xx b/Documentation/video4linux/CARDLIST.em28xx index abdcc68d47f9..f947dca5ac48 100644 --- a/Documentation/video4linux/CARDLIST.em28xx +++ b/Documentation/video4linux/CARDLIST.em28xx @@ -1,9 +1,9 @@ 0 -> Unknown EM2800 video grabber (em2800) [eb1a:2800] - 1 -> Unknown EM2820/2840 video grabber (em2820/em2840) + 1 -> Unknown EM2750/28xx video grabber (em2820/em2840) [eb1a:2750,eb1a:2820,eb1a:2821,eb1a:2860,eb1a:2861,eb1a:2870,eb1a:2881,eb1a:2883] 2 -> Terratec Cinergy 250 USB (em2820/em2840) [0ccd:0036] 3 -> Pinnacle PCTV USB 2 (em2820/em2840) [2304:0208] 4 -> Hauppauge WinTV USB 2 (em2820/em2840) [2040:4200] - 5 -> MSI VOX USB 2.0 (em2820/em2840) [eb1a:2820] + 5 -> MSI VOX USB 2.0 (em2820/em2840) 6 -> Terratec Cinergy 200 USB (em2800) 7 -> Leadtek Winfast USB II (em2800) 8 -> Kworld USB2800 (em2800) diff --git a/drivers/media/video/em28xx/em28xx-cards.c b/drivers/media/video/em28xx/em28xx-cards.c index 9a6a4541e3c7..05264c655b5e 100644 --- a/drivers/media/video/em28xx/em28xx-cards.c +++ b/drivers/media/video/em28xx/em28xx-cards.c @@ -57,22 +57,8 @@ struct em28xx_board em28xx_boards[] = { }}, }, [EM2820_BOARD_UNKNOWN] = { - .name = "Unknown EM2820/2840 video grabber", + .name = "Unknown EM2750/28xx video grabber", .is_em2800 = 0, - .vchannels = 2, - .norm = VIDEO_MODE_PAL, - .tda9887_conf = TDA9887_PRESENT, - .has_tuner = 1, - .decoder = EM28XX_SAA7113, - .input = {{ - .type = EM28XX_VMUX_COMPOSITE1, - .vmux = SAA7115_COMPOSITE0, - .amux = 1, - },{ - .type = EM28XX_VMUX_SVIDEO, - .vmux = SAA7115_SVIDEO3, - .amux = 1, - }}, }, [EM2820_BOARD_KWORLD_PVRTV2800RF] = { .name = "Kworld PVR TV 2800 RF", @@ -337,8 +323,15 @@ const unsigned int em28xx_bcount = ARRAY_SIZE(em28xx_boards); /* table of devices that work with this driver */ struct usb_device_id em28xx_id_table [] = { + { USB_DEVICE(0xeb1a, 0x2750), .driver_info = EM2820_BOARD_UNKNOWN }, { USB_DEVICE(0xeb1a, 0x2800), .driver_info = EM2800_BOARD_UNKNOWN }, - { USB_DEVICE(0xeb1a, 0x2820), .driver_info = EM2820_BOARD_MSI_VOX_USB_2 }, + { USB_DEVICE(0xeb1a, 0x2820), .driver_info = EM2820_BOARD_UNKNOWN }, + { USB_DEVICE(0xeb1a, 0x2821), .driver_info = EM2820_BOARD_UNKNOWN }, + { USB_DEVICE(0xeb1a, 0x2860), .driver_info = EM2820_BOARD_UNKNOWN }, + { USB_DEVICE(0xeb1a, 0x2861), .driver_info = EM2820_BOARD_UNKNOWN }, + { USB_DEVICE(0xeb1a, 0x2870), .driver_info = EM2820_BOARD_UNKNOWN }, + { USB_DEVICE(0xeb1a, 0x2881), .driver_info = EM2820_BOARD_UNKNOWN }, + { USB_DEVICE(0xeb1a, 0x2883), .driver_info = EM2820_BOARD_UNKNOWN }, { USB_DEVICE(0x0ccd, 0x0036), .driver_info = EM2820_BOARD_TERRATEC_CINERGY_250 }, { USB_DEVICE(0x2304, 0x0208), .driver_info = EM2820_BOARD_PINNACLE_USB_2 }, { USB_DEVICE(0x2040, 0x4200), .driver_info = EM2820_BOARD_HAUPPAUGE_WINTV_USB_2 }, diff --git a/drivers/media/video/em28xx/em28xx-video.c b/drivers/media/video/em28xx/em28xx-video.c index bb0933f9d84c..fb0fa57dec92 100644 --- a/drivers/media/video/em28xx/em28xx-video.c +++ b/drivers/media/video/em28xx/em28xx-video.c @@ -1782,15 +1782,11 @@ static int em28xx_usb_probe(struct usb_interface *interface, model=card[nr]; if ((model==EM2800_BOARD_UNKNOWN)||(model==EM2820_BOARD_UNKNOWN)) { - em28xx_errdev( "Your board has no eeprom inside it and thus can't\n" - "%s: be autodetected. Please pass card= insmod option to\n" - "%s: workaround that. Redirect complaints to the vendor of\n" - "%s: the TV card. Generic type will be used." - "%s: Best regards,\n" - "%s: -- tux\n", - dev->name,dev->name,dev->name,dev->name,dev->name); - em28xx_errdev("%s: Here is a list of valid choices for the card= insmod option:\n", - dev->name); + em28xx_errdev("Your board has no unique USB ID and thus can't be autodetected.\n"); + em28xx_errdev("Please pass card= insmod option to workaround that.\n"); + em28xx_errdev("If there isn't any card number for you, please send an email to:\n"); + em28xx_errdev("\tV4L Mailing List \n"); + em28xx_errdev("Here is a list of valid choices for the card= insmod option:\n"); for (i = 0; i < em28xx_bcount; i++) { em28xx_errdev(" card=%d -> %s\n", i, em28xx_boards[i].name); -- cgit v1.2.3 From ea4fd5679b258d8ae85124a47b587a53ba6409de Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Sat, 3 Nov 2007 21:21:57 -0300 Subject: V4L/DVB (6537): Add entry for Pixelview Prolink PlayTV USB 2.0 Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.em28xx | 1 + drivers/media/video/em28xx/em28xx-cards.c | 23 ++++++++++++++++++++++- drivers/media/video/em28xx/em28xx.h | 1 + 3 files changed, 24 insertions(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.em28xx b/Documentation/video4linux/CARDLIST.em28xx index f947dca5ac48..b3a8181343a4 100644 --- a/Documentation/video4linux/CARDLIST.em28xx +++ b/Documentation/video4linux/CARDLIST.em28xx @@ -12,3 +12,4 @@ 11 -> Terratec Hybrid XS (em2880) 12 -> Kworld PVR TV 2800 RF (em2820/em2840) 13 -> Terratec Prodigy XS (em2880) + 14 -> Pixelview Prolink PlayTV USB 2.0 (em2820/em2840) diff --git a/drivers/media/video/em28xx/em28xx-cards.c b/drivers/media/video/em28xx/em28xx-cards.c index 4cd49415eef2..e1e7510146e1 100644 --- a/drivers/media/video/em28xx/em28xx-cards.c +++ b/drivers/media/video/em28xx/em28xx-cards.c @@ -328,6 +328,26 @@ struct em28xx_board em28xx_boards[] = { .amux = 1, }}, }, + [EM2820_BOARD_PROLINK_PLAYTV_USB2] = { + .name = "Pixelview Prolink PlayTV USB 2.0", + .vchannels = 3, + .tda9887_conf = TDA9887_PRESENT, + .has_tuner = 1, + .decoder = EM28XX_SAA7113, + .input = {{ + .type = EM28XX_VMUX_TELEVISION, + .vmux = SAA7115_COMPOSITE2, + .amux = 1, + },{ + .type = EM28XX_VMUX_COMPOSITE1, + .vmux = SAA7115_COMPOSITE0, + .amux = 1, + },{ + .type = EM28XX_VMUX_SVIDEO, + .vmux = SAA7115_SVIDEO3, + .amux = 1, + }}, + }, }; const unsigned int em28xx_bcount = ARRAY_SIZE(em28xx_boards); @@ -355,7 +375,8 @@ struct usb_device_id em28xx_id_table [] = { MODULE_DEVICE_TABLE (usb, em28xx_id_table); static struct em28xx_hash_table em28xx_hash [] = { - { 0, 0, 0 }, + /* P/N: SA 60002070465 Tuner: TVF7533-MF */ + { 0x6ce05a8f, EM2820_BOARD_PROLINK_PLAYTV_USB2, TUNER_YMEC_TVF_5533MF }, }; /* Since em28xx_pre_card_setup() requires a proper dev->model, diff --git a/drivers/media/video/em28xx/em28xx.h b/drivers/media/video/em28xx/em28xx.h index e98dac7fc05b..209f6f9d5581 100644 --- a/drivers/media/video/em28xx/em28xx.h +++ b/drivers/media/video/em28xx/em28xx.h @@ -46,6 +46,7 @@ #define EM2880_BOARD_TERRATEC_HYBRID_XS 11 #define EM2820_BOARD_KWORLD_PVRTV2800RF 12 #define EM2880_BOARD_TERRATEC_PRODIGY_XS 13 +#define EM2820_BOARD_PROLINK_PLAYTV_USB2 14 #define UNSET -1 -- cgit v1.2.3 From 9b15c0251ba9d20dd88a69b2c607a2f2e5f0133a Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Sat, 3 Nov 2007 22:04:33 -0300 Subject: V4L/DVB (6541): Add V-Gear PocketTV to Cardlist.em28xx Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.em28xx | 1 + 1 file changed, 1 insertion(+) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.em28xx b/Documentation/video4linux/CARDLIST.em28xx index b3a8181343a4..f809834af12d 100644 --- a/Documentation/video4linux/CARDLIST.em28xx +++ b/Documentation/video4linux/CARDLIST.em28xx @@ -13,3 +13,4 @@ 12 -> Kworld PVR TV 2800 RF (em2820/em2840) 13 -> Terratec Prodigy XS (em2880) 14 -> Pixelview Prolink PlayTV USB 2.0 (em2820/em2840) + 15 -> V-Gear PocketTV (em2800) -- cgit v1.2.3 From ddef2dcc4ead7a9412533202a42c40648e620f43 Mon Sep 17 00:00:00 2001 From: Luca Risolia Date: Mon, 12 Nov 2007 11:42:54 -0300 Subject: V4L/DVB (6591): Adds support for MT9V111 on sn9c102 Adds a new image sensor to the sn9c102 driver. Signed-off-by: Luca Risolia Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/sn9c102.txt | 1 + drivers/media/video/sn9c102/Makefile | 1 + drivers/media/video/sn9c102/sn9c102_core.c | 4 +- drivers/media/video/sn9c102/sn9c102_devtable.h | 2 + drivers/media/video/sn9c102/sn9c102_mt9v111.c | 259 +++++++++++++++++++++++++ 5 files changed, 265 insertions(+), 2 deletions(-) create mode 100644 drivers/media/video/sn9c102/sn9c102_mt9v111.c (limited to 'Documentation') diff --git a/Documentation/video4linux/sn9c102.txt b/Documentation/video4linux/sn9c102.txt index 1ffad19ce891..b26f5195af51 100644 --- a/Documentation/video4linux/sn9c102.txt +++ b/Documentation/video4linux/sn9c102.txt @@ -568,6 +568,7 @@ the fingerprint is: '88E8 F32F 7244 68BA 3958 5D40 99DA 5D2A FCE6 35A4'. Many thanks to following persons for their contribute (listed in alphabetical order): +- David Anderson for the donation of a webcam; - Luca Capello for the donation of a webcam; - Philippe Coval for having helped testing the PAS202BCA image sensor; - Joao Rodrigo Fuzaro, Joao Limirio, Claudio Filho and Caio Begotti for the diff --git a/drivers/media/video/sn9c102/Makefile b/drivers/media/video/sn9c102/Makefile index a56d16f69c71..7ecd5a90c7c9 100644 --- a/drivers/media/video/sn9c102/Makefile +++ b/drivers/media/video/sn9c102/Makefile @@ -3,6 +3,7 @@ sn9c102-objs := sn9c102_core.o \ sn9c102_hv7131r.o \ sn9c102_mi0343.o \ sn9c102_mi0360.o \ + sn9c102_mt9v111.o \ sn9c102_ov7630.o \ sn9c102_ov7660.o \ sn9c102_pas106b.o \ diff --git a/drivers/media/video/sn9c102/sn9c102_core.c b/drivers/media/video/sn9c102/sn9c102_core.c index 511847912c48..c40ba3adab21 100644 --- a/drivers/media/video/sn9c102/sn9c102_core.c +++ b/drivers/media/video/sn9c102/sn9c102_core.c @@ -47,7 +47,7 @@ #define SN9C102_MODULE_AUTHOR "(C) 2004-2007 Luca Risolia" #define SN9C102_AUTHOR_EMAIL "" #define SN9C102_MODULE_LICENSE "GPL" -#define SN9C102_MODULE_VERSION "1:1.47" +#define SN9C102_MODULE_VERSION "1:1.47pre49" #define SN9C102_MODULE_VERSION_CODE KERNEL_VERSION(1, 1, 47) /*****************************************************************************/ @@ -3322,7 +3322,6 @@ sn9c102_usb_probe(struct usb_interface* intf, const struct usb_device_id* id) cam->v4ldev->fops = &sn9c102_fops; cam->v4ldev->minor = video_nr[dev_nr]; cam->v4ldev->release = video_device_release; - video_set_drvdata(cam->v4ldev, cam); init_completion(&cam->probe); @@ -3340,6 +3339,7 @@ sn9c102_usb_probe(struct usb_interface* intf, const struct usb_device_id* id) DBG(2, "V4L2 device registered as /dev/video%d", cam->v4ldev->minor); + video_set_drvdata(cam->v4ldev, cam); cam->module_param.force_munmap = force_munmap[dev_nr]; cam->module_param.frame_timeout = frame_timeout[dev_nr]; diff --git a/drivers/media/video/sn9c102/sn9c102_devtable.h b/drivers/media/video/sn9c102/sn9c102_devtable.h index 916054faf9be..35223e0d7e49 100644 --- a/drivers/media/video/sn9c102/sn9c102_devtable.h +++ b/drivers/media/video/sn9c102/sn9c102_devtable.h @@ -126,6 +126,7 @@ extern int sn9c102_probe_hv7131d(struct sn9c102_device* cam); extern int sn9c102_probe_hv7131r(struct sn9c102_device* cam); extern int sn9c102_probe_mi0343(struct sn9c102_device* cam); extern int sn9c102_probe_mi0360(struct sn9c102_device* cam); +extern int sn9c102_probe_mt9v111(struct sn9c102_device *cam); extern int sn9c102_probe_ov7630(struct sn9c102_device* cam); extern int sn9c102_probe_ov7660(struct sn9c102_device* cam); extern int sn9c102_probe_pas106b(struct sn9c102_device* cam); @@ -144,6 +145,7 @@ static int (*sn9c102_sensor_table[])(struct sn9c102_device*) = { &sn9c102_probe_hv7131r, /* strong detection based on SENSOR ids */ &sn9c102_probe_mi0343, /* strong detection based on SENSOR ids */ &sn9c102_probe_mi0360, /* strong detection based on SENSOR ids */ + &sn9c102_probe_mt9v111, /* strong detection based on SENSOR ids */ &sn9c102_probe_pas106b, /* strong detection based on SENSOR ids */ &sn9c102_probe_pas202bcb, /* strong detection based on SENSOR ids */ &sn9c102_probe_ov7630, /* strong detection based on SENSOR ids */ diff --git a/drivers/media/video/sn9c102/sn9c102_mt9v111.c b/drivers/media/video/sn9c102/sn9c102_mt9v111.c new file mode 100644 index 000000000000..3b98ac3bbc38 --- /dev/null +++ b/drivers/media/video/sn9c102/sn9c102_mt9v111.c @@ -0,0 +1,259 @@ +/*************************************************************************** + * Plug-in for MT9V111 image sensor connected to the SN9C1xx PC Camera * + * Controllers * + * * + * Copyright (C) 2007 by Luca Risolia * + * * + * This program is free software; you can redistribute it and/or modify * + * it under the terms of the GNU General Public License as published by * + * the Free Software Foundation; either version 2 of the License, or * + * (at your option) any later version. * + * * + * This program is distributed in the hope that it will be useful, * + * but WITHOUT ANY WARRANTY; without even the implied warranty of * + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * + * GNU General Public License for more details. * + * * + * You should have received a copy of the GNU General Public License * + * along with this program; if not, write to the Free Software * + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. * + ***************************************************************************/ + +#include "sn9c102_sensor.h" + + +static int mt9v111_init(struct sn9c102_device *cam) +{ + struct sn9c102_sensor *s = sn9c102_get_sensor(cam); + int err = 0; + + err = sn9c102_write_const_regs(cam, {0x44, 0x01}, {0x40, 0x02}, + {0x00, 0x03}, {0x1a, 0x04}, + {0x1f, 0x05}, {0x20, 0x06}, + {0x1f, 0x07}, {0x81, 0x08}, + {0x5c, 0x09}, {0x00, 0x0a}, + {0x00, 0x0b}, {0x00, 0x0c}, + {0x00, 0x0d}, {0x00, 0x0e}, + {0x00, 0x0f}, {0x03, 0x10}, + {0x00, 0x11}, {0x00, 0x12}, + {0x02, 0x13}, {0x14, 0x14}, + {0x28, 0x15}, {0x1e, 0x16}, + {0xe2, 0x17}, {0x06, 0x18}, + {0x00, 0x19}, {0x00, 0x1a}, + {0x00, 0x1b}, {0x08, 0x20}, + {0x39, 0x21}, {0x51, 0x22}, + {0x63, 0x23}, {0x73, 0x24}, + {0x82, 0x25}, {0x8f, 0x26}, + {0x9b, 0x27}, {0xa7, 0x28}, + {0xb1, 0x29}, {0xbc, 0x2a}, + {0xc6, 0x2b}, {0xcf, 0x2c}, + {0xd8, 0x2d}, {0xe1, 0x2e}, + {0xea, 0x2f}, {0xf2, 0x30}, + {0x13, 0x84}, {0x00, 0x85}, + {0x25, 0x86}, {0x00, 0x87}, + {0x07, 0x88}, {0x00, 0x89}, + {0xee, 0x8a}, {0x0f, 0x8b}, + {0xe5, 0x8c}, {0x0f, 0x8d}, + {0x2e, 0x8e}, {0x00, 0x8f}, + {0x30, 0x90}, {0x00, 0x91}, + {0xd4, 0x92}, {0x0f, 0x93}, + {0xfc, 0x94}, {0x0f, 0x95}, + {0x14, 0x96}, {0x00, 0x97}, + {0x00, 0x98}, {0x60, 0x99}, + {0x07, 0x9a}, {0x40, 0x9b}, + {0x20, 0x9c}, {0x00, 0x9d}, + {0x00, 0x9e}, {0x00, 0x9f}, + {0x2d, 0xc0}, {0x2d, 0xc1}, + {0x3a, 0xc2}, {0x05, 0xc3}, + {0x04, 0xc4}, {0x3f, 0xc5}, + {0x00, 0xc6}, {0x00, 0xc7}, + {0x50, 0xc8}, {0x3c, 0xc9}, + {0x28, 0xca}, {0xd8, 0xcb}, + {0x14, 0xcc}, {0xec, 0xcd}, + {0x32, 0xce}, {0xdd, 0xcf}, + {0x2d, 0xd0}, {0xdd, 0xd1}, + {0x6a, 0xd2}, {0x50, 0xd3}, + {0x60, 0xd4}, {0x00, 0xd5}, + {0x00, 0xd6}); + + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x01, + 0x00, 0x01, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x0d, + 0x00, 0x01, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x0d, + 0x00, 0x00, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x08, + 0x04, 0x80, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x01, + 0x00, 0x04, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x08, + 0x00, 0x08, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x02, + 0x00, 0x16, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x03, + 0x01, 0xe7, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x04, + 0x02, 0x87, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x06, + 0x00, 0x40, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x05, + 0x00, 0x09, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x07, + 0x30, 0x02, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x0c, + 0x00, 0x00, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x12, + 0x00, 0xb0, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x13, + 0x00, 0x7c, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x1e, + 0x00, 0x00, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x20, + 0x00, 0x00, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x20, + 0x00, 0x00, 0, 0); + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, 0x01, + 0x00, 0x04, 0, 0); + + return err; +} + +static int mt9v111_get_ctrl(struct sn9c102_device *cam, + struct v4l2_control *ctrl) +{ + struct sn9c102_sensor *s = sn9c102_get_sensor(cam); + u8 data[2]; + int err = 0; + + switch (ctrl->id) { + case V4L2_CID_VFLIP: + if (sn9c102_i2c_try_raw_read(cam, s, s->i2c_slave_id, 0x20, 2, + data) < 0) + return -EIO; + ctrl->value = data[1] & 0x80 ? 1 : 0; + return 0; + default: + return -EINVAL; + } + + return err ? -EIO : 0; +} + +static int mt9v111_set_ctrl(struct sn9c102_device *cam, + const struct v4l2_control *ctrl) +{ + struct sn9c102_sensor *s = sn9c102_get_sensor(cam); + int err = 0; + + switch (ctrl->id) { + case V4L2_CID_VFLIP: + err += sn9c102_i2c_try_raw_write(cam, s, 4, s->i2c_slave_id, + 0x20, + ctrl->value ? 0x80 : 0x00, + ctrl->value ? 0x80 : 0x00, 0, + 0); + break; + default: + return -EINVAL; + } + + return err ? -EIO : 0; +} + +static int mt9v111_set_crop(struct sn9c102_device *cam, + const struct v4l2_rect *rect) +{ + struct sn9c102_sensor *s = sn9c102_get_sensor(cam); + int err = 0; + u8 v_start = (u8) (rect->top - s->cropcap.bounds.top) + 2; + + err += sn9c102_write_reg(cam, v_start, 0x13); + + return err; +} + +static int mt9v111_set_pix_format(struct sn9c102_device *cam, + const struct v4l2_pix_format *pix) +{ + int err = 0; + + if (pix->pixelformat == V4L2_PIX_FMT_SBGGR8) { + err += sn9c102_write_reg(cam, 0xb4, 0x17); + } else { + err += sn9c102_write_reg(cam, 0xe2, 0x17); + } + + return err; +} + + +static const struct sn9c102_sensor mt9v111 = { + .name = "MT9V111", + .maintainer = "Luca Risolia ", + .supported_bridge = BRIDGE_SN9C105 | BRIDGE_SN9C120, + .frequency = SN9C102_I2C_100KHZ, + .interface = SN9C102_I2C_2WIRES, + .i2c_slave_id = 0x5c, + .init = &mt9v111_init, + .qctrl = { + { + .id = V4L2_CID_VFLIP, + .type = V4L2_CTRL_TYPE_BOOLEAN, + .name = "vertical mirror", + .minimum = 0, + .maximum = 1, + .step = 1, + .default_value = 0, + .flags = 0, + }, + }, + .get_ctrl = &mt9v111_get_ctrl, + .set_ctrl = &mt9v111_set_ctrl, + .cropcap = { + .bounds = { + .left = 0, + .top = 0, + .width = 640, + .height = 480, + }, + .defrect = { + .left = 0, + .top = 0, + .width = 640, + .height = 480, + }, + }, + .set_crop = &mt9v111_set_crop, + .pix_format = { + .width = 640, + .height = 480, + .pixelformat = V4L2_PIX_FMT_SBGGR8, + .priv = 8, + }, + .set_pix_format = &mt9v111_set_pix_format +}; + + +int sn9c102_probe_mt9v111(struct sn9c102_device *cam) +{ + u8 data[2]; + int err = 0; + + err += sn9c102_write_const_regs(cam, {0x01, 0xf1}, {0x00, 0xf1}, + {0x29, 0x01}, {0x42, 0x17}, + {0x62, 0x17}, {0x08, 0x01}); + err += sn9c102_i2c_try_raw_write(cam, &mt9v111, 4, + mt9v111.i2c_slave_id, 0x01, 0x00, + 0x04, 0, 0); + if (err || sn9c102_i2c_try_raw_read(cam, &mt9v111, + mt9v111.i2c_slave_id, 0x36, 2, + data) < 0) + return -EIO; + + if (data[0] != 0x82 || data[1] != 0x3a) + return -ENODEV; + + sn9c102_attach_sensor(cam, &mt9v111); + + return 0; +} -- cgit v1.2.3 From ddf12227c3f8413d09c54f85e6642f79312f1430 Mon Sep 17 00:00:00 2001 From: Michael Krufky Date: Thu, 15 Nov 2007 10:01:11 -0300 Subject: V4L/DVB (6606): saa7134: add autodetection support for alternate subids of Hauppauge HVR1110 Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.saa7134 | 2 +- drivers/media/video/saa7134/saa7134-cards.c | 30 +++++++++++++++++++++++++++++ 2 files changed, 31 insertions(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.saa7134 b/Documentation/video4linux/CARDLIST.saa7134 index a14545300e4c..29ca16cdb713 100644 --- a/Documentation/video4linux/CARDLIST.saa7134 +++ b/Documentation/video4linux/CARDLIST.saa7134 @@ -102,7 +102,7 @@ 101 -> Pinnacle PCTV 310i [11bd:002f] 102 -> Avermedia AVerTV Studio 507 [1461:9715] 103 -> Compro Videomate DVB-T200A -104 -> Hauppauge WinTV-HVR1110 DVB-T/Hybrid [0070:6701] +104 -> Hauppauge WinTV-HVR1110 DVB-T/Hybrid [0070:6700,0070:6701,0070:6702,0070:6703,0070:6704,0070:6705] 105 -> Terratec Cinergy HT PCMCIA [153b:1172] 106 -> Encore ENLTV [1131:2342,1131:2341,3016:2344] 107 -> Encore ENLTV-FM [1131:230f] diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c index 78d7d7059a9e..138cec8bb023 100644 --- a/drivers/media/video/saa7134/saa7134-cards.c +++ b/drivers/media/video/saa7134/saa7134-cards.c @@ -4203,12 +4203,42 @@ struct pci_device_id saa7134_pci_tbl[] = { .subvendor = 0x1043, .subdevice = 0x4876, .driver_data = SAA7134_BOARD_ASUSTeK_P7131_HYBRID_LNA, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x0070, + .subdevice = 0x6700, + .driver_data = SAA7134_BOARD_HAUPPAUGE_HVR1110, },{ .vendor = PCI_VENDOR_ID_PHILIPS, .device = PCI_DEVICE_ID_PHILIPS_SAA7133, .subvendor = 0x0070, .subdevice = 0x6701, .driver_data = SAA7134_BOARD_HAUPPAUGE_HVR1110, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x0070, + .subdevice = 0x6702, + .driver_data = SAA7134_BOARD_HAUPPAUGE_HVR1110, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x0070, + .subdevice = 0x6703, + .driver_data = SAA7134_BOARD_HAUPPAUGE_HVR1110, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x0070, + .subdevice = 0x6704, + .driver_data = SAA7134_BOARD_HAUPPAUGE_HVR1110, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x0070, + .subdevice = 0x6705, + .driver_data = SAA7134_BOARD_HAUPPAUGE_HVR1110, },{ .vendor = PCI_VENDOR_ID_PHILIPS, .device = PCI_DEVICE_ID_PHILIPS_SAA7133, -- cgit v1.2.3 From 122d15888a8aa95fb6ca32eb1307c3e3337f1a38 Mon Sep 17 00:00:00 2001 From: Aidan Thornton Date: Tue, 20 Nov 2007 15:25:08 -0300 Subject: V4L/DVB (6619): Use MTS firmware for the HVR-900 The HVR-900 requires the MTS version of the xc3028 firmware in order to get any sound. The below patch selects this firmware variant on HVR-900 cards, as well as splitting the HVR-950 into its own entry (since I don't know if it uses the MTS variant and it will have to be split off eventually anyway). Signed-off-by: Aidan Thornton Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.em28xx | 3 ++- drivers/media/video/em28xx/em28xx-cards.c | 29 +++++++++++++++++++++++++++-- drivers/media/video/em28xx/em28xx.h | 2 ++ 3 files changed, 31 insertions(+), 3 deletions(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.em28xx b/Documentation/video4linux/CARDLIST.em28xx index f809834af12d..94b53bd71301 100644 --- a/Documentation/video4linux/CARDLIST.em28xx +++ b/Documentation/video4linux/CARDLIST.em28xx @@ -8,9 +8,10 @@ 7 -> Leadtek Winfast USB II (em2800) 8 -> Kworld USB2800 (em2800) 9 -> Pinnacle Dazzle DVC 90 (em2820/em2840) [2304:0207] - 10 -> Hauppauge WinTV HVR 900/950 (em2880) + 10 -> Hauppauge WinTV HVR 900 (em2880) 11 -> Terratec Hybrid XS (em2880) 12 -> Kworld PVR TV 2800 RF (em2820/em2840) 13 -> Terratec Prodigy XS (em2880) 14 -> Pixelview Prolink PlayTV USB 2.0 (em2820/em2840) 15 -> V-Gear PocketTV (em2800) + 16 -> Hauppauge WinTV HVR 950 (em2880) diff --git a/drivers/media/video/em28xx/em28xx-cards.c b/drivers/media/video/em28xx/em28xx-cards.c index fd64058a04e2..c40b9d9b3070 100644 --- a/drivers/media/video/em28xx/em28xx-cards.c +++ b/drivers/media/video/em28xx/em28xx-cards.c @@ -149,7 +149,29 @@ struct em28xx_board em28xx_boards[] = { }}, }, [EM2880_BOARD_HAUPPAUGE_WINTV_HVR_900] = { - .name = "Hauppauge WinTV HVR 900/950", + .name = "Hauppauge WinTV HVR 900", + .vchannels = 3, + .tda9887_conf = TDA9887_PRESENT, + .tuner_type = TUNER_XC2028, + .has_tuner = 1, + .xc2028_type = XC2028_FIRM_MTS, + .decoder = EM28XX_TVP5150, + .input = {{ + .type = EM28XX_VMUX_TELEVISION, + .vmux = TVP5150_COMPOSITE0, + .amux = 0, + },{ + .type = EM28XX_VMUX_COMPOSITE1, + .vmux = TVP5150_COMPOSITE1, + .amux = 1, + },{ + .type = EM28XX_VMUX_SVIDEO, + .vmux = TVP5150_SVIDEO, + .amux = 1, + }}, + }, + [EM2880_BOARD_HAUPPAUGE_WINTV_HVR_950] = { + .name = "Hauppauge WinTV HVR 950", .vchannels = 3, .tda9887_conf = TDA9887_PRESENT, .tuner_type = TUNER_XC2028, @@ -376,7 +398,7 @@ struct usb_device_id em28xx_id_table [] = { { USB_DEVICE(0x2040, 0x4200), .driver_info = EM2820_BOARD_HAUPPAUGE_WINTV_USB_2 }, { USB_DEVICE(0x2304, 0x0207), .driver_info = EM2820_BOARD_PINNACLE_DVC_90 }, { USB_DEVICE(0x2040, 0x6500), .driver_info = EM2880_BOARD_HAUPPAUGE_WINTV_HVR_900 }, - { USB_DEVICE(0x2040, 0x6513), .driver_info = EM2880_BOARD_HAUPPAUGE_WINTV_HVR_900 }, + { USB_DEVICE(0x2040, 0x6513), .driver_info = EM2880_BOARD_HAUPPAUGE_WINTV_HVR_950 }, { USB_DEVICE(0x0ccd, 0x0042), .driver_info = EM2880_BOARD_TERRATEC_HYBRID_XS }, { USB_DEVICE(0x0ccd, 0x0047), .driver_info = EM2880_BOARD_TERRATEC_PRODIGY_XS }, { }, @@ -404,6 +426,7 @@ void em28xx_pre_card_setup(struct em28xx *dev) switch(dev->model){ case EM2880_BOARD_TERRATEC_PRODIGY_XS: case EM2880_BOARD_HAUPPAUGE_WINTV_HVR_900: + case EM2880_BOARD_HAUPPAUGE_WINTV_HVR_950: case EM2880_BOARD_TERRATEC_HYBRID_XS: /* reset through GPIO? */ em28xx_write_regs_req(dev, 0x00, 0x08, "\x7d", 1); @@ -453,6 +476,7 @@ static void em28xx_config_tuner (struct em28xx *dev) ctl.fname = XC2028_DEFAULT_FIRMWARE; ctl.max_len = 64; + ctl.type = em28xx_boards[dev->model].xc2028_type; xc2028_cfg.tuner = TUNER_XC2028; xc2028_cfg.priv = &ctl; @@ -574,6 +598,7 @@ void em28xx_card_setup(struct em28xx *dev) switch (dev->model) { case EM2820_BOARD_HAUPPAUGE_WINTV_USB_2: case EM2880_BOARD_HAUPPAUGE_WINTV_HVR_900: + case EM2880_BOARD_HAUPPAUGE_WINTV_HVR_950: { struct tveeprom tv; #ifdef CONFIG_MODULES diff --git a/drivers/media/video/em28xx/em28xx.h b/drivers/media/video/em28xx/em28xx.h index 8d045867dac7..db82b51c85ff 100644 --- a/drivers/media/video/em28xx/em28xx.h +++ b/drivers/media/video/em28xx/em28xx.h @@ -48,6 +48,7 @@ #define EM2880_BOARD_TERRATEC_PRODIGY_XS 13 #define EM2820_BOARD_PROLINK_PLAYTV_USB2 14 #define EM2800_BOARD_VGEAR_POCKETTV 15 +#define EM2880_BOARD_HAUPPAUGE_WINTV_HVR_950 16 #define UNSET -1 @@ -177,6 +178,7 @@ struct em28xx_board { unsigned int has_msp34xx:1; enum em28xx_decoder decoder; + int xc2028_type; struct em28xx_input input[MAX_EM28XX_INPUT]; }; -- cgit v1.2.3 From 5a9dfc34fcee89ca399cf201a0904675e87c9576 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 23 Nov 2007 14:20:24 -0300 Subject: V4L/DVB (6664): Add missing USB ID's at CARDLIST.em28xx Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.em28xx | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.em28xx b/Documentation/video4linux/CARDLIST.em28xx index 94b53bd71301..ceb277543f51 100644 --- a/Documentation/video4linux/CARDLIST.em28xx +++ b/Documentation/video4linux/CARDLIST.em28xx @@ -8,10 +8,10 @@ 7 -> Leadtek Winfast USB II (em2800) 8 -> Kworld USB2800 (em2800) 9 -> Pinnacle Dazzle DVC 90 (em2820/em2840) [2304:0207] - 10 -> Hauppauge WinTV HVR 900 (em2880) - 11 -> Terratec Hybrid XS (em2880) + 10 -> Hauppauge WinTV HVR 900 (em2880) [2040:6500] + 11 -> Terratec Hybrid XS (em2880) [0ccd:0042] 12 -> Kworld PVR TV 2800 RF (em2820/em2840) - 13 -> Terratec Prodigy XS (em2880) + 13 -> Terratec Prodigy XS (em2880) [0ccd:0047] 14 -> Pixelview Prolink PlayTV USB 2.0 (em2820/em2840) 15 -> V-Gear PocketTV (em2800) - 16 -> Hauppauge WinTV HVR 950 (em2880) + 16 -> Hauppauge WinTV HVR 950 (em2880) [2040:6513] -- cgit v1.2.3 From 4c5efc1778387de98d5348666e7bc4396c9a5404 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Sun, 25 Nov 2007 19:04:47 -0300 Subject: V4L/DVB (6665a): finish the VID_HARDWARE_* removal This patch removes a few remainders of the VID_HARDWARE_* removal. Signed-off-by: Adrian Bunk Signed-off-by: Mauro Carvalho Chehab 643d01fb38b6f376cced035549f4e193018776e7 --- Documentation/DocBook/videobook.tmpl | 9 --------- drivers/media/video/usbvision/usbvision.h | 4 ---- 2 files changed, 13 deletions(-) (limited to 'Documentation') diff --git a/Documentation/DocBook/videobook.tmpl b/Documentation/DocBook/videobook.tmpl index b629da33951d..b3d93ee27693 100644 --- a/Documentation/DocBook/videobook.tmpl +++ b/Documentation/DocBook/videobook.tmpl @@ -96,7 +96,6 @@ static struct video_device my_radio { "My radio", VID_TYPE_TUNER, - VID_HARDWARE_MYRADIO, radio_open. radio_close, NULL, /* no read */ @@ -118,13 +117,6 @@ static struct video_device my_radio indicates that the device can be tuned. Clearly our radio is going to have some way to change channel so it is tuneable. - - The VID_HARDWARE_ types are unique to each device. Numbers are assigned by - alan@redhat.com when device drivers are going to be released. Until then you - can pull a suitably large number out of your hat and use it. 10000 should be - safe for a very long time even allowing for the huge number of vendors - making new and different radio cards at the moment. - We declare an open and close routine, but we do not need read or write, which are used to read and write video data to or from the card itself. As @@ -844,7 +836,6 @@ static struct video_device my_camera "My Camera", VID_TYPE_OVERLAY|VID_TYPE_SCALES|\ VID_TYPE_CAPTURE|VID_TYPE_CHROMAKEY, - VID_HARDWARE_MYCAMERA, camera_open. camera_close, camera_read, /* no read */ diff --git a/drivers/media/video/usbvision/usbvision.h b/drivers/media/video/usbvision/usbvision.h index c32f68566bf9..dd80a98780f2 100644 --- a/drivers/media/video/usbvision/usbvision.h +++ b/drivers/media/video/usbvision/usbvision.h @@ -41,10 +41,6 @@ #define USBVISION_DEBUG /* Turn on debug messages */ -#ifndef VID_HARDWARE_USBVISION - #define VID_HARDWARE_USBVISION 34 /* USBVision Video Grabber */ -#endif - #define USBVISION_PWR_REG 0x00 #define USBVISION_SSPND_EN (1 << 1) #define USBVISION_RES2 (1 << 2) -- cgit v1.2.3 From e08323f099b5aba28610a856fa7d21d0d86fd4f0 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Fri, 7 Dec 2007 20:40:16 -0300 Subject: V4L/DVB (6763): ivtv: add AVerMedia EZMaker PCI Deluxe support Add support for the AVerMedia EZMaker PCI Deluxe and update the ivtv cardlist. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.ivtv | 6 ++++++ drivers/media/video/ivtv/ivtv-cards.c | 29 +++++++++++++++++++++++++++++ drivers/media/video/ivtv/ivtv-cards.h | 3 ++- drivers/media/video/ivtv/ivtv-driver.c | 1 + 4 files changed, 38 insertions(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.ivtv b/Documentation/video4linux/CARDLIST.ivtv index ddd76a0eb100..a019e27e42b3 100644 --- a/Documentation/video4linux/CARDLIST.ivtv +++ b/Documentation/video4linux/CARDLIST.ivtv @@ -16,3 +16,9 @@ 16 -> GOTVIEW PCI DVD2 Deluxe [ffac:0600] 17 -> Yuan MPC622 [ff01:d998] 18 -> Digital Cowboy DCT-MTVP1 [1461:bfff] +19 -> Yuan PG600V2/GotView PCI DVD Lite [ffab:0600,ffad:0600] +20 -> Club3D ZAP-TV1x01 [ffab:0600] +21 -> AverTV MCE 116 Plus [1461:c439] +22 -> ASUS Falcon2 [1043:4b66,1043:462e,1043:4b2e] +23 -> AverMedia PVR-150 Plus [1461:c035] +24 -> AverMedia EZMaker PCI Deluxe [1461:c03f] diff --git a/drivers/media/video/ivtv/ivtv-cards.c b/drivers/media/video/ivtv/ivtv-cards.c index 4bb2fe8bce0b..715285e60089 100644 --- a/drivers/media/video/ivtv/ivtv-cards.c +++ b/drivers/media/video/ivtv/ivtv-cards.c @@ -956,6 +956,34 @@ static const struct ivtv_card ivtv_card_aver_pvr150 = { /* ------------------------------------------------------------------------- */ +/* AVerMedia EZMaker PCI Deluxe card */ + +static const struct ivtv_card_pci_info ivtv_pci_aver_ezmaker[] = { + { PCI_DEVICE_ID_IVTV16, IVTV_PCI_ID_AVERMEDIA, 0xc03f }, + { 0, 0, 0 } +}; + +static const struct ivtv_card ivtv_card_aver_ezmaker = { + .type = IVTV_CARD_AVER_EZMAKER, + .name = "AVerMedia EZMaker PCI Deluxe", + .v4l2_capabilities = IVTV_CAP_ENCODER, + .hw_video = IVTV_HW_CX25840, + .hw_audio = IVTV_HW_CX25840, + .hw_audio_ctrl = IVTV_HW_CX25840, + .hw_all = IVTV_HW_CX25840 | IVTV_HW_WM8739, + .video_inputs = { + { IVTV_CARD_INPUT_SVIDEO1, 0, CX25840_SVIDEO3 }, + { IVTV_CARD_INPUT_COMPOSITE1, 0, CX25840_COMPOSITE1 }, + }, + .audio_inputs = { + { IVTV_CARD_INPUT_LINE_IN1, CX25840_AUDIO_SERIAL, 0 }, + }, + .gpio_init = { .direction = 0x4000, .initial_value = 0x4000 }, + .pci_list = ivtv_pci_aver_ezmaker, +}; + +/* ------------------------------------------------------------------------- */ + /* ASUS Falcon2 */ static const struct ivtv_card_pci_info ivtv_pci_asus_falcon2[] = { @@ -1016,6 +1044,7 @@ static const struct ivtv_card *ivtv_card_list[] = { &ivtv_card_avertv_mce116, &ivtv_card_asus_falcon2, &ivtv_card_aver_pvr150, + &ivtv_card_aver_ezmaker, /* Variations of standard cards but with the same PCI IDs. These cards must come last in this list. */ diff --git a/drivers/media/video/ivtv/ivtv-cards.h b/drivers/media/video/ivtv/ivtv-cards.h index 881b04473163..b982fb06c584 100644 --- a/drivers/media/video/ivtv/ivtv-cards.h +++ b/drivers/media/video/ivtv/ivtv-cards.h @@ -47,7 +47,8 @@ #define IVTV_CARD_AVERTV_MCE116 20 /* AVerTV MCE 116 Plus */ #define IVTV_CARD_ASUS_FALCON2 21 /* ASUS Falcon2 */ #define IVTV_CARD_AVER_PVR150PLUS 22 /* AVerMedia PVR-150 Plus */ -#define IVTV_CARD_LAST 22 +#define IVTV_CARD_AVER_EZMAKER 23 /* AVerMedia EZMaker PCI Deluxe */ +#define IVTV_CARD_LAST 23 /* Variants of existing cards but with the same PCI IDs. The driver detects these based on other device information. diff --git a/drivers/media/video/ivtv/ivtv-driver.c b/drivers/media/video/ivtv/ivtv-driver.c index 7c600d0f48cd..4595c7807f2c 100644 --- a/drivers/media/video/ivtv/ivtv-driver.c +++ b/drivers/media/video/ivtv/ivtv-driver.c @@ -187,6 +187,7 @@ MODULE_PARM_DESC(cardtype, "\t\t\t21 = AverTV MCE 116 Plus\n" "\t\t\t22 = ASUS Falcon2\n" "\t\t\t23 = AverMedia PVR-150 Plus\n" + "\t\t\t24 = AverMedia EZMaker PCI Deluxe\n" "\t\t\t 0 = Autodetect (default)\n" "\t\t\t-1 = Ignore this card\n\t\t"); MODULE_PARM_DESC(pal, "Set PAL standard: B, G, H, D, K, I, M, N, Nc, 60"); -- cgit v1.2.3 From aec2aef267203d9ac06ce3508682c0aa81a069dc Mon Sep 17 00:00:00 2001 From: Aurelien Jarno Date: Mon, 17 Dec 2007 13:49:17 -0300 Subject: V4L/DVB (em28xx): Add support for Pinnacle Dazzle DVC 100 The patch below adds the "Pinnacle Dazzle DVC 100" to the list of cards supported by the em28xx driver. As the configuration is the same as the DVC 90 one, it simply adds a new USB ID to the list of devices supported by the DVC 90 configuration. Signed-off-by: Aurelien Jarno Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.em28xx | 2 +- drivers/media/video/em28xx/em28xx-cards.c | 4 +++- 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.em28xx b/Documentation/video4linux/CARDLIST.em28xx index ceb277543f51..dbc19ca9e483 100644 --- a/Documentation/video4linux/CARDLIST.em28xx +++ b/Documentation/video4linux/CARDLIST.em28xx @@ -7,7 +7,7 @@ 6 -> Terratec Cinergy 200 USB (em2800) 7 -> Leadtek Winfast USB II (em2800) 8 -> Kworld USB2800 (em2800) - 9 -> Pinnacle Dazzle DVC 90 (em2820/em2840) [2304:0207] + 9 -> Pinnacle Dazzle DVC 90/DVC 100 (em2820/em2840) [2304:0207,2304:021a] 10 -> Hauppauge WinTV HVR 900 (em2880) [2040:6500] 11 -> Terratec Hybrid XS (em2880) [0ccd:0042] 12 -> Kworld PVR TV 2800 RF (em2820/em2840) diff --git a/drivers/media/video/em28xx/em28xx-cards.c b/drivers/media/video/em28xx/em28xx-cards.c index c4204c90fb01..e00de3286055 100644 --- a/drivers/media/video/em28xx/em28xx-cards.c +++ b/drivers/media/video/em28xx/em28xx-cards.c @@ -329,7 +329,7 @@ struct em28xx_board em28xx_boards[] = { } }, }, [EM2820_BOARD_PINNACLE_DVC_90] = { - .name = "Pinnacle Dazzle DVC 90", + .name = "Pinnacle Dazzle DVC 90/DVC 100", .vchannels = 3, .has_tuner = 0, .decoder = EM28XX_SAA7113, @@ -416,6 +416,8 @@ struct usb_device_id em28xx_id_table [] = { .driver_info = EM2820_BOARD_HAUPPAUGE_WINTV_USB_2 }, { USB_DEVICE(0x2304, 0x0207), .driver_info = EM2820_BOARD_PINNACLE_DVC_90 }, + { USB_DEVICE(0x2304, 0x021a), + .driver_info = EM2820_BOARD_PINNACLE_DVC_90 }, { USB_DEVICE(0x2040, 0x6500), .driver_info = EM2880_BOARD_HAUPPAUGE_WINTV_HVR_900 }, { USB_DEVICE(0x2040, 0x6513), -- cgit v1.2.3 From d1987d55a1eda774dfbab240a432607c17241d07 Mon Sep 17 00:00:00 2001 From: Steven Toth Date: Tue, 18 Dec 2007 01:57:06 -0300 Subject: V4L/DVB (6885): Add support for the Hauppauge HVR1500Q The express card ATSC/QAM tuner. Signed-off-by: Steven Toth Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.cx23885 | 1 + drivers/media/video/cx23885/cx23885-cards.c | 17 +++++++++ drivers/media/video/cx23885/cx23885-dvb.c | 54 +++++++++++++++++++++++++++++ drivers/media/video/cx23885/cx23885-i2c.c | 2 +- drivers/media/video/cx23885/cx23885.h | 1 + 5 files changed, 74 insertions(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.cx23885 b/Documentation/video4linux/CARDLIST.cx23885 index 00cb646a4bde..54d06e4ff743 100644 --- a/Documentation/video4linux/CARDLIST.cx23885 +++ b/Documentation/video4linux/CARDLIST.cx23885 @@ -3,3 +3,4 @@ 2 -> Hauppauge WinTV-HVR1800 [0070:7800,0070:7801] 3 -> Hauppauge WinTV-HVR1250 [0070:7911] 4 -> DViCO FusionHDTV5 Express [18ac:d500] + 5 -> Hauppauge WinTV-HVR1500Q [0070:7797] diff --git a/drivers/media/video/cx23885/cx23885-cards.c b/drivers/media/video/cx23885/cx23885-cards.c index 134b931b160e..515f415564d5 100644 --- a/drivers/media/video/cx23885/cx23885-cards.c +++ b/drivers/media/video/cx23885/cx23885-cards.c @@ -113,6 +113,11 @@ struct cx23885_board cx23885_boards[] = { .name = "DViCO FusionHDTV5 Express", .portb = CX23885_MPEG_DVB, }, + [CX23885_BOARD_HAUPPAUGE_HVR1500Q] = { + .name = "Hauppauge WinTV-HVR1500Q", + .portc = CX23885_MPEG_DVB, + }, + }; const unsigned int cx23885_bcount = ARRAY_SIZE(cx23885_boards); @@ -144,6 +149,10 @@ struct cx23885_subid cx23885_subids[] = { .subvendor = 0x18ac, .subdevice = 0xd500, .card = CX23885_BOARD_DVICO_FUSIONHDTV_5_EXP, + },{ + .subvendor = 0x0070, + .subdevice = 0x7797, + .card = CX23885_BOARD_HAUPPAUGE_HVR1500Q, }, }; const unsigned int cx23885_idcount = ARRAY_SIZE(cx23885_subids); @@ -204,6 +213,11 @@ void cx23885_gpio_setup(struct cx23885_dev *dev) /* GPIO-0 cx24227 demodulator reset */ cx_set(GP0_IO, 0x00010001); /* Bring the part out of reset */ break; + case CX23885_BOARD_HAUPPAUGE_HVR1500Q: + /* GPIO-0 cx24227 demodulator reset */ + /* GPIO-2 xc5000 tuner reset */ + cx_set(GP0_IO, 0x00050005); /* Bring the part out of reset */ + break; case CX23885_BOARD_HAUPPAUGE_HVR1800: /* GPIO-0 656_CLK */ /* GPIO-1 656_D0 */ @@ -221,6 +235,7 @@ int cx23885_ir_init(struct cx23885_dev *dev) { switch (dev->board) { case CX23885_BOARD_HAUPPAUGE_HVR1250: + case CX23885_BOARD_HAUPPAUGE_HVR1500Q: case CX23885_BOARD_HAUPPAUGE_HVR1800: /* FIXME: Implement me */ break; @@ -244,6 +259,7 @@ void cx23885_card_setup(struct cx23885_dev *dev) switch (dev->board) { case CX23885_BOARD_HAUPPAUGE_HVR1250: + case CX23885_BOARD_HAUPPAUGE_HVR1500Q: case CX23885_BOARD_HAUPPAUGE_HVR1800: case CX23885_BOARD_HAUPPAUGE_HVR1800lp: if (dev->i2c_bus[0].i2c_rc == 0) @@ -258,6 +274,7 @@ void cx23885_card_setup(struct cx23885_dev *dev) ts1->src_sel_val = CX23885_SRC_SEL_PARALLEL_MPEG_VIDEO; break; case CX23885_BOARD_HAUPPAUGE_HVR1250: + case CX23885_BOARD_HAUPPAUGE_HVR1500Q: case CX23885_BOARD_HAUPPAUGE_HVR1800: case CX23885_BOARD_HAUPPAUGE_HVR1800lp: default: diff --git a/drivers/media/video/cx23885/cx23885-dvb.c b/drivers/media/video/cx23885/cx23885-dvb.c index 96d732098a0d..5fa65ef08bfd 100644 --- a/drivers/media/video/cx23885/cx23885-dvb.c +++ b/drivers/media/video/cx23885/cx23885-dvb.c @@ -33,6 +33,7 @@ #include "s5h1409.h" #include "mt2131.h" #include "lgdt330x.h" +#include "xc5000.h" #include "dvb-pll.h" static unsigned int debug = 0; @@ -74,6 +75,32 @@ static void dvb_buf_release(struct videobuf_queue *q, cx23885_free_buffer(q, (struct cx23885_buffer*)vb); } +static int cx23885_request_firmware(struct dvb_frontend *fe, + const struct firmware **fw, char *name) +{ + struct cx23885_tsport *port = fe->dvb->priv; + struct cx23885_dev *dev = port->dev; + + dprintk(1, "%s(?,?,%s)\n", __FUNCTION__, name); + + return request_firmware(fw, name, &dev->pci->dev); +} + +static int hauppauge_hvr1500q_tuner_reset(struct dvb_frontend *fe) +{ + struct cx23885_tsport *port = fe->dvb->priv; + struct cx23885_dev *dev = port->dev; + + dprintk(1, "%s()\n", __FUNCTION__); + + /* Drive the tuner into reset back back */ + cx_clear(GP0_IO, 0x00000004); + mdelay(200); + cx_set(GP0_IO, 0x00000004); + + return 0; +} + static struct videobuf_queue_ops dvb_qops = { .buf_setup = dvb_buf_setup, .buf_prepare = dvb_buf_prepare, @@ -109,6 +136,22 @@ static struct lgdt330x_config fusionhdtv_5_express = { .serial_mpeg = 0x40, }; +static struct s5h1409_config hauppauge_hvr1500q_config = { + .demod_address = 0x32 >> 1, + .output_mode = S5H1409_SERIAL_OUTPUT, + .gpio = S5H1409_GPIO_ON, + .qam_if = 44000, + .inversion = S5H1409_INVERSION_OFF, + .status_mode = S5H1409_DEMODLOCKING +}; + +static struct xc5000_config hauppauge_hvr1500q_tunerconfig = { + .i2c_address = 0x61, + .if_frequency = 4570000, + .request_firmware = cx23885_request_firmware, + .tuner_reset = hauppauge_hvr1500q_tuner_reset +}; + static int dvb_register(struct cx23885_tsport *port) { struct cx23885_dev *dev = port->dev; @@ -152,6 +195,17 @@ static int dvb_register(struct cx23885_tsport *port) &i2c_bus->i2c_adap, DVB_PLL_LG_TDVS_H06XF); } break; + case CX23885_BOARD_HAUPPAUGE_HVR1500Q: + i2c_bus = &dev->i2c_bus[1]; + port->dvb.frontend = dvb_attach(s5h1409_attach, + &hauppauge_hvr1500q_config, + &dev->i2c_bus[0].i2c_adap); + if (port->dvb.frontend != NULL) { + dvb_attach(xc5000_attach, port->dvb.frontend, + &i2c_bus->i2c_adap, + &hauppauge_hvr1500q_tunerconfig); + } + break; default: printk("%s: The frontend of your DVB/ATSC card isn't supported yet\n", dev->name); diff --git a/drivers/media/video/cx23885/cx23885-i2c.c b/drivers/media/video/cx23885/cx23885-i2c.c index 20f3fb450f88..b2ffbf04ef2c 100644 --- a/drivers/media/video/cx23885/cx23885-i2c.c +++ b/drivers/media/video/cx23885/cx23885-i2c.c @@ -332,7 +332,7 @@ static char *i2c_devs[128] = { [ 0x84 >> 1 ] = "tda8295", [ 0xa0 >> 1 ] = "eeprom", [ 0xc0 >> 1 ] = "tuner/mt2131/tda8275", - [ 0xc2 >> 1 ] = "tuner/mt2131/tda8275", + [ 0xc2 >> 1 ] = "tuner/mt2131/tda8275/xc5000", }; static void do_i2c_scan(char *name, struct i2c_client *c) diff --git a/drivers/media/video/cx23885/cx23885.h b/drivers/media/video/cx23885/cx23885.h index 205640cc48f7..3f019f3cb292 100644 --- a/drivers/media/video/cx23885/cx23885.h +++ b/drivers/media/video/cx23885/cx23885.h @@ -53,6 +53,7 @@ #define CX23885_BOARD_HAUPPAUGE_HVR1800 2 #define CX23885_BOARD_HAUPPAUGE_HVR1250 3 #define CX23885_BOARD_DVICO_FUSIONHDTV_5_EXP 4 +#define CX23885_BOARD_HAUPPAUGE_HVR1500Q 5 enum cx23885_itype { CX23885_VMUX_COMPOSITE1 = 1, -- cgit v1.2.3 From 07b4a835d42d6d59e84cbafdc8b7090f97d7b67a Mon Sep 17 00:00:00 2001 From: Michael Krufky Date: Tue, 18 Dec 2007 01:09:11 -0300 Subject: V4L/DVB (6898): cx23885: add support for Hauppauge WinTV HVR-1500 Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.cx23885 | 1 + drivers/media/video/cx23885/cx23885-cards.c | 24 ++++++++++- drivers/media/video/cx23885/cx23885-dvb.c | 65 +++++++++++++++++++++++++++++ drivers/media/video/cx23885/cx23885.h | 1 + 4 files changed, 90 insertions(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.cx23885 b/Documentation/video4linux/CARDLIST.cx23885 index 54d06e4ff743..698971d23f4e 100644 --- a/Documentation/video4linux/CARDLIST.cx23885 +++ b/Documentation/video4linux/CARDLIST.cx23885 @@ -4,3 +4,4 @@ 3 -> Hauppauge WinTV-HVR1250 [0070:7911] 4 -> DViCO FusionHDTV5 Express [18ac:d500] 5 -> Hauppauge WinTV-HVR1500Q [0070:7797] + 6 -> Hauppauge WinTV-HVR1500 [0070:7717] diff --git a/drivers/media/video/cx23885/cx23885-cards.c b/drivers/media/video/cx23885/cx23885-cards.c index d40232cda462..e11fa10a13a6 100644 --- a/drivers/media/video/cx23885/cx23885-cards.c +++ b/drivers/media/video/cx23885/cx23885-cards.c @@ -117,7 +117,10 @@ struct cx23885_board cx23885_boards[] = { .name = "Hauppauge WinTV-HVR1500Q", .portc = CX23885_MPEG_DVB, }, - + [CX23885_BOARD_HAUPPAUGE_HVR1500] = { + .name = "Hauppauge WinTV-HVR1500", + .portc = CX23885_MPEG_DVB, + }, }; const unsigned int cx23885_bcount = ARRAY_SIZE(cx23885_boards); @@ -153,6 +156,10 @@ struct cx23885_subid cx23885_subids[] = { .subvendor = 0x0070, .subdevice = 0x7797, .card = CX23885_BOARD_HAUPPAUGE_HVR1500Q, + },{ + .subvendor = 0x0070, + .subdevice = 0x7717, + .card = CX23885_BOARD_HAUPPAUGE_HVR1500, }, }; const unsigned int cx23885_idcount = ARRAY_SIZE(cx23885_subids); @@ -214,6 +221,18 @@ void cx23885_gpio_setup(struct cx23885_dev *dev) /* GPIO-0 cx24227 demodulator reset */ cx_set(GP0_IO, 0x00010001); /* Bring the part out of reset */ break; + case CX23885_BOARD_HAUPPAUGE_HVR1500: + /* GPIO-0 cx24227 demodulator */ + /* GPIO-2 xc3028 tuner */ + + /* Put the parts into reset */ + cx_set(GP0_IO, 0x00050000); + cx_clear(GP0_IO, 0x00000005); + msleep(5); + + /* Bring the parts out of reset */ + cx_set(GP0_IO, 0x00050005); + break; case CX23885_BOARD_HAUPPAUGE_HVR1500Q: /* GPIO-0 cx24227 demodulator reset */ /* GPIO-2 xc5000 tuner reset */ @@ -236,6 +255,7 @@ int cx23885_ir_init(struct cx23885_dev *dev) { switch (dev->board) { case CX23885_BOARD_HAUPPAUGE_HVR1250: + case CX23885_BOARD_HAUPPAUGE_HVR1500: case CX23885_BOARD_HAUPPAUGE_HVR1500Q: case CX23885_BOARD_HAUPPAUGE_HVR1800: /* FIXME: Implement me */ @@ -260,6 +280,7 @@ void cx23885_card_setup(struct cx23885_dev *dev) switch (dev->board) { case CX23885_BOARD_HAUPPAUGE_HVR1250: + case CX23885_BOARD_HAUPPAUGE_HVR1500: case CX23885_BOARD_HAUPPAUGE_HVR1500Q: case CX23885_BOARD_HAUPPAUGE_HVR1800: case CX23885_BOARD_HAUPPAUGE_HVR1800lp: @@ -275,6 +296,7 @@ void cx23885_card_setup(struct cx23885_dev *dev) ts1->src_sel_val = CX23885_SRC_SEL_PARALLEL_MPEG_VIDEO; break; case CX23885_BOARD_HAUPPAUGE_HVR1250: + case CX23885_BOARD_HAUPPAUGE_HVR1500: case CX23885_BOARD_HAUPPAUGE_HVR1500Q: case CX23885_BOARD_HAUPPAUGE_HVR1800: case CX23885_BOARD_HAUPPAUGE_HVR1800lp: diff --git a/drivers/media/video/cx23885/cx23885-dvb.c b/drivers/media/video/cx23885/cx23885-dvb.c index f0882dd30d3f..c1309118ea87 100644 --- a/drivers/media/video/cx23885/cx23885-dvb.c +++ b/drivers/media/video/cx23885/cx23885-dvb.c @@ -35,6 +35,8 @@ #include "lgdt330x.h" #include "xc5000.h" #include "dvb-pll.h" +#include "tuner-xc2028.h" +#include "tuner-xc2028-types.h" static unsigned int debug = 0; @@ -126,6 +128,14 @@ static struct s5h1409_config hauppauge_hvr1800lp_config = { .status_mode = S5H1409_DEMODLOCKING }; +static struct s5h1409_config hauppauge_hvr1500_config = { + .demod_address = 0x32 >> 1, + .output_mode = S5H1409_SERIAL_OUTPUT, + .gpio = S5H1409_GPIO_OFF, + .inversion = S5H1409_INVERSION_OFF, + .status_mode = S5H1409_DEMODLOCKING +}; + static struct mt2131_config hauppauge_generic_tunerconfig = { 0x61 }; @@ -152,6 +162,36 @@ static struct xc5000_config hauppauge_hvr1500q_tunerconfig = { .tuner_reset = hauppauge_hvr1500q_tuner_reset }; +static int cx23885_hvr1500_xc3028_callback(void *ptr, int command, int arg) +{ + struct cx23885_tsport *port = ptr; + struct cx23885_dev *dev = port->dev; + + switch (command) { + case XC2028_TUNER_RESET: + /* Send the tuner in then out of reset */ + /* GPIO-2 xc3028 tuner */ + dprintk(1, "%s: XC2028_TUNER_RESET %d\n", __FUNCTION__, arg); + + cx_set(GP0_IO, 0x00040000); + cx_clear(GP0_IO, 0x00000004); + msleep(5); + + cx_set(GP0_IO, 0x00040004); + msleep(5); + break; + case XC2028_RESET_CLK: + dprintk(1, "%s: XC2028_RESET_CLK %d\n", __FUNCTION__, arg); + break; + default: + dprintk(1, "%s: unknown command %d, arg %d\n", __FUNCTION__, + command, arg); + return -EINVAL; + } + + return 0; +} + static int dvb_register(struct cx23885_tsport *port) { struct cx23885_dev *dev = port->dev; @@ -206,6 +246,31 @@ static int dvb_register(struct cx23885_tsport *port) &hauppauge_hvr1500q_tunerconfig); } break; + case CX23885_BOARD_HAUPPAUGE_HVR1500: + i2c_bus = &dev->i2c_bus[1]; + port->dvb.frontend = dvb_attach(s5h1409_attach, + &hauppauge_hvr1500_config, + &dev->i2c_bus[0].i2c_adap); + if (port->dvb.frontend != NULL) { + struct dvb_frontend *fe; + struct xc2028_config cfg = { + .i2c_adap = &i2c_bus->i2c_adap, + .i2c_addr = 0x61, + .video_dev = port, + .callback = cx23885_hvr1500_xc3028_callback, + }; + static struct xc2028_ctrl ctl = { + .fname = "xc3028-v27.fw", + .max_len = 64, + .scode_table = OREN538, + }; + + fe = dvb_attach(xc2028_attach, + port->dvb.frontend, &cfg); + if (fe != NULL && fe->ops.tuner_ops.set_config != NULL) + fe->ops.tuner_ops.set_config(fe, &ctl); + } + break; default: printk("%s: The frontend of your DVB/ATSC card isn't supported yet\n", dev->name); diff --git a/drivers/media/video/cx23885/cx23885.h b/drivers/media/video/cx23885/cx23885.h index 3f019f3cb292..974ec14782d7 100644 --- a/drivers/media/video/cx23885/cx23885.h +++ b/drivers/media/video/cx23885/cx23885.h @@ -54,6 +54,7 @@ #define CX23885_BOARD_HAUPPAUGE_HVR1250 3 #define CX23885_BOARD_DVICO_FUSIONHDTV_5_EXP 4 #define CX23885_BOARD_HAUPPAUGE_HVR1500Q 5 +#define CX23885_BOARD_HAUPPAUGE_HVR1500 6 enum cx23885_itype { CX23885_VMUX_COMPOSITE1 = 1, -- cgit v1.2.3 From 6ccb8cfbb94cc82144a92a6a28a754d9772b3aa8 Mon Sep 17 00:00:00 2001 From: Michael Krufky Date: Thu, 27 Dec 2007 21:46:34 -0300 Subject: V4L/DVB (6932): cx23885: add missing subsystem ID for Hauppauge HVR1800 Retail Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.cx23885 | 2 +- drivers/media/video/cx23885/cx23885-cards.c | 4 ++++ 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.cx23885 b/Documentation/video4linux/CARDLIST.cx23885 index 698971d23f4e..302c84f45edb 100644 --- a/Documentation/video4linux/CARDLIST.cx23885 +++ b/Documentation/video4linux/CARDLIST.cx23885 @@ -1,6 +1,6 @@ 0 -> UNKNOWN/GENERIC [0070:3400] 1 -> Hauppauge WinTV-HVR1800lp [0070:7600] - 2 -> Hauppauge WinTV-HVR1800 [0070:7800,0070:7801] + 2 -> Hauppauge WinTV-HVR1800 [0070:7800,0070:7801,0070:7809] 3 -> Hauppauge WinTV-HVR1250 [0070:7911] 4 -> DViCO FusionHDTV5 Express [18ac:d500] 5 -> Hauppauge WinTV-HVR1500Q [0070:7797] diff --git a/drivers/media/video/cx23885/cx23885-cards.c b/drivers/media/video/cx23885/cx23885-cards.c index fbf93c63dcaa..3ddcab305778 100644 --- a/drivers/media/video/cx23885/cx23885-cards.c +++ b/drivers/media/video/cx23885/cx23885-cards.c @@ -144,6 +144,10 @@ struct cx23885_subid cx23885_subids[] = { .subvendor = 0x0070, .subdevice = 0x7801, .card = CX23885_BOARD_HAUPPAUGE_HVR1800, + },{ + .subvendor = 0x0070, + .subdevice = 0x7809, + .card = CX23885_BOARD_HAUPPAUGE_HVR1800, },{ .subvendor = 0x0070, .subdevice = 0x7911, -- cgit v1.2.3 From b00fff0be5655614bb18a4b1d6e33b27c56c6f7e Mon Sep 17 00:00:00 2001 From: Michael Krufky Date: Thu, 27 Dec 2007 22:19:31 -0300 Subject: V4L/DVB (6935): cx23885: add missing subsystem IDs for Hauppauge HVR1500 Retail Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.cx23885 | 4 ++-- drivers/media/video/cx23885/cx23885-cards.c | 8 ++++++++ 2 files changed, 10 insertions(+), 2 deletions(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.cx23885 b/Documentation/video4linux/CARDLIST.cx23885 index 302c84f45edb..0924e6e142c4 100644 --- a/Documentation/video4linux/CARDLIST.cx23885 +++ b/Documentation/video4linux/CARDLIST.cx23885 @@ -3,5 +3,5 @@ 2 -> Hauppauge WinTV-HVR1800 [0070:7800,0070:7801,0070:7809] 3 -> Hauppauge WinTV-HVR1250 [0070:7911] 4 -> DViCO FusionHDTV5 Express [18ac:d500] - 5 -> Hauppauge WinTV-HVR1500Q [0070:7797] - 6 -> Hauppauge WinTV-HVR1500 [0070:7717] + 5 -> Hauppauge WinTV-HVR1500Q [0070:7790,0070:7797] + 6 -> Hauppauge WinTV-HVR1500 [0070:7710,0070:7717] diff --git a/drivers/media/video/cx23885/cx23885-cards.c b/drivers/media/video/cx23885/cx23885-cards.c index 4068695383ee..82af44884f0b 100644 --- a/drivers/media/video/cx23885/cx23885-cards.c +++ b/drivers/media/video/cx23885/cx23885-cards.c @@ -156,10 +156,18 @@ struct cx23885_subid cx23885_subids[] = { .subvendor = 0x18ac, .subdevice = 0xd500, .card = CX23885_BOARD_DVICO_FUSIONHDTV_5_EXP, + },{ + .subvendor = 0x0070, + .subdevice = 0x7790, + .card = CX23885_BOARD_HAUPPAUGE_HVR1500Q, },{ .subvendor = 0x0070, .subdevice = 0x7797, .card = CX23885_BOARD_HAUPPAUGE_HVR1500Q, + },{ + .subvendor = 0x0070, + .subdevice = 0x7710, + .card = CX23885_BOARD_HAUPPAUGE_HVR1500, },{ .subvendor = 0x0070, .subdevice = 0x7717, -- cgit v1.2.3 From a594dcba943a2edc484ed83f694c27306bf5e28e Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Mon, 31 Dec 2007 02:29:46 -0300 Subject: V4L/DVB (6938): Add a tool for extracting xc3028 version 2.7 firmware This tool allows the extraction of xc3028 v2.7 firmware from HVR 12x0 file. Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/extract_xc3028.pl | 926 ++++++++++++++++++++++++++++ 1 file changed, 926 insertions(+) create mode 100644 Documentation/video4linux/extract_xc3028.pl (limited to 'Documentation') diff --git a/Documentation/video4linux/extract_xc3028.pl b/Documentation/video4linux/extract_xc3028.pl new file mode 100644 index 000000000000..cced8ac5c543 --- /dev/null +++ b/Documentation/video4linux/extract_xc3028.pl @@ -0,0 +1,926 @@ +#!/usr/bin/perl + +# Copyright (c) Mauro Carvalho Chehab +# Released under GPLv2 +# +# In order to use, you need to: +# 1) Download the windows driver with something like: +# wget http://www.steventoth.net/linux/xc5000/HVR-12x0-14x0-17x0_1_25_25271_WHQL.zip +# 2) Extract the file hcw85bda.sys from the zip into the current dir: +# unzip -j HVR-12x0-14x0-17x0_1_25_25271_WHQL.zip Driver85/hcw85bda.sys +# 3) run the script: +# ./extract_xc3028.pl +# 4) copy the generated file: +# cp xc3028-v27.fw /lib/firmware + +#use strict; +use IO::Handle; + +my $debug=0; + +sub verify ($$) +{ + my ($filename, $hash) = @_; + my ($testhash); + + if (system("which md5sum > /dev/null 2>&1")) { + die "This firmware requires the md5sum command - see http://www.gnu.org/software/coreutils/\n"; + } + + open(CMD, "md5sum ".$filename."|"); + $testhash = ; + $testhash =~ /([a-zA-Z0-9]*)/; + $testhash = $1; + close CMD; + die "Hash of extracted file does not match (found $testhash, expected $hash!\n" if ($testhash ne $hash); +} + +sub get_hunk ($$) +{ + my ($offset, $length) = @_; + my ($chunklength, $buf, $rcount, $out); + + sysseek(INFILE, $offset, SEEK_SET); + while ($length > 0) { + # Calc chunk size + $chunklength = 2048; + $chunklength = $length if ($chunklength > $length); + + $rcount = sysread(INFILE, $buf, $chunklength); + die "Ran out of data\n" if ($rcount != $chunklength); + $out .= $buf; + $length -= $rcount; + } + return $out; +} + +sub write_le16($) +{ + my $val = shift; + my $msb = ($val >> 8) &0xff; + my $lsb = $val & 0xff; + + syswrite(OUTFILE, chr($lsb).chr($msb)); +} + +sub write_le32($) +{ + my $val = shift; + my $l3 = ($val >> 24) & 0xff; + my $l2 = ($val >> 16) & 0xff; + my $l1 = ($val >> 8) & 0xff; + my $l0 = $val & 0xff; + + syswrite(OUTFILE, chr($l0).chr($l1).chr($l2).chr($l3)); +} + +sub write_le64($$) +{ + my $msb_val = shift; + my $lsb_val = shift; + my $l7 = ($msb_val >> 24) & 0xff; + my $l6 = ($msb_val >> 16) & 0xff; + my $l5 = ($msb_val >> 8) & 0xff; + my $l4 = $msb_val & 0xff; + + my $l3 = ($lsb_val >> 24) & 0xff; + my $l2 = ($lsb_val >> 16) & 0xff; + my $l1 = ($lsb_val >> 8) & 0xff; + my $l0 = $lsb_val & 0xff; + + syswrite(OUTFILE, + chr($l0).chr($l1).chr($l2).chr($l3). + chr($l4).chr($l5).chr($l6).chr($l7)); +} + +sub write_hunk($$) +{ + my ($offset, $length) = @_; + my $out = get_hunk($offset, $length); + + printf "(len %d) ",$length if ($debug); + + for (my $i=0;$i<$length;$i++) { + printf "%02x ",ord(substr($out,$i,1)) if ($debug); + } + printf "\n" if ($debug); + + syswrite(OUTFILE, $out); +} + +sub write_hunk_fix_endian($$) +{ + my ($offset, $length) = @_; + my $out = get_hunk($offset, $length); + + printf "(len_fix %d) ",$length if ($debug); + + for (my $i=0;$i<$length;$i++) { + printf "%02x ",ord(substr($out,$i,1)) if ($debug); + } + printf "\n" if ($debug); + + my $i=0; + while ($i<$length) { + my $size = ord(substr($out,$i,1))*256+ord(substr($out,$i+1,1)); + syswrite(OUTFILE, substr($out,$i+1,1)); + syswrite(OUTFILE, substr($out,$i,1)); + $i+=2; + if ($size>0 && $size <0x8000) { + for (my $j=0;$j<$size;$j++) { + syswrite(OUTFILE, substr($out,$j+$i,1)); + } + $i+=$size; + } + } +} + +sub main_firmware($$$$) +{ + my $out; + my $j=0; + my $outfile = shift; + my $name = shift; + my $version = shift; + my $nr_desc = shift; + + for ($j = length($name); $j <32; $j++) { + $name = $name.chr(0); +} + + open OUTFILE, ">$outfile"; + syswrite(OUTFILE, $name); + write_le16($version); + write_le16($nr_desc); + + # + # Firmware 0, type: BASE FW F8MHZ (0x00000003), id: (0000000000000000), size: 8718 + # + + write_le32(0x00000003); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(8718); # Size + write_hunk_fix_endian(813432, 8718); + + # + # Firmware 1, type: BASE FW F8MHZ MTS (0x00000007), id: (0000000000000000), size: 8712 + # + + write_le32(0x00000007); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(8712); # Size + write_hunk_fix_endian(822152, 8712); + + # + # Firmware 2, type: BASE FW FM (0x00000401), id: (0000000000000000), size: 8562 + # + + write_le32(0x00000401); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(8562); # Size + write_hunk_fix_endian(830872, 8562); + + # + # Firmware 3, type: BASE FW FM INPUT1 (0x00000c01), id: (0000000000000000), size: 8576 + # + + write_le32(0x00000c01); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(8576); # Size + write_hunk_fix_endian(839440, 8576); + + # + # Firmware 4, type: BASE FW (0x00000001), id: (0000000000000000), size: 8706 + # + + write_le32(0x00000001); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(8706); # Size + write_hunk_fix_endian(848024, 8706); + + # + # Firmware 5, type: BASE FW MTS (0x00000005), id: (0000000000000000), size: 8682 + # + + write_le32(0x00000005); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(8682); # Size + write_hunk_fix_endian(856736, 8682); + + # + # Firmware 6, type: STD FW (0x00000000), id: PAL/BG A2/A (0000000100000007), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x00000001, 0x00000007); # ID + write_le32(161); # Size + write_hunk_fix_endian(865424, 161); + + # + # Firmware 7, type: STD FW MTS (0x00000004), id: PAL/BG A2/A (0000000100000007), size: 169 + # + + write_le32(0x00000004); # Type + write_le64(0x00000001, 0x00000007); # ID + write_le32(169); # Size + write_hunk_fix_endian(865592, 169); + + # + # Firmware 8, type: STD FW (0x00000000), id: PAL/BG A2/B (0000000200000007), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x00000002, 0x00000007); # ID + write_le32(161); # Size + write_hunk_fix_endian(865424, 161); + + # + # Firmware 9, type: STD FW MTS (0x00000004), id: PAL/BG A2/B (0000000200000007), size: 169 + # + + write_le32(0x00000004); # Type + write_le64(0x00000002, 0x00000007); # ID + write_le32(169); # Size + write_hunk_fix_endian(865592, 169); + + # + # Firmware 10, type: STD FW (0x00000000), id: PAL/BG NICAM/A (0000000400000007), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x00000004, 0x00000007); # ID + write_le32(161); # Size + write_hunk_fix_endian(866112, 161); + + # + # Firmware 11, type: STD FW MTS (0x00000004), id: PAL/BG NICAM/A (0000000400000007), size: 169 + # + + write_le32(0x00000004); # Type + write_le64(0x00000004, 0x00000007); # ID + write_le32(169); # Size + write_hunk_fix_endian(866280, 169); + + # + # Firmware 12, type: STD FW (0x00000000), id: PAL/BG NICAM/B (0000000800000007), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x00000008, 0x00000007); # ID + write_le32(161); # Size + write_hunk_fix_endian(866112, 161); + + # + # Firmware 13, type: STD FW MTS (0x00000004), id: PAL/BG NICAM/B (0000000800000007), size: 169 + # + + write_le32(0x00000004); # Type + write_le64(0x00000008, 0x00000007); # ID + write_le32(169); # Size + write_hunk_fix_endian(866280, 169); + + # + # Firmware 14, type: STD FW (0x00000000), id: PAL/DK A2 (00000003000000e0), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x00000003, 0x000000e0); # ID + write_le32(161); # Size + write_hunk_fix_endian(866800, 161); + + # + # Firmware 15, type: STD FW MTS (0x00000004), id: PAL/DK A2 (00000003000000e0), size: 169 + # + + write_le32(0x00000004); # Type + write_le64(0x00000003, 0x000000e0); # ID + write_le32(169); # Size + write_hunk_fix_endian(866968, 169); + + # + # Firmware 16, type: STD FW (0x00000000), id: PAL/DK NICAM (0000000c000000e0), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x0000000c, 0x000000e0); # ID + write_le32(161); # Size + write_hunk_fix_endian(867144, 161); + + # + # Firmware 17, type: STD FW MTS (0x00000004), id: PAL/DK NICAM (0000000c000000e0), size: 169 + # + + write_le32(0x00000004); # Type + write_le64(0x0000000c, 0x000000e0); # ID + write_le32(169); # Size + write_hunk_fix_endian(867312, 169); + + # + # Firmware 18, type: STD FW (0x00000000), id: SECAM/K1 (0000000000200000), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x00000000, 0x00200000); # ID + write_le32(161); # Size + write_hunk_fix_endian(867488, 161); + + # + # Firmware 19, type: STD FW MTS (0x00000004), id: SECAM/K1 (0000000000200000), size: 169 + # + + write_le32(0x00000004); # Type + write_le64(0x00000000, 0x00200000); # ID + write_le32(169); # Size + write_hunk_fix_endian(867656, 169); + + # + # Firmware 20, type: STD FW (0x00000000), id: SECAM/K3 (0000000004000000), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x00000000, 0x04000000); # ID + write_le32(161); # Size + write_hunk_fix_endian(867832, 161); + + # + # Firmware 21, type: STD FW MTS (0x00000004), id: SECAM/K3 (0000000004000000), size: 169 + # + + write_le32(0x00000004); # Type + write_le64(0x00000000, 0x04000000); # ID + write_le32(169); # Size + write_hunk_fix_endian(868000, 169); + + # + # Firmware 22, type: STD FW D2633 DTV6 ATSC (0x00010030), id: (0000000000000000), size: 149 + # + + write_le32(0x00010030); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(149); # Size + write_hunk_fix_endian(868176, 149); + + # + # Firmware 23, type: STD FW D2620 DTV6 QAM (0x00000068), id: (0000000000000000), size: 149 + # + + write_le32(0x00000068); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(149); # Size + write_hunk_fix_endian(868336, 149); + + # + # Firmware 24, type: STD FW D2633 DTV6 QAM (0x00000070), id: (0000000000000000), size: 149 + # + + write_le32(0x00000070); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(149); # Size + write_hunk_fix_endian(868488, 149); + + # + # Firmware 25, type: STD FW D2620 DTV7 (0x00000088), id: (0000000000000000), size: 149 + # + + write_le32(0x00000088); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(149); # Size + write_hunk_fix_endian(868648, 149); + + # + # Firmware 26, type: STD FW D2633 DTV7 (0x00000090), id: (0000000000000000), size: 149 + # + + write_le32(0x00000090); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(149); # Size + write_hunk_fix_endian(868800, 149); + + # + # Firmware 27, type: STD FW D2620 DTV78 (0x00000108), id: (0000000000000000), size: 149 + # + + write_le32(0x00000108); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(149); # Size + write_hunk_fix_endian(868960, 149); + + # + # Firmware 28, type: STD FW D2633 DTV78 (0x00000110), id: (0000000000000000), size: 149 + # + + write_le32(0x00000110); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(149); # Size + write_hunk_fix_endian(869112, 149); + + # + # Firmware 29, type: STD FW D2620 DTV8 (0x00000208), id: (0000000000000000), size: 149 + # + + write_le32(0x00000208); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(149); # Size + write_hunk_fix_endian(868648, 149); + + # + # Firmware 30, type: STD FW D2633 DTV8 (0x00000210), id: (0000000000000000), size: 149 + # + + write_le32(0x00000210); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(149); # Size + write_hunk_fix_endian(868800, 149); + + # + # Firmware 31, type: STD FW FM (0x00000400), id: (0000000000000000), size: 135 + # + + write_le32(0x00000400); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le32(135); # Size + write_hunk_fix_endian(869584, 135); + + # + # Firmware 32, type: STD FW (0x00000000), id: PAL/I (0000000000000010), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x00000000, 0x00000010); # ID + write_le32(161); # Size + write_hunk_fix_endian(869728, 161); + + # + # Firmware 33, type: STD FW MTS (0x00000004), id: PAL/I (0000000000000010), size: 169 + # + + write_le32(0x00000004); # Type + write_le64(0x00000000, 0x00000010); # ID + write_le32(169); # Size + write_hunk_fix_endian(869896, 169); + + # + # Firmware 34, type: STD FW (0x00000000), id: SECAM/L AM (0000001000400000), size: 169 + # + + write_le32(0x00000000); # Type + write_le64(0x00000010, 0x00400000); # ID + write_le32(169); # Size + write_hunk_fix_endian(870072, 169); + + # + # Firmware 35, type: STD FW (0x00000000), id: SECAM/L NICAM (0000000c00400000), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x0000000c, 0x00400000); # ID + write_le32(161); # Size + write_hunk_fix_endian(870248, 161); + + # + # Firmware 36, type: STD FW (0x00000000), id: SECAM/Lc (0000000000800000), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x00000000, 0x00800000); # ID + write_le32(161); # Size + write_hunk_fix_endian(870416, 161); + + # + # Firmware 37, type: STD FW (0x00000000), id: NTSC/M Kr (0000000000008000), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x00000000, 0x00008000); # ID + write_le32(161); # Size + write_hunk_fix_endian(870584, 161); + + # + # Firmware 38, type: STD FW LCD (0x00001000), id: NTSC/M Kr (0000000000008000), size: 161 + # + + write_le32(0x00001000); # Type + write_le64(0x00000000, 0x00008000); # ID + write_le32(161); # Size + write_hunk_fix_endian(870752, 161); + + # + # Firmware 39, type: STD FW LCD NOGD (0x00003000), id: NTSC/M Kr (0000000000008000), size: 161 + # + + write_le32(0x00003000); # Type + write_le64(0x00000000, 0x00008000); # ID + write_le32(161); # Size + write_hunk_fix_endian(870920, 161); + + # + # Firmware 40, type: STD FW MTS (0x00000004), id: NTSC/M Kr (0000000000008000), size: 169 + # + + write_le32(0x00000004); # Type + write_le64(0x00000000, 0x00008000); # ID + write_le32(169); # Size + write_hunk_fix_endian(871088, 169); + + # + # Firmware 41, type: STD FW (0x00000000), id: NTSC PAL/M PAL/N (000000000000b700), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x00000000, 0x0000b700); # ID + write_le32(161); # Size + write_hunk_fix_endian(871264, 161); + + # + # Firmware 42, type: STD FW LCD (0x00001000), id: NTSC PAL/M PAL/N (000000000000b700), size: 161 + # + + write_le32(0x00001000); # Type + write_le64(0x00000000, 0x0000b700); # ID + write_le32(161); # Size + write_hunk_fix_endian(871432, 161); + + # + # Firmware 43, type: STD FW LCD NOGD (0x00003000), id: NTSC PAL/M PAL/N (000000000000b700), size: 161 + # + + write_le32(0x00003000); # Type + write_le64(0x00000000, 0x0000b700); # ID + write_le32(161); # Size + write_hunk_fix_endian(871600, 161); + + # + # Firmware 44, type: STD FW (0x00000000), id: NTSC/M Jp (0000000000002000), size: 161 + # + + write_le32(0x00000000); # Type + write_le64(0x00000000, 0x00002000); # ID + write_le32(161); # Size + write_hunk_fix_endian(871264, 161); + + # + # Firmware 45, type: STD FW MTS (0x00000004), id: NTSC PAL/M PAL/N (000000000000b700), size: 169 + # + + write_le32(0x00000004); # Type + write_le64(0x00000000, 0x0000b700); # ID + write_le32(169); # Size + write_hunk_fix_endian(871936, 169); + + # + # Firmware 46, type: STD FW MTS LCD (0x00001004), id: NTSC PAL/M PAL/N (000000000000b700), size: 169 + # + + write_le32(0x00001004); # Type + write_le64(0x00000000, 0x0000b700); # ID + write_le32(169); # Size + write_hunk_fix_endian(872112, 169); + + # + # Firmware 47, type: STD FW MTS LCD NOGD (0x00003004), id: NTSC PAL/M PAL/N (000000000000b700), size: 169 + # + + write_le32(0x00003004); # Type + write_le64(0x00000000, 0x0000b700); # ID + write_le32(169); # Size + write_hunk_fix_endian(872288, 169); + + # + # Firmware 48, type: SCODE FW HAS IF (0x60000000), IF = 3.28 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(3280); # IF + write_le32(192); # Size + write_hunk(811896, 192); + + # + # Firmware 49, type: SCODE FW HAS IF (0x60000000), IF = 3.30 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(3300); # IF + write_le32(192); # Size + write_hunk(813048, 192); + + # + # Firmware 50, type: SCODE FW HAS IF (0x60000000), IF = 3.44 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(3440); # IF + write_le32(192); # Size + write_hunk(812280, 192); + + # + # Firmware 51, type: SCODE FW HAS IF (0x60000000), IF = 3.46 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(3460); # IF + write_le32(192); # Size + write_hunk(812472, 192); + + # + # Firmware 52, type: SCODE FW DTV6 ATSC OREN36 HAS IF (0x60210020), IF = 3.80 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60210020); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(3800); # IF + write_le32(192); # Size + write_hunk(809784, 192); + + # + # Firmware 53, type: SCODE FW HAS IF (0x60000000), IF = 4.00 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(4000); # IF + write_le32(192); # Size + write_hunk(812088, 192); + + # + # Firmware 54, type: SCODE FW DTV6 ATSC TOYOTA388 HAS IF (0x60410020), IF = 4.08 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60410020); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(4080); # IF + write_le32(192); # Size + write_hunk(809976, 192); + + # + # Firmware 55, type: SCODE FW HAS IF (0x60000000), IF = 4.20 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(4200); # IF + write_le32(192); # Size + write_hunk(811704, 192); + + # + # Firmware 56, type: SCODE FW MONO HAS IF (0x60008000), IF = 4.32 MHz id: NTSC/M Kr (0000000000008000), size: 192 + # + + write_le32(0x60008000); # Type + write_le64(0x00000000, 0x00008000); # ID + write_le16(4320); # IF + write_le32(192); # Size + write_hunk(808056, 192); + + # + # Firmware 57, type: SCODE FW HAS IF (0x60000000), IF = 4.45 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(4450); # IF + write_le32(192); # Size + write_hunk(812664, 192); + + # + # Firmware 58, type: SCODE FW HAS IF (0x60000000), IF = 4.50 MHz id: NTSC/M Jp (0000000000002000), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000000, 0x00002000); # ID + write_le16(4500); # IF + write_le32(192); # Size + write_hunk(807672, 192); + + # + # Firmware 59, type: SCODE FW LCD NOGD IF HAS IF (0x60023000), IF = 4.60 MHz id: NTSC/M Kr (0000000000008000), size: 192 + # + + write_le32(0x60023000); # Type + write_le64(0x00000000, 0x00008000); # ID + write_le16(4600); # IF + write_le32(192); # Size + write_hunk(807864, 192); + + # + # Firmware 60, type: SCODE FW DTV78 ZARLINK456 HAS IF (0x62000100), IF = 4.76 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x62000100); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(4760); # IF + write_le32(192); # Size + write_hunk(807288, 192); + + # + # Firmware 61, type: SCODE FW HAS IF (0x60000000), IF = 4.94 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(4940); # IF + write_le32(192); # Size + write_hunk(811512, 192); + + # + # Firmware 62, type: SCODE FW DTV7 ZARLINK456 HAS IF (0x62000080), IF = 5.26 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x62000080); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(5260); # IF + write_le32(192); # Size + write_hunk(810552, 192); + + # + # Firmware 63, type: SCODE FW MONO HAS IF (0x60008000), IF = 5.32 MHz id: PAL/BG NICAM/B (0000000800000007), size: 192 + # + + write_le32(0x60008000); # Type + write_le64(0x00000008, 0x00000007); # ID + write_le16(5320); # IF + write_le32(192); # Size + write_hunk(810744, 192); + + # + # Firmware 64, type: SCODE FW DTV8 CHINA HAS IF (0x64000200), IF = 5.40 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x64000200); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(5400); # IF + write_le32(192); # Size + write_hunk(807096, 192); + + # + # Firmware 65, type: SCODE FW DTV6 ATSC OREN538 HAS IF (0x60110020), IF = 5.58 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60110020); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(5580); # IF + write_le32(192); # Size + write_hunk(809592, 192); + + # + # Firmware 66, type: SCODE FW HAS IF (0x60000000), IF = 5.64 MHz id: PAL/BG A2/B (0000000200000007), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000002, 0x00000007); # ID + write_le16(5640); # IF + write_le32(192); # Size + write_hunk(808440, 192); + + # + # Firmware 67, type: SCODE FW HAS IF (0x60000000), IF = 5.74 MHz id: PAL/BG NICAM/B (0000000800000007), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000008, 0x00000007); # ID + write_le16(5740); # IF + write_le32(192); # Size + write_hunk(808632, 192); + + # + # Firmware 68, type: SCODE FW DTV7 DIBCOM52 HAS IF (0x61000080), IF = 5.90 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x61000080); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(5900); # IF + write_le32(192); # Size + write_hunk(810360, 192); + + # + # Firmware 69, type: SCODE FW MONO HAS IF (0x60008000), IF = 6.00 MHz id: PAL/I (0000000000000010), size: 192 + # + + write_le32(0x60008000); # Type + write_le64(0x00000000, 0x00000010); # ID + write_le16(6000); # IF + write_le32(192); # Size + write_hunk(808824, 192); + + # + # Firmware 70, type: SCODE FW DTV6 QAM F6MHZ HAS IF (0x68000060), IF = 6.20 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x68000060); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(6200); # IF + write_le32(192); # Size + write_hunk(809400, 192); + + # + # Firmware 71, type: SCODE FW HAS IF (0x60000000), IF = 6.24 MHz id: PAL/I (0000000000000010), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000000, 0x00000010); # ID + write_le16(6240); # IF + write_le32(192); # Size + write_hunk(808248, 192); + + # + # Firmware 72, type: SCODE FW MONO HAS IF (0x60008000), IF = 6.32 MHz id: SECAM/K1 (0000000000200000), size: 192 + # + + write_le32(0x60008000); # Type + write_le64(0x00000000, 0x00200000); # ID + write_le16(6320); # IF + write_le32(192); # Size + write_hunk(811320, 192); + + # + # Firmware 73, type: SCODE FW HAS IF (0x60000000), IF = 6.34 MHz id: SECAM/K1 (0000000000200000), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000000, 0x00200000); # ID + write_le16(6340); # IF + write_le32(192); # Size + write_hunk(809208, 192); + + # + # Firmware 74, type: SCODE FW MONO HAS IF (0x60008000), IF = 6.50 MHz id: SECAM/K3 (0000000004000000), size: 192 + # + + write_le32(0x60008000); # Type + write_le64(0x00000000, 0x04000000); # ID + write_le16(6500); # IF + write_le32(192); # Size + write_hunk(811128, 192); + + # + # Firmware 75, type: SCODE FW DTV6 ATSC ATI638 HAS IF (0x60090020), IF = 6.58 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60090020); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(6580); # IF + write_le32(192); # Size + write_hunk(807480, 192); + + # + # Firmware 76, type: SCODE FW HAS IF (0x60000000), IF = 6.60 MHz id: PAL/DK A2 (00000003000000e0), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000003, 0x000000e0); # ID + write_le16(6600); # IF + write_le32(192); # Size + write_hunk(809016, 192); + + # + # Firmware 77, type: SCODE FW MONO HAS IF (0x60008000), IF = 6.68 MHz id: PAL/DK A2 (00000003000000e0), size: 192 + # + + write_le32(0x60008000); # Type + write_le64(0x00000003, 0x000000e0); # ID + write_le16(6680); # IF + write_le32(192); # Size + write_hunk(810936, 192); + + # + # Firmware 78, type: SCODE FW DTV6 ATSC TOYOTA794 HAS IF (0x60810020), IF = 8.14 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60810020); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(8140); # IF + write_le32(192); # Size + write_hunk(810168, 192); + + # + # Firmware 79, type: SCODE FW HAS IF (0x60000000), IF = 8.20 MHz id: (0000000000000000), size: 192 + # + + write_le32(0x60000000); # Type + write_le64(0x00000000, 0x00000000); # ID + write_le16(8200); # IF + write_le32(192); # Size + write_hunk(812856, 192); +} + +sub extract_firmware { + my $sourcefile = "hcw85bda.sys"; + my $hash = "0e44dbf63bb0169d57446aec21881ff2"; + my $outfile = "xc3028-v27.fw"; + my $name = "xc2028 firmware"; + my $version = 519; + my $nr_desc = 80; + my $out; + + verify($sourcefile, $hash); + + open INFILE, "<$sourcefile"; + main_firmware($outfile, $name, $version, $nr_desc); + close INFILE; +} + +extract_firmware; +printf "Firmwares generated.\n"; -- cgit v1.2.3 From e8018c9e78b0ff4bb0290e46f4045fb4ea589ae8 Mon Sep 17 00:00:00 2001 From: "Andrey J. Melnikoff (TEMHOTA)" Date: Mon, 7 Jan 2008 05:17:39 -0300 Subject: V4L/DVB (6973): Add Beholder TV 401/405/407/409/505/507/609/M6 support This patch updates cardlist for Beholder TV tuners: old models (with GPIO ir) 401, 403, 405, 407, 409, 505, 507 and add support for 607, 609, M6 cards with new i2c-ir. Signed-off-by: Igor Kuznetsov Signed-off-by: Andrey J. Melnikov Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.saa7134 | 13 + drivers/media/common/ir-keymaps.c | 104 ++++++ drivers/media/video/ir-kbd-i2c.c | 3 +- drivers/media/video/saa7134/saa7134-cards.c | 504 ++++++++++++++++++++++++++++ drivers/media/video/saa7134/saa7134-i2c.c | 2 + drivers/media/video/saa7134/saa7134-input.c | 60 ++++ drivers/media/video/saa7134/saa7134.h | 13 + include/media/ir-common.h | 1 + 8 files changed, 699 insertions(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.saa7134 b/Documentation/video4linux/CARDLIST.saa7134 index 29ca16cdb713..387c35f2ecd7 100644 --- a/Documentation/video4linux/CARDLIST.saa7134 +++ b/Documentation/video4linux/CARDLIST.saa7134 @@ -116,3 +116,16 @@ 115 -> Sabrent PCMCIA TV-PCB05 [0919:2003] 116 -> 10MOONS TM300 TV Card [1131:2304] 117 -> Avermedia Super 007 [1461:f01d] +118 -> Beholder BeholdTV 401 [1131:0000,0000:4016] +119 -> Beholder BeholdTV 403 [0000:4036] +120 -> Beholder BeholdTV 403 FM [1131:0000,0000:4037] +121 -> Beholder BeholdTV 405 [0000:4050] +122 -> Beholder BeholdTV 405 FM [0000:4051] +123 -> [0000:4070] +124 -> Beholder BeholdTV 407 FM [0000:4071] +125 -> Beholder BeholdTV 409 [0000:4090] +126 -> Beholder BeholdTV 505 FM/RDS [0000:5051,0000:505B,5ace:5050] +127 -> Beholder BeholdTV 507 FM/RDS / BeholdTV 509 FM [0000:5071,0000:507B,5ace:5070,5ace:5090] +128 -> Beholder BeholdTV Columbus TVFM [0000:5201] +129 -> Beholder BeholdTV 607 / BeholdTV 609 [5ace:6070,5ace:6071,5ace:6072,5ace:6073,5ace:6090,5ace:6091,5ace:6092,5ace:6093] +130 -> Beholder BeholdTV M6 / BeholdTV M6 Extra [5ace:6190,5ace:6193] diff --git a/drivers/media/common/ir-keymaps.c b/drivers/media/common/ir-keymaps.c index 42762dfb7387..ee2c69e3152a 100644 --- a/drivers/media/common/ir-keymaps.c +++ b/drivers/media/common/ir-keymaps.c @@ -1848,3 +1848,107 @@ IR_KEYTAB_TYPE ir_codes_fusionhdtv_mce[IR_KEYTAB_SIZE] = { }; EXPORT_SYMBOL_GPL(ir_codes_fusionhdtv_mce); + +/* + * Igor Kuznetsov + * Andrey J. Melnikov + * + * Keytable is used by BeholdTV 60x series, M6 series at + * least, and probably other cards too. + * The "ascii-art picture" below (in comments, first row + * is the keycode in hex, and subsequent row(s) shows + * the button labels (several variants when appropriate) + * helps to descide which keycodes to assign to the buttons. + */ +IR_KEYTAB_TYPE ir_codes_behold[IR_KEYTAB_SIZE] = { + + /* 0x1c 0x12 * + * TV/FM POWER * + * */ + [ 0x1c ] = KEY_TUNER, /*XXX KEY_TV KEY_RADIO */ + [ 0x12 ] = KEY_POWER, + + /* 0x01 0x02 0x03 * + * 1 2 3 * + * * + * 0x04 0x05 0x06 * + * 4 5 6 * + * * + * 0x07 0x08 0x09 * + * 7 8 9 * + * */ + [ 0x01 ] = KEY_1, + [ 0x02 ] = KEY_2, + [ 0x03 ] = KEY_3, + [ 0x04 ] = KEY_4, + [ 0x05 ] = KEY_5, + [ 0x06 ] = KEY_6, + [ 0x07 ] = KEY_7, + [ 0x08 ] = KEY_8, + [ 0x09 ] = KEY_9, + + /* 0x0a 0x00 0x17 * + * RECALL 0 MODE * + * */ + [ 0x0a ] = KEY_AGAIN, + [ 0x00 ] = KEY_0, + [ 0x17 ] = KEY_MODE, + + /* 0x14 0x10 * + * ASPECT FULLSCREEN * + * */ + [ 0x14 ] = KEY_SCREEN, + [ 0x10 ] = KEY_ZOOM, + + /* 0x0b * + * Up * + * * + * 0x18 0x16 0x0c * + * Left Ok Right * + * * + * 0x015 * + * Down * + * */ + [ 0x0b ] = KEY_CHANNELUP, /*XXX KEY_UP */ + [ 0x18 ] = KEY_VOLUMEDOWN, /*XXX KEY_LEFT */ + [ 0x16 ] = KEY_OK, /*XXX KEY_ENTER */ + [ 0x0c ] = KEY_VOLUMEUP, /*XXX KEY_RIGHT */ + [ 0x15 ] = KEY_CHANNELDOWN, /*XXX KEY_DOWN */ + + /* 0x11 0x0d * + * MUTE INFO * + * */ + [ 0x11 ] = KEY_MUTE, + [ 0x0d ] = KEY_INFO, + + /* 0x0f 0x1b 0x1a * + * RECORD PLAY/PAUSE STOP * + * * + * 0x0e 0x1f 0x1e * + *TELETEXT AUDIO SOURCE * + * RED YELLOW * + * */ + [ 0x0f ] = KEY_RECORD, + [ 0x1b ] = KEY_PLAYPAUSE, + [ 0x1a ] = KEY_STOP, + [ 0x0e ] = KEY_TEXT, + [ 0x1f ] = KEY_RED, /*XXX KEY_AUDIO */ + [ 0x1e ] = KEY_YELLOW, /*XXX KEY_SOURCE */ + + /* 0x1d 0x13 0x19 * + * SLEEP PREVIEW DVB * + * GREEN BLUE * + * */ + [ 0x1d ] = KEY_SLEEP, + [ 0x13 ] = KEY_GREEN, + [ 0x19 ] = KEY_BLUE, /*XXX KEY_SAT */ + + /* 0x58 0x5c * + * FREEZE SNAPSHOT * + * */ + [ 0x58 ] = KEY_SLOW, + [ 0x5c ] = KEY_SAVE, + +}; + +EXPORT_SYMBOL_GPL(ir_codes_behold); diff --git a/drivers/media/video/ir-kbd-i2c.c b/drivers/media/video/ir-kbd-i2c.c index 29779d8bf7fb..9851987b95fb 100644 --- a/drivers/media/video/ir-kbd-i2c.c +++ b/drivers/media/video/ir-kbd-i2c.c @@ -398,6 +398,7 @@ static int ir_attach(struct i2c_adapter *adap, int addr, case 0x7a: case 0x47: case 0x71: + case 0x2d: if (adap->id == I2C_HW_B_CX2388x) { /* Handled by cx88-input */ name = "CX2388x remote"; @@ -504,7 +505,7 @@ static int ir_probe(struct i2c_adapter *adap) */ static const int probe_bttv[] = { 0x1a, 0x18, 0x4b, 0x64, 0x30, -1}; - static const int probe_saa7134[] = { 0x7a, 0x47, 0x71, -1 }; + static const int probe_saa7134[] = { 0x7a, 0x47, 0x71, 0x2d, -1 }; static const int probe_em28XX[] = { 0x30, 0x47, -1 }; static const int probe_cx88[] = { 0x18, 0x6b, 0x71, -1 }; static const int probe_cx23885[] = { 0x6b, -1 }; diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c index b29427aaa88d..6a35d88fdbb2 100644 --- a/drivers/media/video/saa7134/saa7134-cards.c +++ b/drivers/media/video/saa7134/saa7134-cards.c @@ -3562,6 +3562,329 @@ struct saa7134_board saa7134_boards[] = { .tv = 1, }}, }, + [SAA7134_BOARD_BEHOLD_401] = { + .name = "Beholder BeholdTV 401", + .audio_clock = 0x00187de7, + .tuner_type = TUNER_PHILIPS_FQ1216ME, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .inputs = {{ + .name = name_svideo, + .vmux = 8, + .amux = LINE1, + },{ + .name = name_comp1, + .vmux = 1, + .amux = LINE1, + },{ + .name = name_tv, + .vmux = 3, + .amux = LINE2, + .tv = 1, + }}, + .mute = { + .name = name_mute, + .amux = LINE1, + }, + }, + [SAA7134_BOARD_BEHOLD_403] = { + .name = "Beholder BeholdTV 403", + .audio_clock = 0x00187de7, + .tuner_type = TUNER_PHILIPS_FQ1216ME, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .inputs = {{ + .name = name_svideo, + .vmux = 8, + .amux = LINE1, + },{ + .name = name_comp1, + .vmux = 1, + .amux = LINE1, + },{ + .name = name_tv, + .vmux = 3, + .amux = LINE2, + .tv = 1, + }}, + }, + [SAA7134_BOARD_BEHOLD_403FM] = { + .name = "Beholder BeholdTV 403 FM", + .audio_clock = 0x00187de7, + .tuner_type = TUNER_PHILIPS_FQ1216ME, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .inputs = {{ + .name = name_svideo, + .vmux = 8, + .amux = LINE1, + },{ + .name = name_comp1, + .vmux = 1, + .amux = LINE1, + },{ + .name = name_tv, + .vmux = 3, + .amux = LINE2, + .tv = 1, + }}, + .radio = { + .name = name_radio, + .amux = LINE2, + }, + }, + [SAA7134_BOARD_BEHOLD_405] = { + .name = "Beholder BeholdTV 405", + .audio_clock = 0x00187de7, + .tuner_type = TUNER_PHILIPS_FM1216ME_MK3, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .tda9887_conf = TDA9887_PRESENT, + .inputs = {{ + .name = name_svideo, + .vmux = 8, + .amux = LINE1, + },{ + .name = name_comp1, + .vmux = 3, + .amux = LINE1, + },{ + .name = name_tv, + .vmux = 3, + .amux = LINE2, + .tv = 1, + }}, + }, + [SAA7134_BOARD_BEHOLD_405FM] = { + /* Sergey */ + .name = "Beholder BeholdTV 405 FM", + .audio_clock = 0x00187de7, + .tuner_type = TUNER_PHILIPS_FM1216ME_MK3, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .tda9887_conf = TDA9887_PRESENT, + .inputs = {{ + .name = name_svideo, + .vmux = 8, + .amux = LINE1, + },{ + .name = name_comp1, + .vmux = 3, + .amux = LINE1, + },{ + .name = name_tv, + .vmux = 3, + .amux = LINE2, + .tv = 1, + }}, + .radio = { + .name = name_radio, + .amux = LINE2, + }, + }, + [SAA7134_BOARD_BEHOLD_407FM] = { + .name = "Beholder BeholdTV 407 FM", + .audio_clock = 0x00187de7, + .tuner_type = TUNER_PHILIPS_FM1216ME_MK3, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .tda9887_conf = TDA9887_PRESENT, + .gpiomask = 0xc0c000, + .inputs = {{ + .name = name_svideo, + .vmux = 8, + .amux = LINE1, + .gpio = 0xc0c000, + },{ + .name = name_comp1, + .vmux = 1, + .amux = LINE1, + .gpio = 0xc0c000, + },{ + .name = name_tv, + .vmux = 3, + .amux = TV, + .tv = 1, + .gpio = 0xc0c000, + }}, + .radio = { + .name = name_radio, + .amux = LINE2, + .gpio = 0xc0c000, + }, + }, + [SAA7134_BOARD_BEHOLD_409] = { + .name = "Beholder BeholdTV 409", + .audio_clock = 0x00187de7, + .tuner_type = TUNER_PHILIPS_FM1216ME_MK3, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .tda9887_conf = TDA9887_PRESENT, + .inputs = {{ + .name = name_tv, + .vmux = 3, + .amux = TV, + .tv = 1, + },{ + .name = name_comp1, + .vmux = 1, + .amux = LINE1, + },{ + .name = name_svideo, + .vmux = 8, + .amux = LINE1, + }}, + }, + [SAA7134_BOARD_BEHOLD_505FM] = { + .name = "Beholder BeholdTV 505 FM/RDS", + .audio_clock = 0x00200000, + .tuner_type = TUNER_PHILIPS_FM1216ME_MK3, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .tda9887_conf = TDA9887_PRESENT, + .inputs = {{ + .name = name_tv, + .vmux = 3, + .amux = LINE2, + .tv = 1, + },{ + .name = name_comp1, + .vmux = 1, + .amux = LINE1, + },{ + .name = name_svideo, + .vmux = 8, + .amux = LINE1, + }}, + .mute = { + .name = name_mute, + .amux = LINE1, + }, + .radio = { + .name = name_radio, + .amux = LINE2, + }, + }, + [SAA7134_BOARD_BEHOLD_507_9FM] = { + .name = "Beholder BeholdTV 507 FM/RDS / BeholdTV 509 FM", + .audio_clock = 0x00187de7, + .tuner_type = TUNER_PHILIPS_FM1216ME_MK3, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .tda9887_conf = TDA9887_PRESENT, + .inputs = {{ + .name = name_tv, + .vmux = 3, + .amux = TV, + .tv = 1, + },{ + .name = name_comp1, + .vmux = 1, + .amux = LINE1, + },{ + .name = name_svideo, + .vmux = 8, + .amux = LINE1, + }}, + .radio = { + .name = name_radio, + .amux = LINE2, + }, + }, + [SAA7134_BOARD_BEHOLD_COLUMBUS_TVFM] = { + .name = "Beholder BeholdTV Columbus TVFM", + .audio_clock = 0x00187de7, + .tuner_type = TUNER_ALPS_TSBE5_PAL, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .tda9887_conf = TDA9887_PRESENT, + .inputs = {{ + .name = name_tv, + .vmux = 3, + .amux = TV, + .tv = 1, + },{ + .name = name_comp1, + .vmux = 1, + .amux = LINE1, + },{ + .name = name_svideo, + .vmux = 8, + .amux = LINE1, + }}, + .radio = { + .name = name_radio, + .amux = LINE2, + }, + }, + [SAA7134_BOARD_BEHOLD_607_9FM] = { + /* Andrey Melnikoff */ + .name = "Beholder BeholdTV 607 / BeholdTV 609", + .audio_clock = 0x00187de7, + .tuner_type = TUNER_PHILIPS_FM1216ME_MK3, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .tda9887_conf = TDA9887_PRESENT, + .inputs = {{ + .name = name_tv, + .vmux = 3, + .amux = TV, + .tv = 1, + },{ + .name = name_comp1, + .vmux = 1, + .amux = LINE1, + },{ + .name = name_svideo, + .vmux = 8, + .amux = LINE1, + }}, + .radio = { + .name = name_radio, + .amux = LINE2, + }, + }, + [SAA7134_BOARD_BEHOLD_M6] = { + /* Igor Kuznetsov */ + /* Andrey Melnikoff */ + .name = "Beholder BeholdTV M6 / BeholdTV M6 Extra", + .audio_clock = 0x00187de7, + .tuner_type = TUNER_PHILIPS_FM1216ME_MK3, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .tda9887_conf = TDA9887_PRESENT, + .inputs = {{ + .name = name_tv, + .vmux = 3, + .amux = TV, + .tv = 1, + },{ + .name = name_comp1, + .vmux = 1, + .amux = LINE1, + },{ + .name = name_svideo, + .vmux = 8, + .amux = LINE1, + }}, + .radio = { + .name = name_radio, + .amux = LINE2, + }, + .mpeg = SAA7134_MPEG_EMPRESS, + }, }; const unsigned int saa7134_bcount = ARRAY_SIZE(saa7134_boards); @@ -4326,6 +4649,174 @@ struct pci_device_id saa7134_pci_tbl[] = { .subvendor = 0x1461, /* Avermedia Technologies Inc */ .subdevice = 0xf01d, /* AVerTV DVB-T Super 007 */ .driver_data = SAA7134_BOARD_AVERMEDIA_SUPER_007, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7130, + .subvendor = 0x1131, + .subdevice = 0x0000, + .driver_data = SAA7134_BOARD_BEHOLD_401, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7130, + .subvendor = 0x0000, + .subdevice = 0x4016, + .driver_data = SAA7134_BOARD_BEHOLD_401, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7134, + .subvendor = 0x1131, + .subdevice = 0x0000, + .driver_data = SAA7134_BOARD_BEHOLD_403FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7134, + .subvendor = 0x0000, + .subdevice = 0x4036, + .driver_data = SAA7134_BOARD_BEHOLD_403, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7134, + .subvendor = 0x0000, + .subdevice = 0x4037, + .driver_data = SAA7134_BOARD_BEHOLD_403FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7130, + .subvendor = 0x0000, + .subdevice = 0x4050, + .driver_data = SAA7134_BOARD_BEHOLD_405, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7130, + .subvendor = 0x0000, + .subdevice = 0x4051, + .driver_data = SAA7134_BOARD_BEHOLD_405FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7134, + .subvendor = 0x0000, + .subdevice = 0x4070, + .driver_data = SAA7134_BOARD_BEHOLD_407, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7134, + .subvendor = 0x0000, + .subdevice = 0x4071, + .driver_data = SAA7134_BOARD_BEHOLD_407FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x0000, + .subdevice = 0x4090, + .driver_data = SAA7134_BOARD_BEHOLD_409, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7130, + .subvendor = 0x0000, + .subdevice = 0x5051, + .driver_data = SAA7134_BOARD_BEHOLD_505FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7130, + .subvendor = 0x0000, + .subdevice = 0x505B, + .driver_data = SAA7134_BOARD_BEHOLD_505FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7130, + .subvendor = 0x5ace, + .subdevice = 0x5050, + .driver_data = SAA7134_BOARD_BEHOLD_505FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x0000, + .subdevice = 0x5071, + .driver_data = SAA7134_BOARD_BEHOLD_507_9FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x0000, + .subdevice = 0x507B, + .driver_data = SAA7134_BOARD_BEHOLD_507_9FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7134, + .subvendor = 0x5ace, + .subdevice = 0x5070, + .driver_data = SAA7134_BOARD_BEHOLD_507_9FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x5ace, + .subdevice = 0x5090, + .driver_data = SAA7134_BOARD_BEHOLD_507_9FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x0000, + .subdevice = 0x5201, + .driver_data = SAA7134_BOARD_BEHOLD_COLUMBUS_TVFM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7134, + .subvendor = 0x5ace, + .subdevice = 0x6070, + .driver_data = SAA7134_BOARD_BEHOLD_607_9FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7134, + .subvendor = 0x5ace, + .subdevice = 0x6071, + .driver_data = SAA7134_BOARD_BEHOLD_607_9FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7134, + .subvendor = 0x5ace, + .subdevice = 0x6072, + .driver_data = SAA7134_BOARD_BEHOLD_607_9FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7134, + .subvendor = 0x5ace, + .subdevice = 0x6073, + .driver_data = SAA7134_BOARD_BEHOLD_607_9FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x5ace, + .subdevice = 0x6090, + .driver_data = SAA7134_BOARD_BEHOLD_607_9FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x5ace, + .subdevice = 0x6091, + .driver_data = SAA7134_BOARD_BEHOLD_607_9FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x5ace, + .subdevice = 0x6092, + .driver_data = SAA7134_BOARD_BEHOLD_607_9FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x5ace, + .subdevice = 0x6093, + .driver_data = SAA7134_BOARD_BEHOLD_607_9FM, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x5ace, + .subdevice = 0x6190, + .driver_data = SAA7134_BOARD_BEHOLD_M6, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x5ace, + .subdevice = 0x6193, + .driver_data = SAA7134_BOARD_BEHOLD_M6, },{ .vendor = PCI_VENDOR_ID_PHILIPS, .device = PCI_DEVICE_ID_PHILIPS_SAA7133, @@ -4473,6 +4964,16 @@ int saa7134_board_init1(struct saa7134_dev *dev) case SAA7134_BOARD_ENCORE_ENLTV: case SAA7134_BOARD_ENCORE_ENLTV_FM: case SAA7134_BOARD_10MOONSTVMASTER3: + case SAA7134_BOARD_BEHOLD_401: + case SAA7134_BOARD_BEHOLD_403: + case SAA7134_BOARD_BEHOLD_403FM: + case SAA7134_BOARD_BEHOLD_405: + case SAA7134_BOARD_BEHOLD_405FM: + case SAA7134_BOARD_BEHOLD_407: + case SAA7134_BOARD_BEHOLD_407FM: + case SAA7134_BOARD_BEHOLD_409: + case SAA7134_BOARD_BEHOLD_505FM: + case SAA7134_BOARD_BEHOLD_507_9FM: dev->has_remote = SAA7134_REMOTE_GPIO; break; case SAA7134_BOARD_FLYDVBS_LR300: @@ -4512,6 +5013,7 @@ int saa7134_board_init1(struct saa7134_dev *dev) saa_andorl(SAA7134_GPIO_GPSTATUS0 >> 2, 0x08000000, 0x00000000); break; case SAA7134_BOARD_AVERMEDIA_CARDBUS: + case SAA7134_BOARD_BEHOLD_COLUMBUS_TVFM: /* power-up tuner chip */ saa_andorl(SAA7134_GPIO_GPMODE0 >> 2, 0xffffffff, 0xffffffff); saa_andorl(SAA7134_GPIO_GPSTATUS0 >> 2, 0xffffffff, 0xffffffff); @@ -4533,6 +5035,8 @@ int saa7134_board_init1(struct saa7134_dev *dev) case SAA7134_BOARD_PINNACLE_PCTV_310i: case SAA7134_BOARD_UPMOST_PURPLE_TV: case SAA7134_BOARD_HAUPPAUGE_HVR1110: + case SAA7134_BOARD_BEHOLD_607_9FM: + case SAA7134_BOARD_BEHOLD_M6: dev->has_remote = SAA7134_REMOTE_I2C; break; case SAA7134_BOARD_AVERMEDIA_A169_B: diff --git a/drivers/media/video/saa7134/saa7134-i2c.c b/drivers/media/video/saa7134/saa7134-i2c.c index 800b397f8f16..d3322c3018f2 100644 --- a/drivers/media/video/saa7134/saa7134-i2c.c +++ b/drivers/media/video/saa7134/saa7134-i2c.c @@ -334,6 +334,7 @@ static int attach_inform(struct i2c_client *client) case 0x7a: case 0x47: case 0x71: + case 0x2d: { struct IR_i2c *ir = i2c_get_clientdata(client); d1printk("%s i2c IR detected (%s).\n", @@ -439,6 +440,7 @@ static char *i2c_devs[128] = { [ 0xa0 >> 1 ] = "eeprom", [ 0xc0 >> 1 ] = "tuner (analog)", [ 0x86 >> 1 ] = "tda9887", + [ 0x5a >> 1 ] = "remote control", }; static void do_i2c_scan(char *name, struct i2c_client *c) diff --git a/drivers/media/video/saa7134/saa7134-input.c b/drivers/media/video/saa7134/saa7134-input.c index 8cfeb2bde90f..7f4212928e9f 100644 --- a/drivers/media/video/saa7134/saa7134-input.c +++ b/drivers/media/video/saa7134/saa7134-input.c @@ -52,6 +52,11 @@ module_param(repeat_period, int, 0644); MODULE_PARM_DESC(repeat_period, "repeat period between " "keypresses when key is down"); +static unsigned int disable_other_ir; +module_param(disable_other_ir, int, 0644); +MODULE_PARM_DESC(disable_other_ir, "disable full codes of " + "alternative remotes from other manufacturers"); + #define dprintk(fmt, arg...) if (ir_debug) \ printk(KERN_DEBUG "%s/ir: " fmt, dev->name , ## arg) #define i2cdprintk(fmt, arg...) if (ir_debug) \ @@ -154,6 +159,45 @@ static int get_key_hvr1110(struct IR_i2c *ir, u32 *ir_key, u32 *ir_raw) return 1; } + +static int get_key_beholdm6xx(struct IR_i2c *ir, u32 *ir_key, u32 *ir_raw) +{ + unsigned char data[12]; + u32 gpio; + + struct saa7134_dev *dev = ir->c.adapter->algo_data; + + /* rising SAA7134_GPIO_GPRESCAN reads the status */ + saa_clearb(SAA7134_GPIO_GPMODE3, SAA7134_GPIO_GPRESCAN); + saa_setb(SAA7134_GPIO_GPMODE3, SAA7134_GPIO_GPRESCAN); + + gpio = saa_readl(SAA7134_GPIO_GPSTATUS0 >> 2); + + if (0x400000 &~ gpio) + return 0; /* No button press */ + + ir->c.addr = 0x5a >> 1; + + if (12 != i2c_master_recv(&ir->c, data, 12)) { + i2cdprintk("read error\n"); + return -EIO; + } + /* IR of this card normally decode signals NEC-standard from + * - Sven IHOO MT 5.1R remote. xxyye718 + * - Sven DVD HD-10xx remote. xxyyf708 + * - BBK ... + * - mayby others + * So, skip not our, if disable full codes mode. + */ + if (data[10] != 0x6b && data[11] != 0x86 && disable_other_ir) + return 0; + + *ir_key = data[9]; + *ir_raw = data[9]; + + return 1; +} + void saa7134_input_irq(struct saa7134_dev *dev) { struct card_ir *ir = dev->remote; @@ -288,6 +332,16 @@ int saa7134_input_init1(struct saa7134_dev *dev) case SAA7134_BOARD_MANLI_MTV001: case SAA7134_BOARD_MANLI_MTV002: case SAA7134_BOARD_BEHOLD_409FM: + case SAA7134_BOARD_BEHOLD_401: + case SAA7134_BOARD_BEHOLD_403: + case SAA7134_BOARD_BEHOLD_403FM: + case SAA7134_BOARD_BEHOLD_405: + case SAA7134_BOARD_BEHOLD_405FM: + case SAA7134_BOARD_BEHOLD_407: + case SAA7134_BOARD_BEHOLD_407FM: + case SAA7134_BOARD_BEHOLD_409: + case SAA7134_BOARD_BEHOLD_505FM: + case SAA7134_BOARD_BEHOLD_507_9FM: ir_codes = ir_codes_manli; mask_keycode = 0x001f00; mask_keyup = 0x004000; @@ -458,6 +512,12 @@ void saa7134_set_i2c_ir(struct saa7134_dev *dev, struct IR_i2c *ir) ir->get_key = get_key_hvr1110; ir->ir_codes = ir_codes_hauppauge_new; break; + case SAA7134_BOARD_BEHOLD_607_9FM: + case SAA7134_BOARD_BEHOLD_M6: + snprintf(ir->c.name, sizeof(ir->c.name), "BeholdTV"); + ir->get_key = get_key_beholdm6xx; + ir->ir_codes = ir_codes_behold; + break; default: dprintk("Shouldn't get here: Unknown board %x for I2C IR?\n",dev->board); break; diff --git a/drivers/media/video/saa7134/saa7134.h b/drivers/media/video/saa7134/saa7134.h index 3e77aeb7974f..ce450304fb53 100644 --- a/drivers/media/video/saa7134/saa7134.h +++ b/drivers/media/video/saa7134/saa7134.h @@ -240,6 +240,19 @@ struct saa7134_format { #define SAA7134_BOARD_SABRENT_TV_PCB05 115 #define SAA7134_BOARD_10MOONSTVMASTER3 116 #define SAA7134_BOARD_AVERMEDIA_SUPER_007 117 +#define SAA7134_BOARD_BEHOLD_401 118 +#define SAA7134_BOARD_BEHOLD_403 119 +#define SAA7134_BOARD_BEHOLD_403FM 120 +#define SAA7134_BOARD_BEHOLD_405 121 +#define SAA7134_BOARD_BEHOLD_405FM 122 +#define SAA7134_BOARD_BEHOLD_407 123 +#define SAA7134_BOARD_BEHOLD_407FM 124 +#define SAA7134_BOARD_BEHOLD_409 125 +#define SAA7134_BOARD_BEHOLD_505FM 126 +#define SAA7134_BOARD_BEHOLD_507_9FM 127 +#define SAA7134_BOARD_BEHOLD_COLUMBUS_TVFM 128 +#define SAA7134_BOARD_BEHOLD_607_9FM 129 +#define SAA7134_BOARD_BEHOLD_M6 130 #define SAA7134_MAXBOARDS 8 #define SAA7134_INPUT_MAX 8 diff --git a/include/media/ir-common.h b/include/media/ir-common.h index b904d3257260..9c2fc09a7133 100644 --- a/include/media/ir-common.h +++ b/include/media/ir-common.h @@ -140,6 +140,7 @@ extern IR_KEYTAB_TYPE ir_codes_asus_pc39[IR_KEYTAB_SIZE]; extern IR_KEYTAB_TYPE ir_codes_encore_enltv[IR_KEYTAB_SIZE]; extern IR_KEYTAB_TYPE ir_codes_tt_1500[IR_KEYTAB_SIZE]; extern IR_KEYTAB_TYPE ir_codes_fusionhdtv_mce[IR_KEYTAB_SIZE]; +extern IR_KEYTAB_TYPE ir_codes_behold[IR_KEYTAB_SIZE]; #endif -- cgit v1.2.3 From 1434bc0cbb842ed71eada244ac46185e3b17e4e2 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Mon, 7 Jan 2008 05:20:04 -0300 Subject: V4L/DVB (6975): Add missing entry for Beholder BeholdTV 407 Acked-by: Andrey J. Melnikov Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.saa7134 | 2 +- drivers/media/video/saa7134/saa7134-cards.c | 27 +++++++++++++++++++++++++++ 2 files changed, 28 insertions(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.saa7134 b/Documentation/video4linux/CARDLIST.saa7134 index 387c35f2ecd7..fda570e50fa8 100644 --- a/Documentation/video4linux/CARDLIST.saa7134 +++ b/Documentation/video4linux/CARDLIST.saa7134 @@ -121,7 +121,7 @@ 120 -> Beholder BeholdTV 403 FM [1131:0000,0000:4037] 121 -> Beholder BeholdTV 405 [0000:4050] 122 -> Beholder BeholdTV 405 FM [0000:4051] -123 -> [0000:4070] +123 -> Beholder BeholdTV 407 [0000:4070] 124 -> Beholder BeholdTV 407 FM [0000:4071] 125 -> Beholder BeholdTV 409 [0000:4090] 126 -> Beholder BeholdTV 505 FM/RDS [0000:5051,0000:505B,5ace:5050] diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c index 6a35d88fdbb2..46b6ac4b3105 100644 --- a/drivers/media/video/saa7134/saa7134-cards.c +++ b/drivers/media/video/saa7134/saa7134-cards.c @@ -3687,6 +3687,33 @@ struct saa7134_board saa7134_boards[] = { .amux = LINE2, }, }, + [SAA7134_BOARD_BEHOLD_407] = { + .name = "Beholder BeholdTV 407", + .audio_clock = 0x00187de7, + .tuner_type = TUNER_PHILIPS_FM1216ME_MK3, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .tda9887_conf = TDA9887_PRESENT, + .gpiomask = 0xc0c000, + .inputs = {{ + .name = name_svideo, + .vmux = 8, + .amux = LINE1, + .gpio = 0xc0c000, + },{ + .name = name_comp1, + .vmux = 1, + .amux = LINE1, + .gpio = 0xc0c000, + },{ + .name = name_tv, + .vmux = 3, + .amux = TV, + .tv = 1, + .gpio = 0xc0c000, + }}, + }, [SAA7134_BOARD_BEHOLD_407FM] = { .name = "Beholder BeholdTV 407 FM", .audio_clock = 0x00187de7, -- cgit v1.2.3 From e9ca1c79f2868d26c3e74ff3f20d5fa7d30724c5 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Wed, 9 Jan 2008 21:35:56 -0300 Subject: V4L/DVB (6993): Fix CARDLIST.saa7134 Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.saa7134 | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.saa7134 b/Documentation/video4linux/CARDLIST.saa7134 index fda570e50fa8..fe8514e06d26 100644 --- a/Documentation/video4linux/CARDLIST.saa7134 +++ b/Documentation/video4linux/CARDLIST.saa7134 @@ -116,9 +116,9 @@ 115 -> Sabrent PCMCIA TV-PCB05 [0919:2003] 116 -> 10MOONS TM300 TV Card [1131:2304] 117 -> Avermedia Super 007 [1461:f01d] -118 -> Beholder BeholdTV 401 [1131:0000,0000:4016] +118 -> Beholder BeholdTV 401 [0000:4016] 119 -> Beholder BeholdTV 403 [0000:4036] -120 -> Beholder BeholdTV 403 FM [1131:0000,0000:4037] +120 -> Beholder BeholdTV 403 FM [0000:4037] 121 -> Beholder BeholdTV 405 [0000:4050] 122 -> Beholder BeholdTV 405 FM [0000:4051] 123 -> Beholder BeholdTV 407 [0000:4070] -- cgit v1.2.3 From eeec4b3853a8af0cd2b42af24b9e8fd0e0ef1f8a Mon Sep 17 00:00:00 2001 From: Thierry MERLE Date: Wed, 9 Jan 2008 17:33:53 -0300 Subject: V4L/DVB (6994): usbvision: add Pinnacle Studio PCTV USB (NTSC) FM V3 Add the "Pinnacle Studio PCTV USB (NTSC) FM" device. This is the third occurrence of the same device designation... Signed-off-by: Thierry MERLE Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.usbvision | 1 + drivers/media/video/usbvision/usbvision-cards.c | 18 +++++++++++++++++- drivers/media/video/usbvision/usbvision-cards.h | 1 + 3 files changed, 19 insertions(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.usbvision b/Documentation/video4linux/CARDLIST.usbvision index 3d6850ef0245..0b72d3fee17e 100644 --- a/Documentation/video4linux/CARDLIST.usbvision +++ b/Documentation/video4linux/CARDLIST.usbvision @@ -62,3 +62,4 @@ 61 -> Pinnacle Studio Linx Video input cable (PAL) [2304:0301] 62 -> Pinnacle PCTV Bungee USB (PAL) FM [2304:0419] 63 -> Hauppauge WinTv-USB [2400:4200] + 64 -> Pinnacle Studio PCTV USB (NTSC) FM V3 [2304:0113] diff --git a/drivers/media/video/usbvision/usbvision-cards.c b/drivers/media/video/usbvision/usbvision-cards.c index f09eb102731b..503b13beb922 100644 --- a/drivers/media/video/usbvision/usbvision-cards.c +++ b/drivers/media/video/usbvision/usbvision-cards.c @@ -901,6 +901,20 @@ struct usbvision_device_data_st usbvision_device_data[] = { .Y_Offset = -1, .ModelString = "Pinnacle Studio PCTV USB (NTSC) FM", }, + [PINNA_PCTV_USB_NTSC_FM_V3] = { + .Interface = -1, + .Codec = CODEC_SAA7111, + .VideoChannels = 3, + .VideoNorm = V4L2_STD_NTSC, + .AudioChannels = 1, + .Radio = 1, + .vbi = 1, + .Tuner = 1, + .TunerType = TUNER_PHILIPS_NTSC_M, + .X_Offset = -1, + .Y_Offset = -1, + .ModelString = "Pinnacle Studio PCTV USB (NTSC) FM V3", + }, [PINNA_PCTV_USB_PAL_FM_V2] = { .Interface = -1, .Codec = CODEC_SAA7113, @@ -1044,7 +1058,7 @@ struct usb_device_id usbvision_table [] = { { USB_DEVICE(0x0573, 0x4d2a), .driver_info=HPG_WINTV_PRO_NTSC_MN }, { USB_DEVICE(0x0573, 0x4d2b), .driver_info=HPG_WINTV_PRO_NTSC_MN_V2 }, { USB_DEVICE(0x0573, 0x4d2c), .driver_info=HPG_WINTV_PRO_PAL }, - { USB_DEVICE(0x0573, 0x4d20), .driver_info=HPG_WINTV_PRO_NTSC_MN_V3 }, + { USB_DEVICE(0x0573, 0x4d20), .driver_info = HPG_WINTV_PRO_NTSC_MN_V3 }, { USB_DEVICE(0x0573, 0x4d21), .driver_info=HPG_WINTV_PRO_PAL_BG }, { USB_DEVICE(0x0573, 0x4d22), .driver_info=HPG_WINTV_PRO_PAL_I }, { USB_DEVICE(0x0573, 0x4d23), .driver_info=HPG_WINTV_PRO_PAL_SECAM_L }, @@ -1074,6 +1088,8 @@ struct usb_device_id usbvision_table [] = { { USB_DEVICE(0x2304, 0x0110), .driver_info=PINNA_PCTV_USB_PAL_FM }, { USB_DEVICE(0x2304, 0x0111), .driver_info=MIRO_PCTV_USB }, { USB_DEVICE(0x2304, 0x0112), .driver_info=PINNA_PCTV_USB_NTSC_FM }, + { USB_DEVICE(0x2304, 0x0113), + .driver_info = PINNA_PCTV_USB_NTSC_FM_V3 }, { USB_DEVICE(0x2304, 0x0210), .driver_info=PINNA_PCTV_USB_PAL_FM_V2 }, { USB_DEVICE(0x2304, 0x0212), .driver_info=PINNA_PCTV_USB_NTSC_FM_V2 }, { USB_DEVICE(0x2304, 0x0214), .driver_info=PINNA_PCTV_USB_PAL_FM_V3 }, diff --git a/drivers/media/video/usbvision/usbvision-cards.h b/drivers/media/video/usbvision/usbvision-cards.h index 512c5cee4145..9c6ad22960d8 100644 --- a/drivers/media/video/usbvision/usbvision-cards.h +++ b/drivers/media/video/usbvision/usbvision-cards.h @@ -62,5 +62,6 @@ #define PINNA_LINX_VD_IN_CAB_PAL 61 #define PINNA_PCTV_BUNGEE_PAL_FM 62 #define HPG_WINTV 63 +#define PINNA_PCTV_USB_NTSC_FM_V3 64 extern const int usbvision_device_data_size; -- cgit v1.2.3 From b6667e5fb8c478e65768fdb18fa6576ac1634068 Mon Sep 17 00:00:00 2001 From: Hermann Pitton Date: Sat, 12 Jan 2008 14:43:48 -0300 Subject: V4L/DVB (7017): saa7134: add MSI TV@nywhere Plus with the older tda8275 tuner Thanks go to Mark Schultz for the initial contribution and to Dean Hilkewich for testing it again. Signed-off-by: Hermann Pitton Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.saa7134 | 2 +- drivers/media/video/saa7134/saa7134-cards.c | 8 +++++++- 2 files changed, 8 insertions(+), 2 deletions(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.saa7134 b/Documentation/video4linux/CARDLIST.saa7134 index fe8514e06d26..5d3b6b4d2515 100644 --- a/Documentation/video4linux/CARDLIST.saa7134 +++ b/Documentation/video4linux/CARDLIST.saa7134 @@ -80,7 +80,7 @@ 79 -> Sedna/MuchTV PC TV Cardbus TV/Radio (ITO25 Rev:2B) 80 -> ASUS Digimatrix TV [1043:0210] 81 -> Philips Tiger reference design [1131:2018] - 82 -> MSI TV@Anywhere plus [1462:6231] + 82 -> MSI TV@Anywhere plus [1462:6231,1462:8624] 83 -> Terratec Cinergy 250 PCI TV [153b:1160] 84 -> LifeView FlyDVB Trio [5168:0319] 85 -> AverTV DVB-T 777 [1461:2c05,1461:2c05] diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c index 4a5a1459f025..7d7f383b404f 100644 --- a/drivers/media/video/saa7134/saa7134-cards.c +++ b/drivers/media/video/saa7134/saa7134-cards.c @@ -4392,7 +4392,13 @@ struct pci_device_id saa7134_pci_tbl[] = { .vendor = PCI_VENDOR_ID_PHILIPS, .device = PCI_DEVICE_ID_PHILIPS_SAA7133, .subvendor = 0x1462, - .subdevice = 0x6231, + .subdevice = 0x6231, /* tda8275a, ks003 IR */ + .driver_data = SAA7134_BOARD_MSI_TVATANYWHERE_PLUS, + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x1462, + .subdevice = 0x8624, /* tda8275, ks003 IR */ .driver_data = SAA7134_BOARD_MSI_TVATANYWHERE_PLUS, },{ .vendor = PCI_VENDOR_ID_PHILIPS, -- cgit v1.2.3 From 15b9becc68793209a2afd6c580bf71a71fee90a6 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Sun, 13 Jan 2008 12:02:20 -0300 Subject: V4L/DVB (7020): Add USB ID for a newer variant of Hauppauge WinTV USB2 Thanks to Jeroen Janssen Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.em28xx | 2 +- drivers/media/video/em28xx/em28xx-cards.c | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.em28xx b/Documentation/video4linux/CARDLIST.em28xx index dbc19ca9e483..6a8469f2bcae 100644 --- a/Documentation/video4linux/CARDLIST.em28xx +++ b/Documentation/video4linux/CARDLIST.em28xx @@ -2,7 +2,7 @@ 1 -> Unknown EM2750/28xx video grabber (em2820/em2840) [eb1a:2750,eb1a:2820,eb1a:2821,eb1a:2860,eb1a:2861,eb1a:2870,eb1a:2881,eb1a:2883] 2 -> Terratec Cinergy 250 USB (em2820/em2840) [0ccd:0036] 3 -> Pinnacle PCTV USB 2 (em2820/em2840) [2304:0208] - 4 -> Hauppauge WinTV USB 2 (em2820/em2840) [2040:4200] + 4 -> Hauppauge WinTV USB 2 (em2820/em2840) [2040:4200,2040:4201] 5 -> MSI VOX USB 2.0 (em2820/em2840) 6 -> Terratec Cinergy 200 USB (em2800) 7 -> Leadtek Winfast USB II (em2800) diff --git a/drivers/media/video/em28xx/em28xx-cards.c b/drivers/media/video/em28xx/em28xx-cards.c index 5ffdbaab3ad3..fe0ee4518479 100644 --- a/drivers/media/video/em28xx/em28xx-cards.c +++ b/drivers/media/video/em28xx/em28xx-cards.c @@ -419,6 +419,8 @@ struct usb_device_id em28xx_id_table [] = { .driver_info = EM2820_BOARD_PINNACLE_USB_2 }, { USB_DEVICE(0x2040, 0x4200), .driver_info = EM2820_BOARD_HAUPPAUGE_WINTV_USB_2 }, + { USB_DEVICE(0x2040, 0x4201), + .driver_info = EM2820_BOARD_HAUPPAUGE_WINTV_USB_2 }, { USB_DEVICE(0x2304, 0x0207), .driver_info = EM2820_BOARD_PINNACLE_DVC_90 }, { USB_DEVICE(0x2304, 0x021a), -- cgit v1.2.3 From 27c685a4b09b5e391023d769cddb97e4fcb3b9e1 Mon Sep 17 00:00:00 2001 From: Steven Toth Date: Sat, 5 Jan 2008 16:50:14 -0300 Subject: V4L/DVB (7042): xc5000: Tuner analog support From Zhang: This an updated patch that adds analog support for the xc5000 tuner driver. it was tested on a Pinnacle PCTV HD 800i card (patches to follow). Patch commited as-is, cleanup to follow ... Steve. Signed-off-by: Chaogui Zhang Signed-off-by: Steven Toth Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.tuner | 1 + drivers/media/dvb/frontends/xc5000.c | 213 ++++++++++++++++++++++++++----- drivers/media/dvb/frontends/xc5000.h | 12 +- drivers/media/video/tuner-core.c | 20 +++ drivers/media/video/tuner-types.c | 4 + include/media/tuner.h | 1 + 6 files changed, 213 insertions(+), 38 deletions(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.tuner b/Documentation/video4linux/CARDLIST.tuner index 2211c8eac643..0e2394695bb8 100644 --- a/Documentation/video4linux/CARDLIST.tuner +++ b/Documentation/video4linux/CARDLIST.tuner @@ -73,3 +73,4 @@ tuner=71 - Xceive xc2028/xc3028 tuner tuner=72 - Thomson FE6600 tuner=73 - Samsung TCPG 6121P30A tuner=75 - Philips TEA5761 FM Radio +tuner=76 - Xceive 5000 tuner diff --git a/drivers/media/dvb/frontends/xc5000.c b/drivers/media/dvb/frontends/xc5000.c index a036530ee4b4..28048d5307a2 100644 --- a/drivers/media/dvb/frontends/xc5000.c +++ b/drivers/media/dvb/frontends/xc5000.c @@ -22,6 +22,7 @@ #include #include +#include #include #include #include @@ -56,6 +57,10 @@ MODULE_PARM_DESC(debug, "Turn on/off debugging (default:off)."); #define XC_RESULT_I2C_READ_FAILURE 3 #define XC_RESULT_OUT_OF_RANGE 5 +/* Product id */ +#define XC_PRODUCT_ID_FW_NOT_LOADED 0x2000 +#define XC_PRODUCT_ID_FW_LOADED 0x1388 + /* Registers */ #define XREG_INIT 0x00 #define XREG_VIDEO_MODE 0x01 @@ -122,7 +127,29 @@ typedef struct { } XC_TV_STANDARD; /* Tuner standards */ -#define DTV6 17 +#define MN_NTSC_PAL_BTSC 0 +#define MN_NTSC_PAL_A2 1 +#define MN_NTSC_PAL_EIAJ 2 +#define MN_NTSC_PAL_Mono 3 +#define BG_PAL_A2 4 +#define BG_PAL_NICAM 5 +#define BG_PAL_MONO 6 +#define I_PAL_NICAM 7 +#define I_PAL_NICAM_MONO 8 +#define DK_PAL_A2 9 +#define DK_PAL_NICAM 10 +#define DK_PAL_MONO 11 +#define DK_SECAM_A2DK1 12 +#define DK_SECAM_A2LDK3 13 +#define DK_SECAM_A2MONO 14 +#define L_SECAM_NICAM 15 +#define LC_SECAM_NICAM 16 +#define DTV6 17 +#define DTV8 18 +#define DTV7_8 19 +#define DTV7 20 +#define FM_Radio_INPUT2 21 +#define FM_Radio_INPUT1 22 XC_TV_STANDARD XC5000_Standard[MAX_TV_STANDARD] = { {"M/N-NTSC/PAL-BTSC", 0x0400, 0x8020}, @@ -184,12 +211,13 @@ static void xc5000_TunerReset(struct dvb_frontend *fe) dprintk(1, "%s()\n", __FUNCTION__); - if (priv->cfg->tuner_reset) { - ret = priv->cfg->tuner_reset(fe); + if (priv->cfg->tuner_callback) { + ret = priv->cfg->tuner_callback(priv->cfg->video_dev, + XC5000_TUNER_RESET, 0); if (ret) printk(KERN_ERR "xc5000: reset failed\n"); } else - printk(KERN_ERR "xc5000: no tuner reset function, fatal\n"); + printk(KERN_ERR "xc5000: no tuner reset callback function, fatal\n"); } static int xc_write_reg(struct xc5000_priv *priv, u16 regAddr, u16 i2cData) @@ -259,7 +287,6 @@ static int xc_load_i2c_sequence(struct dvb_frontend *fe, u8 i2c_sequence[]) index=0; while ((i2c_sequence[index]!=0xFF) || (i2c_sequence[index+1]!=0xFF)) { - len = i2c_sequence[index]* 256 + i2c_sequence[index+1]; if (len == 0x0000) { /* RESET command */ @@ -311,7 +338,7 @@ static int xc_SetTVStandard(struct xc5000_priv *priv, u16 VideoMode, u16 AudioMode) { int ret; - dprintk(1, "%s(%d,%d)\n", __FUNCTION__, VideoMode, AudioMode); + dprintk(1, "%s(0x%04x,0x%04x)\n", __FUNCTION__, VideoMode, AudioMode); dprintk(1, "%s() Standard = %s\n", __FUNCTION__, XC5000_Standard[priv->video_standard].Name); @@ -325,7 +352,11 @@ static int xc_SetTVStandard(struct xc5000_priv *priv, static int xc_shutdown(struct xc5000_priv *priv) { - return xc_write_reg(priv, XREG_POWER_DOWN, 0); + return 0; + /* Fixme: cannot bring tuner back alive once shutdown + * without reloading the driver modules. + * return xc_write_reg(priv, XREG_POWER_DOWN, 0); + */ } static int xc_SetSignalSource(struct xc5000_priv *priv, u16 rf_mode) @@ -349,7 +380,7 @@ static int xc_set_RF_frequency(struct xc5000_priv *priv, u32 freq_hz) { u16 freq_code; - dprintk(1, "%s(%d)\n", __FUNCTION__, freq_hz); + dprintk(1, "%s(%u)\n", __FUNCTION__, freq_hz); if ((freq_hz > xc5000_tuner_ops.info.frequency_max) || (freq_hz < xc5000_tuner_ops.info.frequency_min)) @@ -457,7 +488,7 @@ static int xc_tune_channel(struct xc5000_priv *priv, u32 freq_hz) { int found = 0; - dprintk(1, "%s(%d)\n", __FUNCTION__, freq_hz); + dprintk(1, "%s(%u)\n", __FUNCTION__, freq_hz); if (xc_set_RF_frequency(priv, freq_hz) != XC_RESULT_SUCCESS) return 0; @@ -480,7 +511,7 @@ static int xc5000_readreg(struct xc5000_priv *priv, u16 reg, u16 *val) }; if (i2c_transfer(priv->i2c, msg, 2) != 2) { - printk(KERN_WARNING "xc5000 I2C read failed\n"); + printk(KERN_WARNING "xc5000: I2C read failed\n"); return -EREMOTEIO; } @@ -494,7 +525,7 @@ static int xc5000_writeregs(struct xc5000_priv *priv, u8 *buf, u8 len) .flags = 0, .buf = buf, .len = len }; if (i2c_transfer(priv->i2c, &msg, 1) != 1) { - printk(KERN_ERR "xc5000 I2C write failed (len=%i)\n", + printk(KERN_ERR "xc5000: I2C write failed (len=%i)\n", (int)len); return -EREMOTEIO; } @@ -519,16 +550,11 @@ static int xc5000_fwupload(struct dvb_frontend* fe) const struct firmware *fw; int ret; - if (!priv->cfg->request_firmware) { - printk(KERN_ERR "xc5000: no firmware callback, fatal\n"); - return -EIO; - } - /* request the firmware, this will block and timeout */ printk(KERN_INFO "xc5000: waiting for firmware upload (%s)...\n", XC5000_DEFAULT_FIRMWARE); - ret = priv->cfg->request_firmware(fe, &fw, XC5000_DEFAULT_FIRMWARE); + ret = request_firmware(&fw, XC5000_DEFAULT_FIRMWARE, &priv->i2c->dev); if (ret) { printk(KERN_ERR "xc5000: Upload failed. (file not found?)\n"); ret = XC_RESULT_RESET_FAILURE; @@ -601,7 +627,6 @@ static int xc5000_set_params(struct dvb_frontend *fe, dprintk(1, "%s() frequency=%d (Hz)\n", __FUNCTION__, params->frequency); - switch(params->u.vsb.modulation) { case VSB_8: case VSB_16: @@ -658,6 +683,93 @@ static int xc5000_set_params(struct dvb_frontend *fe, return 0; } +static int xc_load_fw_and_init_tuner(struct dvb_frontend *fe); + +static int xc5000_set_analog_params(struct dvb_frontend *fe, + struct analog_parameters *params) +{ + struct xc5000_priv *priv = fe->tuner_priv; + int ret; + + if(priv->fwloaded == 0) + xc_load_fw_and_init_tuner(fe); + + dprintk(1, "%s() frequency=%d (in units of 62.5khz)\n", + __FUNCTION__, params->frequency); + + priv->rf_mode = XC_RF_MODE_CABLE; /* Fix me: it could be air. */ + + /* params->frequency is in units of 62.5khz */ + priv->freq_hz = params->frequency * 62500; + + /* FIX ME: Some video standards may have several possible audio + standards. We simply default to one of them here. + */ + if(params->std & V4L2_STD_MN) { + /* default to BTSC audio standard */ + priv->video_standard = MN_NTSC_PAL_BTSC; + goto tune_channel; + } + + if(params->std & V4L2_STD_PAL_BG) { + /* default to NICAM audio standard */ + priv->video_standard = BG_PAL_NICAM; + goto tune_channel; + } + + if(params->std & V4L2_STD_PAL_I) { + /* default to NICAM audio standard */ + priv->video_standard = I_PAL_NICAM; + goto tune_channel; + } + + if(params->std & V4L2_STD_PAL_DK) { + /* default to NICAM audio standard */ + priv->video_standard = DK_PAL_NICAM; + goto tune_channel; + } + + if(params->std & V4L2_STD_SECAM_DK) { + /* default to A2 DK1 audio standard */ + priv->video_standard = DK_SECAM_A2DK1; + goto tune_channel; + } + + if(params->std & V4L2_STD_SECAM_L) { + priv->video_standard = L_SECAM_NICAM; + goto tune_channel; + } + + if(params->std & V4L2_STD_SECAM_LC) { + priv->video_standard = LC_SECAM_NICAM; + goto tune_channel; + } + +tune_channel: + ret = xc_SetSignalSource(priv, priv->rf_mode); + if (ret != XC_RESULT_SUCCESS) { + printk(KERN_ERR + "xc5000: xc_SetSignalSource(%d) failed\n", + priv->rf_mode); + return -EREMOTEIO; + } + + ret = xc_SetTVStandard(priv, + XC5000_Standard[priv->video_standard].VideoMode, + XC5000_Standard[priv->video_standard].AudioMode); + if (ret != XC_RESULT_SUCCESS) { + printk(KERN_ERR "xc5000: xc_SetTVStandard failed\n"); + return -EREMOTEIO; + } + + xc_tune_channel(priv, priv->freq_hz); + + if (debug) + xc_debug_dump(priv); + + return 0; +} + static int xc5000_get_frequency(struct dvb_frontend *fe, u32 *freq) { struct xc5000_priv *priv = fe->tuner_priv; @@ -670,6 +782,7 @@ static int xc5000_get_bandwidth(struct dvb_frontend *fe, u32 *bw) { struct xc5000_priv *priv = fe->tuner_priv; dprintk(1, "%s()\n", __FUNCTION__); + *bw = priv->bandwidth; return 0; } @@ -691,13 +804,12 @@ static int xc5000_get_status(struct dvb_frontend *fe, u32 *status) static int xc_load_fw_and_init_tuner(struct dvb_frontend *fe) { struct xc5000_priv *priv = fe->tuner_priv; - int ret; + int ret = 0; if (priv->fwloaded == 0) { ret = xc5000_fwupload(fe); if (ret != XC_RESULT_SUCCESS) return ret; - priv->fwloaded = 1; } @@ -720,9 +832,27 @@ static int xc_load_fw_and_init_tuner(struct dvb_frontend *fe) static int xc5000_sleep(struct dvb_frontend *fe) { struct xc5000_priv *priv = fe->tuner_priv; + int ret; + dprintk(1, "%s()\n", __FUNCTION__); - return xc_shutdown(priv); + /* On Pinnacle PCTV HD 800i, the tuner cannot be reinitialized + * once shutdown without reloading the driver. Maybe I am not + * doing something right. + * + */ + + ret = xc_shutdown(priv); + if(ret != XC_RESULT_SUCCESS) { + printk(KERN_ERR + "xc5000: %s() unable to shutdown tuner\n", + __FUNCTION__); + return -EREMOTEIO; + } + else { + /* priv->fwloaded = 0; */ + return XC_RESULT_SUCCESS; + } } static int xc5000_init(struct dvb_frontend *fe) @@ -757,14 +887,15 @@ static const struct dvb_tuner_ops xc5000_tuner_ops = { .frequency_step = 50000, }, - .release = xc5000_release, - .init = xc5000_init, - .sleep = xc5000_sleep, + .release = xc5000_release, + .init = xc5000_init, + .sleep = xc5000_sleep, - .set_params = xc5000_set_params, - .get_frequency = xc5000_get_frequency, - .get_bandwidth = xc5000_get_bandwidth, - .get_status = xc5000_get_status + .set_params = xc5000_set_params, + .set_analog_params = xc5000_set_analog_params, + .get_frequency = xc5000_get_frequency, + .get_bandwidth = xc5000_get_bandwidth, + .get_status = xc5000_get_status }; struct dvb_frontend * xc5000_attach(struct dvb_frontend *fe, @@ -783,14 +914,33 @@ struct dvb_frontend * xc5000_attach(struct dvb_frontend *fe, priv->cfg = cfg; priv->bandwidth = BANDWIDTH_6_MHZ; priv->i2c = i2c; - priv->fwloaded = 0; + /* Check if firmware has been loaded. It is possible that another + instance of the driver has loaded the firmware. + */ if (xc5000_readreg(priv, XREG_PRODUCT_ID, &id) != 0) { kfree(priv); return NULL; } - if ((id != 0x2000) && (id != 0x1388)) { + switch(id) { + case XC_PRODUCT_ID_FW_LOADED: + printk(KERN_INFO + "xc5000: Successfully identified at address 0x%02x\n", + cfg->i2c_address); + printk(KERN_INFO + "xc5000: Firmware has been loaded previously\n"); + priv->fwloaded = 1; + break; + case XC_PRODUCT_ID_FW_NOT_LOADED: + printk(KERN_INFO + "xc5000: Successfully identified at address 0x%02x\n", + cfg->i2c_address); + printk(KERN_INFO + "xc5000: Firmware has not been loaded previously\n"); + priv->fwloaded = 0; + break; + default: printk(KERN_ERR "xc5000: Device not found at addr 0x%02x (0x%x)\n", cfg->i2c_address, id); @@ -798,9 +948,6 @@ struct dvb_frontend * xc5000_attach(struct dvb_frontend *fe, return NULL; } - printk(KERN_INFO "xc5000: successfully identified at address 0x%02x\n", - cfg->i2c_address); - memcpy(&fe->ops.tuner_ops, &xc5000_tuner_ops, sizeof(struct dvb_tuner_ops)); diff --git a/drivers/media/dvb/frontends/xc5000.h b/drivers/media/dvb/frontends/xc5000.h index 941b31948133..85b2d438fa41 100644 --- a/drivers/media/dvb/frontends/xc5000.h +++ b/drivers/media/dvb/frontends/xc5000.h @@ -28,13 +28,15 @@ struct dvb_frontend; struct i2c_adapter; struct xc5000_config { - u8 i2c_address; - u32 if_khz; - int (*request_firmware)(struct dvb_frontend *fe, - const struct firmware **fw, char *name); - int (*tuner_reset)(struct dvb_frontend* fe); + u8 i2c_address; + u32 if_khz; + void *video_dev; + int (*tuner_callback) (void *dev, int command, int arg); }; +/* xc5000 callback command */ +#define XC5000_TUNER_RESET 0 + #if defined(CONFIG_DVB_TUNER_XC5000) || defined(CONFIG_DVB_TUNER_XC5000_MODULE) extern struct dvb_frontend* xc5000_attach(struct dvb_frontend *fe, struct i2c_adapter *i2c, diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c index 35976e6cb1b8..16cdeeafeb6c 100644 --- a/drivers/media/video/tuner-core.c +++ b/drivers/media/video/tuner-core.c @@ -27,6 +27,7 @@ #include "tuner-xc2028.h" #include "tuner-simple.h" #include "tda9887.h" +#include "xc5000.h" #define UNSET (-1U) @@ -329,6 +330,8 @@ static void attach_tda829x(struct tuner *t) tda829x_attach(&t->fe, t->i2c->adapter, t->i2c->addr, &cfg); } +static struct xc5000_config xc5000_cfg; + static void set_type(struct i2c_client *c, unsigned int type, unsigned int new_mode_mask, unsigned int new_config, int (*tuner_callback) (void *dev, int command,int arg)) @@ -428,6 +431,23 @@ static void set_type(struct i2c_client *c, unsigned int type, case TUNER_TDA9887: tda9887_attach(&t->fe, t->i2c->adapter, t->i2c->addr); break; + case TUNER_XC5000: + xc5000_cfg.i2c_address = t->i2c->addr; + xc5000_cfg.if_khz = 5380; + xc5000_cfg.video_dev = c->adapter->algo_data; + xc5000_cfg.tuner_callback = t->tuner_callback; + if (!xc5000_attach(&t->fe, t->i2c->adapter, &xc5000_cfg)) { + t->type = TUNER_ABSENT; + t->mode_mask = T_UNINITIALIZED; + return; + } + { + struct dvb_tuner_ops *xc_tuner_ops; + xc_tuner_ops = &t->fe.ops.tuner_ops; + if(xc_tuner_ops->init != NULL) + xc_tuner_ops->init(&t->fe); + } + break; default: attach_simple_tuner(t); break; diff --git a/drivers/media/video/tuner-types.c b/drivers/media/video/tuner-types.c index e2cd05a05802..883047f9c28c 100644 --- a/drivers/media/video/tuner-types.c +++ b/drivers/media/video/tuner-types.c @@ -1475,6 +1475,10 @@ struct tunertype tuners[] = { .name = "Philips TEA5761 FM Radio", /* see tea5767.c for details */ }, + [TUNER_XC5000] = { /* Xceive 5000 */ + .name = "Xceive 5000 tuner", + /* see xc5000.c for details */ + }, }; unsigned const int tuner_count = ARRAY_SIZE(tuners); diff --git a/include/media/tuner.h b/include/media/tuner.h index 97be269afbee..1bf24a6ed8f1 100644 --- a/include/media/tuner.h +++ b/include/media/tuner.h @@ -121,6 +121,7 @@ #define TUNER_SAMSUNG_TCPG_6121P30A 73 /* Hauppauge PVR-500 PAL */ #define TUNER_TDA9887 74 /* This tuner should be used only internally */ #define TUNER_TEA5761 75 /* Only FM Radio Tuner */ +#define TUNER_XC5000 76 /* Xceive Silicon Tuner */ /* tv card specific */ #define TDA9887_PRESENT (1<<0) -- cgit v1.2.3 From 60464da8b1ed8d7c19563f58cadb0ca990448a36 Mon Sep 17 00:00:00 2001 From: Steven Toth Date: Sat, 5 Jan 2008 16:53:01 -0300 Subject: V4L/DVB (7043): New card supported(partially): Pinnacle 800i From Zhang: This patch continues the support for the Pinnacle HD 800i. Patch committed as-is, cleanups to follow ... Steve Signed-off-by: Chaogui Zhang Signed-off-by: Steven Toth Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/CARDLIST.cx88 | 1 + drivers/media/video/cx88/cx88-cards.c | 63 +++++++++++++++++++++++++++++++++ drivers/media/video/cx88/cx88-dvb.c | 36 +++++++++++++++++++ drivers/media/video/cx88/cx88-i2c.c | 17 ++++----- drivers/media/video/cx88/cx88-mpeg.c | 7 +++- drivers/media/video/cx88/cx88.h | 2 ++ 6 files changed, 117 insertions(+), 9 deletions(-) (limited to 'Documentation') diff --git a/Documentation/video4linux/CARDLIST.cx88 b/Documentation/video4linux/CARDLIST.cx88 index 82ac8250e978..bc5593bd9704 100644 --- a/Documentation/video4linux/CARDLIST.cx88 +++ b/Documentation/video4linux/CARDLIST.cx88 @@ -56,3 +56,4 @@ 55 -> Shenzhen Tungsten Ages Tech TE-DTV-250 / Swann OEM [c180:c980] 56 -> Hauppauge WinTV-HVR1300 DVB-T/Hybrid MPEG Encoder [0070:9600,0070:9601,0070:9602] 57 -> ADS Tech Instant Video PCI [1421:0390] + 58 -> Pinnacle PCTV HD 800i [11bd:0051] diff --git a/drivers/media/video/cx88/cx88-cards.c b/drivers/media/video/cx88/cx88-cards.c index d484e5ce579c..3ba31572c657 100644 --- a/drivers/media/video/cx88/cx88-cards.c +++ b/drivers/media/video/cx88/cx88-cards.c @@ -1375,6 +1375,32 @@ static const struct cx88_board cx88_boards[] = { .gpio0 = 0x07fa, }}, }, + [CX88_BOARD_PINNACLE_PCTV_HD_800i] = { + .name = "Pinnacle PCTV HD 800i", + .tuner_type = TUNER_XC5000, + .radio_type = UNSET, + .tuner_addr = ADDR_UNSET, + .radio_addr = ADDR_UNSET, + .input = {{ + .type = CX88_VMUX_TELEVISION, + .vmux = 0, + .gpio0 = 0x04fb, + .gpio1 = 0x10ff, + },{ + .type = CX88_VMUX_COMPOSITE1, + .vmux = 1, + .gpio0 = 0x04fb, + .gpio1 = 0x10ef, + .audioroute = 1, + },{ + .type = CX88_VMUX_SVIDEO, + .vmux = 2, + .gpio0 = 0x04fb, + .gpio1 = 0x10ef, + .audioroute = 1, + }}, + .mpeg = CX88_MPEG_DVB, + }, }; /* ------------------------------------------------------------------ */ @@ -1684,6 +1710,10 @@ static const struct cx88_subid cx88_subids[] = { .subvendor = 0x1421, .subdevice = 0x0390, .card = CX88_BOARD_ADSTECH_PTV_390, + },{ + .subvendor = 0x11bd, + .subdevice = 0x0051, + .card = CX88_BOARD_PINNACLE_PCTV_HD_800i, }, }; @@ -1850,6 +1880,39 @@ static void dvico_fusionhdtv_hybrid_init(struct cx88_core *core) } } +/* ----------------------------------------------------------------------- */ +/* Tuner callback function. Currently only needed for the Pinnacle * + * PCTV HD 800i with an xc5000 sillicon tuner. This is used for both * + * analog tuner attach (tuner-core.c) and dvb tuner attach (cx88-dvb.c) */ + +int cx88_tuner_callback(void *i2c_algo, int command, int arg) +{ + struct i2c_algo_bit_data *algo = i2c_algo; + struct cx88_core *core = algo->data; + + switch(core->boardnr) { + case CX88_BOARD_PINNACLE_PCTV_HD_800i: + if(command == 0) { /* This is the reset command from xc5000 */ + /* Reset XC5000 tuner via GPIO pin #2 */ + cx_set(MO_GP0_IO, 0x00000400); + cx_clear(MO_GP0_IO, 0x00000004); + mdelay(200); + cx_set(MO_GP0_IO, 0x00000004); + printk(KERN_ERR "xc5000: in reset for xc5000\n"); + mdelay(200); + return 0; + } + else { + printk(KERN_ERR + "xc5000: unknown tuner callback command.\n"); + return -EINVAL; + } + break; + } + return 0; /* Should never be here */ +} +EXPORT_SYMBOL(cx88_tuner_callback); + /* ----------------------------------------------------------------------- */ static void cx88_card_list(struct cx88_core *core, struct pci_dev *pci) diff --git a/drivers/media/video/cx88/cx88-dvb.c b/drivers/media/video/cx88/cx88-dvb.c index fce19caf9d04..0246f97a9987 100644 --- a/drivers/media/video/cx88/cx88-dvb.c +++ b/drivers/media/video/cx88/cx88-dvb.c @@ -40,6 +40,8 @@ #include "cx22702.h" #include "or51132.h" #include "lgdt330x.h" +#include "s5h1409.h" +#include "xc5000.h" #include "nxt200x.h" #include "cx24123.h" #include "isl6421.h" @@ -371,6 +373,22 @@ static struct cx24123_config kworld_dvbs_100_config = { .lnb_polarity = 1, }; +static struct s5h1409_config pinnacle_pctv_hd_800i_config = { + .demod_address = 0x32 >> 1, + .output_mode = S5H1409_PARALLEL_OUTPUT, + .gpio = S5H1409_GPIO_ON, + .qam_if = 44000, + .inversion = S5H1409_INVERSION_OFF, + .status_mode = S5H1409_DEMODLOCKING, +}; + +static struct xc5000_config pinnacle_pctv_hd_800i_tuner_config = { + .i2c_address = 0x64, + .if_khz = 5380, + /* cannot set .video_dev here, do it right before attach */ + .tuner_callback = cx88_tuner_callback, +}; + static int dvb_register(struct cx8802_dev *dev) { /* init struct videobuf_dvb */ @@ -625,6 +643,24 @@ static int dvb_register(struct cx8802_dev *dev) dev->dvb.frontend->ops.set_voltage = geniatech_dvbs_set_voltage; } break; + case CX88_BOARD_PINNACLE_PCTV_HD_800i: + /* Parallel mpeg data port and punctured clock mode */ + dev->ts_gen_cntrl = 0x04; + + dev->dvb.frontend = dvb_attach(s5h1409_attach, + &pinnacle_pctv_hd_800i_config, + &dev->core->i2c_adap); + if (dev->dvb.frontend != NULL) { + /* tuner_config.video_dev must point to + * i2c_adap.algo_data + */ + pinnacle_pctv_hd_800i_tuner_config.video_dev = + dev->core->i2c_adap.algo_data; + dvb_attach(xc5000_attach, dev->dvb.frontend, + &dev->core->i2c_adap, + &pinnacle_pctv_hd_800i_tuner_config); + } + break; default: printk(KERN_ERR "%s/2: The frontend of your DVB/ATSC card isn't supported yet\n", dev->core->name); diff --git a/drivers/media/video/cx88/cx88-i2c.c b/drivers/media/video/cx88/cx88-i2c.c index 937497c86247..566b26af523e 100644 --- a/drivers/media/video/cx88/cx88-i2c.c +++ b/drivers/media/video/cx88/cx88-i2c.c @@ -109,20 +109,20 @@ static int attach_inform(struct i2c_client *client) if (core->board.radio_type != UNSET) { if ((core->board.radio_addr==ADDR_UNSET)||(core->board.radio_addr==client->addr)) { - tun_setup.mode_mask = T_RADIO; - tun_setup.type = core->board.radio_type; - tun_setup.addr = core->board.radio_addr; - + tun_setup.mode_mask = T_RADIO; + tun_setup.type = core->board.radio_type; + tun_setup.addr = core->board.radio_addr; + tun_setup.tuner_callback = cx88_tuner_callback; client->driver->command (client, TUNER_SET_TYPE_ADDR, &tun_setup); } } if (core->board.tuner_type != UNSET) { if ((core->board.tuner_addr==ADDR_UNSET)||(core->board.tuner_addr==client->addr)) { - tun_setup.mode_mask = T_ANALOG_TV; - tun_setup.type = core->board.tuner_type; - tun_setup.addr = core->board.tuner_addr; - + tun_setup.mode_mask = T_ANALOG_TV; + tun_setup.type = core->board.tuner_type; + tun_setup.addr = core->board.tuner_addr; + tun_setup.tuner_callback = cx88_tuner_callback; client->driver->command (client,TUNER_SET_TYPE_ADDR, &tun_setup); } } @@ -182,6 +182,7 @@ static char *i2c_devs[128] = { [ 0xa0 >> 1 ] = "eeprom", [ 0xc0 >> 1 ] = "tuner (analog)", [ 0xc2 >> 1 ] = "tuner (analog/dvb)", + [ 0xc8 >> 1 ] = "xc5000", }; static void do_i2c_scan(char *name, struct i2c_client *c) diff --git a/drivers/media/video/cx88/cx88-mpeg.c b/drivers/media/video/cx88/cx88-mpeg.c index 339a88a64f54..a21b50357a89 100644 --- a/drivers/media/video/cx88/cx88-mpeg.c +++ b/drivers/media/video/cx88/cx88-mpeg.c @@ -102,7 +102,7 @@ static int cx8802_start_dma(struct cx8802_dev *dev, cx_write(TS_GEN_CNTRL, 0x0040 | dev->ts_gen_cntrl); udelay(100); cx_write(MO_PINMUX_IO, 0x00); - cx_write(TS_HW_SOP_CNTRL,0x47<<16|188<<4|0x01); + cx_write(TS_HW_SOP_CNTRL, 0x47<<16|188<<4|0x01); switch (core->boardnr) { case CX88_BOARD_DVICO_FUSIONHDTV_3_GOLD_Q: case CX88_BOARD_DVICO_FUSIONHDTV_3_GOLD_T: @@ -117,6 +117,11 @@ static int cx8802_start_dma(struct cx8802_dev *dev, break; case CX88_BOARD_HAUPPAUGE_HVR1300: break; + case CX88_BOARD_PINNACLE_PCTV_HD_800i: + /* Enable MPEG parallel port */ + cx_write(MO_PINMUX_IO, 0x80); + udelay(100); + break; default: cx_write(TS_SOP_STAT, 0x00); break; diff --git a/drivers/media/video/cx88/cx88.h b/drivers/media/video/cx88/cx88.h index 0cd2946a2789..4e823f2a539a 100644 --- a/drivers/media/video/cx88/cx88.h +++ b/drivers/media/video/cx88/cx88.h @@ -210,6 +210,7 @@ extern struct sram_channel cx88_sram_channels[]; #define CX88_BOARD_TE_DTV_250_OEM_SWANN 55 #define CX88_BOARD_HAUPPAUGE_HVR1300 56 #define CX88_BOARD_ADSTECH_PTV_390 57 +#define CX88_BOARD_PINNACLE_PCTV_HD_800i 58 enum cx88_itype { CX88_VMUX_COMPOSITE1 = 1, @@ -589,6 +590,7 @@ extern void cx88_call_i2c_clients(struct cx88_core *core, /* ----------------------------------------------------------- */ /* cx88-cards.c */ +extern int cx88_tuner_callback(void *dev, int command, int arg); extern int cx88_get_resources(const struct cx88_core *core, struct pci_dev *pci); extern struct cx88_core *cx88_core_create(struct pci_dev *pci, int nr); -- cgit v1.2.3 From 386900781205d203c1141d3e2dae759f1b531193 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 21 Jan 2008 12:15:19 -0300 Subject: V4L/DVB (7077): bt878: remove handcrafted PCI subsystem ID check This patch moves the subsystem ID and subsystem vendor ID check from probing function to the PCI generic function by describing subsystem IDs in pci_device_id table. This enables to add new PCI IDs to a device driver pci_ids table at runtime by new_id file in sysfs pci driver tree. Signed-off-by: Akinobu Mita Signed-off-by: Manu Abraham Signed-off-by: Mauro Carvalho Chehab --- Documentation/dvb/bt8xx.txt | 12 +++++++ drivers/media/dvb/bt8xx/bt878.c | 76 ++++++++++++++++++----------------------- drivers/media/dvb/bt8xx/bt878.h | 6 ---- 3 files changed, 45 insertions(+), 49 deletions(-) (limited to 'Documentation') diff --git a/Documentation/dvb/bt8xx.txt b/Documentation/dvb/bt8xx.txt index ecb47adda063..b7b1d1b1da46 100644 --- a/Documentation/dvb/bt8xx.txt +++ b/Documentation/dvb/bt8xx.txt @@ -78,6 +78,18 @@ Example: For a full list of card ID's please see Documentation/video4linux/CARDLIST.bttv. In case of further problems please subscribe and send questions to the mailing list: linux-dvb@linuxtv.org. +2c) Probing the cards with broken PCI subsystem ID +-------------------------------------------------- +There are some TwinHan cards that the EEPROM has become corrupted for some +reason. The cards do not have correct PCI subsystem ID. But we can force +probing the cards with broken PCI subsystem ID + + $ echo 109e 0878 $subvendor $subdevice > \ + /sys/bus/pci/drivers/bt878/new_id + +109e: PCI_VENDOR_ID_BROOKTREE +0878: PCI_DEVICE_ID_BROOKTREE_878 + Authors: Richard Walker, Jamie Honan, Michael Hunold, diff --git a/drivers/media/dvb/bt8xx/bt878.c b/drivers/media/dvb/bt8xx/bt878.c index 85e36a1d6d78..c7bbb40223f5 100644 --- a/drivers/media/dvb/bt8xx/bt878.c +++ b/drivers/media/dvb/bt8xx/bt878.c @@ -378,23 +378,37 @@ bt878_device_control(struct bt878 *bt, unsigned int cmd, union dst_gpio_packet * EXPORT_SYMBOL(bt878_device_control); +#define BROOKTREE_878_DEVICE(vend, dev, name) \ + { \ + .vendor = PCI_VENDOR_ID_BROOKTREE, \ + .device = PCI_DEVICE_ID_BROOKTREE_878, \ + .subvendor = (vend), .subdevice = (dev), \ + .driver_data = (unsigned long) name \ + } -static struct cards card_list[] __devinitdata = { - - { 0x01010071, BTTV_BOARD_NEBULA_DIGITV, "Nebula Electronics DigiTV" }, - { 0x07611461, BTTV_BOARD_AVDVBT_761, "AverMedia AverTV DVB-T 761" }, - { 0x001c11bd, BTTV_BOARD_PINNACLESAT, "Pinnacle PCTV Sat" }, - { 0x002611bd, BTTV_BOARD_TWINHAN_DST, "Pinnacle PCTV SAT CI" }, - { 0x00011822, BTTV_BOARD_TWINHAN_DST, "Twinhan VisionPlus DVB" }, - { 0xfc00270f, BTTV_BOARD_TWINHAN_DST, "ChainTech digitop DST-1000 DVB-S" }, - { 0x07711461, BTTV_BOARD_AVDVBT_771, "AVermedia AverTV DVB-T 771" }, - { 0xdb1018ac, BTTV_BOARD_DVICO_DVBT_LITE, "DViCO FusionHDTV DVB-T Lite" }, - { 0xdb1118ac, BTTV_BOARD_DVICO_DVBT_LITE, "Ultraview DVB-T Lite" }, - { 0xd50018ac, BTTV_BOARD_DVICO_FUSIONHDTV_5_LITE, "DViCO FusionHDTV 5 Lite" }, - { 0x20007063, BTTV_BOARD_PC_HDTV, "pcHDTV HD-2000 TV" }, - { 0x00261822, BTTV_BOARD_TWINHAN_DST, "DNTV Live! Mini" } +static struct pci_device_id bt878_pci_tbl[] __devinitdata = { + BROOKTREE_878_DEVICE(0x0071, 0x0101, "Nebula Electronics DigiTV"), + BROOKTREE_878_DEVICE(0x1461, 0x0761, "AverMedia AverTV DVB-T 761"), + BROOKTREE_878_DEVICE(0x11bd, 0x001c, "Pinnacle PCTV Sat"), + BROOKTREE_878_DEVICE(0x11bd, 0x0026, "Pinnacle PCTV SAT CI"), + BROOKTREE_878_DEVICE(0x1822, 0x0001, "Twinhan VisionPlus DVB"), + BROOKTREE_878_DEVICE(0x270f, 0xfc00, + "ChainTech digitop DST-1000 DVB-S"), + BROOKTREE_878_DEVICE(0x1461, 0x0771, "AVermedia AverTV DVB-T 771"), + BROOKTREE_878_DEVICE(0x18ac, 0xdb10, "DViCO FusionHDTV DVB-T Lite"), + BROOKTREE_878_DEVICE(0x18ac, 0xdb11, "Ultraview DVB-T Lite"), + BROOKTREE_878_DEVICE(0x18ac, 0xd500, "DViCO FusionHDTV 5 Lite"), + BROOKTREE_878_DEVICE(0x7063, 0x2000, "pcHDTV HD-2000 TV"), + BROOKTREE_878_DEVICE(0x1822, 0x0026, "DNTV Live! Mini"), + { } }; +MODULE_DEVICE_TABLE(pci, bt878_pci_tbl); + +static const char * __devinit card_name(const struct pci_device_id *id) +{ + return id->driver_data ? (const char *)id->driver_data : "Unknown"; +} /***********************/ /* PCI device handling */ @@ -403,15 +417,13 @@ static struct cards card_list[] __devinitdata = { static int __devinit bt878_probe(struct pci_dev *dev, const struct pci_device_id *pci_id) { - int result = 0, has_dvb = 0, i; + int result = 0; unsigned char lat; struct bt878 *bt; #if defined(__powerpc__) unsigned int cmd; #endif unsigned int cardid; - unsigned short id; - struct cards *dvb_cards; printk(KERN_INFO "bt878: Bt878 AUDIO function found (%d).\n", bt878_num); @@ -423,25 +435,11 @@ static int __devinit bt878_probe(struct pci_dev *dev, if (pci_enable_device(dev)) return -EIO; - pci_read_config_word(dev, PCI_SUBSYSTEM_ID, &id); - cardid = id << 16; - pci_read_config_word(dev, PCI_SUBSYSTEM_VENDOR_ID, &id); - cardid |= id; - - for (i = 0, dvb_cards = card_list; i < ARRAY_SIZE(card_list); i++, dvb_cards++) { - if (cardid == dvb_cards->pci_id) { - printk("%s: card id=[0x%x],[ %s ] has DVB functions.\n", - __func__, cardid, dvb_cards->name); - has_dvb = 1; - } - } + cardid = dev->subsystem_device << 16; + cardid |= dev->subsystem_vendor; - if (!has_dvb) { - printk("%s: card id=[0x%x], Unknown card.\nExiting..\n", __func__, cardid); - result = -EINVAL; - - goto fail0; - } + printk(KERN_INFO "%s: card id=[0x%x],[ %s ] has DVB functions.\n", + __func__, cardid, card_name(pci_id)); bt = &bt878[bt878_num]; bt->dev = dev; @@ -572,14 +570,6 @@ static void __devexit bt878_remove(struct pci_dev *pci_dev) return; } -static struct pci_device_id bt878_pci_tbl[] __devinitdata = { - {PCI_VENDOR_ID_BROOKTREE, PCI_DEVICE_ID_BROOKTREE_878, - PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, - {0,} -}; - -MODULE_DEVICE_TABLE(pci, bt878_pci_tbl); - static struct pci_driver bt878_pci_driver = { .name = "bt878", .id_table = bt878_pci_tbl, diff --git a/drivers/media/dvb/bt8xx/bt878.h b/drivers/media/dvb/bt8xx/bt878.h index d593bc145628..375fd2892a11 100644 --- a/drivers/media/dvb/bt8xx/bt878.h +++ b/drivers/media/dvb/bt8xx/bt878.h @@ -101,12 +101,6 @@ #define BTTV_BOARD_DVICO_DVBT_LITE 0x80 #define BTTV_BOARD_DVICO_FUSIONHDTV_5_LITE 0x87 -struct cards { - __u32 pci_id; - __u16 card_id; - char *name; -}; - extern int bt878_num; struct bt878 { -- cgit v1.2.3 From e57895d38991036f9ccf193b70fc5ebd5f6e6dc9 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 1 Jan 2008 15:47:10 +0200 Subject: IB/mthca: Remove MSI support as scheduled Remove MSI support from the mthca driver, as scheduled. There is no reason to use MSI instead of MSI-X, since MSI-X performs better. No one has spoken up since MSI support was deprecated in commit f6be6fbe ("IB/mthca: Schedule MSI support for removal"), so apparently the MSI support is unused. Signed-off-by: Adrian Bunk Signed-off-by: Roland Dreier --- Documentation/feature-removal-schedule.txt | 10 -------- drivers/infiniband/hw/mthca/mthca_dev.h | 13 +++++----- drivers/infiniband/hw/mthca/mthca_eq.c | 6 ++--- drivers/infiniband/hw/mthca/mthca_main.c | 38 ++++-------------------------- 4 files changed, 13 insertions(+), 54 deletions(-) (limited to 'Documentation') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 20c4c8bac9d7..9b8291f4c211 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -295,16 +295,6 @@ Who: linuxppc-dev@ozlabs.org --------------------------- -What: mthca driver's MSI support -When: January 2008 -Files: drivers/infiniband/hw/mthca/*.[ch] -Why: All mthca hardware also supports MSI-X, which provides - strictly more functionality than MSI. So there is no point in - having both MSI-X and MSI support in the driver. -Who: Roland Dreier - ---------------------------- - What: sk98lin network driver When: Feburary 2008 Why: In kernel tree version of driver is unmaintained. Sk98lin driver diff --git a/drivers/infiniband/hw/mthca/mthca_dev.h b/drivers/infiniband/hw/mthca/mthca_dev.h index 15aa32eb78b6..7bbdd1f4e6c7 100644 --- a/drivers/infiniband/hw/mthca/mthca_dev.h +++ b/drivers/infiniband/hw/mthca/mthca_dev.h @@ -60,13 +60,12 @@ enum { MTHCA_FLAG_DDR_HIDDEN = 1 << 1, MTHCA_FLAG_SRQ = 1 << 2, - MTHCA_FLAG_MSI = 1 << 3, - MTHCA_FLAG_MSI_X = 1 << 4, - MTHCA_FLAG_NO_LAM = 1 << 5, - MTHCA_FLAG_FMR = 1 << 6, - MTHCA_FLAG_MEMFREE = 1 << 7, - MTHCA_FLAG_PCIE = 1 << 8, - MTHCA_FLAG_SINAI_OPT = 1 << 9 + MTHCA_FLAG_MSI_X = 1 << 3, + MTHCA_FLAG_NO_LAM = 1 << 4, + MTHCA_FLAG_FMR = 1 << 5, + MTHCA_FLAG_MEMFREE = 1 << 6, + MTHCA_FLAG_PCIE = 1 << 7, + MTHCA_FLAG_SINAI_OPT = 1 << 8 }; enum { diff --git a/drivers/infiniband/hw/mthca/mthca_eq.c b/drivers/infiniband/hw/mthca/mthca_eq.c index b29de51b7f35..b60eb5df96e8 100644 --- a/drivers/infiniband/hw/mthca/mthca_eq.c +++ b/drivers/infiniband/hw/mthca/mthca_eq.c @@ -827,8 +827,7 @@ int mthca_init_eq_table(struct mthca_dev *dev) if (err) goto err_out_free; - if (dev->mthca_flags & MTHCA_FLAG_MSI || - dev->mthca_flags & MTHCA_FLAG_MSI_X) { + if (dev->mthca_flags & MTHCA_FLAG_MSI_X) { dev->eq_table.clr_mask = 0; } else { dev->eq_table.clr_mask = @@ -839,8 +838,7 @@ int mthca_init_eq_table(struct mthca_dev *dev) dev->eq_table.arm_mask = 0; - intr = (dev->mthca_flags & MTHCA_FLAG_MSI) ? - 128 : dev->eq_table.inta_pin; + intr = dev->eq_table.inta_pin; err = mthca_create_eq(dev, dev->limits.num_cqs + MTHCA_NUM_SPARE_EQE, (dev->mthca_flags & MTHCA_FLAG_MSI_X) ? 128 : intr, diff --git a/drivers/infiniband/hw/mthca/mthca_main.c b/drivers/infiniband/hw/mthca/mthca_main.c index 60de6f93869e..2fc36ca134a6 100644 --- a/drivers/infiniband/hw/mthca/mthca_main.c +++ b/drivers/infiniband/hw/mthca/mthca_main.c @@ -65,14 +65,9 @@ static int msi_x = 1; module_param(msi_x, int, 0444); MODULE_PARM_DESC(msi_x, "attempt to use MSI-X if nonzero"); -static int msi = 0; -module_param(msi, int, 0444); -MODULE_PARM_DESC(msi, "attempt to use MSI if nonzero (deprecated, use MSI-X instead)"); - #else /* CONFIG_PCI_MSI */ #define msi_x (0) -#define msi (0) #endif /* CONFIG_PCI_MSI */ @@ -816,13 +811,11 @@ static int mthca_setup_hca(struct mthca_dev *dev) err = mthca_NOP(dev, &status); if (err || status) { - if (dev->mthca_flags & (MTHCA_FLAG_MSI | MTHCA_FLAG_MSI_X)) { + if (dev->mthca_flags & MTHCA_FLAG_MSI_X) { mthca_warn(dev, "NOP command failed to generate interrupt " "(IRQ %d).\n", - dev->mthca_flags & MTHCA_FLAG_MSI_X ? - dev->eq_table.eq[MTHCA_EQ_CMD].msi_x_vector : - dev->pdev->irq); - mthca_warn(dev, "Trying again with MSI/MSI-X disabled.\n"); + dev->eq_table.eq[MTHCA_EQ_CMD].msi_x_vector); + mthca_warn(dev, "Trying again with MSI-X disabled.\n"); } else { mthca_err(dev, "NOP command failed to generate interrupt " "(IRQ %d), aborting.\n", @@ -1128,29 +1121,12 @@ static int __mthca_init_one(struct pci_dev *pdev, int hca_type) if (msi_x && !mthca_enable_msi_x(mdev)) mdev->mthca_flags |= MTHCA_FLAG_MSI_X; - else if (msi) { - static int warned; - - if (!warned) { - printk(KERN_WARNING PFX "WARNING: MSI support will be " - "removed from the ib_mthca driver in January 2008.\n"); - printk(KERN_WARNING " If you are using MSI and cannot " - "switch to MSI-X, please tell " - ".\n"); - ++warned; - } - - if (!pci_enable_msi(pdev)) - mdev->mthca_flags |= MTHCA_FLAG_MSI; - } err = mthca_setup_hca(mdev); - if (err == -EBUSY && (mdev->mthca_flags & (MTHCA_FLAG_MSI | MTHCA_FLAG_MSI_X))) { + if (err == -EBUSY && (mdev->mthca_flags & MTHCA_FLAG_MSI_X)) { if (mdev->mthca_flags & MTHCA_FLAG_MSI_X) pci_disable_msix(pdev); - if (mdev->mthca_flags & MTHCA_FLAG_MSI) - pci_disable_msi(pdev); - mdev->mthca_flags &= ~(MTHCA_FLAG_MSI_X | MTHCA_FLAG_MSI); + mdev->mthca_flags &= ~MTHCA_FLAG_MSI_X; err = mthca_setup_hca(mdev); } @@ -1192,8 +1168,6 @@ err_cleanup: err_close: if (mdev->mthca_flags & MTHCA_FLAG_MSI_X) pci_disable_msix(pdev); - if (mdev->mthca_flags & MTHCA_FLAG_MSI) - pci_disable_msi(pdev); mthca_close_hca(mdev); @@ -1246,8 +1220,6 @@ static void __mthca_remove_one(struct pci_dev *pdev) if (mdev->mthca_flags & MTHCA_FLAG_MSI_X) pci_disable_msix(pdev); - if (mdev->mthca_flags & MTHCA_FLAG_MSI) - pci_disable_msi(pdev); ib_dealloc_device(&mdev->ib_dev); mthca_release_regions(pdev, mdev->mthca_flags & -- cgit v1.2.3 From 1252c434e39dc60ca9e8ed682f3e04930e2c08de Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 30 Oct 2007 12:09:03 -0700 Subject: ocfs2: Documentation update Remove 'readpages' from the list in ocfs2.txt. Instead of having two identical lists, I just removed the list in the OCFS2 section of fs/Kconfig and added a pointer to Documentation/filesystems/ocfs2.txt. Signed-off-by: Mark Fasheh --- Documentation/filesystems/ocfs2.txt | 1 - fs/Kconfig | 10 ++-------- 2 files changed, 2 insertions(+), 9 deletions(-) (limited to 'Documentation') diff --git a/Documentation/filesystems/ocfs2.txt b/Documentation/filesystems/ocfs2.txt index ed55238023a9..e78abdcc59ee 100644 --- a/Documentation/filesystems/ocfs2.txt +++ b/Documentation/filesystems/ocfs2.txt @@ -35,7 +35,6 @@ Features which OCFS2 does not support yet: - Directory change notification (F_NOTIFY) - Distributed Caching (F_SETLEASE/F_GETLEASE/break_lease) - POSIX ACLs - - readpages / writepages (not user visible) Mount options ============= diff --git a/fs/Kconfig b/fs/Kconfig index 781b47d2f9f2..16598a417423 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -440,14 +440,8 @@ config OCFS2_FS Tools web page: http://oss.oracle.com/projects/ocfs2-tools OCFS2 mailing lists: http://oss.oracle.com/projects/ocfs2/mailman/ - Note: Features which OCFS2 does not support yet: - - extended attributes - - quotas - - cluster aware flock - - Directory change notification (F_NOTIFY) - - Distributed Caching (F_SETLEASE/F_GETLEASE/break_lease) - - POSIX ACLs - - readpages / writepages (not user visible) + For more information on OCFS2, see the file + . config OCFS2_DEBUG_MASKLOG bool "OCFS2 logging support" -- cgit v1.2.3 From 7f68fc28219be3b44ef4132f95c6506ff3e806b5 Mon Sep 17 00:00:00 2001 From: Tao Ma Date: Tue, 18 Dec 2007 15:46:37 +0800 Subject: ocfs2: Reserve ioctl range We need to reserve at least two ioctls for online-resize. Reserve a small range of ioctls for Ocfs2 use in Documentation/ioctl-number.txt. This should give us enough room for future growth. Signed-off-by: Tao Ma Signed-off-by: Mark Fasheh --- Documentation/ioctl-number.txt | 1 + 1 file changed, 1 insertion(+) (limited to 'Documentation') diff --git a/Documentation/ioctl-number.txt b/Documentation/ioctl-number.txt index 5c7fbf9d96b4..c18363bd8d11 100644 --- a/Documentation/ioctl-number.txt +++ b/Documentation/ioctl-number.txt @@ -138,6 +138,7 @@ Code Seq# Include File Comments 'm' 00-1F net/irda/irmod.h conflict! 'n' 00-7F linux/ncp_fs.h 'n' E0-FF video/matrox.h matroxfb +'o' 00-1F fs/ocfs2/ocfs2_fs.h OCFS2 'p' 00-0F linux/phantom.h conflict! (OpenHaptics needs this) 'p' 00-3F linux/mc146818rtc.h conflict! 'p' 40-7F linux/nvram.h -- cgit v1.2.3 From d147b3d630edef1d34de6ea819787a1ac1b8603b Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Wed, 7 Nov 2007 14:40:36 -0800 Subject: ocfs2: Support commit= mount option Mostly taken from ext3. This allows the user to set the jbd commit interval, in seconds. The default of 5 seconds stays the same, but now users can easily increase the commit interval. Typically, this would be increased in order to benefit performance at the expense of data-safety. Signed-off-by: Mark Fasheh --- Documentation/filesystems/ocfs2.txt | 11 +++++++++++ fs/ocfs2/journal.c | 8 ++++++-- fs/ocfs2/ocfs2.h | 1 + fs/ocfs2/super.c | 23 +++++++++++++++++++++++ 4 files changed, 41 insertions(+), 2 deletions(-) (limited to 'Documentation') diff --git a/Documentation/filesystems/ocfs2.txt b/Documentation/filesystems/ocfs2.txt index e78abdcc59ee..b63bd2d7fcd3 100644 --- a/Documentation/filesystems/ocfs2.txt +++ b/Documentation/filesystems/ocfs2.txt @@ -61,3 +61,14 @@ data=writeback Data ordering is not preserved, data may be written preferred_slot=0(*) During mount, try to use this filesystem slot first. If it is in use by another node, the first empty one found will be chosen. Invalid values will be ignored. +commit=nrsec (*) Ocfs2 can be told to sync all its data and metadata + every 'nrsec' seconds. The default value is 5 seconds. + This means that if you lose your power, you will lose + as much as the latest 5 seconds of work (your + filesystem will not be damaged though, thanks to the + journaling). This default value (or any low value) + will hurt performance, but it's good for data-safety. + Setting it to 0 will have the same effect as leaving + it at the default (5 seconds). + Setting it to very large values will improve + performance. diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index 4f440a88bf53..8b9ce2a729ab 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -313,14 +313,18 @@ int ocfs2_journal_dirty_data(handle_t *handle, return err; } -#define OCFS2_DEFAULT_COMMIT_INTERVAL (HZ * 5) +#define OCFS2_DEFAULT_COMMIT_INTERVAL (HZ * JBD_DEFAULT_MAX_COMMIT_AGE) void ocfs2_set_journal_params(struct ocfs2_super *osb) { journal_t *journal = osb->journal->j_journal; + unsigned long commit_interval = OCFS2_DEFAULT_COMMIT_INTERVAL; + + if (osb->osb_commit_interval) + commit_interval = osb->osb_commit_interval; spin_lock(&journal->j_state_lock); - journal->j_commit_interval = OCFS2_DEFAULT_COMMIT_INTERVAL; + journal->j_commit_interval = commit_interval; if (osb->s_mount_opt & OCFS2_MOUNT_BARRIER) journal->j_flags |= JFS_BARRIER; else diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index f8f866144c6a..82802f5672a1 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -229,6 +229,7 @@ struct ocfs2_super wait_queue_head_t checkpoint_event; atomic_t needs_checkpoint; struct ocfs2_journal *journal; + unsigned long osb_commit_interval; enum ocfs2_local_alloc_state local_alloc_state; struct buffer_head *local_alloc_bh; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 479ac50c86d9..8044ed97d362 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -83,6 +83,7 @@ MODULE_LICENSE("GPL"); struct mount_options { + unsigned long commit_interval; unsigned long mount_opt; unsigned int atime_quantum; signed short slot; @@ -149,6 +150,7 @@ enum { Opt_data_writeback, Opt_atime_quantum, Opt_slot, + Opt_commit, Opt_err, }; @@ -164,6 +166,7 @@ static match_table_t tokens = { {Opt_data_writeback, "data=writeback"}, {Opt_atime_quantum, "atime_quantum=%u"}, {Opt_slot, "preferred_slot=%u"}, + {Opt_commit, "commit=%u"}, {Opt_err, NULL} }; @@ -442,6 +445,8 @@ unlock_osb: osb->s_mount_opt = parsed_options.mount_opt; osb->s_atime_quantum = parsed_options.atime_quantum; osb->preferred_slot = parsed_options.slot; + if (parsed_options.commit_interval) + osb->osb_commit_interval = parsed_options.commit_interval; if (!ocfs2_is_hard_readonly(osb)) ocfs2_set_journal_params(osb); @@ -596,6 +601,7 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent) osb->s_mount_opt = parsed_options.mount_opt; osb->s_atime_quantum = parsed_options.atime_quantum; osb->preferred_slot = parsed_options.slot; + osb->osb_commit_interval = parsed_options.commit_interval; sb->s_magic = OCFS2_SUPER_MAGIC; @@ -746,6 +752,7 @@ static int ocfs2_parse_options(struct super_block *sb, mlog_entry("remount: %d, options: \"%s\"\n", is_remount, options ? options : "(none)"); + mopt->commit_interval = 0; mopt->mount_opt = 0; mopt->atime_quantum = OCFS2_DEFAULT_ATIME_QUANTUM; mopt->slot = OCFS2_INVALID_SLOT; @@ -815,6 +822,18 @@ static int ocfs2_parse_options(struct super_block *sb, if (option) mopt->slot = (s16)option; break; + case Opt_commit: + option = 0; + if (match_int(&args[0], &option)) { + status = 0; + goto bail; + } + if (option < 0) + return 0; + if (option == 0) + option = JBD_DEFAULT_MAX_COMMIT_AGE; + mopt->commit_interval = HZ * option; + break; default: mlog(ML_ERROR, "Unrecognized mount option \"%s\" " @@ -863,6 +882,10 @@ static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt) if (osb->s_atime_quantum != OCFS2_DEFAULT_ATIME_QUANTUM) seq_printf(s, ",atime_quantum=%u", osb->s_atime_quantum); + if (osb->osb_commit_interval) + seq_printf(s, ",commit=%u", + (unsigned) (osb->osb_commit_interval / HZ)); + return 0; } -- cgit v1.2.3 From 2fbe8d1ebe004425b4f7b8bba345623d2280be82 Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Thu, 20 Dec 2007 14:58:11 -0800 Subject: ocfs2: Local alloc window size changeable via mount option Local alloc is a performance optimization in ocfs2 in which a node takes a window of bits from the global bitmap and then uses that for all small local allocations. This window size is fixed to 8MB currently. This patch allows users to specify the window size in MB including disabling it by passing in 0. If the number specified is too large, the fs will use the default value of 8MB. mount -o localalloc=X /dev/sdX /mntpoint Signed-off-by: Sunil Mushran Signed-off-by: Mark Fasheh --- Documentation/filesystems/ocfs2.txt | 3 +++ fs/ocfs2/localalloc.c | 42 ++++++++++++++++++++++++++----------- fs/ocfs2/ocfs2.h | 1 + fs/ocfs2/ocfs2_fs.h | 8 +++++++ fs/ocfs2/suballoc.c | 5 +++-- fs/ocfs2/super.c | 17 +++++++++++++++ 6 files changed, 62 insertions(+), 14 deletions(-) (limited to 'Documentation') diff --git a/Documentation/filesystems/ocfs2.txt b/Documentation/filesystems/ocfs2.txt index b63bd2d7fcd3..071fad137eb5 100644 --- a/Documentation/filesystems/ocfs2.txt +++ b/Documentation/filesystems/ocfs2.txt @@ -72,3 +72,6 @@ commit=nrsec (*) Ocfs2 can be told to sync all its data and metadata it at the default (5 seconds). Setting it to very large values will improve performance. +localalloc=8(*) Allows custom localalloc size in MB. If the value is too + large, the fs will silently revert it to the default. + Localalloc is not enabled for local mounts. diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index 0de0792fce7f..add1ffdc5c6c 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -75,18 +75,12 @@ static int ocfs2_local_alloc_new_window(struct ocfs2_super *osb, static int ocfs2_local_alloc_slide_window(struct ocfs2_super *osb, struct inode *local_alloc_inode); -/* - * Determine how large our local alloc window should be, in bits. - * - * These values (and the behavior in ocfs2_alloc_should_use_local) have - * been chosen so that most allocations, including new block groups go - * through local alloc. - */ static inline int ocfs2_local_alloc_window_bits(struct ocfs2_super *osb) { - BUG_ON(osb->s_clustersize_bits < 12); + BUG_ON(osb->s_clustersize_bits > 20); - return 2048 >> (osb->s_clustersize_bits - 12); + /* Size local alloc windows by the megabyte */ + return osb->local_alloc_size << (20 - osb->s_clustersize_bits); } /* @@ -96,18 +90,23 @@ static inline int ocfs2_local_alloc_window_bits(struct ocfs2_super *osb) int ocfs2_alloc_should_use_local(struct ocfs2_super *osb, u64 bits) { int la_bits = ocfs2_local_alloc_window_bits(osb); + int ret = 0; if (osb->local_alloc_state != OCFS2_LA_ENABLED) - return 0; + goto bail; /* la_bits should be at least twice the size (in clusters) of * a new block group. We want to be sure block group * allocations go through the local alloc, so allow an * allocation to take up to half the bitmap. */ if (bits > (la_bits / 2)) - return 0; + goto bail; - return 1; + ret = 1; +bail: + mlog(0, "state=%d, bits=%llu, la_bits=%d, ret=%d\n", + osb->local_alloc_state, (unsigned long long)bits, la_bits, ret); + return ret; } int ocfs2_load_local_alloc(struct ocfs2_super *osb) @@ -121,6 +120,19 @@ int ocfs2_load_local_alloc(struct ocfs2_super *osb) mlog_entry_void(); + if (ocfs2_mount_local(osb)) + goto bail; + + if (osb->local_alloc_size == 0) + goto bail; + + if (ocfs2_local_alloc_window_bits(osb) >= osb->bitmap_cpg) { + mlog(ML_NOTICE, "Requested local alloc window %d is larger " + "than max possible %u. Using defaults.\n", + ocfs2_local_alloc_window_bits(osb), (osb->bitmap_cpg - 1)); + osb->local_alloc_size = OCFS2_DEFAULT_LOCAL_ALLOC_SIZE; + } + /* read the alloc off disk */ inode = ocfs2_get_system_file_inode(osb, LOCAL_ALLOC_SYSTEM_INODE, osb->slot_num); @@ -181,6 +193,9 @@ bail: if (inode) iput(inode); + mlog(0, "Local alloc window bits = %d\n", + ocfs2_local_alloc_window_bits(osb)); + mlog_exit(status); return status; } @@ -521,6 +536,9 @@ bail: iput(local_alloc_inode); } + mlog(0, "bits=%d, slot=%d, ret=%d\n", bits_wanted, osb->slot_num, + status); + mlog_exit(status); return status; } diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index 82802f5672a1..d12bd7036da7 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -231,6 +231,7 @@ struct ocfs2_super struct ocfs2_journal *journal; unsigned long osb_commit_interval; + int local_alloc_size; enum ocfs2_local_alloc_state local_alloc_state; struct buffer_head *local_alloc_bh; u64 la_last_gd; diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h index 425551737f1f..3633edd3982f 100644 --- a/fs/ocfs2/ocfs2_fs.h +++ b/fs/ocfs2/ocfs2_fs.h @@ -270,6 +270,14 @@ struct ocfs2_new_group_input { /* Journal limits (in bytes) */ #define OCFS2_MIN_JOURNAL_SIZE (4 * 1024 * 1024) +/* + * Default local alloc size (in megabytes) + * + * The value chosen should be such that most allocations, including new + * block groups, use local alloc. + */ +#define OCFS2_DEFAULT_LOCAL_ALLOC_SIZE 8 + struct ocfs2_system_inode_info { char *si_name; int si_iflags; diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index 4391744e80f8..7e397e2c25dd 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -1516,8 +1516,9 @@ int __ocfs2_claim_clusters(struct ocfs2_super *osb, if (min_clusters > (osb->bitmap_cpg - 1)) { /* The only paths asking for contiguousness * should know about this already. */ - mlog(ML_ERROR, "minimum allocation requested exceeds " - "group bitmap size!"); + mlog(ML_ERROR, "minimum allocation requested %u exceeds " + "group bitmap size %u!\n", min_clusters, + osb->bitmap_cpg); status = -ENOSPC; goto bail; } diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 8044ed97d362..1104f14c3183 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -87,6 +87,7 @@ struct mount_options unsigned long mount_opt; unsigned int atime_quantum; signed short slot; + unsigned int localalloc_opt; }; static int ocfs2_parse_options(struct super_block *sb, char *options, @@ -151,6 +152,7 @@ enum { Opt_atime_quantum, Opt_slot, Opt_commit, + Opt_localalloc, Opt_err, }; @@ -167,6 +169,7 @@ static match_table_t tokens = { {Opt_atime_quantum, "atime_quantum=%u"}, {Opt_slot, "preferred_slot=%u"}, {Opt_commit, "commit=%u"}, + {Opt_localalloc, "localalloc=%d"}, {Opt_err, NULL} }; @@ -602,6 +605,7 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent) osb->s_atime_quantum = parsed_options.atime_quantum; osb->preferred_slot = parsed_options.slot; osb->osb_commit_interval = parsed_options.commit_interval; + osb->local_alloc_size = parsed_options.localalloc_opt; sb->s_magic = OCFS2_SUPER_MAGIC; @@ -756,6 +760,7 @@ static int ocfs2_parse_options(struct super_block *sb, mopt->mount_opt = 0; mopt->atime_quantum = OCFS2_DEFAULT_ATIME_QUANTUM; mopt->slot = OCFS2_INVALID_SLOT; + mopt->localalloc_opt = OCFS2_DEFAULT_LOCAL_ALLOC_SIZE; if (!options) { status = 1; @@ -834,6 +839,15 @@ static int ocfs2_parse_options(struct super_block *sb, option = JBD_DEFAULT_MAX_COMMIT_AGE; mopt->commit_interval = HZ * option; break; + case Opt_localalloc: + option = 0; + if (match_int(&args[0], &option)) { + status = 0; + goto bail; + } + if (option >= 0 && (option <= ocfs2_local_alloc_size(sb) * 8)) + mopt->localalloc_opt = option; + break; default: mlog(ML_ERROR, "Unrecognized mount option \"%s\" " @@ -886,6 +900,9 @@ static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt) seq_printf(s, ",commit=%u", (unsigned) (osb->osb_commit_interval / HZ)); + if (osb->local_alloc_size != OCFS2_DEFAULT_LOCAL_ALLOC_SIZE) + seq_printf(s, ",localalloc=%d", osb->local_alloc_size); + return 0; } -- cgit v1.2.3 From 53fc622b9e829c8e632e45ef8c14f054388759c1 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Thu, 20 Dec 2007 16:49:04 -0800 Subject: [PATCH 2/2] ocfs2: cluster aware flock() Hook up ocfs2_flock(), using the new flock lock type in dlmglue.c. A new mount option, "localflocks" is added so that users can revert to old functionality as need be. Signed-off-by: Mark Fasheh --- Documentation/filesystems/ocfs2.txt | 1 + fs/ocfs2/Makefile | 1 + fs/ocfs2/file.c | 60 ++++++++++++++++- fs/ocfs2/locks.c | 125 ++++++++++++++++++++++++++++++++++++ fs/ocfs2/locks.h | 31 +++++++++ fs/ocfs2/ocfs2.h | 1 + fs/ocfs2/super.c | 19 ++++++ 7 files changed, 237 insertions(+), 1 deletion(-) create mode 100644 fs/ocfs2/locks.c create mode 100644 fs/ocfs2/locks.h (limited to 'Documentation') diff --git a/Documentation/filesystems/ocfs2.txt b/Documentation/filesystems/ocfs2.txt index 071fad137eb5..c318a8bbb1ef 100644 --- a/Documentation/filesystems/ocfs2.txt +++ b/Documentation/filesystems/ocfs2.txt @@ -75,3 +75,4 @@ commit=nrsec (*) Ocfs2 can be told to sync all its data and metadata localalloc=8(*) Allows custom localalloc size in MB. If the value is too large, the fs will silently revert it to the default. Localalloc is not enabled for local mounts. +localflocks This disables cluster aware flock. diff --git a/fs/ocfs2/Makefile b/fs/ocfs2/Makefile index 3591890b32c6..4d4ce48bb42c 100644 --- a/fs/ocfs2/Makefile +++ b/fs/ocfs2/Makefile @@ -19,6 +19,7 @@ ocfs2-objs := \ ioctl.o \ journal.o \ localalloc.o \ + locks.o \ mmap.o \ namei.o \ resize.o \ diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 432e5f3c4784..caefd571782e 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -51,6 +51,7 @@ #include "inode.h" #include "ioctl.h" #include "journal.h" +#include "locks.h" #include "mmap.h" #include "suballoc.h" #include "super.h" @@ -63,6 +64,35 @@ static int ocfs2_sync_inode(struct inode *inode) return sync_mapping_buffers(inode->i_mapping); } +static int ocfs2_init_file_private(struct inode *inode, struct file *file) +{ + struct ocfs2_file_private *fp; + + fp = kzalloc(sizeof(struct ocfs2_file_private), GFP_KERNEL); + if (!fp) + return -ENOMEM; + + fp->fp_file = file; + mutex_init(&fp->fp_mutex); + ocfs2_file_lock_res_init(&fp->fp_flock, fp); + file->private_data = fp; + + return 0; +} + +static void ocfs2_free_file_private(struct inode *inode, struct file *file) +{ + struct ocfs2_file_private *fp = file->private_data; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + + if (fp) { + ocfs2_simple_drop_lockres(osb, &fp->fp_flock); + ocfs2_lock_res_free(&fp->fp_flock); + kfree(fp); + file->private_data = NULL; + } +} + static int ocfs2_file_open(struct inode *inode, struct file *file) { int status; @@ -89,7 +119,18 @@ static int ocfs2_file_open(struct inode *inode, struct file *file) oi->ip_open_count++; spin_unlock(&oi->ip_lock); - status = 0; + + status = ocfs2_init_file_private(inode, file); + if (status) { + /* + * We want to set open count back if we're failing the + * open. + */ + spin_lock(&oi->ip_lock); + oi->ip_open_count--; + spin_unlock(&oi->ip_lock); + } + leave: mlog_exit(status); return status; @@ -108,11 +149,24 @@ static int ocfs2_file_release(struct inode *inode, struct file *file) oi->ip_flags &= ~OCFS2_INODE_OPEN_DIRECT; spin_unlock(&oi->ip_lock); + ocfs2_free_file_private(inode, file); + mlog_exit(0); return 0; } +static int ocfs2_dir_open(struct inode *inode, struct file *file) +{ + return ocfs2_init_file_private(inode, file); +} + +static int ocfs2_dir_release(struct inode *inode, struct file *file) +{ + ocfs2_free_file_private(inode, file); + return 0; +} + static int ocfs2_sync_file(struct file *file, struct dentry *dentry, int datasync) @@ -2191,6 +2245,7 @@ const struct file_operations ocfs2_fops = { #ifdef CONFIG_COMPAT .compat_ioctl = ocfs2_compat_ioctl, #endif + .flock = ocfs2_flock, .splice_read = ocfs2_file_splice_read, .splice_write = ocfs2_file_splice_write, }; @@ -2199,8 +2254,11 @@ const struct file_operations ocfs2_dops = { .read = generic_read_dir, .readdir = ocfs2_readdir, .fsync = ocfs2_sync_file, + .release = ocfs2_dir_release, + .open = ocfs2_dir_open, .ioctl = ocfs2_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ocfs2_compat_ioctl, #endif + .flock = ocfs2_flock, }; diff --git a/fs/ocfs2/locks.c b/fs/ocfs2/locks.c new file mode 100644 index 000000000000..203f87143877 --- /dev/null +++ b/fs/ocfs2/locks.c @@ -0,0 +1,125 @@ +/* -*- mode: c; c-basic-offset: 8; -*- + * vim: noexpandtab sw=8 ts=8 sts=0: + * + * locks.c + * + * Userspace file locking support + * + * Copyright (C) 2007 Oracle. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public + * License along with this program; if not, write to the + * Free Software Foundation, Inc., 59 Temple Place - Suite 330, + * Boston, MA 021110-1307, USA. + */ + +#include + +#define MLOG_MASK_PREFIX ML_INODE +#include + +#include "ocfs2.h" + +#include "dlmglue.h" +#include "file.h" +#include "locks.h" + +static int ocfs2_do_flock(struct file *file, struct inode *inode, + int cmd, struct file_lock *fl) +{ + int ret = 0, level = 0, trylock = 0; + struct ocfs2_file_private *fp = file->private_data; + struct ocfs2_lock_res *lockres = &fp->fp_flock; + + if (fl->fl_type == F_WRLCK) + level = 1; + if (!IS_SETLKW(cmd)) + trylock = 1; + + mutex_lock(&fp->fp_mutex); + + if (lockres->l_flags & OCFS2_LOCK_ATTACHED && + lockres->l_level > LKM_NLMODE) { + int old_level = 0; + + if (lockres->l_level == LKM_EXMODE) + old_level = 1; + + if (level == old_level) + goto out; + + /* + * Converting an existing lock is not guaranteed to be + * atomic, so we can get away with simply unlocking + * here and allowing the lock code to try at the new + * level. + */ + + flock_lock_file_wait(file, + &(struct file_lock){.fl_type = F_UNLCK}); + + ocfs2_file_unlock(file); + } + + ret = ocfs2_file_lock(file, level, trylock); + if (ret) { + if (ret == -EAGAIN && trylock) + ret = -EWOULDBLOCK; + else + mlog_errno(ret); + goto out; + } + + ret = flock_lock_file_wait(file, fl); + +out: + mutex_unlock(&fp->fp_mutex); + + return ret; +} + +static int ocfs2_do_funlock(struct file *file, int cmd, struct file_lock *fl) +{ + int ret; + struct ocfs2_file_private *fp = file->private_data; + + mutex_lock(&fp->fp_mutex); + ocfs2_file_unlock(file); + ret = flock_lock_file_wait(file, fl); + mutex_unlock(&fp->fp_mutex); + + return ret; +} + +/* + * Overall flow of ocfs2_flock() was influenced by gfs2_flock(). + */ +int ocfs2_flock(struct file *file, int cmd, struct file_lock *fl) +{ + struct inode *inode = file->f_mapping->host; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + + if (!(fl->fl_flags & FL_FLOCK)) + return -ENOLCK; + if (__mandatory_lock(inode)) + return -ENOLCK; + + if ((osb->s_mount_opt & OCFS2_MOUNT_LOCALFLOCKS) || + ocfs2_mount_local(osb)) + return flock_lock_file_wait(file, fl); + + if (fl->fl_type == F_UNLCK) + return ocfs2_do_funlock(file, cmd, fl); + else + return ocfs2_do_flock(file, inode, cmd, fl); +} diff --git a/fs/ocfs2/locks.h b/fs/ocfs2/locks.h new file mode 100644 index 000000000000..9743ef2324ec --- /dev/null +++ b/fs/ocfs2/locks.h @@ -0,0 +1,31 @@ +/* -*- mode: c; c-basic-offset: 8; -*- + * vim: noexpandtab sw=8 ts=8 sts=0: + * + * locks.h + * + * Function prototypes for Userspace file locking support + * + * Copyright (C) 2002, 2004 Oracle. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public + * License along with this program; if not, write to the + * Free Software Foundation, Inc., 59 Temple Place - Suite 330, + * Boston, MA 021110-1307, USA. + */ + +#ifndef OCFS2_LOCKS_H +#define OCFS2_LOCKS_H + +int ocfs2_flock(struct file *file, int cmd, struct file_lock *fl); + +#endif /* OCFS2_LOCKS_H */ diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index 63c131e1cc77..22e334d125d0 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -171,6 +171,7 @@ enum ocfs2_mount_options OCFS2_MOUNT_NOINTR = 1 << 2, /* Don't catch signals */ OCFS2_MOUNT_ERRORS_PANIC = 1 << 3, /* Panic on errors */ OCFS2_MOUNT_DATA_WRITEBACK = 1 << 4, /* No data ordering */ + OCFS2_MOUNT_LOCALFLOCKS = 1 << 5, /* No cluster aware user file locks */ }; #define OCFS2_OSB_SOFT_RO 0x0001 diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 1104f14c3183..4a091f586646 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -153,6 +153,7 @@ enum { Opt_slot, Opt_commit, Opt_localalloc, + Opt_localflocks, Opt_err, }; @@ -170,6 +171,7 @@ static match_table_t tokens = { {Opt_slot, "preferred_slot=%u"}, {Opt_commit, "commit=%u"}, {Opt_localalloc, "localalloc=%d"}, + {Opt_localflocks, "localflocks"}, {Opt_err, NULL} }; @@ -848,6 +850,20 @@ static int ocfs2_parse_options(struct super_block *sb, if (option >= 0 && (option <= ocfs2_local_alloc_size(sb) * 8)) mopt->localalloc_opt = option; break; + case Opt_localflocks: + /* + * Changing this during remount could race + * flock() requests, or "unbalance" existing + * ones (e.g., a lock is taken in one mode but + * dropped in the other). If users care enough + * to flip locking modes during remount, we + * could add a "local" flag to individual + * flock structures for proper tracking of + * state. + */ + if (!is_remount) + mopt->mount_opt |= OCFS2_MOUNT_LOCALFLOCKS; + break; default: mlog(ML_ERROR, "Unrecognized mount option \"%s\" " @@ -903,6 +919,9 @@ static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt) if (osb->local_alloc_size != OCFS2_DEFAULT_LOCAL_ALLOC_SIZE) seq_printf(s, ",localalloc=%d", osb->local_alloc_size); + if (opts & OCFS2_MOUNT_LOCALFLOCKS) + seq_printf(s, ",localflocks,"); + return 0; } -- cgit v1.2.3 From 661ca0da3e1fa60c9e56f995fd83772ffa9ffe90 Mon Sep 17 00:00:00 2001 From: Sebastian Ott Date: Sat, 26 Jan 2008 14:10:36 +0100 Subject: [S390] Cleanup in Documentation/kernel-parameters.txt. Signed-off-by: Sebastian Ott Signed-off-by: Martin Schwidefsky --- Documentation/kernel-parameters.txt | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'Documentation') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 65de5ba7b74c..d29b6d926a42 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -370,8 +370,6 @@ and is between 256 and 4096 characters. It is defined in the file configured. Potentially dangerous and should only be used if you are entirely sure of the consequences. - chandev= [HW,NET] Generic channel device initialisation - checkreqprot [SELINUX] Set initial checkreqprot flag value. Format: { "0" | "1" } See security/selinux/Kconfig help text. @@ -382,6 +380,12 @@ and is between 256 and 4096 characters. It is defined in the file Value can be changed at runtime via /selinux/checkreqprot. + cio_ignore= [S390] + See Documentation/s390/CommonIO for details. + + cio_msg= [S390] + See Documentation/s390/CommonIO for details. + clock= [BUGS=X86-32, HW] gettimeofday clocksource override. [Deprecated] Forces specified clocksource (if available) to be used -- cgit v1.2.3 From 14ff56bbb3422cc465a12e81f5e1b5f06168aeda Mon Sep 17 00:00:00 2001 From: Sebastian Ott Date: Sat, 26 Jan 2008 14:10:37 +0100 Subject: [S390] cio: Dump ccw device information in case of timeout. Information about a ccw device will be dumped in case of a ccw timeout. This can be enabled with the kernel parameter ccw_timeout_log. Signed-off-by: Sebastian Ott Signed-off-by: Martin Schwidefsky --- Documentation/kernel-parameters.txt | 3 +++ Documentation/s390/CommonIO | 5 ++++ drivers/s390/cio/device_fsm.c | 50 +++++++++++++++++++++++++++++++++++++ 3 files changed, 58 insertions(+) (limited to 'Documentation') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index d29b6d926a42..880f882160e2 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -370,6 +370,9 @@ and is between 256 and 4096 characters. It is defined in the file configured. Potentially dangerous and should only be used if you are entirely sure of the consequences. + ccw_timeout_log [S390] + See Documentation/s390/CommonIO for details. + checkreqprot [SELINUX] Set initial checkreqprot flag value. Format: { "0" | "1" } See security/selinux/Kconfig help text. diff --git a/Documentation/s390/CommonIO b/Documentation/s390/CommonIO index 86320aa3fb0b..8fbc0a852870 100644 --- a/Documentation/s390/CommonIO +++ b/Documentation/s390/CommonIO @@ -4,6 +4,11 @@ S/390 common I/O-Layer - command line parameters, procfs and debugfs entries Command line parameters ----------------------- +* ccw_timeout_log + + Enable logging of debug information in case of ccw device timeouts. + + * cio_msg = yes | no Determines whether information on found devices and sensed device diff --git a/drivers/s390/cio/device_fsm.c b/drivers/s390/cio/device_fsm.c index bfad421cda66..31c0f7885161 100644 --- a/drivers/s390/cio/device_fsm.c +++ b/drivers/s390/cio/device_fsm.c @@ -25,6 +25,8 @@ #include "ioasm.h" #include "chp.h" +static int timeout_log_enabled; + int device_is_online(struct subchannel *sch) { @@ -82,6 +84,52 @@ int device_trigger_verify(struct subchannel *sch) return 0; } +static int __init ccw_timeout_log_setup(char *unused) +{ + timeout_log_enabled = 1; + return 1; +} + +__setup("ccw_timeout_log", ccw_timeout_log_setup); + +static void ccw_timeout_log(struct ccw_device *cdev) +{ + struct schib schib; + struct subchannel *sch; + int cc; + + sch = to_subchannel(cdev->dev.parent); + cc = stsch(sch->schid, &schib); + + printk(KERN_WARNING "cio: ccw device timeout occurred at %llx, " + "device information:\n", get_clock()); + printk(KERN_WARNING "cio: orb:\n"); + print_hex_dump(KERN_WARNING, "cio: ", DUMP_PREFIX_NONE, 16, 1, + &sch->orb, sizeof(sch->orb), 0); + printk(KERN_WARNING "cio: ccw device bus id: %s\n", cdev->dev.bus_id); + printk(KERN_WARNING "cio: subchannel bus id: %s\n", sch->dev.bus_id); + printk(KERN_WARNING "cio: subchannel lpm: %02x, opm: %02x, " + "vpm: %02x\n", sch->lpm, sch->opm, sch->vpm); + + if ((void *)(addr_t)sch->orb.cpa == &sch->sense_ccw || + (void *)(addr_t)sch->orb.cpa == cdev->private->iccws) + printk(KERN_WARNING "cio: last channel program (intern):\n"); + else + printk(KERN_WARNING "cio: last channel program:\n"); + + print_hex_dump(KERN_WARNING, "cio: ", DUMP_PREFIX_NONE, 16, 1, + (void *)(addr_t)sch->orb.cpa, sizeof(struct ccw1), 0); + printk(KERN_WARNING "cio: ccw device state: %d\n", + cdev->private->state); + printk(KERN_WARNING "cio: store subchannel returned: cc=%d\n", cc); + printk(KERN_WARNING "cio: schib:\n"); + print_hex_dump(KERN_WARNING, "cio: ", DUMP_PREFIX_NONE, 16, 1, + &schib, sizeof(schib), 0); + printk(KERN_WARNING "cio: ccw device flags:\n"); + print_hex_dump(KERN_WARNING, "cio: ", DUMP_PREFIX_NONE, 16, 1, + &cdev->private->flags, sizeof(cdev->private->flags), 0); +} + /* * Timeout function. It just triggers a DEV_EVENT_TIMEOUT. */ @@ -92,6 +140,8 @@ ccw_device_timeout(unsigned long data) cdev = (struct ccw_device *) data; spin_lock_irq(cdev->ccwlock); + if (timeout_log_enabled) + ccw_timeout_log(cdev); dev_fsm_event(cdev, DEV_EVENT_TIMEOUT); spin_unlock_irq(cdev->ccwlock); } -- cgit v1.2.3 From 4e8e56c6713398f417317d449f50c08bf2756c66 Mon Sep 17 00:00:00 2001 From: Peter Oberparleiter Date: Sat, 26 Jan 2008 14:10:44 +0100 Subject: [S390] cio: Extend adapter interrupt interface. From: Cornelia Huck Change the adapter interrupt interface in order to allow multiple adapter interrupt handlers to be registered. Indicators are now allocated by cio instead of the device driver. The qdio parts have been Acked-by: Ursula Braun Signed-off-by: Peter Oberparleiter Signed-off-by: Cornelia Huck Signed-off-by: Martin Schwidefsky --- Documentation/DocBook/s390-drivers.tmpl | 1 + drivers/s390/cio/airq.c | 171 +++++++++++++++++++++----------- drivers/s390/cio/airq.h | 10 -- drivers/s390/cio/cio.c | 2 +- drivers/s390/cio/cio.h | 1 + drivers/s390/cio/qdio.c | 35 ++++--- include/asm-s390/airq.h | 19 ++++ 7 files changed, 154 insertions(+), 85 deletions(-) delete mode 100644 drivers/s390/cio/airq.h create mode 100644 include/asm-s390/airq.h (limited to 'Documentation') diff --git a/Documentation/DocBook/s390-drivers.tmpl b/Documentation/DocBook/s390-drivers.tmpl index 254e769282a4..3d2f31b99dd9 100644 --- a/Documentation/DocBook/s390-drivers.tmpl +++ b/Documentation/DocBook/s390-drivers.tmpl @@ -116,6 +116,7 @@ !Iinclude/asm-s390/ccwdev.h !Edrivers/s390/cio/device.c !Edrivers/s390/cio/device_ops.c +!Edrivers/s390/cio/airq.c The channel-measurement facility diff --git a/drivers/s390/cio/airq.c b/drivers/s390/cio/airq.c index 5287631fbfc8..b7a07a866291 100644 --- a/drivers/s390/cio/airq.c +++ b/drivers/s390/cio/airq.c @@ -1,12 +1,12 @@ /* * drivers/s390/cio/airq.c - * S/390 common I/O routines -- support for adapter interruptions + * Support for adapter interruptions * - * Copyright (C) 1999-2002 IBM Deutschland Entwicklung GmbH, - * IBM Corporation - * Author(s): Ingo Adlung (adlung@de.ibm.com) - * Cornelia Huck (cornelia.huck@de.ibm.com) - * Arnd Bergmann (arndb@de.ibm.com) + * Copyright IBM Corp. 1999,2007 + * Author(s): Ingo Adlung + * Cornelia Huck + * Arnd Bergmann + * Peter Oberparleiter */ #include @@ -14,72 +14,131 @@ #include #include +#include + +#include "cio.h" #include "cio_debug.h" -#include "airq.h" -static adapter_int_handler_t adapter_handler; +#define NR_AIRQS 32 +#define NR_AIRQS_PER_WORD sizeof(unsigned long) +#define NR_AIRQ_WORDS (NR_AIRQS / NR_AIRQS_PER_WORD) -/* - * register for adapter interrupts - * - * With HiperSockets the zSeries architecture provides for - * means of adapter interrups, pseudo I/O interrupts that are - * not tied to an I/O subchannel, but to an adapter. However, - * it doesn't disclose the info how to enable/disable them, but - * to recognize them only. Perhaps we should consider them - * being shared interrupts, and thus build a linked list - * of adapter handlers ... to be evaluated ... - */ -int -s390_register_adapter_interrupt (adapter_int_handler_t handler) -{ - int ret; - char dbf_txt[15]; +union indicator_t { + unsigned long word[NR_AIRQ_WORDS]; + unsigned char byte[NR_AIRQS]; +} __attribute__((packed)); - CIO_TRACE_EVENT (4, "rgaint"); +struct airq_t { + adapter_int_handler_t handler; + void *drv_data; +}; - if (handler == NULL) - ret = -EINVAL; - else - ret = (cmpxchg(&adapter_handler, NULL, handler) ? -EBUSY : 0); - if (!ret) - synchronize_sched(); /* Allow interrupts to complete. */ +static union indicator_t indicators; +static struct airq_t *airqs[NR_AIRQS]; - sprintf (dbf_txt, "ret:%d", ret); - CIO_TRACE_EVENT (4, dbf_txt); +static int register_airq(struct airq_t *airq) +{ + int i; - return ret; + for (i = 0; i < NR_AIRQS; i++) + if (!cmpxchg(&airqs[i], NULL, airq)) + return i; + return -ENOMEM; } -int -s390_unregister_adapter_interrupt (adapter_int_handler_t handler) +/** + * s390_register_adapter_interrupt() - register adapter interrupt handler + * @handler: adapter handler to be registered + * @drv_data: driver data passed with each call to the handler + * + * Returns: + * Pointer to the indicator to be used on success + * ERR_PTR() if registration failed + */ +void *s390_register_adapter_interrupt(adapter_int_handler_t handler, + void *drv_data) { + struct airq_t *airq; + char dbf_txt[16]; int ret; - char dbf_txt[15]; - CIO_TRACE_EVENT (4, "urgaint"); - - if (handler == NULL) - ret = -EINVAL; - else { - adapter_handler = NULL; - synchronize_sched(); /* Allow interrupts to complete. */ - ret = 0; + airq = kmalloc(sizeof(struct airq_t), GFP_KERNEL); + if (!airq) { + ret = -ENOMEM; + goto out; } - sprintf (dbf_txt, "ret:%d", ret); - CIO_TRACE_EVENT (4, dbf_txt); - - return ret; + airq->handler = handler; + airq->drv_data = drv_data; + ret = register_airq(airq); + if (ret < 0) + kfree(airq); +out: + snprintf(dbf_txt, sizeof(dbf_txt), "rairq:%d", ret); + CIO_TRACE_EVENT(4, dbf_txt); + if (ret < 0) + return ERR_PTR(ret); + else + return &indicators.byte[ret]; } +EXPORT_SYMBOL(s390_register_adapter_interrupt); -void -do_adapter_IO (void) +/** + * s390_unregister_adapter_interrupt - unregister adapter interrupt handler + * @ind: indicator for which the handler is to be unregistered + */ +void s390_unregister_adapter_interrupt(void *ind) { - CIO_TRACE_EVENT (6, "doaio"); + struct airq_t *airq; + char dbf_txt[16]; + int i; - if (adapter_handler) - (*adapter_handler) (); + i = (int) ((addr_t) ind) - ((addr_t) &indicators.byte[0]); + snprintf(dbf_txt, sizeof(dbf_txt), "urairq:%d", i); + CIO_TRACE_EVENT(4, dbf_txt); + indicators.byte[i] = 0; + airq = xchg(&airqs[i], NULL); + /* + * Allow interrupts to complete. This will ensure that the airq handle + * is no longer referenced by any interrupt handler. + */ + synchronize_sched(); + kfree(airq); } +EXPORT_SYMBOL(s390_unregister_adapter_interrupt); + +#define INDICATOR_MASK (0xffUL << ((NR_AIRQS_PER_WORD - 1) * 8)) -EXPORT_SYMBOL (s390_register_adapter_interrupt); -EXPORT_SYMBOL (s390_unregister_adapter_interrupt); +void do_adapter_IO(void) +{ + int w; + int i; + unsigned long word; + struct airq_t *airq; + + /* + * Access indicator array in word-sized chunks to minimize storage + * fetch operations. + */ + for (w = 0; w < NR_AIRQ_WORDS; w++) { + word = indicators.word[w]; + i = w * NR_AIRQS_PER_WORD; + /* + * Check bytes within word for active indicators. + */ + while (word) { + if (word & INDICATOR_MASK) { + airq = airqs[i]; + if (likely(airq)) + airq->handler(&indicators.byte[i], + airq->drv_data); + else + /* + * Reset ill-behaved indicator. + */ + indicators.byte[i] = 0; + } + word <<= 8; + i++; + } + } +} diff --git a/drivers/s390/cio/airq.h b/drivers/s390/cio/airq.h deleted file mode 100644 index 7d6be3fdcd66..000000000000 --- a/drivers/s390/cio/airq.h +++ /dev/null @@ -1,10 +0,0 @@ -#ifndef S390_AINTERRUPT_H -#define S390_AINTERRUPT_H - -typedef int (*adapter_int_handler_t)(void); - -extern int s390_register_adapter_interrupt(adapter_int_handler_t handler); -extern int s390_unregister_adapter_interrupt(adapter_int_handler_t handler); -extern void do_adapter_IO (void); - -#endif diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c index d0bcebde3fa2..89ced3408138 100644 --- a/drivers/s390/cio/cio.c +++ b/drivers/s390/cio/cio.c @@ -23,7 +23,7 @@ #include #include #include -#include "airq.h" +#include #include "cio.h" #include "css.h" #include "chsc.h" diff --git a/drivers/s390/cio/cio.h b/drivers/s390/cio/cio.h index a6ef218defbe..d1483d65049c 100644 --- a/drivers/s390/cio/cio.h +++ b/drivers/s390/cio/cio.h @@ -99,6 +99,7 @@ extern int cio_get_options (struct subchannel *); extern int cio_modify (struct subchannel *); int cio_create_sch_lock(struct subchannel *); +void do_adapter_IO(void); /* Use with care. */ #ifdef CONFIG_CCW_CONSOLE diff --git a/drivers/s390/cio/qdio.c b/drivers/s390/cio/qdio.c index 40a3208c7cf3..7a353db8564f 100644 --- a/drivers/s390/cio/qdio.c +++ b/drivers/s390/cio/qdio.c @@ -48,11 +48,11 @@ #include #include #include +#include #include "cio.h" #include "css.h" #include "device.h" -#include "airq.h" #include "qdio.h" #include "ioasm.h" #include "chsc.h" @@ -96,7 +96,7 @@ static debug_info_t *qdio_dbf_slsb_in; static volatile struct qdio_q *tiq_list=NULL; /* volatile as it could change during a while loop */ static DEFINE_SPINLOCK(ttiq_list_lock); -static int register_thinint_result; +static void *tiqdio_ind; static void tiqdio_tl(unsigned long); static DECLARE_TASKLET(tiqdio_tasklet,tiqdio_tl,0); @@ -399,7 +399,7 @@ qdio_get_indicator(void) { int i; - for (i=1;idev_st_chg_ind); } @@ -3740,23 +3738,25 @@ static void tiqdio_register_thinints(void) { char dbf_text[20]; - register_thinint_result= - s390_register_adapter_interrupt(&tiqdio_thinint_handler); - if (register_thinint_result) { - sprintf(dbf_text,"regthn%x",(register_thinint_result&0xff)); + + tiqdio_ind = + s390_register_adapter_interrupt(&tiqdio_thinint_handler, NULL); + if (IS_ERR(tiqdio_ind)) { + sprintf(dbf_text, "regthn%lx", PTR_ERR(tiqdio_ind)); QDIO_DBF_TEXT0(0,setup,dbf_text); QDIO_PRINT_ERR("failed to register adapter handler " \ - "(rc=%i).\nAdapter interrupts might " \ + "(rc=%li).\nAdapter interrupts might " \ "not work. Continuing.\n", - register_thinint_result); + PTR_ERR(tiqdio_ind)); + tiqdio_ind = NULL; } } static void tiqdio_unregister_thinints(void) { - if (!register_thinint_result) - s390_unregister_adapter_interrupt(&tiqdio_thinint_handler); + if (tiqdio_ind) + s390_unregister_adapter_interrupt(tiqdio_ind); } static int @@ -3768,8 +3768,8 @@ qdio_get_qdio_memory(void) for (i=1;i + * Cornelia Huck + * Arnd Bergmann + * Peter Oberparleiter + */ + +#ifndef _ASM_S390_AIRQ_H +#define _ASM_S390_AIRQ_H + +typedef void (*adapter_int_handler_t)(void *, void *); + +void *s390_register_adapter_interrupt(adapter_int_handler_t, void *); +void s390_unregister_adapter_interrupt(void *); + +#endif /* _ASM_S390_AIRQ_H */ -- cgit v1.2.3 From 48483b3290988952a593c6e66ca354c19f1a4350 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Sat, 26 Jan 2008 14:11:05 +0100 Subject: [S390] Get rid of additional_cpus kernel parameter. It caused only a lot of confusion. From now on cpu hotplug of up to NR_CPUS will work by default. If somebody wants to limit that then the possible_cpus parameter can be used. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- Documentation/cpu-hotplug.txt | 2 +- arch/s390/kernel/early.c | 1 - arch/s390/kernel/setup.c | 1 - arch/s390/kernel/smp.c | 180 ++++++++++++++++++------------------------ drivers/s390/char/sclp_cmd.c | 46 +---------- include/asm-s390/sclp.h | 1 - include/asm-s390/smp.h | 3 - 7 files changed, 79 insertions(+), 155 deletions(-) (limited to 'Documentation') diff --git a/Documentation/cpu-hotplug.txt b/Documentation/cpu-hotplug.txt index fb94f5a71b68..ba0aacde94fb 100644 --- a/Documentation/cpu-hotplug.txt +++ b/Documentation/cpu-hotplug.txt @@ -50,7 +50,7 @@ additional_cpus=n (*) Use this to limit hotpluggable cpus. This option sets cpu_possible_map = cpu_present_map + additional_cpus (*) Option valid only for following architectures -- x86_64, ia64, s390 +- x86_64, ia64 ia64 and x86_64 use the number of disabled local apics in ACPI tables MADT to determine the number of potentially hot-pluggable cpus. The implementation diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index c7cbb0114147..9f7b73b180f0 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -278,7 +278,6 @@ void __init startup_init(void) setup_lowcore_early(); sclp_read_info_early(); sclp_facilities_detect(); - sclp_read_cpu_info_early(); memsize = sclp_memory_detect(); #ifndef CONFIG_64BIT /* diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 3a61bfc2c4fb..cbdf3fb05e81 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -922,7 +922,6 @@ setup_arch(char **cmdline_p) cpu_init(); __cpu_logical_map[0] = S390_lowcore.cpu_data.cpu_addr; - smp_setup_cpu_possible_map(); /* * Setup capabilities (ELF_HWCAP & ELF_PLATFORM). diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 66fe28930d82..320e4e97bf52 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -54,7 +54,7 @@ EXPORT_SYMBOL(lowcore_ptr); cpumask_t cpu_online_map = CPU_MASK_NONE; EXPORT_SYMBOL(cpu_online_map); -cpumask_t cpu_possible_map = CPU_MASK_NONE; +cpumask_t cpu_possible_map = CPU_MASK_ALL; EXPORT_SYMBOL(cpu_possible_map); static struct task_struct *current_set[NR_CPUS]; @@ -399,7 +399,7 @@ static void __init smp_get_save_area(unsigned int cpu, unsigned int phy_cpu) "kernel was compiled with NR_CPUS=%i\n", cpu, NR_CPUS); return; } - zfcpdump_save_areas[cpu] = alloc_bootmem(sizeof(union save_area)); + zfcpdump_save_areas[cpu] = kmalloc(sizeof(union save_area), GFP_KERNEL); __cpu_logical_map[CPU_INIT_NO] = (__u16) phy_cpu; while (signal_processor(CPU_INIT_NO, sigp_stop_and_store_status) == sigp_busy) @@ -435,67 +435,6 @@ static int cpu_stopped(int cpu) return 0; } -/* - * Lets check how many CPUs we have. - */ -static void __init smp_count_cpus(unsigned int *configured_cpus, - unsigned int *standby_cpus) -{ - unsigned int cpu; - struct sclp_cpu_info *info; - u16 boot_cpu_addr, cpu_addr; - - boot_cpu_addr = S390_lowcore.cpu_data.cpu_addr; - current_thread_info()->cpu = 0; - *configured_cpus = 1; - *standby_cpus = 0; - - info = alloc_bootmem_pages(sizeof(*info)); - if (!info) - disabled_wait((unsigned long) __builtin_return_address(0)); - - /* Use sigp detection algorithm if sclp doesn't work. */ - if (sclp_get_cpu_info(info)) { - smp_use_sigp_detection = 1; - for (cpu = 0; cpu <= 65535; cpu++) { - if (cpu == boot_cpu_addr) - continue; - __cpu_logical_map[CPU_INIT_NO] = cpu; - if (cpu_stopped(CPU_INIT_NO)) - (*configured_cpus)++; - } - goto out; - } - - if (info->has_cpu_type) { - for (cpu = 0; cpu < info->combined; cpu++) { - if (info->cpu[cpu].address == boot_cpu_addr) { - smp_cpu_type = info->cpu[cpu].type; - break; - } - } - } - /* Count cpus. */ - for (cpu = 0; cpu < info->combined; cpu++) { - if (info->has_cpu_type && info->cpu[cpu].type != smp_cpu_type) - continue; - cpu_addr = info->cpu[cpu].address; - if (cpu_addr == boot_cpu_addr) - continue; - __cpu_logical_map[CPU_INIT_NO] = cpu_addr; - if (!cpu_stopped(CPU_INIT_NO)) { - (*standby_cpus)++; - continue; - } - smp_get_save_area(*configured_cpus, cpu_addr); - (*configured_cpus)++; - } -out: - printk(KERN_INFO "CPUs: %d configured, %d standby\n", - *configured_cpus, *standby_cpus); - free_bootmem((unsigned long) info, sizeof(*info)); -} - static int cpu_known(int cpu_id) { int cpu; @@ -529,7 +468,7 @@ static int smp_rescan_cpus_sigp(cpumask_t avail) return 0; } -static int __init_refok smp_rescan_cpus_sclp(cpumask_t avail) +static int smp_rescan_cpus_sclp(cpumask_t avail) { struct sclp_cpu_info *info; int cpu_id, logical_cpu, cpu; @@ -538,10 +477,7 @@ static int __init_refok smp_rescan_cpus_sclp(cpumask_t avail) logical_cpu = first_cpu(avail); if (logical_cpu == NR_CPUS) return 0; - if (slab_is_available()) - info = kmalloc(sizeof(*info), GFP_KERNEL); - else - info = alloc_bootmem(sizeof(*info)); + info = kmalloc(sizeof(*info), GFP_KERNEL); if (!info) return -ENOMEM; rc = sclp_get_cpu_info(info); @@ -564,10 +500,7 @@ static int __init_refok smp_rescan_cpus_sclp(cpumask_t avail) break; } out: - if (slab_is_available()) - kfree(info); - else - free_bootmem((unsigned long) info, sizeof(*info)); + kfree(info); return rc; } @@ -575,15 +508,71 @@ static int smp_rescan_cpus(void) { cpumask_t avail; - cpus_setall(avail); - cpus_and(avail, avail, cpu_possible_map); - cpus_andnot(avail, avail, cpu_present_map); + cpus_xor(avail, cpu_possible_map, cpu_present_map); if (smp_use_sigp_detection) return smp_rescan_cpus_sigp(avail); else return smp_rescan_cpus_sclp(avail); } +static void __init smp_detect_cpus(void) +{ + unsigned int cpu, c_cpus, s_cpus; + struct sclp_cpu_info *info; + u16 boot_cpu_addr, cpu_addr; + + c_cpus = 1; + s_cpus = 0; + boot_cpu_addr = S390_lowcore.cpu_data.cpu_addr; + info = kmalloc(sizeof(*info), GFP_KERNEL); + if (!info) + panic("smp_detect_cpus failed to allocate memory\n"); + /* Use sigp detection algorithm if sclp doesn't work. */ + if (sclp_get_cpu_info(info)) { + smp_use_sigp_detection = 1; + for (cpu = 0; cpu <= 65535; cpu++) { + if (cpu == boot_cpu_addr) + continue; + __cpu_logical_map[CPU_INIT_NO] = cpu; + if (!cpu_stopped(CPU_INIT_NO)) + continue; + smp_get_save_area(c_cpus, cpu); + c_cpus++; + } + goto out; + } + + if (info->has_cpu_type) { + for (cpu = 0; cpu < info->combined; cpu++) { + if (info->cpu[cpu].address == boot_cpu_addr) { + smp_cpu_type = info->cpu[cpu].type; + break; + } + } + } + + for (cpu = 0; cpu < info->combined; cpu++) { + if (info->has_cpu_type && info->cpu[cpu].type != smp_cpu_type) + continue; + cpu_addr = info->cpu[cpu].address; + if (cpu_addr == boot_cpu_addr) + continue; + __cpu_logical_map[CPU_INIT_NO] = cpu_addr; + if (!cpu_stopped(CPU_INIT_NO)) { + s_cpus++; + continue; + } + smp_get_save_area(c_cpus, cpu_addr); + c_cpus++; + } +out: + kfree(info); + printk(KERN_INFO "CPUs: %d configured, %d standby\n", c_cpus, s_cpus); + lock_cpu_hotplug(); + smp_rescan_cpus(); + unlock_cpu_hotplug(); +} + /* * Activate a secondary processor. */ @@ -674,41 +663,20 @@ int __cpu_up(unsigned int cpu) return 0; } -static unsigned int __initdata additional_cpus; -static unsigned int __initdata possible_cpus; - -void __init smp_setup_cpu_possible_map(void) -{ - unsigned int pos_cpus, cpu; - unsigned int configured_cpus, standby_cpus; - - smp_count_cpus(&configured_cpus, &standby_cpus); - pos_cpus = min(configured_cpus + standby_cpus + additional_cpus, - (unsigned int) NR_CPUS); - if (possible_cpus) - pos_cpus = min(possible_cpus, (unsigned int) NR_CPUS); - for (cpu = 0; cpu < pos_cpus; cpu++) - cpu_set(cpu, cpu_possible_map); - cpu_present_map = cpumask_of_cpu(0); - smp_rescan_cpus(); -} - -#ifdef CONFIG_HOTPLUG_CPU - -static int __init setup_additional_cpus(char *s) -{ - additional_cpus = simple_strtoul(s, NULL, 0); - return 0; -} -early_param("additional_cpus", setup_additional_cpus); - static int __init setup_possible_cpus(char *s) { - possible_cpus = simple_strtoul(s, NULL, 0); + int pcpus, cpu; + + pcpus = simple_strtoul(s, NULL, 0); + cpu_possible_map = cpumask_of_cpu(0); + for (cpu = 1; cpu < pcpus && cpu < NR_CPUS; cpu++) + cpu_set(cpu, cpu_possible_map); return 0; } early_param("possible_cpus", setup_possible_cpus); +#ifdef CONFIG_HOTPLUG_CPU + int __cpu_disable(void) { struct ec_creg_mask_parms cr_parms; @@ -768,6 +736,8 @@ void __init smp_prepare_cpus(unsigned int max_cpus) unsigned int cpu; int i; + smp_detect_cpus(); + /* request the 0x1201 emergency signal external interrupt */ if (register_external_interrupt(0x1201, do_ext_call_interrupt) != 0) panic("Couldn't request external interrupt 0x1201"); @@ -816,6 +786,8 @@ void __init smp_prepare_boot_cpu(void) { BUG_ON(smp_processor_id() != 0); + current_thread_info()->cpu = 0; + cpu_set(0, cpu_present_map); cpu_set(0, cpu_online_map); S390_lowcore.percpu_offset = __per_cpu_offset[0]; current_set[0] = current; diff --git a/drivers/s390/char/sclp_cmd.c b/drivers/s390/char/sclp_cmd.c index d7e6f4d65b78..b5c23396f8fe 100644 --- a/drivers/s390/char/sclp_cmd.c +++ b/drivers/s390/char/sclp_cmd.c @@ -191,9 +191,6 @@ struct read_cpu_info_sccb { u8 reserved[4096 - 16]; } __attribute__((packed, aligned(PAGE_SIZE))); -static struct read_cpu_info_sccb __initdata early_read_cpu_info_sccb; -static struct sclp_cpu_info __initdata sclp_cpu_info; - static void sclp_fill_cpu_info(struct sclp_cpu_info *info, struct read_cpu_info_sccb *sccb) { @@ -208,48 +205,16 @@ static void sclp_fill_cpu_info(struct sclp_cpu_info *info, info->combined * sizeof(struct sclp_cpu_entry)); } -void __init sclp_read_cpu_info_early(void) -{ - int rc; - struct read_cpu_info_sccb *sccb; - - if (!SCLP_HAS_CPU_INFO) - return; - - sccb = &early_read_cpu_info_sccb; - do { - memset(sccb, 0, sizeof(*sccb)); - sccb->header.length = sizeof(*sccb); - rc = sclp_cmd_sync_early(SCLP_CMDW_READ_CPU_INFO, sccb); - } while (rc == -EBUSY); - - if (rc) - return; - if (sccb->header.response_code != 0x10) - return; - sclp_fill_cpu_info(&sclp_cpu_info, sccb); -} - -static int __init sclp_get_cpu_info_early(struct sclp_cpu_info *info) -{ - if (!SCLP_HAS_CPU_INFO) - return -EOPNOTSUPP; - *info = sclp_cpu_info; - return 0; -} - -static int sclp_get_cpu_info_late(struct sclp_cpu_info *info) +int sclp_get_cpu_info(struct sclp_cpu_info *info) { int rc; struct read_cpu_info_sccb *sccb; if (!SCLP_HAS_CPU_INFO) return -EOPNOTSUPP; - sccb = (struct read_cpu_info_sccb *) __get_free_page(GFP_KERNEL - | GFP_DMA); + sccb = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA); if (!sccb) return -ENOMEM; - memset(sccb, 0, sizeof(*sccb)); sccb->header.length = sizeof(*sccb); rc = do_sync_request(SCLP_CMDW_READ_CPU_INFO, sccb); if (rc) @@ -266,13 +231,6 @@ out: return rc; } -int __init_refok sclp_get_cpu_info(struct sclp_cpu_info *info) -{ - if (slab_is_available()) - return sclp_get_cpu_info_late(info); - return sclp_get_cpu_info_early(info); -} - struct cpu_configure_sccb { struct sccb_header header; } __attribute__((packed, aligned(8))); diff --git a/include/asm-s390/sclp.h b/include/asm-s390/sclp.h index b8c7695cd4c8..b5f2843013a3 100644 --- a/include/asm-s390/sclp.h +++ b/include/asm-s390/sclp.h @@ -46,7 +46,6 @@ int sclp_get_cpu_info(struct sclp_cpu_info *info); int sclp_cpu_configure(u8 cpu); int sclp_cpu_deconfigure(u8 cpu); void sclp_read_info_early(void); -void sclp_read_cpu_info_early(void); void sclp_facilities_detect(void); unsigned long long sclp_memory_detect(void); int sclp_sdias_blk_count(void); diff --git a/include/asm-s390/smp.h b/include/asm-s390/smp.h index 07708c07701e..218454b91862 100644 --- a/include/asm-s390/smp.h +++ b/include/asm-s390/smp.h @@ -35,8 +35,6 @@ extern void machine_restart_smp(char *); extern void machine_halt_smp(void); extern void machine_power_off_smp(void); -extern void smp_setup_cpu_possible_map(void); - #define NO_PROC_ID 0xFF /* No processor magic marker */ /* @@ -103,7 +101,6 @@ static inline void smp_send_stop(void) #define hard_smp_processor_id() 0 #define smp_cpu_not_running(cpu) 1 -#define smp_setup_cpu_possible_map() do { } while (0) #endif extern union save_area *zfcpdump_save_areas[NR_CPUS + 1]; -- cgit v1.2.3 From 9e2697ff371b4380dca108a66860868c19d8c4b6 Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 14 Dec 2007 13:30:14 +0000 Subject: [ARM] pxa: add cpufreq support There have been patches hanging around for ages to add support for cpufreq to PXA255 processors. It's about time we applied one. Signed-off-by: Russell King --- Documentation/cpu-freq/user-guide.txt | 1 + arch/arm/Kconfig | 8 +- arch/arm/mach-pxa/Makefile | 1 + arch/arm/mach-pxa/cpu-pxa.c | 293 ++++++++++++++++++++++++++++++++++ 4 files changed, 302 insertions(+), 1 deletion(-) create mode 100644 arch/arm/mach-pxa/cpu-pxa.c (limited to 'Documentation') diff --git a/Documentation/cpu-freq/user-guide.txt b/Documentation/cpu-freq/user-guide.txt index 555c8cf3650a..af3b925ece08 100644 --- a/Documentation/cpu-freq/user-guide.txt +++ b/Documentation/cpu-freq/user-guide.txt @@ -45,6 +45,7 @@ The following ARM processors are supported by cpufreq: ARM Integrator ARM-SA1100 ARM-SA1110 +Intel PXA 1.2 x86 diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 1be718208704..4ea3d9e97f7f 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -868,7 +868,7 @@ config KEXEC endmenu -if (ARCH_SA1100 || ARCH_INTEGRATOR || ARCH_OMAP || ARCH_IMX ) +if (ARCH_SA1100 || ARCH_INTEGRATOR || ARCH_OMAP || ARCH_IMX || ARCH_PXA) menu "CPU Frequency scaling" @@ -904,6 +904,12 @@ config CPU_FREQ_IMX If in doubt, say N. +config CPU_FREQ_PXA + bool + depends on CPU_FREQ && ARCH_PXA && PXA25x + default y + select CPU_FREQ_DEFAULT_GOV_USERSPACE + endmenu endif diff --git a/arch/arm/mach-pxa/Makefile b/arch/arm/mach-pxa/Makefile index 4263527e5123..3133dc4eaa7e 100644 --- a/arch/arm/mach-pxa/Makefile +++ b/arch/arm/mach-pxa/Makefile @@ -42,6 +42,7 @@ obj-$(CONFIG_LEDS) += $(led-y) # Misc features obj-$(CONFIG_PM) += pm.o sleep.o +obj-$(CONFIG_CPU_FREQ) += cpu-pxa.o obj-$(CONFIG_PXA_SSP) += ssp.o ifeq ($(CONFIG_PXA27x),y) diff --git a/arch/arm/mach-pxa/cpu-pxa.c b/arch/arm/mach-pxa/cpu-pxa.c new file mode 100644 index 000000000000..18d042bdf243 --- /dev/null +++ b/arch/arm/mach-pxa/cpu-pxa.c @@ -0,0 +1,293 @@ +/* + * linux/arch/arm/mach-pxa/cpu-pxa.c + * + * Copyright (C) 2002,2003 Intrinsyc Software + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * History: + * 31-Jul-2002 : Initial version [FB] + * 29-Jan-2003 : added PXA255 support [FB] + * 20-Apr-2003 : ported to v2.5 (Dustin McIntire, Sensoria Corp.) + * + * Note: + * This driver may change the memory bus clock rate, but will not do any + * platform specific access timing changes... for example if you have flash + * memory connected to CS0, you will need to register a platform specific + * notifier which will adjust the memory access strobes to maintain a + * minimum strobe width. + * + */ + +#include +#include +#include +#include +#include + +#include +#include + +#ifdef DEBUG +static unsigned int freq_debug; +MODULE_PARM(freq_debug, "i"); +MODULE_PARM_DESC(freq_debug, "Set the debug messages to on=1/off=0"); +#else +#define freq_debug 0 +#endif + +typedef struct { + unsigned int khz; + unsigned int membus; + unsigned int cccr; + unsigned int div2; +} pxa_freqs_t; + +/* Define the refresh period in mSec for the SDRAM and the number of rows */ +#define SDRAM_TREF 64 /* standard 64ms SDRAM */ +#define SDRAM_ROWS 4096 /* 64MB=8192 32MB=4096 */ +#define MDREFR_DRI(x) (((x) * SDRAM_TREF) / (SDRAM_ROWS * 32)) + +#define CCLKCFG_TURBO 0x1 +#define CCLKCFG_FCS 0x2 +#define PXA25x_MIN_FREQ 99500 +#define PXA25x_MAX_FREQ 398100 +#define MDREFR_DB2_MASK (MDREFR_K2DB2 | MDREFR_K1DB2) +#define MDREFR_DRI_MASK 0xFFF + + +/* Use the run mode frequencies for the CPUFREQ_POLICY_PERFORMANCE policy */ +static pxa_freqs_t pxa255_run_freqs[] = +{ + /* CPU MEMBUS CCCR DIV2*/ + { 99500, 99500, 0x121, 1}, /* run= 99, turbo= 99, PXbus=50, SDRAM=50 */ + {132700, 132700, 0x123, 1}, /* run=133, turbo=133, PXbus=66, SDRAM=66 */ + {199100, 99500, 0x141, 0}, /* run=199, turbo=199, PXbus=99, SDRAM=99 */ + {265400, 132700, 0x143, 1}, /* run=265, turbo=265, PXbus=133, SDRAM=66 */ + {331800, 165900, 0x145, 1}, /* run=331, turbo=331, PXbus=166, SDRAM=83 */ + {398100, 99500, 0x161, 0}, /* run=398, turbo=398, PXbus=196, SDRAM=99 */ + {0,} +}; +#define NUM_RUN_FREQS ARRAY_SIZE(pxa255_run_freqs) + +static struct cpufreq_frequency_table pxa255_run_freq_table[NUM_RUN_FREQS+1]; + +/* Use the turbo mode frequencies for the CPUFREQ_POLICY_POWERSAVE policy */ +static pxa_freqs_t pxa255_turbo_freqs[] = +{ + /* CPU MEMBUS CCCR DIV2*/ + { 99500, 99500, 0x121, 1}, /* run=99, turbo= 99, PXbus=50, SDRAM=50 */ + {199100, 99500, 0x221, 0}, /* run=99, turbo=199, PXbus=50, SDRAM=99 */ + {298500, 99500, 0x321, 0}, /* run=99, turbo=287, PXbus=50, SDRAM=99 */ + {298600, 99500, 0x1c1, 0}, /* run=199, turbo=287, PXbus=99, SDRAM=99 */ + {398100, 99500, 0x241, 0}, /* run=199, turbo=398, PXbus=99, SDRAM=99 */ + {0,} +}; +#define NUM_TURBO_FREQS ARRAY_SIZE(pxa255_turbo_freqs) + +static struct cpufreq_frequency_table pxa255_turbo_freq_table[NUM_TURBO_FREQS+1]; + +extern unsigned get_clk_frequency_khz(int info); + +/* find a valid frequency point */ +static int pxa_verify_policy(struct cpufreq_policy *policy) +{ + struct cpufreq_frequency_table *pxa_freqs_table; + int ret; + + if (policy->policy == CPUFREQ_POLICY_PERFORMANCE) { + pxa_freqs_table = pxa255_run_freq_table; + } else if (policy->policy == CPUFREQ_POLICY_POWERSAVE) { + pxa_freqs_table = pxa255_turbo_freq_table; + } else { + printk("CPU PXA: Unknown policy found. " + "Using CPUFREQ_POLICY_PERFORMANCE\n"); + pxa_freqs_table = pxa255_run_freq_table; + } + + ret = cpufreq_frequency_table_verify(policy, pxa_freqs_table); + + if (freq_debug) + pr_debug("Verified CPU policy: %dKhz min to %dKhz max\n", + policy->min, policy->max); + + return ret; +} + +static int pxa_set_target(struct cpufreq_policy *policy, + unsigned int target_freq, + unsigned int relation) +{ + struct cpufreq_frequency_table *pxa_freqs_table; + pxa_freqs_t *pxa_freq_settings; + struct cpufreq_freqs freqs; + int idx; + unsigned long flags; + unsigned int unused, preset_mdrefr, postset_mdrefr; + void *ramstart = phys_to_virt(0xa0000000); + + /* Get the current policy */ + if (policy->policy == CPUFREQ_POLICY_PERFORMANCE) { + pxa_freq_settings = pxa255_run_freqs; + pxa_freqs_table = pxa255_run_freq_table; + } else if (policy->policy == CPUFREQ_POLICY_POWERSAVE) { + pxa_freq_settings = pxa255_turbo_freqs; + pxa_freqs_table = pxa255_turbo_freq_table; + } else { + printk("CPU PXA: Unknown policy found. " + "Using CPUFREQ_POLICY_PERFORMANCE\n"); + pxa_freq_settings = pxa255_run_freqs; + pxa_freqs_table = pxa255_run_freq_table; + } + + /* Lookup the next frequency */ + if (cpufreq_frequency_table_target(policy, pxa_freqs_table, + target_freq, relation, &idx)) { + return -EINVAL; + } + + freqs.old = policy->cur; + freqs.new = pxa_freq_settings[idx].khz; + freqs.cpu = policy->cpu; + + if (freq_debug) + pr_debug(KERN_INFO "Changing CPU frequency to %d Mhz, (SDRAM %d Mhz)\n", + freqs.new / 1000, (pxa_freq_settings[idx].div2) ? + (pxa_freq_settings[idx].membus / 2000) : + (pxa_freq_settings[idx].membus / 1000)); + + /* + * Tell everyone what we're about to do... + * you should add a notify client with any platform specific + * Vcc changing capability + */ + cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); + + /* Calculate the next MDREFR. If we're slowing down the SDRAM clock + * we need to preset the smaller DRI before the change. If we're speeding + * up we need to set the larger DRI value after the change. + */ + preset_mdrefr = postset_mdrefr = MDREFR; + if ((MDREFR & MDREFR_DRI_MASK) > MDREFR_DRI(pxa_freq_settings[idx].membus)) { + preset_mdrefr = (preset_mdrefr & ~MDREFR_DRI_MASK) | + MDREFR_DRI(pxa_freq_settings[idx].membus); + } + postset_mdrefr = (postset_mdrefr & ~MDREFR_DRI_MASK) | + MDREFR_DRI(pxa_freq_settings[idx].membus); + + /* If we're dividing the memory clock by two for the SDRAM clock, this + * must be set prior to the change. Clearing the divide must be done + * after the change. + */ + if (pxa_freq_settings[idx].div2) { + preset_mdrefr |= MDREFR_DB2_MASK; + postset_mdrefr |= MDREFR_DB2_MASK; + } else { + postset_mdrefr &= ~MDREFR_DB2_MASK; + } + + local_irq_save(flags); + + /* Set new the CCCR */ + CCCR = pxa_freq_settings[idx].cccr; + + asm volatile(" \n\ + ldr r4, [%1] /* load MDREFR */ \n\ + b 2f \n\ + .align 5 \n\ +1: \n\ + str %4, [%1] /* preset the MDREFR */ \n\ + mcr p14, 0, %2, c6, c0, 0 /* set CCLKCFG[FCS] */ \n\ + str %5, [%1] /* postset the MDREFR */ \n\ + \n\ + b 3f \n\ +2: b 1b \n\ +3: nop \n\ + " + : "=&r" (unused) + : "r" (&MDREFR), "r" (CCLKCFG_TURBO|CCLKCFG_FCS), "r" (ramstart), + "r" (preset_mdrefr), "r" (postset_mdrefr) + : "r4", "r5"); + local_irq_restore(flags); + + /* + * Tell everyone what we've just done... + * you should add a notify client with any platform specific + * SDRAM refresh timer adjustments + */ + cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); + + return 0; +} + +static int pxa_cpufreq_init(struct cpufreq_policy *policy) +{ + int i; + + /* set default policy and cpuinfo */ + policy->governor = CPUFREQ_DEFAULT_GOVERNOR; + policy->policy = CPUFREQ_POLICY_PERFORMANCE; + policy->cpuinfo.max_freq = PXA25x_MAX_FREQ; + policy->cpuinfo.min_freq = PXA25x_MIN_FREQ; + policy->cpuinfo.transition_latency = 1000; /* FIXME: 1 ms, assumed */ + policy->cur = get_clk_frequency_khz(0); /* current freq */ + policy->min = policy->max = policy->cur; + + /* Generate the run cpufreq_frequency_table struct */ + for (i = 0; i < NUM_RUN_FREQS; i++) { + pxa255_run_freq_table[i].frequency = pxa255_run_freqs[i].khz; + pxa255_run_freq_table[i].index = i; + } + + pxa255_run_freq_table[i].frequency = CPUFREQ_TABLE_END; + /* Generate the turbo cpufreq_frequency_table struct */ + for (i = 0; i < NUM_TURBO_FREQS; i++) { + pxa255_turbo_freq_table[i].frequency = pxa255_turbo_freqs[i].khz; + pxa255_turbo_freq_table[i].index = i; + } + pxa255_turbo_freq_table[i].frequency = CPUFREQ_TABLE_END; + + printk(KERN_INFO "PXA CPU frequency change support initialized\n"); + + return 0; +} + +static struct cpufreq_driver pxa_cpufreq_driver = { + .verify = pxa_verify_policy, + .target = pxa_set_target, + .init = pxa_cpufreq_init, + .name = "PXA25x", +}; + +static int __init pxa_cpu_init(void) +{ + int ret = -ENODEV; + if (cpu_is_pxa25x()) + ret = cpufreq_register_driver(&pxa_cpufreq_driver); + return ret; +} + +static void __exit pxa_cpu_exit(void) +{ + if (cpu_is_pxa25x()) + cpufreq_unregister_driver(&pxa_cpufreq_driver); +} + + +MODULE_AUTHOR ("Intrinsyc Software Inc."); +MODULE_DESCRIPTION ("CPU frequency changing driver for the PXA architecture"); +MODULE_LICENSE("GPL"); +module_init(pxa_cpu_init); +module_exit(pxa_cpu_exit); -- cgit v1.2.3 From 5de865b4c5af253db19bdae3c0553952dc7eda1d Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Mon, 3 Dec 2007 17:15:52 -0500 Subject: ARM kprobes: let's enable it Signed-off-by: Nicolas Pitre --- Documentation/kprobes.txt | 1 + arch/arm/Kconfig.instrumentation | 10 ++++++++++ 2 files changed, 11 insertions(+) (limited to 'Documentation') diff --git a/Documentation/kprobes.txt b/Documentation/kprobes.txt index cb12ae175aa2..53a63890aea4 100644 --- a/Documentation/kprobes.txt +++ b/Documentation/kprobes.txt @@ -141,6 +141,7 @@ architectures: - ppc64 - ia64 (Does not support probes on instruction slot1.) - sparc64 (Return probes not yet implemented.) +- arm 3. Configuring Kprobes diff --git a/arch/arm/Kconfig.instrumentation b/arch/arm/Kconfig.instrumentation index 63b8c6d5606a..453ad8e15d69 100644 --- a/arch/arm/Kconfig.instrumentation +++ b/arch/arm/Kconfig.instrumentation @@ -43,6 +43,16 @@ config OPROFILE_MPCORE config OPROFILE_ARM11_CORE bool +config KPROBES + bool "Kprobes" + depends on KALLSYMS && MODULES && !UML && !XIP_KERNEL + help + Kprobes allows you to trap at almost any kernel address and + execute a callback function. register_kprobe() establishes + a probepoint and specifies the callback. Kprobes is useful + for kernel debugging, non-intrusive instrumentation and testing. + If in doubt, say "N". + config MARKERS bool "Activate markers" help -- cgit v1.2.3 From ade2daf9c6e57845fe83a24e0a9fa1c03c6e91b1 Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Sat, 26 Jan 2008 20:13:07 +0100 Subject: ide: make remaining built-in only IDE host drivers modular (take 2) * Make remaining built-in only IDE host drivers modular, add ide-scan-pci.c file for probing PCI host drivers registered with IDE core (special case for built-in IDE and CONFIG_IDEPCI_PCIBUS_ORDER=y) and then take care of the ordering in which all IDE host drivers are probed when IDE is built-in during link time. * Move probing of gayle, falconide, macide, q40ide and buddha (m68k arch specific) host drivers, before PCI ones (no PCI on m68k), ide-cris (cris arch specific), cmd640 (x86 arch specific) and pmac (ppc arch specific). * Move probing of ide-cris (cris arch specific) host driver before cmd640 (x86 arch specific). * Move probing of mpc8xx (ppc specific) host driver before ide-pnp (depends on ISA and none of ppc platform that use mpc8xx supports ISA) and ide-h8300 (h8300 arch specific). * Add "probe_vlb" kernel parameter to cmd640 host driver and update Documentation/ide.txt accordingly. * Make IDE_ARM config option visible so it can also be disabled if needed. * Remove bogus comment from ide.c while at it. v2: * Fix two issues spotted by Sergei: - replace ENOMEM error value by ENOENT in ide-h8300 host driver - fix MODULE_PARM_DESC() in cmd640 host driver Cc: Sergei Shtylyov Cc: Mikael Starvik Cc: Geert Uytterhoeven Cc: Roman Zippel Cc: Benjamin Herrenschmidt Signed-off-by: Bartlomiej Zolnierkiewicz --- Documentation/ide.txt | 9 +-- drivers/ide/Kconfig | 26 ++++---- drivers/ide/Makefile | 58 ++++++++++-------- drivers/ide/arm/Makefile | 4 ++ drivers/ide/arm/ide_arm.c | 6 +- drivers/ide/cris/Makefile | 2 +- drivers/ide/cris/ide-cris.c | 7 ++- drivers/ide/h8300/Makefile | 2 + drivers/ide/h8300/ide-h8300.c | 10 +++- drivers/ide/ide-pnp.c | 9 ++- drivers/ide/ide-scan-pci.c | 11 ++++ drivers/ide/ide.c | 131 +---------------------------------------- drivers/ide/legacy/Makefile | 19 ++++-- drivers/ide/legacy/ali14xx.c | 5 +- drivers/ide/legacy/buddha.c | 6 +- drivers/ide/legacy/dtc2278.c | 5 +- drivers/ide/legacy/falconide.c | 7 ++- drivers/ide/legacy/gayle.c | 10 +++- drivers/ide/legacy/ht6560b.c | 5 +- drivers/ide/legacy/macide.c | 8 ++- drivers/ide/legacy/q40ide.c | 7 ++- drivers/ide/legacy/qd65xx.c | 5 +- drivers/ide/legacy/umc8672.c | 5 +- drivers/ide/pci/Makefile | 4 ++ drivers/ide/pci/cmd640.c | 8 ++- drivers/ide/ppc/Makefile | 3 + drivers/ide/ppc/mpc8xx.c | 6 +- drivers/ide/ppc/pmac.c | 2 + drivers/ide/setup-pci.c | 9 +-- include/linux/ide.h | 3 +- 30 files changed, 171 insertions(+), 221 deletions(-) create mode 100644 drivers/ide/h8300/Makefile create mode 100644 drivers/ide/ide-scan-pci.c create mode 100644 drivers/ide/ppc/Makefile (limited to 'Documentation') diff --git a/Documentation/ide.txt b/Documentation/ide.txt index 1d50f23a5cab..b29ccb43d6cc 100644 --- a/Documentation/ide.txt +++ b/Documentation/ide.txt @@ -30,7 +30,7 @@ *** *** The CMD640 is also used on some Vesa Local Bus (VLB) cards, and is *NOT* *** automatically detected by Linux. For safe, reliable operation with such -*** interfaces, one *MUST* use the "ide0=cmd640_vlb" kernel option. +*** interfaces, one *MUST* use the "cmd640.probe_vlb" kernel option. *** *** Use of the "serialize" option is no longer necessary. @@ -292,9 +292,6 @@ The following are valid ONLY on ide0, which usually corresponds to the first ATA interface found on the particular host, and the defaults for the base,ctl ports must not be altered. - "ide0=cmd640_vlb" : *REQUIRED* for VLB cards with the CMD640 chip - (not for PCI -- automatically detected) - "ide=doubler" : probe/support IDE doublers on Amiga There may be more options than shown -- use the source, Luke! @@ -310,6 +307,10 @@ i.e. to enable probing for ALI M14xx chipsets (ali14xx host driver) use: * "probe" module parameter when ali14xx driver is compiled as module ("modprobe ali14xx probe") +Also for legacy CMD640 host driver (cmd640) you need to use "probe_vlb" +kernel paremeter to enable probing for VLB version of the chipset (PCI ones +are detected automatically). + ================================================================================ IDE ATAPI streaming tape driver diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig index 7c419e87a4a4..e92128a87f26 100644 --- a/drivers/ide/Kconfig +++ b/drivers/ide/Kconfig @@ -325,7 +325,7 @@ config BLK_DEV_PLATFORM If unsure, say N. config BLK_DEV_CMD640 - bool "CMD640 chipset bugfix/support" + tristate "CMD640 chipset bugfix/support" depends on X86 ---help--- The CMD-Technologies CMD640 IDE chip is used on many common 486 and @@ -359,7 +359,7 @@ config BLK_DEV_CMD640_ENHANCED Otherwise say N. config BLK_DEV_IDEPNP - bool "PNP EIDE support" + tristate "PNP EIDE support" depends on PNP help If you have a PnP (Plug and Play) compatible EIDE card and @@ -788,7 +788,7 @@ config BLK_DEV_CELLEB endif config BLK_DEV_IDE_PMAC - bool "Builtin PowerMac IDE support" + tristate "Builtin PowerMac IDE support" depends on PPC_PMAC && IDE=y && BLK_DEV_IDE=y help This driver provides support for the built-in IDE controller on @@ -842,7 +842,9 @@ config BLK_DEV_IDE_AU1XXX_SEQTS_PER_RQ depends on BLK_DEV_IDE_AU1XXX config IDE_ARM - def_bool ARM && (ARCH_CLPS7500 || ARCH_RPC || ARCH_SHARK) + tristate "ARM IDE support" + depends on ARM && (ARCH_CLPS7500 || ARCH_RPC || ARCH_SHARK) + default y config BLK_DEV_IDE_ICSIDE tristate "ICS IDE interface support" @@ -874,7 +876,7 @@ config BLK_DEV_IDE_BAST Simtec BAST or the Thorcom VR1000 config ETRAX_IDE - bool "ETRAX IDE support" + tristate "ETRAX IDE support" depends on CRIS && BROKEN select BLK_DEV_IDEDMA help @@ -908,14 +910,14 @@ config ETRAX_IDE_G27_RESET endchoice config IDE_H8300 - bool "H8300 IDE support" + tristate "H8300 IDE support" depends on H8300 default y help Enables the H8300 IDE driver. config BLK_DEV_GAYLE - bool "Amiga Gayle IDE interface support" + tristate "Amiga Gayle IDE interface support" depends on AMIGA help This is the IDE driver for the Amiga Gayle IDE interface. It supports @@ -946,7 +948,7 @@ config BLK_DEV_IDEDOUBLER runtime using the "ide=doubler" kernel boot parameter. config BLK_DEV_BUDDHA - bool "Buddha/Catweasel/X-Surf IDE interface support (EXPERIMENTAL)" + tristate "Buddha/Catweasel/X-Surf IDE interface support (EXPERIMENTAL)" depends on ZORRO && EXPERIMENTAL help This is the IDE driver for the IDE interfaces on the Buddha, @@ -958,7 +960,7 @@ config BLK_DEV_BUDDHA to one of its IDE interfaces. config BLK_DEV_FALCON_IDE - bool "Falcon IDE interface support" + tristate "Falcon IDE interface support" depends on ATARI help This is the IDE driver for the builtin IDE interface on the Atari @@ -967,7 +969,7 @@ config BLK_DEV_FALCON_IDE interface. config BLK_DEV_MAC_IDE - bool "Macintosh Quadra/Powerbook IDE interface support" + tristate "Macintosh Quadra/Powerbook IDE interface support" depends on MAC help This is the IDE driver for the builtin IDE interface on some m68k @@ -980,7 +982,7 @@ config BLK_DEV_MAC_IDE builtin IDE interface. config BLK_DEV_Q40IDE - bool "Q40/Q60 IDE interface support" + tristate "Q40/Q60 IDE interface support" depends on Q40 help Enable the on-board IDE controller in the Q40/Q60. This should @@ -988,7 +990,7 @@ config BLK_DEV_Q40IDE drive subsystem through an expansion card. config BLK_DEV_MPC8xx_IDE - bool "MPC8xx IDE support" + tristate "MPC8xx IDE support" depends on 8xx && (LWMON || IVMS8 || IVML24 || TQM8xxL) && IDE=y && BLK_DEV_IDE=y && !PPC_MERGE help This option provides support for IDE on Motorola MPC8xx Systems. diff --git a/drivers/ide/Makefile b/drivers/ide/Makefile index b181fc672057..0d2da89d15cf 100644 --- a/drivers/ide/Makefile +++ b/drivers/ide/Makefile @@ -7,41 +7,37 @@ # Note : at this point, these files are compiled on all systems. # In the future, some of these should be built conditionally. # -# First come modules that register themselves with the core +# link order is important here EXTRA_CFLAGS += -Idrivers/ide -obj-$(CONFIG_BLK_DEV_IDE) += pci/ - ide-core-y += ide.o ide-io.o ide-iops.o ide-lib.o ide-probe.o ide-taskfile.o -ide-core-$(CONFIG_BLK_DEV_CMD640) += pci/cmd640.o - -# Core IDE code - must come before legacy +# core IDE code ide-core-$(CONFIG_BLK_DEV_IDEPCI) += setup-pci.o ide-core-$(CONFIG_BLK_DEV_IDEDMA) += ide-dma.o ide-core-$(CONFIG_IDE_PROC_FS) += ide-proc.o -ide-core-$(CONFIG_BLK_DEV_IDEPNP) += ide-pnp.o ide-core-$(CONFIG_BLK_DEV_IDEACPI) += ide-acpi.o -# built-in only drivers from arm/ -ide-core-$(CONFIG_IDE_ARM) += arm/ide_arm.o +obj-$(CONFIG_BLK_DEV_IDE) += ide-core.o -# built-in only drivers from legacy/ -ide-core-$(CONFIG_BLK_DEV_BUDDHA) += legacy/buddha.o -ide-core-$(CONFIG_BLK_DEV_FALCON_IDE) += legacy/falconide.o -ide-core-$(CONFIG_BLK_DEV_GAYLE) += legacy/gayle.o -ide-core-$(CONFIG_BLK_DEV_MAC_IDE) += legacy/macide.o -ide-core-$(CONFIG_BLK_DEV_Q40IDE) += legacy/q40ide.o +ifeq ($(CONFIG_IDE_ARM), y) + ide-arm-core-y += arm/ide_arm.o + obj-y += ide-arm-core.o +endif -# built-in only drivers from ppc/ -ide-core-$(CONFIG_BLK_DEV_MPC8xx_IDE) += ppc/mpc8xx.o -ide-core-$(CONFIG_BLK_DEV_IDE_PMAC) += ppc/pmac.o +obj-$(CONFIG_BLK_DEV_IDE) += legacy/ pci/ -# built-in only drivers from h8300/ -ide-core-$(CONFIG_IDE_H8300) += h8300/ide-h8300.o +obj-$(CONFIG_IDEPCI_PCIBUS_ORDER) += ide-scan-pci.o -obj-$(CONFIG_BLK_DEV_IDE) += ide-core.o +ifeq ($(CONFIG_BLK_DEV_CMD640), y) + cmd640-core-y += pci/cmd640.o + obj-y += cmd640-core.o +endif + +obj-$(CONFIG_BLK_DEV_IDE) += cris/ ppc/ +obj-$(CONFIG_BLK_DEV_IDEPNP) += ide-pnp.o +obj-$(CONFIG_IDE_H8300) += h8300/ obj-$(CONFIG_IDE_GENERIC) += ide-generic.o obj-$(CONFIG_BLK_DEV_IDEDISK) += ide-disk.o @@ -49,6 +45,20 @@ obj-$(CONFIG_BLK_DEV_IDECD) += ide-cd.o obj-$(CONFIG_BLK_DEV_IDETAPE) += ide-tape.o obj-$(CONFIG_BLK_DEV_IDEFLOPPY) += ide-floppy.o -obj-$(CONFIG_BLK_DEV_IDE) += legacy/ arm/ mips/ -obj-$(CONFIG_BLK_DEV_HD) += legacy/ -obj-$(CONFIG_ETRAX_IDE) += cris/ +ifeq ($(CONFIG_BLK_DEV_IDECS), y) + ide-cs-core-y += legacy/ide-cs.o + obj-y += ide-cs-core.o +endif + +ifeq ($(CONFIG_BLK_DEV_PLATFORM), y) + ide-platform-core-y += legacy/ide_platform.o + obj-y += ide-platform-core.o +endif + +obj-$(CONFIG_BLK_DEV_IDE) += arm/ mips/ + +# old hd driver must be last +ifeq ($(CONFIG_BLK_DEV_HD), y) + hd-core-y += legacy/hd.o + obj-y += hd-core.o +endif diff --git a/drivers/ide/arm/Makefile b/drivers/ide/arm/Makefile index 6a78f0755f26..5f63ad216862 100644 --- a/drivers/ide/arm/Makefile +++ b/drivers/ide/arm/Makefile @@ -3,4 +3,8 @@ obj-$(CONFIG_BLK_DEV_IDE_ICSIDE) += icside.o obj-$(CONFIG_BLK_DEV_IDE_RAPIDE) += rapide.o obj-$(CONFIG_BLK_DEV_IDE_BAST) += bast-ide.o +ifeq ($(CONFIG_IDE_ARM), m) + obj-m += ide_arm.o +endif + EXTRA_CFLAGS := -Idrivers/ide diff --git a/drivers/ide/arm/ide_arm.c b/drivers/ide/arm/ide_arm.c index a1b5ddab6a48..60f2497542c0 100644 --- a/drivers/ide/arm/ide_arm.c +++ b/drivers/ide/arm/ide_arm.c @@ -24,7 +24,7 @@ # define IDE_ARM_IRQ IRQ_HARDDISK #endif -void __init ide_arm_init(void) +static int __init ide_arm_init(void) { ide_hwif_t *hwif; hw_regs_t hw; @@ -41,4 +41,8 @@ void __init ide_arm_init(void) ide_device_add(idx); } + + return 0; } + +module_init(ide_arm_init); diff --git a/drivers/ide/cris/Makefile b/drivers/ide/cris/Makefile index 6176e8d6b2e6..20b95960531f 100644 --- a/drivers/ide/cris/Makefile +++ b/drivers/ide/cris/Makefile @@ -1,3 +1,3 @@ EXTRA_CFLAGS += -Idrivers/ide -obj-y += ide-cris.o +obj-$(CONFIG_IDE_ETRAX) += ide-cris.o diff --git a/drivers/ide/cris/ide-cris.c b/drivers/ide/cris/ide-cris.c index 924536297036..8c3294c4d23e 100644 --- a/drivers/ide/cris/ide-cris.c +++ b/drivers/ide/cris/ide-cris.c @@ -754,8 +754,7 @@ static void cris_set_dma_mode(ide_drive_t *drive, const u8 speed) cris_ide_set_speed(TYPE_DMA, 0, strobe, hold); } -void __init -init_e100_ide (void) +static int __init init_e100_ide(void) { hw_regs_t hw; int ide_offsets[IDE_NR_PORTS], h, i; @@ -823,6 +822,8 @@ init_e100_ide (void) cris_ide_set_speed(TYPE_UDMA, ATA_UDMA2_CYC, ATA_UDMA2_DVS, 0); ide_device_add(idx); + + return 0; } static cris_dma_descr_type mydescr __attribute__ ((__aligned__(16))); @@ -1056,3 +1057,5 @@ static void cris_dma_start(ide_drive_t *drive) LED_DISK_READ(1); } } + +module_init(init_e100_ide); diff --git a/drivers/ide/h8300/Makefile b/drivers/ide/h8300/Makefile new file mode 100644 index 000000000000..5eba16f423f4 --- /dev/null +++ b/drivers/ide/h8300/Makefile @@ -0,0 +1,2 @@ + +obj-$(CONFIG_IDE_H8300) += ide-h8300.o diff --git a/drivers/ide/h8300/ide-h8300.c b/drivers/ide/h8300/ide-h8300.c index 9fa78e98d1ba..4f6d0191cf6c 100644 --- a/drivers/ide/h8300/ide-h8300.c +++ b/drivers/ide/h8300/ide-h8300.c @@ -84,7 +84,7 @@ static inline void hwif_setup(ide_hwif_t *hwif) hwif->INSL = NULL; } -void __init h8300_ide_init(void) +static int __init h8300_ide_init(void) { hw_regs_t hw; ide_hwif_t *hwif; @@ -104,7 +104,7 @@ void __init h8300_ide_init(void) hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]); if (hwif == NULL) { printk(KERN_ERR "ide-h8300: IDE I/F register failed\n"); - return; + return -ENOENT; } index = hwif->index; @@ -117,8 +117,12 @@ void __init h8300_ide_init(void) ide_device_add(idx); - return; + return 0; out_busy: printk(KERN_ERR "ide-h8300: IDE I/F resource already used.\n"); + + return -EBUSY; } + +module_init(h8300_ide_init); diff --git a/drivers/ide/ide-pnp.c b/drivers/ide/ide-pnp.c index 802efd4d9760..cbbb0f75be92 100644 --- a/drivers/ide/ide-pnp.c +++ b/drivers/ide/ide-pnp.c @@ -75,12 +75,15 @@ static struct pnp_driver idepnp_driver = { .remove = idepnp_remove, }; -void __init pnpide_init(void) +static int __init pnpide_init(void) { - pnp_register_driver(&idepnp_driver); + return pnp_register_driver(&idepnp_driver); } -void __exit pnpide_exit(void) +static void __exit pnpide_exit(void) { pnp_unregister_driver(&idepnp_driver); } + +module_init(pnpide_init); +module_exit(pnpide_exit); diff --git a/drivers/ide/ide-scan-pci.c b/drivers/ide/ide-scan-pci.c new file mode 100644 index 000000000000..23015d89e735 --- /dev/null +++ b/drivers/ide/ide-scan-pci.c @@ -0,0 +1,11 @@ +#include +#include +#include +#include + +static int __init ide_scan_pci(void) +{ + return ide_scan_pcibus(); +} + +module_init(ide_scan_pci); diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c index 6f99f5c90062..5f3e53ec5838 100644 --- a/drivers/ide/ide.c +++ b/drivers/ide/ide.c @@ -95,7 +95,7 @@ DEFINE_MUTEX(ide_cfg_mtx); __cacheline_aligned_in_smp DEFINE_SPINLOCK(ide_lock); #ifdef CONFIG_IDEPCI_PCIBUS_ORDER -static int ide_scan_direction; /* THIS was formerly 2.2.x pci=reverse */ +int ide_scan_direction; /* THIS was formerly 2.2.x pci=reverse */ #endif int noautodma = 0; @@ -178,8 +178,6 @@ static void init_hwif_default(ide_hwif_t *hwif, unsigned int index) #endif } -extern void ide_arm_init(void); - /* * init_ide_data() sets reasonable default values into all fields * of all instances of the hwifs and drives, but only on the first call. @@ -1223,26 +1221,12 @@ static int __init match_parm (char *s, const char *keywords[], int vals[], int m return 0; /* zero = nothing matched */ } -#ifdef CONFIG_BLK_DEV_ALI14XX extern int probe_ali14xx; -extern int ali14xx_init(void); -#endif -#ifdef CONFIG_BLK_DEV_UMC8672 extern int probe_umc8672; -extern int umc8672_init(void); -#endif -#ifdef CONFIG_BLK_DEV_DTC2278 extern int probe_dtc2278; -extern int dtc2278_init(void); -#endif -#ifdef CONFIG_BLK_DEV_HT6560B extern int probe_ht6560b; -extern int ht6560b_init(void); -#endif -#ifdef CONFIG_BLK_DEV_QD65XX extern int probe_qd65xx; -extern int qd65xx_init(void); -#endif +extern int cmd640_vlb; static int __initdata is_chipset_set[MAX_HWIFS]; @@ -1458,11 +1442,8 @@ static int __init ide_setup(char *s) #endif #ifdef CONFIG_BLK_DEV_CMD640 case -14: /* "cmd640_vlb" */ - { - extern int cmd640_vlb; /* flag for cmd640.c */ cmd640_vlb = 1; goto done; - } #endif #ifdef CONFIG_BLK_DEV_HT6560B case -13: /* "ht6560b" */ @@ -1552,83 +1533,6 @@ done: return 1; } -extern void __init pnpide_init(void); -extern void __exit pnpide_exit(void); -extern void __init h8300_ide_init(void); -extern void __init mpc8xx_ide_probe(void); - -/* - * probe_for_hwifs() finds/initializes "known" IDE interfaces - */ -static void __init probe_for_hwifs (void) -{ -#ifdef CONFIG_IDEPCI_PCIBUS_ORDER - ide_scan_pcibus(ide_scan_direction); -#endif - -#ifdef CONFIG_ETRAX_IDE - { - extern void init_e100_ide(void); - init_e100_ide(); - } -#endif /* CONFIG_ETRAX_IDE */ -#ifdef CONFIG_BLK_DEV_CMD640 - { - extern void ide_probe_for_cmd640x(void); - ide_probe_for_cmd640x(); - } -#endif /* CONFIG_BLK_DEV_CMD640 */ -#ifdef CONFIG_BLK_DEV_IDE_PMAC - { - extern int pmac_ide_probe(void); - (void)pmac_ide_probe(); - } -#endif /* CONFIG_BLK_DEV_IDE_PMAC */ -#ifdef CONFIG_BLK_DEV_GAYLE - { - extern void gayle_init(void); - gayle_init(); - } -#endif /* CONFIG_BLK_DEV_GAYLE */ -#ifdef CONFIG_BLK_DEV_FALCON_IDE - { - extern void falconide_init(void); - falconide_init(); - } -#endif /* CONFIG_BLK_DEV_FALCON_IDE */ -#ifdef CONFIG_BLK_DEV_MAC_IDE - { - extern void macide_init(void); - macide_init(); - } -#endif /* CONFIG_BLK_DEV_MAC_IDE */ -#ifdef CONFIG_BLK_DEV_Q40IDE - { - extern void q40ide_init(void); - q40ide_init(); - } -#endif /* CONFIG_BLK_DEV_Q40IDE */ -#ifdef CONFIG_BLK_DEV_BUDDHA - { - extern void buddha_init(void); - buddha_init(); - } -#endif /* CONFIG_BLK_DEV_BUDDHA */ -#ifdef CONFIG_BLK_DEV_IDEPNP - pnpide_init(); -#endif -#ifdef CONFIG_H8300 - h8300_ide_init(); -#endif -#ifdef BLK_DEV_MPC8xx_IDE - mpc8xx_ide_probe(); -#endif -} - -/* - * Probe module - */ - EXPORT_SYMBOL(ide_lock); static int ide_bus_match(struct device *dev, struct device_driver *drv) @@ -1775,33 +1679,6 @@ static int __init ide_init(void) proc_ide_create(); -#ifdef CONFIG_IDE_ARM - ide_arm_init(); -#endif -#ifdef CONFIG_BLK_DEV_ALI14XX - if (probe_ali14xx) - (void)ali14xx_init(); -#endif -#ifdef CONFIG_BLK_DEV_UMC8672 - if (probe_umc8672) - (void)umc8672_init(); -#endif -#ifdef CONFIG_BLK_DEV_DTC2278 - if (probe_dtc2278) - (void)dtc2278_init(); -#endif -#ifdef CONFIG_BLK_DEV_HT6560B - if (probe_ht6560b) - (void)ht6560b_init(); -#endif -#ifdef CONFIG_BLK_DEV_QD65XX - if (probe_qd65xx) - (void)qd65xx_init(); -#endif - - /* Probe for special PCI and other "known" interface chipsets. */ - probe_for_hwifs(); - return 0; } @@ -1837,10 +1714,6 @@ void __exit cleanup_module (void) for (index = 0; index < MAX_HWIFS; ++index) ide_unregister(index); -#ifdef CONFIG_BLK_DEV_IDEPNP - pnpide_exit(); -#endif - proc_ide_destroy(); bus_unregister(&ide_bus_type); diff --git a/drivers/ide/legacy/Makefile b/drivers/ide/legacy/Makefile index 409822349f10..7043ec7d1e05 100644 --- a/drivers/ide/legacy/Makefile +++ b/drivers/ide/legacy/Makefile @@ -1,15 +1,24 @@ +# link order is important here + obj-$(CONFIG_BLK_DEV_ALI14XX) += ali14xx.o +obj-$(CONFIG_BLK_DEV_UMC8672) += umc8672.o obj-$(CONFIG_BLK_DEV_DTC2278) += dtc2278.o obj-$(CONFIG_BLK_DEV_HT6560B) += ht6560b.o obj-$(CONFIG_BLK_DEV_QD65XX) += qd65xx.o -obj-$(CONFIG_BLK_DEV_UMC8672) += umc8672.o -obj-$(CONFIG_BLK_DEV_IDECS) += ide-cs.o +obj-$(CONFIG_BLK_DEV_GAYLE) += gayle.o +obj-$(CONFIG_BLK_DEV_FALCON_IDE) += falconide.o +obj-$(CONFIG_BLK_DEV_MAC_IDE) += macide.o +obj-$(CONFIG_BLK_DEV_Q40IDE) += q40ide.o +obj-$(CONFIG_BLK_DEV_BUDDHA) += buddha.o -obj-$(CONFIG_BLK_DEV_PLATFORM) += ide_platform.o +ifeq ($(CONFIG_BLK_DEV_IDECS), m) + obj-m += ide-cs.o +endif -# Last of all -obj-$(CONFIG_BLK_DEV_HD) += hd.o +ifeq ($(CONFIG_BLK_DEV_PLATFORM), m) + obj-m += ide_platform.o +endif EXTRA_CFLAGS := -Idrivers/ide diff --git a/drivers/ide/legacy/ali14xx.c b/drivers/ide/legacy/ali14xx.c index 38c3a6d63f30..5ec0be4cbad7 100644 --- a/drivers/ide/legacy/ali14xx.c +++ b/drivers/ide/legacy/ali14xx.c @@ -231,8 +231,7 @@ int probe_ali14xx = 0; module_param_named(probe, probe_ali14xx, bool, 0); MODULE_PARM_DESC(probe, "probe for ALI M14xx chipsets"); -/* Can be called directly from ide.c. */ -int __init ali14xx_init(void) +static int __init ali14xx_init(void) { if (probe_ali14xx == 0) goto out; @@ -248,9 +247,7 @@ out: return -ENODEV; } -#ifdef MODULE module_init(ali14xx_init); -#endif MODULE_AUTHOR("see local file"); MODULE_DESCRIPTION("support of ALI 14XX IDE chipsets"); diff --git a/drivers/ide/legacy/buddha.c b/drivers/ide/legacy/buddha.c index ba64c4b9f918..e97766aef374 100644 --- a/drivers/ide/legacy/buddha.c +++ b/drivers/ide/legacy/buddha.c @@ -143,7 +143,7 @@ static int xsurf_ack_intr(ide_hwif_t *hwif) * Probe for a Buddha or Catweasel IDE interface */ -void __init buddha_init(void) +static int __init buddha_init(void) { hw_regs_t hw; ide_hwif_t *hwif; @@ -243,4 +243,8 @@ fail_base2: ide_device_add(idx); } + + return 0; } + +module_init(buddha_init); diff --git a/drivers/ide/legacy/dtc2278.c b/drivers/ide/legacy/dtc2278.c index 24a845d45bd2..13eee6da2806 100644 --- a/drivers/ide/legacy/dtc2278.c +++ b/drivers/ide/legacy/dtc2278.c @@ -150,8 +150,7 @@ int probe_dtc2278 = 0; module_param_named(probe, probe_dtc2278, bool, 0); MODULE_PARM_DESC(probe, "probe for DTC2278xx chipsets"); -/* Can be called directly from ide.c. */ -int __init dtc2278_init(void) +static int __init dtc2278_init(void) { if (probe_dtc2278 == 0) return -ENODEV; @@ -163,9 +162,7 @@ int __init dtc2278_init(void) return 0; } -#ifdef MODULE module_init(dtc2278_init); -#endif MODULE_AUTHOR("See Local File"); MODULE_DESCRIPTION("support of DTC-2278 VLB IDE chipsets"); diff --git a/drivers/ide/legacy/falconide.c b/drivers/ide/legacy/falconide.c index c1a84540beb1..dec2ef99c778 100644 --- a/drivers/ide/legacy/falconide.c +++ b/drivers/ide/legacy/falconide.c @@ -62,7 +62,7 @@ EXPORT_SYMBOL(falconide_intr_lock); * Probe for a Falcon IDE interface */ -void __init falconide_init(void) +static int __init falconide_init(void) { if (MACH_IS_ATARI && ATARIHW_PRESENT(IDE)) { hw_regs_t hw; @@ -84,4 +84,9 @@ void __init falconide_init(void) ide_device_add(idx); } + } + + return 0; } + +module_init(falconide_init); diff --git a/drivers/ide/legacy/gayle.c b/drivers/ide/legacy/gayle.c index ec53dc9b483c..e21ef75c9055 100644 --- a/drivers/ide/legacy/gayle.c +++ b/drivers/ide/legacy/gayle.c @@ -110,13 +110,13 @@ static int gayle_ack_intr_a1200(ide_hwif_t *hwif) * Probe for a Gayle IDE interface (and optionally for an IDE doubler) */ -void __init gayle_init(void) +static int __init gayle_init(void) { int a4000, i; u8 idx[4] = { 0xff, 0xff, 0xff, 0xff }; if (!MACH_IS_AMIGA) - return; + return -ENODEV; if ((a4000 = AMIGAHW_PRESENT(A4000_IDE)) || AMIGAHW_PRESENT(A1200_IDE)) goto found; @@ -126,7 +126,7 @@ void __init gayle_init(void) NULL)) goto found; #endif - return; + return -ENODEV; found: for (i = 0; i < GAYLE_NUM_PROBE_HWIFS; i++) { @@ -191,4 +191,8 @@ found: } ide_device_add(idx); + + return 0; } + +module_init(gayle_init); diff --git a/drivers/ide/legacy/ht6560b.c b/drivers/ide/legacy/ht6560b.c index a4245d13f11b..8da5031a6d05 100644 --- a/drivers/ide/legacy/ht6560b.c +++ b/drivers/ide/legacy/ht6560b.c @@ -307,8 +307,7 @@ int probe_ht6560b = 0; module_param_named(probe, probe_ht6560b, bool, 0); MODULE_PARM_DESC(probe, "probe for HT6560B chipset"); -/* Can be called directly from ide.c. */ -int __init ht6560b_init(void) +static int __init ht6560b_init(void) { ide_hwif_t *hwif, *mate; static u8 idx[4] = { 0, 1, 0xff, 0xff }; @@ -369,9 +368,7 @@ release_region: return -ENODEV; } -#ifdef MODULE module_init(ht6560b_init); -#endif MODULE_AUTHOR("See Local File"); MODULE_DESCRIPTION("HT-6560B EIDE-controller support"); diff --git a/drivers/ide/legacy/macide.c b/drivers/ide/legacy/macide.c index c1b7881c280a..6b3e960350aa 100644 --- a/drivers/ide/legacy/macide.c +++ b/drivers/ide/legacy/macide.c @@ -81,7 +81,7 @@ int macide_ack_intr(ide_hwif_t* hwif) * Probe for a Macintosh IDE interface */ -void __init macide_init(void) +static int __init macide_init(void) { hw_regs_t hw; ide_hwif_t *hwif; @@ -106,7 +106,7 @@ void __init macide_init(void) IRQ_BABOON_1); break; default: - return; + return -ENODEV; } hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]); @@ -139,4 +139,8 @@ void __init macide_init(void) ide_device_add(idx); } + + return 0; } + +module_init(macide_init); diff --git a/drivers/ide/legacy/q40ide.c b/drivers/ide/legacy/q40ide.c index 2082e9c6efd9..0154c91ee4b1 100644 --- a/drivers/ide/legacy/q40ide.c +++ b/drivers/ide/legacy/q40ide.c @@ -111,7 +111,7 @@ static const char *q40_ide_names[Q40IDE_NUM_HWIFS]={ * Probe for Q40 IDE interfaces */ -void __init q40ide_init(void) +static int __init q40ide_init(void) { int i; ide_hwif_t *hwif; @@ -119,7 +119,7 @@ void __init q40ide_init(void) u8 idx[4] = { 0xff, 0xff, 0xff, 0xff }; if (!MACH_IS_Q40) - return ; + return -ENODEV; for (i = 0; i < Q40IDE_NUM_HWIFS; i++) { hw_regs_t hw; @@ -153,5 +153,8 @@ void __init q40ide_init(void) } ide_device_add(idx); + + return 0; } +module_init(q40ide_init); diff --git a/drivers/ide/legacy/qd65xx.c b/drivers/ide/legacy/qd65xx.c index 912e73853faa..2bac4c1a6532 100644 --- a/drivers/ide/legacy/qd65xx.c +++ b/drivers/ide/legacy/qd65xx.c @@ -478,8 +478,7 @@ int probe_qd65xx = 0; module_param_named(probe, probe_qd65xx, bool, 0); MODULE_PARM_DESC(probe, "probe for QD65xx chipsets"); -/* Can be called directly from ide.c. */ -int __init qd65xx_init(void) +static int __init qd65xx_init(void) { if (probe_qd65xx == 0) return -ENODEV; @@ -492,9 +491,7 @@ int __init qd65xx_init(void) return 0; } -#ifdef MODULE module_init(qd65xx_init); -#endif MODULE_AUTHOR("Samuel Thibault"); MODULE_DESCRIPTION("support of qd65xx vlb ide chipset"); diff --git a/drivers/ide/legacy/umc8672.c b/drivers/ide/legacy/umc8672.c index 79577b916874..a1ae1ae6699d 100644 --- a/drivers/ide/legacy/umc8672.c +++ b/drivers/ide/legacy/umc8672.c @@ -169,8 +169,7 @@ int probe_umc8672 = 0; module_param_named(probe, probe_umc8672, bool, 0); MODULE_PARM_DESC(probe, "probe for UMC8672 chipset"); -/* Can be called directly from ide.c. */ -int __init umc8672_init(void) +static int __init umc8672_init(void) { if (probe_umc8672 == 0) goto out; @@ -181,9 +180,7 @@ out: return -ENODEV;; } -#ifdef MODULE module_init(umc8672_init); -#endif MODULE_AUTHOR("Wolfram Podien"); MODULE_DESCRIPTION("Support for UMC 8672 IDE chipset"); diff --git a/drivers/ide/pci/Makefile b/drivers/ide/pci/Makefile index 95d1ea8f1f14..94803253e8af 100644 --- a/drivers/ide/pci/Makefile +++ b/drivers/ide/pci/Makefile @@ -36,4 +36,8 @@ obj-$(CONFIG_BLK_DEV_VIA82CXXX) += via82cxxx.o # Must appear at the end of the block obj-$(CONFIG_BLK_DEV_GENERIC) += generic.o +ifeq ($(CONFIG_BLK_DEV_CMD640), m) + obj-m += cmd640.o +endif + EXTRA_CFLAGS := -Idrivers/ide diff --git a/drivers/ide/pci/cmd640.c b/drivers/ide/pci/cmd640.c index 5096e059ac5a..da3565e0071f 100644 --- a/drivers/ide/pci/cmd640.c +++ b/drivers/ide/pci/cmd640.c @@ -706,9 +706,9 @@ static int pci_conf2(void) } /* - * Probe for a cmd640 chipset, and initialize it if found. Called from ide.c + * Probe for a cmd640 chipset, and initialize it if found. */ -int __init ide_probe_for_cmd640x (void) +static int __init cmd640x_init(void) { #ifdef CONFIG_BLK_DEV_CMD640_ENHANCED int second_port_toggled = 0; @@ -883,3 +883,7 @@ int __init ide_probe_for_cmd640x (void) return 1; } +module_param_named(probe_vlb, cmd640_vlb, bool, 0); +MODULE_PARM_DESC(probe_vlb, "probe for VLB version of CMD640 chipset"); + +module_init(cmd640x_init); diff --git a/drivers/ide/ppc/Makefile b/drivers/ide/ppc/Makefile new file mode 100644 index 000000000000..65af5848b28c --- /dev/null +++ b/drivers/ide/ppc/Makefile @@ -0,0 +1,3 @@ + +obj-$(CONFIG_BLK_DEV_IDE_PMAC) += pmac.o +obj-$(CONFIG_BLK_DEV_MPC8xx_IDE) += mpc8xx.o diff --git a/drivers/ide/ppc/mpc8xx.c b/drivers/ide/ppc/mpc8xx.c index 8172e813b034..3fd5d45b5e0e 100644 --- a/drivers/ide/ppc/mpc8xx.c +++ b/drivers/ide/ppc/mpc8xx.c @@ -839,7 +839,7 @@ void m8xx_ide_init(void) ppc_ide_md.ide_init_hwif = m8xx_ide_init_hwif_ports; } -void __init mpc8xx_ide_probe(void) +static int __init mpc8xx_ide_probe(void) { u8 idx[4] = { 0xff, 0xff, 0xff, 0xff }; @@ -851,4 +851,8 @@ void __init mpc8xx_ide_probe(void) #endif ide_device_add(idx); + + return 0; } + +module_init(mpc8xx_ide_probe); diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c index 36e4b9570746..cd514743df96 100644 --- a/drivers/ide/ppc/pmac.c +++ b/drivers/ide/ppc/pmac.c @@ -1786,3 +1786,5 @@ pmac_ide_setup_dma(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif) } #endif /* CONFIG_BLK_DEV_IDEDMA_PMAC */ + +module_init(pmac_ide_probe); diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c index d89f84d41b08..63ef8aaa7b90 100644 --- a/drivers/ide/setup-pci.c +++ b/drivers/ide/setup-pci.c @@ -766,21 +766,20 @@ static int __init ide_scan_pcidev(struct pci_dev *dev) /** * ide_scan_pcibus - perform the initial IDE driver scan - * @scan_direction: set for reverse order scanning * * Perform the initial bus rather than driver ordered scan of the * PCI drivers. After this all IDE pci handling becomes standard * module ordering not traditionally ordered. */ - -void __init ide_scan_pcibus (int scan_direction) + +int __init ide_scan_pcibus(void) { struct pci_dev *dev = NULL; struct pci_driver *d; struct list_head *l, *n; pre_init = 0; - if (!scan_direction) + if (!ide_scan_direction) while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev))) ide_scan_pcidev(dev); else @@ -801,5 +800,7 @@ void __init ide_scan_pcibus (int scan_direction) printk(KERN_ERR "%s: failed to register %s driver\n", __FUNCTION__, d->driver.mod_name); } + + return 0; } #endif diff --git a/include/linux/ide.h b/include/linux/ide.h index 9c037a0f2af2..735737500f8f 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -1014,7 +1014,8 @@ extern void do_ide_request(struct request_queue *); void ide_init_disk(struct gendisk *, ide_drive_t *); #ifdef CONFIG_IDEPCI_PCIBUS_ORDER -extern void ide_scan_pcibus(int scan_direction) __init; +extern int ide_scan_direction; +int __init ide_scan_pcibus(void); extern int __ide_pci_register_driver(struct pci_driver *driver, struct module *owner, const char *mod_name); #define ide_pci_register_driver(d) __ide_pci_register_driver(d, THIS_MODULE, KBUILD_MODNAME) #else -- cgit v1.2.3 From 9e47be0c97f7357b80e91dc0632e9cce2eb025e0 Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Sat, 26 Jan 2008 20:13:09 +0100 Subject: ide: remove broken disk byte-swapping support Remove broken disk byte-swapping support: - it can cause a data corruption on SMP (or if using PREEMPT on UP) - all data coming from disk are byte-swapped by taskfile_*_data() which results in incorrect identify data being reported by /proc/ide/ and IOCTLs - "hdx=bswap/byteswap" kernel parameter has been broken on m68k host drivers (including Atari/Q40 ones) since 2.5.x days (because of 'hwif' zero-ing) - byte-swapping is limited to PIO transfers (for working with TiVo disks on x86 machines using user-space solutions or dm-byteswap should result in much better performance because DMA can be used) For previous discussions please see: http://www.ussg.iu.edu/hypermail/linux/kernel/0201.0/0768.html http://lkml.org/lkml/2004/2/28/111 [ I have dm-byteswap device mapper target if somebody is interested (patch is for 2.6.4 though but I'll dust it off if needed). ] Acked-by: Sergei Shtylyov Signed-off-by: Bartlomiej Zolnierkiewicz --- Documentation/ide.txt | 4 ---- drivers/ide/ide-disk.c | 1 - drivers/ide/ide-taskfile.c | 32 ++------------------------------ drivers/ide/ide.c | 6 +----- include/linux/ide.h | 1 - 5 files changed, 3 insertions(+), 41 deletions(-) (limited to 'Documentation') diff --git a/Documentation/ide.txt b/Documentation/ide.txt index b29ccb43d6cc..94e2e3b9e77f 100644 --- a/Documentation/ide.txt +++ b/Documentation/ide.txt @@ -244,10 +244,6 @@ Summary of ide driver parameters for kernel command line "hdx=nodma" : disallow DMA - "hdx=swapdata" : when the drive is a disk, byte swap all data - - "hdx=bswap" : same as above.......... - "hdx=scsi" : the return of the ide-scsi flag, this is useful for allowing ide-floppy, ide-tape, and ide-cdrom|writers to use ide-scsi emulation on a device specific option. diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c index c1fb75c387ea..041be43a62cb 100644 --- a/drivers/ide/ide-disk.c +++ b/drivers/ide/ide-disk.c @@ -761,7 +761,6 @@ static void idedisk_add_settings(ide_drive_t *drive) ide_add_setting(drive, "bios_head", SETTING_RW, TYPE_BYTE, 0, 255, 1, 1, &drive->bios_head, NULL); ide_add_setting(drive, "bios_sect", SETTING_RW, TYPE_BYTE, 0, 63, 1, 1, &drive->bios_sect, NULL); ide_add_setting(drive, "address", SETTING_RW, TYPE_BYTE, 0, 2, 1, 1, &drive->addressing, set_lba_addressing); - ide_add_setting(drive, "bswap", SETTING_READ, TYPE_BYTE, 0, 1, 1, 1, &drive->bswap, NULL); ide_add_setting(drive, "multcount", SETTING_RW, TYPE_BYTE, 0, id->max_multsect, 1, 1, &drive->mult_count, set_multcount); ide_add_setting(drive, "nowerr", SETTING_RW, TYPE_BYTE, 0, 1, 1, 1, &drive->nowerr, set_nowerr); ide_add_setting(drive, "lun", SETTING_RW, TYPE_INT, 0, 7, 1, 1, &drive->lun, NULL); diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c index c58edc86ed3e..3ecafab8f54e 100644 --- a/drivers/ide/ide-taskfile.c +++ b/drivers/ide/ide-taskfile.c @@ -35,34 +35,6 @@ #include #include -static void ata_bswap_data (void *buffer, int wcount) -{ - u16 *p = buffer; - - while (wcount--) { - *p = *p << 8 | *p >> 8; p++; - *p = *p << 8 | *p >> 8; p++; - } -} - -static void taskfile_input_data(ide_drive_t *drive, void *buffer, u32 wcount) -{ - HWIF(drive)->ata_input_data(drive, buffer, wcount); - if (drive->bswap) - ata_bswap_data(buffer, wcount); -} - -static void taskfile_output_data(ide_drive_t *drive, void *buffer, u32 wcount) -{ - if (drive->bswap) { - ata_bswap_data(buffer, wcount); - HWIF(drive)->ata_output_data(drive, buffer, wcount); - ata_bswap_data(buffer, wcount); - } else { - HWIF(drive)->ata_output_data(drive, buffer, wcount); - } -} - void ide_tf_load(ide_drive_t *drive, ide_task_t *task) { ide_hwif_t *hwif = drive->hwif; @@ -352,9 +324,9 @@ static void ide_pio_sector(ide_drive_t *drive, unsigned int write) /* do the actual data transfer */ if (write) - taskfile_output_data(drive, buf, SECTOR_WORDS); + hwif->ata_output_data(drive, buf, SECTOR_WORDS); else - taskfile_input_data(drive, buf, SECTOR_WORDS); + hwif->ata_input_data(drive, buf, SECTOR_WORDS); kunmap_atomic(buf, KM_BIO_SRC_IRQ); #ifdef CONFIG_HIGHMEM diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c index 5f3e53ec5838..52115ef1f01a 100644 --- a/drivers/ide/ide.c +++ b/drivers/ide/ide.c @@ -1303,7 +1303,7 @@ static int __init ide_setup(char *s) if (s[0] == 'h' && s[1] == 'd' && s[2] >= 'a' && s[2] <= max_drive) { const char *hd_words[] = { "none", "noprobe", "nowerr", "cdrom", "nodma", - "autotune", "noautotune", "minus8", "swapdata", "bswap", + "autotune", "noautotune", "-8", "-9", "-10", "noflush", "remap", "remap63", "scsi", NULL }; unit = s[2] - 'a'; hw = unit / MAX_DRIVES; @@ -1339,10 +1339,6 @@ static int __init ide_setup(char *s) case -7: /* "noautotune" */ drive->autotune = IDE_TUNE_NOAUTO; goto obsolete_option; - case -9: /* "swapdata" */ - case -10: /* "bswap" */ - drive->bswap = 1; - goto done; case -11: /* noflush */ drive->noflush = 1; goto done; diff --git a/include/linux/ide.h b/include/linux/ide.h index 74f1ef9c6d94..f94cf036bd95 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -394,7 +394,6 @@ typedef struct ide_drive_s { u8 state; /* retry state */ u8 waiting_for_dma; /* dma currently in progress */ u8 unmask; /* okay to unmask other irqs */ - u8 bswap; /* byte swap data */ u8 noflush; /* don't attempt flushes */ u8 dsc_overlap; /* DSC overlap */ u8 nice1; /* give potential excess bandwidth */ -- cgit v1.2.3 From 5864ae03ca982fb60bedeebfd67562db37c1ee6a Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Sun, 27 Jan 2008 18:14:45 +0100 Subject: i2c: Add support for the PCF8575 chip Signed-off-by: Bart Van Assche Signed-off-by: Jean Delvare --- Documentation/i2c/chips/pcf8575 | 72 ++++++++++++++ drivers/i2c/chips/Kconfig | 16 ++- drivers/i2c/chips/Makefile | 1 + drivers/i2c/chips/pcf8575.c | 214 ++++++++++++++++++++++++++++++++++++++++ 4 files changed, 302 insertions(+), 1 deletion(-) create mode 100644 Documentation/i2c/chips/pcf8575 create mode 100644 drivers/i2c/chips/pcf8575.c (limited to 'Documentation') diff --git a/Documentation/i2c/chips/pcf8575 b/Documentation/i2c/chips/pcf8575 new file mode 100644 index 000000000000..25f5698a61cf --- /dev/null +++ b/Documentation/i2c/chips/pcf8575 @@ -0,0 +1,72 @@ +About the PCF8575 chip and the pcf8575 kernel driver +==================================================== + +The PCF8575 chip is produced by the following manufacturers: + + * Philips NXP + http://www.nxp.com/#/pip/cb=[type=product,path=50807/41735/41850,final=PCF8575_3]|pip=[pip=PCF8575_3][0] + + * Texas Instruments + http://focus.ti.com/docs/prod/folders/print/pcf8575.html + + +Some vendors sell small PCB's with the PCF8575 mounted on it. You can connect +such a board to a Linux host via e.g. an USB to I2C interface. Examples of +PCB boards with a PCF8575: + + * SFE Breakout Board for PCF8575 I2C Expander by RobotShop + http://www.robotshop.ca/home/products/robot-parts/electronics/adapters-converters/sfe-pcf8575-i2c-expander-board.html + + * Breakout Board for PCF8575 I2C Expander by Spark Fun Electronics + http://www.sparkfun.com/commerce/product_info.php?products_id=8130 + + +Description +----------- +The PCF8575 chip is a 16-bit I/O expander for the I2C bus. Up to eight of +these chips can be connected to the same I2C bus. You can find this +chip on some custom designed hardware, but you won't find it on PC +motherboards. + +The PCF8575 chip consists of a 16-bit quasi-bidirectional port and an I2C-bus +interface. Each of the sixteen I/O's can be independently used as an input or +an output. To set up an I/O pin as an input, you have to write a 1 to the +corresponding output. + +For more information please see the datasheet. + + +Detection +--------- + +There is no method known to detect whether a chip on a given I2C address is +a PCF8575 or whether it is any other I2C device. So there are two alternatives +to let the driver find the installed PCF8575 devices: +- Load this driver after any other I2C driver for I2C devices with addresses + in the range 0x20 .. 0x27. +- Pass the I2C bus and address of the installed PCF8575 devices explicitly to + the driver at load time via the probe=... or force=... parameters. + +/sys interface +-------------- + +For each address on which a PCF8575 chip was found or forced the following +files will be created under /sys: +* /sys/bus/i2c/devices/-
/read +* /sys/bus/i2c/devices/-
/write +where bus is the I2C bus number (0, 1, ...) and address is the four-digit +hexadecimal representation of the 7-bit I2C address of the PCF8575 +(0020 .. 0027). + +The read file is read-only. Reading it will trigger an I2C read and will hence +report the current input state for the pins configured as inputs, and the +current output value for the pins configured as outputs. + +The write file is read-write. Writing a value to it will configure all pins +as output for which the corresponding bit is zero. Reading the write file will +return the value last written, or -EAGAIN if no value has yet been written to +the write file. + +On module initialization the configuration of the chip is not changed -- the +chip is left in the state it was already configured in through either power-up +or through previous I2C write actions. diff --git a/drivers/i2c/chips/Kconfig b/drivers/i2c/chips/Kconfig index 2e1c24f671cf..17702b32eab0 100644 --- a/drivers/i2c/chips/Kconfig +++ b/drivers/i2c/chips/Kconfig @@ -57,7 +57,7 @@ config SENSORS_PCF8574 default n help If you say yes here you get support for Philips PCF8574 and - PCF8574A chips. + PCF8574A chips. These chips are 8-bit I/O expanders for the I2C bus. This driver can also be built as a module. If so, the module will be called pcf8574. @@ -65,6 +65,20 @@ config SENSORS_PCF8574 These devices are hard to detect and rarely found on mainstream hardware. If unsure, say N. +config PCF8575 + tristate "Philips PCF8575" + default n + help + If you say yes here you get support for Philips PCF8575 chip. + This chip is a 16-bit I/O expander for the I2C bus. Several other + chip manufacturers sell equivalent chips, e.g. Texas Instruments. + + This driver can also be built as a module. If so, the module + will be called pcf8575. + + This device is hard to detect and is rarely found on mainstream + hardware. If unsure, say N. + config SENSORS_PCA9539 tristate "Philips PCA9539 16-bit I/O port" depends on EXPERIMENTAL diff --git a/drivers/i2c/chips/Makefile b/drivers/i2c/chips/Makefile index ca924e105959..50d4734b5038 100644 --- a/drivers/i2c/chips/Makefile +++ b/drivers/i2c/chips/Makefile @@ -10,6 +10,7 @@ obj-$(CONFIG_SENSORS_MAX6875) += max6875.o obj-$(CONFIG_SENSORS_M41T00) += m41t00.o obj-$(CONFIG_SENSORS_PCA9539) += pca9539.o obj-$(CONFIG_SENSORS_PCF8574) += pcf8574.o +obj-$(CONFIG_PCF8575) += pcf8575.o obj-$(CONFIG_SENSORS_PCF8591) += pcf8591.o obj-$(CONFIG_ISP1301_OMAP) += isp1301_omap.o obj-$(CONFIG_TPS65010) += tps65010.o diff --git a/drivers/i2c/chips/pcf8575.c b/drivers/i2c/chips/pcf8575.c new file mode 100644 index 000000000000..db3c87726978 --- /dev/null +++ b/drivers/i2c/chips/pcf8575.c @@ -0,0 +1,214 @@ +/* + pcf8575.c + + About the PCF8575 chip: the PCF8575 is a 16-bit I/O expander for the I2C bus + produced by a.o. Philips Semiconductors. + + Copyright (C) 2006 Michael Hennerich, Analog Devices Inc. + + Based on pcf8574.c. + + Copyright (c) 2007 Bart Van Assche . + Ported this driver from ucLinux to the mainstream Linux kernel. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. +*/ + +#include +#include +#include +#include /* kzalloc() */ +#include /* sysfs_create_group() */ + +/* Addresses to scan */ +static unsigned short normal_i2c[] = { + 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, + I2C_CLIENT_END +}; + +/* Insmod parameters */ +I2C_CLIENT_INSMOD; + + +/* Each client has this additional data */ +struct pcf8575_data { + struct i2c_client client; + int write; /* last written value, or error code */ +}; + +static int pcf8575_attach_adapter(struct i2c_adapter *adapter); +static int pcf8575_detect(struct i2c_adapter *adapter, int address, int kind); +static int pcf8575_detach_client(struct i2c_client *client); + +/* This is the driver that will be inserted */ +static struct i2c_driver pcf8575_driver = { + .driver = { + .owner = THIS_MODULE, + .name = "pcf8575", + }, + .attach_adapter = pcf8575_attach_adapter, + .detach_client = pcf8575_detach_client, +}; + +/* following are the sysfs callback functions */ +static ssize_t show_read(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct i2c_client *client = to_i2c_client(dev); + u16 val; + u8 iopin_state[2]; + + i2c_master_recv(client, iopin_state, 2); + + val = iopin_state[0]; + val |= iopin_state[1] << 8; + + return sprintf(buf, "%u\n", val); +} + +static DEVICE_ATTR(read, S_IRUGO, show_read, NULL); + +static ssize_t show_write(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct pcf8575_data *data = dev_get_drvdata(dev); + if (data->write < 0) + return data->write; + return sprintf(buf, "%d\n", data->write); +} + +static ssize_t set_write(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct i2c_client *client = to_i2c_client(dev); + struct pcf8575_data *data = i2c_get_clientdata(client); + unsigned long val = simple_strtoul(buf, NULL, 10); + u8 iopin_state[2]; + + if (val > 0xffff) + return -EINVAL; + + data->write = val; + + iopin_state[0] = val & 0xFF; + iopin_state[1] = val >> 8; + + i2c_master_send(client, iopin_state, 2); + + return count; +} + +static DEVICE_ATTR(write, S_IWUSR | S_IRUGO, show_write, set_write); + +static struct attribute *pcf8575_attributes[] = { + &dev_attr_read.attr, + &dev_attr_write.attr, + NULL +}; + +static const struct attribute_group pcf8575_attr_group = { + .attrs = pcf8575_attributes, +}; + +/* + * Real code + */ + +static int pcf8575_attach_adapter(struct i2c_adapter *adapter) +{ + return i2c_probe(adapter, &addr_data, pcf8575_detect); +} + +/* This function is called by i2c_probe */ +static int pcf8575_detect(struct i2c_adapter *adapter, int address, int kind) +{ + struct i2c_client *client; + struct pcf8575_data *data; + int err = 0; + + if (!i2c_check_functionality(adapter, I2C_FUNC_I2C)) + goto exit; + + /* OK. For now, we presume we have a valid client. We now create the + client structure, even though we cannot fill it completely yet. */ + data = kzalloc(sizeof(struct pcf8575_data), GFP_KERNEL); + if (!data) { + err = -ENOMEM; + goto exit; + } + + client = &data->client; + i2c_set_clientdata(client, data); + client->addr = address; + client->adapter = adapter; + client->driver = &pcf8575_driver; + strlcpy(client->name, "pcf8575", I2C_NAME_SIZE); + data->write = -EAGAIN; + + /* This is the place to detect whether the chip at the specified + address really is a PCF8575 chip. However, there is no method known + to detect whether an I2C chip is a PCF8575 or any other I2C chip. */ + + /* Tell the I2C layer a new client has arrived */ + err = i2c_attach_client(client); + if (err) + goto exit_free; + + /* Register sysfs hooks */ + err = sysfs_create_group(&client->dev.kobj, &pcf8575_attr_group); + if (err) + goto exit_detach; + + return 0; + +exit_detach: + i2c_detach_client(client); +exit_free: + kfree(data); +exit: + return err; +} + +static int pcf8575_detach_client(struct i2c_client *client) +{ + int err; + + sysfs_remove_group(&client->dev.kobj, &pcf8575_attr_group); + + err = i2c_detach_client(client); + if (err) + return err; + + kfree(i2c_get_clientdata(client)); + return 0; +} + +static int __init pcf8575_init(void) +{ + return i2c_add_driver(&pcf8575_driver); +} + +static void __exit pcf8575_exit(void) +{ + i2c_del_driver(&pcf8575_driver); +} + +MODULE_AUTHOR("Michael Hennerich , " + "Bart Van Assche "); +MODULE_DESCRIPTION("pcf8575 driver"); +MODULE_LICENSE("GPL"); + +module_init(pcf8575_init); +module_exit(pcf8575_exit); -- cgit v1.2.3 From b3af547e197fa3ca648d148dd8d36befe989e5a0 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sun, 27 Jan 2008 18:14:45 +0100 Subject: i2c-stub: Mention the existence of an helper script There's a new script named i2c-stub-from-dump that can be very helpful when working with the i2c-stub driver. Signed-off-by: Jean Delvare --- Documentation/i2c/i2c-stub | 3 +++ 1 file changed, 3 insertions(+) (limited to 'Documentation') diff --git a/Documentation/i2c/i2c-stub b/Documentation/i2c/i2c-stub index 89e69ad3436c..41889c091a37 100644 --- a/Documentation/i2c/i2c-stub +++ b/Documentation/i2c/i2c-stub @@ -25,6 +25,9 @@ The typical use-case is like this: 3. load the target sensors chip driver module 4. observe its behavior in the kernel log +There's a script named i2c-stub-from-dump in the i2c-tools package which +can load register values automatically from a chip dump. + PARAMETERS: int chip_addr[10]: -- cgit v1.2.3 From 569be443e3c1329fc6725988004f5d8a32fe3be5 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sun, 27 Jan 2008 18:14:45 +0100 Subject: i2c-stub: Use a single array for byte and word operations This mimics the behavior of actual SMBus chips better. Signed-off-by: Jean Delvare Cc: Mark M. Hoffman --- Documentation/i2c/i2c-stub | 3 --- drivers/i2c/busses/i2c-stub.c | 15 ++++++++------- 2 files changed, 8 insertions(+), 10 deletions(-) (limited to 'Documentation') diff --git a/Documentation/i2c/i2c-stub b/Documentation/i2c/i2c-stub index 41889c091a37..0d8be1c20c16 100644 --- a/Documentation/i2c/i2c-stub +++ b/Documentation/i2c/i2c-stub @@ -35,9 +35,6 @@ int chip_addr[10]: CAVEATS: -There are independent arrays for byte/data and word/data commands. Depending -on if/how a target driver mixes them, you'll need to be careful. - If your target driver polls some byte or word waiting for it to change, the stub could lock it up. Use i2cset to unlock it. diff --git a/drivers/i2c/busses/i2c-stub.c b/drivers/i2c/busses/i2c-stub.c index 84df29da1ddc..c2a9f8c94f5e 100644 --- a/drivers/i2c/busses/i2c-stub.c +++ b/drivers/i2c/busses/i2c-stub.c @@ -1,8 +1,8 @@ /* - i2c-stub.c - Part of lm_sensors, Linux kernel modules for hardware - monitoring + i2c-stub.c - I2C/SMBus chip emulator Copyright (c) 2004 Mark M. Hoffman + Copyright (C) 2007 Jean Delvare This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by @@ -37,8 +37,8 @@ MODULE_PARM_DESC(chip_addr, struct stub_chip { u8 pointer; - u8 bytes[256]; - u16 words[256]; + u16 words[256]; /* Byte operations use the LSB as per SMBus + specification */ }; static struct stub_chip *stub_chips; @@ -75,7 +75,7 @@ static s32 stub_xfer(struct i2c_adapter * adap, u16 addr, unsigned short flags, "wrote 0x%02x.\n", addr, command); } else { - data->byte = chip->bytes[chip->pointer++]; + data->byte = chip->words[chip->pointer++] & 0xff; dev_dbg(&adap->dev, "smbus byte - addr 0x%02x, " "read 0x%02x.\n", addr, data->byte); @@ -86,12 +86,13 @@ static s32 stub_xfer(struct i2c_adapter * adap, u16 addr, unsigned short flags, case I2C_SMBUS_BYTE_DATA: if (read_write == I2C_SMBUS_WRITE) { - chip->bytes[command] = data->byte; + chip->words[command] &= 0xff00; + chip->words[command] |= data->byte; dev_dbg(&adap->dev, "smbus byte data - addr 0x%02x, " "wrote 0x%02x at 0x%02x.\n", addr, data->byte, command); } else { - data->byte = chip->bytes[command]; + data->byte = chip->words[command] & 0xff; dev_dbg(&adap->dev, "smbus byte data - addr 0x%02x, " "read 0x%02x at 0x%02x.\n", addr, data->byte, command); -- cgit v1.2.3 From eee87d3196c9a7ac3422f4298e2250ca68d791c1 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Sun, 27 Jan 2008 18:14:45 +0100 Subject: i2c: the scheduled I2C RTC driver removal This patch contains the scheduled removal of legacy I2C RTC drivers with replacement drivers. Signed-off-by: Adrian Bunk Signed-off-by: Jean Delvare --- Documentation/feature-removal-schedule.txt | 7 - arch/ppc/platforms/83xx/mpc834x_sys.c | 20 -- arch/ppc/platforms/85xx/tqm85xx.c | 21 -- arch/ppc/platforms/katana.c | 21 -- drivers/i2c/chips/Kconfig | 38 --- drivers/i2c/chips/Makefile | 3 - drivers/i2c/chips/ds1337.c | 410 ---------------------------- drivers/i2c/chips/ds1374.c | 267 ------------------- drivers/i2c/chips/m41t00.c | 413 ----------------------------- include/linux/i2c-id.h | 2 - include/linux/m41t00.h | 50 ---- 11 files changed, 1252 deletions(-) delete mode 100644 drivers/i2c/chips/ds1337.c delete mode 100644 drivers/i2c/chips/ds1374.c delete mode 100644 drivers/i2c/chips/m41t00.c delete mode 100644 include/linux/m41t00.h (limited to 'Documentation') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 9b8291f4c211..2cd16deb7195 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -266,13 +266,6 @@ Who: Tejun Heo --------------------------- -What: Legacy RTC drivers (under drivers/i2c/chips) -When: November 2007 -Why: Obsolete. We have a RTC subsystem with better drivers. -Who: Jean Delvare - ---------------------------- - What: iptables SAME target When: 1.1. 2008 Files: net/ipv4/netfilter/ipt_SAME.c, include/linux/netfilter_ipv4/ipt_SAME.h diff --git a/arch/ppc/platforms/83xx/mpc834x_sys.c b/arch/ppc/platforms/83xx/mpc834x_sys.c index b84f8df325c4..cb0a7493ff64 100644 --- a/arch/ppc/platforms/83xx/mpc834x_sys.c +++ b/arch/ppc/platforms/83xx/mpc834x_sys.c @@ -224,26 +224,6 @@ mpc834x_sys_init_IRQ(void) ipic_set_default_priority(); } -#if defined(CONFIG_I2C_MPC) && defined(CONFIG_SENSORS_DS1374) -extern ulong ds1374_get_rtc_time(void); -extern int ds1374_set_rtc_time(ulong); - -static int __init -mpc834x_rtc_hookup(void) -{ - struct timespec tv; - - ppc_md.get_rtc_time = ds1374_get_rtc_time; - ppc_md.set_rtc_time = ds1374_set_rtc_time; - - tv.tv_nsec = 0; - tv.tv_sec = (ppc_md.get_rtc_time)(); - do_settimeofday(&tv); - - return 0; -} -late_initcall(mpc834x_rtc_hookup); -#endif static __inline__ void mpc834x_sys_set_bat(void) { diff --git a/arch/ppc/platforms/85xx/tqm85xx.c b/arch/ppc/platforms/85xx/tqm85xx.c index 4ee2bd156dc5..27ce389c1224 100644 --- a/arch/ppc/platforms/85xx/tqm85xx.c +++ b/arch/ppc/platforms/85xx/tqm85xx.c @@ -258,27 +258,6 @@ int tqm85xx_show_cpuinfo(struct seq_file *m) return 0; } -#if defined(CONFIG_I2C) && defined(CONFIG_SENSORS_DS1337) -extern ulong ds1337_get_rtc_time(void); -extern int ds1337_set_rtc_time(unsigned long nowtime); - -static int __init -tqm85xx_rtc_hookup(void) -{ - struct timespec tv; - - ppc_md.set_rtc_time = ds1337_set_rtc_time; - ppc_md.get_rtc_time = ds1337_get_rtc_time; - - tv.tv_nsec = 0; - tv.tv_sec = (ppc_md.get_rtc_time)(); - do_settimeofday(&tv); - - return 0; -} -late_initcall(tqm85xx_rtc_hookup); -#endif - #ifdef CONFIG_PCI /* * interrupt routing diff --git a/arch/ppc/platforms/katana.c b/arch/ppc/platforms/katana.c index 52f63e6f0856..fe6e88cdb1cd 100644 --- a/arch/ppc/platforms/katana.c +++ b/arch/ppc/platforms/katana.c @@ -838,27 +838,6 @@ katana_find_end_of_memory(void) return bdp->bi_memsize; } -#if defined(CONFIG_I2C_MV64XXX) && defined(CONFIG_SENSORS_M41T00) -extern ulong m41t00_get_rtc_time(void); -extern int m41t00_set_rtc_time(ulong); - -static int __init -katana_rtc_hookup(void) -{ - struct timespec tv; - - ppc_md.get_rtc_time = m41t00_get_rtc_time; - ppc_md.set_rtc_time = m41t00_set_rtc_time; - - tv.tv_nsec = 0; - tv.tv_sec = (ppc_md.get_rtc_time)(); - do_settimeofday(&tv); - - return 0; -} -late_initcall(katana_rtc_hookup); -#endif - #if defined(CONFIG_SERIAL_TEXT_DEBUG) && defined(CONFIG_SERIAL_MPSC_CONSOLE) static void __init katana_map_io(void) diff --git a/drivers/i2c/chips/Kconfig b/drivers/i2c/chips/Kconfig index 17702b32eab0..fe3008e59467 100644 --- a/drivers/i2c/chips/Kconfig +++ b/drivers/i2c/chips/Kconfig @@ -4,32 +4,6 @@ menu "Miscellaneous I2C Chip support" -config SENSORS_DS1337 - tristate "Dallas DS1337 and DS1339 Real Time Clock (DEPRECATED)" - depends on EXPERIMENTAL - help - If you say yes here you get support for Dallas Semiconductor - DS1337 and DS1339 real-time clock chips. - - This driver can also be built as a module. If so, the module - will be called ds1337. - - This driver is deprecated and will be dropped soon. Use - rtc-ds1307 instead. - -config SENSORS_DS1374 - tristate "Dallas DS1374 Real Time Clock (DEPRECATED)" - depends on EXPERIMENTAL - help - If you say yes here you get support for Dallas Semiconductor - DS1374 real-time clock chips. - - This driver can also be built as a module. If so, the module - will be called ds1374. - - This driver is deprecated and will be dropped soon. Use - rtc-ds1374 instead. - config DS1682 tristate "Dallas DS1682 Total Elapsed Time Recorder with Alarm" depends on EXPERIMENTAL @@ -130,18 +104,6 @@ config TPS65010 This driver can also be built as a module. If so, the module will be called tps65010. -config SENSORS_M41T00 - tristate "ST M41T00 RTC chip (DEPRECATED)" - depends on PPC32 - help - If you say yes here you get support for the ST M41T00 RTC chip. - - This driver can also be built as a module. If so, the module - will be called m41t00. - - This driver is deprecated and will be dropped soon. Use - rtc-ds1307 or rtc-m41t80 instead. - config SENSORS_MAX6875 tristate "Maxim MAX6875 Power supply supervisor" depends on EXPERIMENTAL diff --git a/drivers/i2c/chips/Makefile b/drivers/i2c/chips/Makefile index 50d4734b5038..501f00cea782 100644 --- a/drivers/i2c/chips/Makefile +++ b/drivers/i2c/chips/Makefile @@ -2,12 +2,9 @@ # Makefile for miscellaneous I2C chip drivers. # -obj-$(CONFIG_SENSORS_DS1337) += ds1337.o -obj-$(CONFIG_SENSORS_DS1374) += ds1374.o obj-$(CONFIG_DS1682) += ds1682.o obj-$(CONFIG_SENSORS_EEPROM) += eeprom.o obj-$(CONFIG_SENSORS_MAX6875) += max6875.o -obj-$(CONFIG_SENSORS_M41T00) += m41t00.o obj-$(CONFIG_SENSORS_PCA9539) += pca9539.o obj-$(CONFIG_SENSORS_PCF8574) += pcf8574.o obj-$(CONFIG_PCF8575) += pcf8575.o diff --git a/drivers/i2c/chips/ds1337.c b/drivers/i2c/chips/ds1337.c deleted file mode 100644 index ec17d6b684a2..000000000000 --- a/drivers/i2c/chips/ds1337.c +++ /dev/null @@ -1,410 +0,0 @@ -/* - * linux/drivers/i2c/chips/ds1337.c - * - * Copyright (C) 2005 James Chapman - * - * based on linux/drivers/acorn/char/pcf8583.c - * Copyright (C) 2000 Russell King - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - * - * Driver for Dallas Semiconductor DS1337 and DS1339 real time clock chip - */ - -#include -#include -#include -#include -#include -#include /* get the user-level API */ -#include -#include - -/* Device registers */ -#define DS1337_REG_HOUR 2 -#define DS1337_REG_DAY 3 -#define DS1337_REG_DATE 4 -#define DS1337_REG_MONTH 5 -#define DS1337_REG_CONTROL 14 -#define DS1337_REG_STATUS 15 - -/* FIXME - how do we export these interface constants? */ -#define DS1337_GET_DATE 0 -#define DS1337_SET_DATE 1 - -/* - * Functions declaration - */ -static unsigned short normal_i2c[] = { 0x68, I2C_CLIENT_END }; - -I2C_CLIENT_INSMOD_1(ds1337); - -static int ds1337_attach_adapter(struct i2c_adapter *adapter); -static int ds1337_detect(struct i2c_adapter *adapter, int address, int kind); -static void ds1337_init_client(struct i2c_client *client); -static int ds1337_detach_client(struct i2c_client *client); -static int ds1337_command(struct i2c_client *client, unsigned int cmd, - void *arg); - -/* - * Driver data (common to all clients) - */ -static struct i2c_driver ds1337_driver = { - .driver = { - .name = "ds1337", - }, - .attach_adapter = ds1337_attach_adapter, - .detach_client = ds1337_detach_client, - .command = ds1337_command, -}; - -/* - * Client data (each client gets its own) - */ -struct ds1337_data { - struct i2c_client client; - struct list_head list; -}; - -/* - * Internal variables - */ -static LIST_HEAD(ds1337_clients); - -static inline int ds1337_read(struct i2c_client *client, u8 reg, u8 *value) -{ - s32 tmp = i2c_smbus_read_byte_data(client, reg); - - if (tmp < 0) - return -EIO; - - *value = tmp; - - return 0; -} - -/* - * Chip access functions - */ -static int ds1337_get_datetime(struct i2c_client *client, struct rtc_time *dt) -{ - int result; - u8 buf[7]; - u8 val; - struct i2c_msg msg[2]; - u8 offs = 0; - - if (!dt) { - dev_dbg(&client->dev, "%s: EINVAL: dt=NULL\n", __FUNCTION__); - return -EINVAL; - } - - msg[0].addr = client->addr; - msg[0].flags = 0; - msg[0].len = 1; - msg[0].buf = &offs; - - msg[1].addr = client->addr; - msg[1].flags = I2C_M_RD; - msg[1].len = sizeof(buf); - msg[1].buf = &buf[0]; - - result = i2c_transfer(client->adapter, msg, 2); - - dev_dbg(&client->dev, "%s: [%d] %02x %02x %02x %02x %02x %02x %02x\n", - __FUNCTION__, result, buf[0], buf[1], buf[2], buf[3], - buf[4], buf[5], buf[6]); - - if (result == 2) { - dt->tm_sec = BCD2BIN(buf[0]); - dt->tm_min = BCD2BIN(buf[1]); - val = buf[2] & 0x3f; - dt->tm_hour = BCD2BIN(val); - dt->tm_wday = BCD2BIN(buf[3]) - 1; - dt->tm_mday = BCD2BIN(buf[4]); - val = buf[5] & 0x7f; - dt->tm_mon = BCD2BIN(val) - 1; - dt->tm_year = BCD2BIN(buf[6]); - if (buf[5] & 0x80) - dt->tm_year += 100; - - dev_dbg(&client->dev, "%s: secs=%d, mins=%d, " - "hours=%d, mday=%d, mon=%d, year=%d, wday=%d\n", - __FUNCTION__, dt->tm_sec, dt->tm_min, - dt->tm_hour, dt->tm_mday, - dt->tm_mon, dt->tm_year, dt->tm_wday); - - return 0; - } - - dev_err(&client->dev, "error reading data! %d\n", result); - return -EIO; -} - -static int ds1337_set_datetime(struct i2c_client *client, struct rtc_time *dt) -{ - int result; - u8 buf[8]; - u8 val; - struct i2c_msg msg[1]; - - if (!dt) { - dev_dbg(&client->dev, "%s: EINVAL: dt=NULL\n", __FUNCTION__); - return -EINVAL; - } - - dev_dbg(&client->dev, "%s: secs=%d, mins=%d, hours=%d, " - "mday=%d, mon=%d, year=%d, wday=%d\n", __FUNCTION__, - dt->tm_sec, dt->tm_min, dt->tm_hour, - dt->tm_mday, dt->tm_mon, dt->tm_year, dt->tm_wday); - - buf[0] = 0; /* reg offset */ - buf[1] = BIN2BCD(dt->tm_sec); - buf[2] = BIN2BCD(dt->tm_min); - buf[3] = BIN2BCD(dt->tm_hour); - buf[4] = BIN2BCD(dt->tm_wday + 1); - buf[5] = BIN2BCD(dt->tm_mday); - buf[6] = BIN2BCD(dt->tm_mon + 1); - val = dt->tm_year; - if (val >= 100) { - val -= 100; - buf[6] |= (1 << 7); - } - buf[7] = BIN2BCD(val); - - msg[0].addr = client->addr; - msg[0].flags = 0; - msg[0].len = sizeof(buf); - msg[0].buf = &buf[0]; - - result = i2c_transfer(client->adapter, msg, 1); - if (result == 1) - return 0; - - dev_err(&client->dev, "error writing data! %d\n", result); - return -EIO; -} - -static int ds1337_command(struct i2c_client *client, unsigned int cmd, - void *arg) -{ - dev_dbg(&client->dev, "%s: cmd=%d\n", __FUNCTION__, cmd); - - switch (cmd) { - case DS1337_GET_DATE: - return ds1337_get_datetime(client, arg); - - case DS1337_SET_DATE: - return ds1337_set_datetime(client, arg); - - default: - return -EINVAL; - } -} - -/* - * Public API for access to specific device. Useful for low-level - * RTC access from kernel code. - */ -int ds1337_do_command(int bus, int cmd, void *arg) -{ - struct list_head *walk; - struct list_head *tmp; - struct ds1337_data *data; - - list_for_each_safe(walk, tmp, &ds1337_clients) { - data = list_entry(walk, struct ds1337_data, list); - if (data->client.adapter->nr == bus) - return ds1337_command(&data->client, cmd, arg); - } - - return -ENODEV; -} - -static int ds1337_attach_adapter(struct i2c_adapter *adapter) -{ - return i2c_probe(adapter, &addr_data, ds1337_detect); -} - -/* - * The following function does more than just detection. If detection - * succeeds, it also registers the new chip. - */ -static int ds1337_detect(struct i2c_adapter *adapter, int address, int kind) -{ - struct i2c_client *new_client; - struct ds1337_data *data; - int err = 0; - const char *name = ""; - - if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA | - I2C_FUNC_I2C)) - goto exit; - - if (!(data = kzalloc(sizeof(struct ds1337_data), GFP_KERNEL))) { - err = -ENOMEM; - goto exit; - } - INIT_LIST_HEAD(&data->list); - - /* The common I2C client data is placed right before the - * DS1337-specific data. - */ - new_client = &data->client; - i2c_set_clientdata(new_client, data); - new_client->addr = address; - new_client->adapter = adapter; - new_client->driver = &ds1337_driver; - new_client->flags = 0; - - /* - * Now we do the remaining detection. A negative kind means that - * the driver was loaded with no force parameter (default), so we - * must both detect and identify the chip. A zero kind means that - * the driver was loaded with the force parameter, the detection - * step shall be skipped. A positive kind means that the driver - * was loaded with the force parameter and a given kind of chip is - * requested, so both the detection and the identification steps - * are skipped. - * - * For detection, we read registers that are most likely to cause - * detection failure, i.e. those that have more bits with fixed - * or reserved values. - */ - - /* Default to an DS1337 if forced */ - if (kind == 0) - kind = ds1337; - - if (kind < 0) { /* detection and identification */ - u8 data; - - /* Check that status register bits 6-2 are zero */ - if ((ds1337_read(new_client, DS1337_REG_STATUS, &data) < 0) || - (data & 0x7c)) - goto exit_free; - - /* Check for a valid day register value */ - if ((ds1337_read(new_client, DS1337_REG_DAY, &data) < 0) || - (data == 0) || (data & 0xf8)) - goto exit_free; - - /* Check for a valid date register value */ - if ((ds1337_read(new_client, DS1337_REG_DATE, &data) < 0) || - (data == 0) || (data & 0xc0) || ((data & 0x0f) > 9) || - (data >= 0x32)) - goto exit_free; - - /* Check for a valid month register value */ - if ((ds1337_read(new_client, DS1337_REG_MONTH, &data) < 0) || - (data == 0) || (data & 0x60) || ((data & 0x0f) > 9) || - ((data >= 0x13) && (data <= 0x19))) - goto exit_free; - - /* Check that control register bits 6-5 are zero */ - if ((ds1337_read(new_client, DS1337_REG_CONTROL, &data) < 0) || - (data & 0x60)) - goto exit_free; - - kind = ds1337; - } - - if (kind == ds1337) - name = "ds1337"; - - /* We can fill in the remaining client fields */ - strlcpy(new_client->name, name, I2C_NAME_SIZE); - - /* Tell the I2C layer a new client has arrived */ - if ((err = i2c_attach_client(new_client))) - goto exit_free; - - /* Initialize the DS1337 chip */ - ds1337_init_client(new_client); - - /* Add client to local list */ - list_add(&data->list, &ds1337_clients); - - return 0; - -exit_free: - kfree(data); -exit: - return err; -} - -static void ds1337_init_client(struct i2c_client *client) -{ - u8 status, control; - - /* On some boards, the RTC isn't configured by boot firmware. - * Handle that case by starting/configuring the RTC now. - */ - status = i2c_smbus_read_byte_data(client, DS1337_REG_STATUS); - control = i2c_smbus_read_byte_data(client, DS1337_REG_CONTROL); - - if ((status & 0x80) || (control & 0x80)) { - /* RTC not running */ - u8 buf[1+16]; /* First byte is interpreted as address */ - struct i2c_msg msg[1]; - - dev_dbg(&client->dev, "%s: RTC not running!\n", __FUNCTION__); - - /* Initialize all, including STATUS and CONTROL to zero */ - memset(buf, 0, sizeof(buf)); - - /* Write valid values in the date/time registers */ - buf[1+DS1337_REG_DAY] = 1; - buf[1+DS1337_REG_DATE] = 1; - buf[1+DS1337_REG_MONTH] = 1; - - msg[0].addr = client->addr; - msg[0].flags = 0; - msg[0].len = sizeof(buf); - msg[0].buf = &buf[0]; - - i2c_transfer(client->adapter, msg, 1); - } else { - /* Running: ensure that device is set in 24-hour mode */ - s32 val; - - val = i2c_smbus_read_byte_data(client, DS1337_REG_HOUR); - if ((val >= 0) && (val & (1 << 6))) - i2c_smbus_write_byte_data(client, DS1337_REG_HOUR, - val & 0x3f); - } -} - -static int ds1337_detach_client(struct i2c_client *client) -{ - int err; - struct ds1337_data *data = i2c_get_clientdata(client); - - if ((err = i2c_detach_client(client))) - return err; - - list_del(&data->list); - kfree(data); - return 0; -} - -static int __init ds1337_init(void) -{ - return i2c_add_driver(&ds1337_driver); -} - -static void __exit ds1337_exit(void) -{ - i2c_del_driver(&ds1337_driver); -} - -MODULE_AUTHOR("James Chapman "); -MODULE_DESCRIPTION("DS1337 RTC driver"); -MODULE_LICENSE("GPL"); - -EXPORT_SYMBOL_GPL(ds1337_do_command); - -module_init(ds1337_init); -module_exit(ds1337_exit); diff --git a/drivers/i2c/chips/ds1374.c b/drivers/i2c/chips/ds1374.c deleted file mode 100644 index 8a2ff0c114d9..000000000000 --- a/drivers/i2c/chips/ds1374.c +++ /dev/null @@ -1,267 +0,0 @@ -/* - * drivers/i2c/chips/ds1374.c - * - * I2C client/driver for the Maxim/Dallas DS1374 Real-Time Clock - * - * Author: Randy Vinson - * - * Based on the m41t00.c by Mark Greer - * - * 2005 (c) MontaVista Software, Inc. This file is licensed under - * the terms of the GNU General Public License version 2. This program - * is licensed "as is" without any warranty of any kind, whether express - * or implied. - */ -/* - * This i2c client/driver wedges between the drivers/char/genrtc.c RTC - * interface and the SMBus interface of the i2c subsystem. - * It would be more efficient to use i2c msgs/i2c_transfer directly but, as - * recommened in .../Documentation/i2c/writing-clients section - * "Sending and receiving", using SMBus level communication is preferred. - */ - -#include -#include -#include -#include -#include -#include -#include -#include - -#define DS1374_REG_TOD0 0x00 -#define DS1374_REG_TOD1 0x01 -#define DS1374_REG_TOD2 0x02 -#define DS1374_REG_TOD3 0x03 -#define DS1374_REG_WDALM0 0x04 -#define DS1374_REG_WDALM1 0x05 -#define DS1374_REG_WDALM2 0x06 -#define DS1374_REG_CR 0x07 -#define DS1374_REG_SR 0x08 -#define DS1374_REG_SR_OSF 0x80 -#define DS1374_REG_TCR 0x09 - -#define DS1374_DRV_NAME "ds1374" - -static DEFINE_MUTEX(ds1374_mutex); - -static struct i2c_driver ds1374_driver; -static struct i2c_client *save_client; - -static unsigned short ignore[] = { I2C_CLIENT_END }; -static unsigned short normal_addr[] = { 0x68, I2C_CLIENT_END }; - -static struct i2c_client_address_data addr_data = { - .normal_i2c = normal_addr, - .probe = ignore, - .ignore = ignore, -}; - -static ulong ds1374_read_rtc(void) -{ - ulong time = 0; - int reg = DS1374_REG_WDALM0; - - while (reg--) { - s32 tmp; - if ((tmp = i2c_smbus_read_byte_data(save_client, reg)) < 0) { - dev_warn(&save_client->dev, - "can't read from rtc chip\n"); - return 0; - } - time = (time << 8) | (tmp & 0xff); - } - return time; -} - -static void ds1374_write_rtc(ulong time) -{ - int reg; - - for (reg = DS1374_REG_TOD0; reg < DS1374_REG_WDALM0; reg++) { - if (i2c_smbus_write_byte_data(save_client, reg, time & 0xff) - < 0) { - dev_warn(&save_client->dev, - "can't write to rtc chip\n"); - break; - } - time = time >> 8; - } -} - -static void ds1374_check_rtc_status(void) -{ - s32 tmp; - - tmp = i2c_smbus_read_byte_data(save_client, DS1374_REG_SR); - if (tmp < 0) { - dev_warn(&save_client->dev, - "can't read status from rtc chip\n"); - return; - } - if (tmp & DS1374_REG_SR_OSF) { - dev_warn(&save_client->dev, - "oscillator discontinuity flagged, time unreliable\n"); - tmp &= ~DS1374_REG_SR_OSF; - tmp = i2c_smbus_write_byte_data(save_client, DS1374_REG_SR, - tmp & 0xff); - if (tmp < 0) - dev_warn(&save_client->dev, - "can't clear discontinuity notification\n"); - } -} - -ulong ds1374_get_rtc_time(void) -{ - ulong t1, t2; - int limit = 10; /* arbitrary retry limit */ - - mutex_lock(&ds1374_mutex); - - /* - * Since the reads are being performed one byte at a time using - * the SMBus vs a 4-byte i2c transfer, there is a chance that a - * carry will occur during the read. To detect this, 2 reads are - * performed and compared. - */ - do { - t1 = ds1374_read_rtc(); - t2 = ds1374_read_rtc(); - } while (t1 != t2 && limit--); - - mutex_unlock(&ds1374_mutex); - - if (t1 != t2) { - dev_warn(&save_client->dev, - "can't get consistent time from rtc chip\n"); - t1 = 0; - } - - return t1; -} - -static ulong new_time; - -static void ds1374_set_work(struct work_struct *work) -{ - ulong t1, t2; - int limit = 10; /* arbitrary retry limit */ - - t1 = new_time; - - mutex_lock(&ds1374_mutex); - - /* - * Since the writes are being performed one byte at a time using - * the SMBus vs a 4-byte i2c transfer, there is a chance that a - * carry will occur during the write. To detect this, the write - * value is read back and compared. - */ - do { - ds1374_write_rtc(t1); - t2 = ds1374_read_rtc(); - } while (t1 != t2 && limit--); - - mutex_unlock(&ds1374_mutex); - - if (t1 != t2) - dev_warn(&save_client->dev, - "can't confirm time set from rtc chip\n"); -} - -static struct workqueue_struct *ds1374_workqueue; - -static DECLARE_WORK(ds1374_work, ds1374_set_work); - -int ds1374_set_rtc_time(ulong nowtime) -{ - new_time = nowtime; - - if (in_interrupt()) - queue_work(ds1374_workqueue, &ds1374_work); - else - ds1374_set_work(NULL); - - return 0; -} - -/* - ***************************************************************************** - * - * Driver Interface - * - ***************************************************************************** - */ -static int ds1374_probe(struct i2c_adapter *adap, int addr, int kind) -{ - struct i2c_client *client; - int rc; - - client = kzalloc(sizeof(struct i2c_client), GFP_KERNEL); - if (!client) - return -ENOMEM; - - strncpy(client->name, DS1374_DRV_NAME, I2C_NAME_SIZE); - client->addr = addr; - client->adapter = adap; - client->driver = &ds1374_driver; - - ds1374_workqueue = create_singlethread_workqueue("ds1374"); - if (!ds1374_workqueue) { - kfree(client); - return -ENOMEM; /* most expected reason */ - } - - if ((rc = i2c_attach_client(client)) != 0) { - kfree(client); - return rc; - } - - save_client = client; - - ds1374_check_rtc_status(); - - return 0; -} - -static int ds1374_attach(struct i2c_adapter *adap) -{ - return i2c_probe(adap, &addr_data, ds1374_probe); -} - -static int ds1374_detach(struct i2c_client *client) -{ - int rc; - - if ((rc = i2c_detach_client(client)) == 0) { - kfree(i2c_get_clientdata(client)); - destroy_workqueue(ds1374_workqueue); - } - return rc; -} - -static struct i2c_driver ds1374_driver = { - .driver = { - .name = DS1374_DRV_NAME, - }, - .id = I2C_DRIVERID_DS1374, - .attach_adapter = ds1374_attach, - .detach_client = ds1374_detach, -}; - -static int __init ds1374_init(void) -{ - return i2c_add_driver(&ds1374_driver); -} - -static void __exit ds1374_exit(void) -{ - i2c_del_driver(&ds1374_driver); -} - -module_init(ds1374_init); -module_exit(ds1374_exit); - -MODULE_AUTHOR("Randy Vinson "); -MODULE_DESCRIPTION("Maxim/Dallas DS1374 RTC I2C Client Driver"); -MODULE_LICENSE("GPL"); diff --git a/drivers/i2c/chips/m41t00.c b/drivers/i2c/chips/m41t00.c deleted file mode 100644 index 3fcb646e2073..000000000000 --- a/drivers/i2c/chips/m41t00.c +++ /dev/null @@ -1,413 +0,0 @@ -/* - * I2C client/driver for the ST M41T00 family of i2c rtc chips. - * - * Author: Mark A. Greer - * - * 2005, 2006 (c) MontaVista Software, Inc. This file is licensed under - * the terms of the GNU General Public License version 2. This program - * is licensed "as is" without any warranty of any kind, whether express - * or implied. - */ -/* - * This i2c client/driver wedges between the drivers/char/genrtc.c RTC - * interface and the SMBus interface of the i2c subsystem. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -static struct i2c_driver m41t00_driver; -static struct i2c_client *save_client; - -static unsigned short ignore[] = { I2C_CLIENT_END }; -static unsigned short normal_addr[] = { I2C_CLIENT_END, I2C_CLIENT_END }; - -static struct i2c_client_address_data addr_data = { - .normal_i2c = normal_addr, - .probe = ignore, - .ignore = ignore, -}; - -struct m41t00_chip_info { - u8 type; - char *name; - u8 read_limit; - u8 sec; /* Offsets for chip regs */ - u8 min; - u8 hour; - u8 day; - u8 mon; - u8 year; - u8 alarm_mon; - u8 alarm_hour; - u8 sqw; - u8 sqw_freq; -}; - -static struct m41t00_chip_info m41t00_chip_info_tbl[] = { - { - .type = M41T00_TYPE_M41T00, - .name = "m41t00", - .read_limit = 5, - .sec = 0, - .min = 1, - .hour = 2, - .day = 4, - .mon = 5, - .year = 6, - }, - { - .type = M41T00_TYPE_M41T81, - .name = "m41t81", - .read_limit = 1, - .sec = 1, - .min = 2, - .hour = 3, - .day = 5, - .mon = 6, - .year = 7, - .alarm_mon = 0xa, - .alarm_hour = 0xc, - .sqw = 0x13, - }, - { - .type = M41T00_TYPE_M41T85, - .name = "m41t85", - .read_limit = 1, - .sec = 1, - .min = 2, - .hour = 3, - .day = 5, - .mon = 6, - .year = 7, - .alarm_mon = 0xa, - .alarm_hour = 0xc, - .sqw = 0x13, - }, -}; -static struct m41t00_chip_info *m41t00_chip; - -ulong -m41t00_get_rtc_time(void) -{ - s32 sec, min, hour, day, mon, year; - s32 sec1, min1, hour1, day1, mon1, year1; - u8 reads = 0; - u8 buf[8], msgbuf[1] = { 0 }; /* offset into rtc's regs */ - struct i2c_msg msgs[] = { - { - .addr = save_client->addr, - .flags = 0, - .len = 1, - .buf = msgbuf, - }, - { - .addr = save_client->addr, - .flags = I2C_M_RD, - .len = 8, - .buf = buf, - }, - }; - - sec = min = hour = day = mon = year = 0; - - do { - if (i2c_transfer(save_client->adapter, msgs, 2) < 0) - goto read_err; - - sec1 = sec; - min1 = min; - hour1 = hour; - day1 = day; - mon1 = mon; - year1 = year; - - sec = buf[m41t00_chip->sec] & 0x7f; - min = buf[m41t00_chip->min] & 0x7f; - hour = buf[m41t00_chip->hour] & 0x3f; - day = buf[m41t00_chip->day] & 0x3f; - mon = buf[m41t00_chip->mon] & 0x1f; - year = buf[m41t00_chip->year]; - } while ((++reads < m41t00_chip->read_limit) && ((sec != sec1) - || (min != min1) || (hour != hour1) || (day != day1) - || (mon != mon1) || (year != year1))); - - if ((m41t00_chip->read_limit > 1) && ((sec != sec1) || (min != min1) - || (hour != hour1) || (day != day1) || (mon != mon1) - || (year != year1))) - goto read_err; - - sec = BCD2BIN(sec); - min = BCD2BIN(min); - hour = BCD2BIN(hour); - day = BCD2BIN(day); - mon = BCD2BIN(mon); - year = BCD2BIN(year); - - year += 1900; - if (year < 1970) - year += 100; - - return mktime(year, mon, day, hour, min, sec); - -read_err: - dev_err(&save_client->dev, "m41t00_get_rtc_time: Read error\n"); - return 0; -} -EXPORT_SYMBOL_GPL(m41t00_get_rtc_time); - -static void -m41t00_set(void *arg) -{ - struct rtc_time tm; - int nowtime = *(int *)arg; - s32 sec, min, hour, day, mon, year; - u8 wbuf[9], *buf = &wbuf[1], msgbuf[1] = { 0 }; - struct i2c_msg msgs[] = { - { - .addr = save_client->addr, - .flags = 0, - .len = 1, - .buf = msgbuf, - }, - { - .addr = save_client->addr, - .flags = I2C_M_RD, - .len = 8, - .buf = buf, - }, - }; - - to_tm(nowtime, &tm); - tm.tm_year = (tm.tm_year - 1900) % 100; - - sec = BIN2BCD(tm.tm_sec); - min = BIN2BCD(tm.tm_min); - hour = BIN2BCD(tm.tm_hour); - day = BIN2BCD(tm.tm_mday); - mon = BIN2BCD(tm.tm_mon); - year = BIN2BCD(tm.tm_year); - - /* Read reg values into buf[0..7]/wbuf[1..8] */ - if (i2c_transfer(save_client->adapter, msgs, 2) < 0) { - dev_err(&save_client->dev, "m41t00_set: Read error\n"); - return; - } - - wbuf[0] = 0; /* offset into rtc's regs */ - buf[m41t00_chip->sec] = (buf[m41t00_chip->sec] & ~0x7f) | (sec & 0x7f); - buf[m41t00_chip->min] = (buf[m41t00_chip->min] & ~0x7f) | (min & 0x7f); - buf[m41t00_chip->hour] = (buf[m41t00_chip->hour] & ~0x3f) | (hour& 0x3f); - buf[m41t00_chip->day] = (buf[m41t00_chip->day] & ~0x3f) | (day & 0x3f); - buf[m41t00_chip->mon] = (buf[m41t00_chip->mon] & ~0x1f) | (mon & 0x1f); - buf[m41t00_chip->year] = year; - - if (i2c_master_send(save_client, wbuf, 9) < 0) - dev_err(&save_client->dev, "m41t00_set: Write error\n"); -} - -static ulong new_time; -/* well, isn't this API just _lovely_? */ -static void -m41t00_barf(struct work_struct *unusable) -{ - m41t00_set(&new_time); -} - -static struct workqueue_struct *m41t00_wq; -static DECLARE_WORK(m41t00_work, m41t00_barf); - -int -m41t00_set_rtc_time(ulong nowtime) -{ - new_time = nowtime; - - if (in_interrupt()) - queue_work(m41t00_wq, &m41t00_work); - else - m41t00_set(&new_time); - - return 0; -} -EXPORT_SYMBOL_GPL(m41t00_set_rtc_time); - -/* - ***************************************************************************** - * - * platform_data Driver Interface - * - ***************************************************************************** - */ -static int __init -m41t00_platform_probe(struct platform_device *pdev) -{ - struct m41t00_platform_data *pdata; - int i; - - if (pdev && (pdata = pdev->dev.platform_data)) { - normal_addr[0] = pdata->i2c_addr; - - for (i=0; itype) { - m41t00_chip = &m41t00_chip_info_tbl[i]; - m41t00_chip->sqw_freq = pdata->sqw_freq; - return 0; - } - } - return -ENODEV; -} - -static int __exit -m41t00_platform_remove(struct platform_device *pdev) -{ - return 0; -} - -static struct platform_driver m41t00_platform_driver = { - .probe = m41t00_platform_probe, - .remove = m41t00_platform_remove, - .driver = { - .owner = THIS_MODULE, - .name = M41T00_DRV_NAME, - }, -}; - -/* - ***************************************************************************** - * - * Driver Interface - * - ***************************************************************************** - */ -static int -m41t00_probe(struct i2c_adapter *adap, int addr, int kind) -{ - struct i2c_client *client; - int rc; - - if (!i2c_check_functionality(adap, I2C_FUNC_I2C - | I2C_FUNC_SMBUS_BYTE_DATA)) - return 0; - - client = kzalloc(sizeof(struct i2c_client), GFP_KERNEL); - if (!client) - return -ENOMEM; - - strlcpy(client->name, m41t00_chip->name, I2C_NAME_SIZE); - client->addr = addr; - client->adapter = adap; - client->driver = &m41t00_driver; - - if ((rc = i2c_attach_client(client))) - goto attach_err; - - if (m41t00_chip->type != M41T00_TYPE_M41T00) { - /* If asked, disable SQW, set SQW frequency & re-enable */ - if (m41t00_chip->sqw_freq) - if (((rc = i2c_smbus_read_byte_data(client, - m41t00_chip->alarm_mon)) < 0) - || ((rc = i2c_smbus_write_byte_data(client, - m41t00_chip->alarm_mon, rc & ~0x40)) <0) - || ((rc = i2c_smbus_write_byte_data(client, - m41t00_chip->sqw, - m41t00_chip->sqw_freq)) < 0) - || ((rc = i2c_smbus_write_byte_data(client, - m41t00_chip->alarm_mon, rc | 0x40)) <0)) - goto sqw_err; - - /* Make sure HT (Halt Update) bit is cleared */ - if ((rc = i2c_smbus_read_byte_data(client, - m41t00_chip->alarm_hour)) < 0) - goto ht_err; - - if (rc & 0x40) - if ((rc = i2c_smbus_write_byte_data(client, - m41t00_chip->alarm_hour, rc & ~0x40))<0) - goto ht_err; - } - - /* Make sure ST (stop) bit is cleared */ - if ((rc = i2c_smbus_read_byte_data(client, m41t00_chip->sec)) < 0) - goto st_err; - - if (rc & 0x80) - if ((rc = i2c_smbus_write_byte_data(client, m41t00_chip->sec, - rc & ~0x80)) < 0) - goto st_err; - - m41t00_wq = create_singlethread_workqueue(m41t00_chip->name); - save_client = client; - return 0; - -st_err: - dev_err(&client->dev, "m41t00_probe: Can't clear ST bit\n"); - goto attach_err; -ht_err: - dev_err(&client->dev, "m41t00_probe: Can't clear HT bit\n"); - goto attach_err; -sqw_err: - dev_err(&client->dev, "m41t00_probe: Can't set SQW Frequency\n"); -attach_err: - kfree(client); - return rc; -} - -static int -m41t00_attach(struct i2c_adapter *adap) -{ - return i2c_probe(adap, &addr_data, m41t00_probe); -} - -static int -m41t00_detach(struct i2c_client *client) -{ - int rc; - - if ((rc = i2c_detach_client(client)) == 0) { - kfree(client); - destroy_workqueue(m41t00_wq); - } - return rc; -} - -static struct i2c_driver m41t00_driver = { - .driver = { - .name = M41T00_DRV_NAME, - }, - .id = I2C_DRIVERID_STM41T00, - .attach_adapter = m41t00_attach, - .detach_client = m41t00_detach, -}; - -static int __init -m41t00_init(void) -{ - int rc; - - if (!(rc = platform_driver_register(&m41t00_platform_driver))) - rc = i2c_add_driver(&m41t00_driver); - return rc; -} - -static void __exit -m41t00_exit(void) -{ - i2c_del_driver(&m41t00_driver); - platform_driver_unregister(&m41t00_platform_driver); -} - -module_init(m41t00_init); -module_exit(m41t00_exit); - -MODULE_AUTHOR("Mark A. Greer "); -MODULE_DESCRIPTION("ST Microelectronics M41T00 RTC I2C Client Driver"); -MODULE_LICENSE("GPL"); diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h index c7a51a196f51..10c8edd7795c 100644 --- a/include/linux/i2c-id.h +++ b/include/linux/i2c-id.h @@ -83,7 +83,6 @@ #define I2C_DRIVERID_SAA7114 49 /* video decoder */ #define I2C_DRIVERID_ZR36120 50 /* Zoran 36120 video encoder */ #define I2C_DRIVERID_24LC32A 51 /* Microchip 24LC32A 32k EEPROM */ -#define I2C_DRIVERID_STM41T00 52 /* real time clock */ #define I2C_DRIVERID_UDA1342 53 /* UDA1342 audio codec */ #define I2C_DRIVERID_ADV7170 54 /* video encoder */ #define I2C_DRIVERID_MAX1617 56 /* temp sensor */ @@ -95,7 +94,6 @@ #define I2C_DRIVERID_TDA7313 62 /* TDA7313 audio processor */ #define I2C_DRIVERID_MAX6900 63 /* MAX6900 real-time clock */ #define I2C_DRIVERID_SAA7114H 64 /* video decoder */ -#define I2C_DRIVERID_DS1374 65 /* DS1374 real time clock */ #define I2C_DRIVERID_TDA9874 66 /* TV sound decoder */ #define I2C_DRIVERID_SAA6752HS 67 /* MPEG2 encoder */ #define I2C_DRIVERID_TVEEPROM 68 /* TV EEPROM */ diff --git a/include/linux/m41t00.h b/include/linux/m41t00.h deleted file mode 100644 index b423360ca38e..000000000000 --- a/include/linux/m41t00.h +++ /dev/null @@ -1,50 +0,0 @@ -/* - * Definitions for the ST M41T00 family of i2c rtc chips. - * - * Author: Mark A. Greer - * - * 2005, 2006 (c) MontaVista Software, Inc. This file is licensed under - * the terms of the GNU General Public License version 2. This program - * is licensed "as is" without any warranty of any kind, whether express - * or implied. - */ - -#ifndef _M41T00_H -#define _M41T00_H - -#define M41T00_DRV_NAME "m41t00" -#define M41T00_I2C_ADDR 0x68 - -#define M41T00_TYPE_M41T00 0 -#define M41T00_TYPE_M41T81 81 -#define M41T00_TYPE_M41T85 85 - -struct m41t00_platform_data { - u8 type; - u8 i2c_addr; - u8 sqw_freq; -}; - -/* SQW output disabled, this is default value by power on */ -#define M41T00_SQW_DISABLE (0) - -#define M41T00_SQW_32KHZ (1<<4) /* 32.768 KHz */ -#define M41T00_SQW_8KHZ (2<<4) /* 8.192 KHz */ -#define M41T00_SQW_4KHZ (3<<4) /* 4.096 KHz */ -#define M41T00_SQW_2KHZ (4<<4) /* 2.048 KHz */ -#define M41T00_SQW_1KHZ (5<<4) /* 1.024 KHz */ -#define M41T00_SQW_512HZ (6<<4) /* 512 Hz */ -#define M41T00_SQW_256HZ (7<<4) /* 256 Hz */ -#define M41T00_SQW_128HZ (8<<4) /* 128 Hz */ -#define M41T00_SQW_64HZ (9<<4) /* 64 Hz */ -#define M41T00_SQW_32HZ (10<<4) /* 32 Hz */ -#define M41T00_SQW_16HZ (11<<4) /* 16 Hz */ -#define M41T00_SQW_8HZ (12<<4) /* 8 Hz */ -#define M41T00_SQW_4HZ (13<<4) /* 4 Hz */ -#define M41T00_SQW_2HZ (14<<4) /* 2 Hz */ -#define M41T00_SQW_1HZ (15<<4) /* 1 Hz */ - -extern ulong m41t00_get_rtc_time(void); -extern int m41t00_set_rtc_time(ulong nowtime); - -#endif /* _M41T00_H */ -- cgit v1.2.3 From 7e8b99251be8b6f992baa88e3a6ba3c4ae01660b Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Sun, 27 Jan 2008 18:14:46 +0100 Subject: i2c: some overdue driver removal This patch contains the overdue removal of three I2C drivers. [JD: In fact only i2c-ixp4xx can be removed at the moment, the other two platforms don't implement the generic GPIO layer yet.] Signed-off-by: Adrian Bunk Signed-off-by: Jean Delvare --- Documentation/feature-removal-schedule.txt | 8 -- drivers/i2c/busses/Kconfig | 14 --- drivers/i2c/busses/Makefile | 1 - drivers/i2c/busses/i2c-ixp4xx.c | 178 ----------------------------- include/asm-arm/arch-ixp4xx/platform.h | 11 -- include/linux/i2c-id.h | 1 - 6 files changed, 213 deletions(-) delete mode 100644 drivers/i2c/busses/i2c-ixp4xx.c (limited to 'Documentation') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 2cd16deb7195..fb51d38e8b46 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -225,14 +225,6 @@ Who: Len Brown --------------------------- -What: i2c-ixp2000, i2c-ixp4xx and scx200_i2c drivers -When: September 2007 -Why: Obsolete. The new i2c-gpio driver replaces all hardware-specific - I2C-over-GPIO drivers. -Who: Jean Delvare - ---------------------------- - What: 'time' kernel boot parameter When: January 2008 Why: replaced by 'printk.time=' so that printk timestamps can be diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig index c466c6cfc2e5..2cef92a0bd0e 100644 --- a/drivers/i2c/busses/Kconfig +++ b/drivers/i2c/busses/Kconfig @@ -259,20 +259,6 @@ config I2C_IOP3XX This driver can also be built as a module. If so, the module will be called i2c-iop3xx. -config I2C_IXP4XX - tristate "IXP4xx GPIO-Based I2C Interface (DEPRECATED)" - depends on ARCH_IXP4XX - select I2C_ALGOBIT - help - Say Y here if you have an Intel IXP4xx(420,421,422,425) based - system and are using GPIO lines for an I2C bus. - - This support is also available as a module. If so, the module - will be called i2c-ixp4xx. - - This driver is deprecated and will be dropped soon. Use i2c-gpio - instead. - config I2C_IXP2000 tristate "IXP2000 GPIO-Based I2C Interface (DEPRECATED)" depends on ARCH_IXP2000 diff --git a/drivers/i2c/busses/Makefile b/drivers/i2c/busses/Makefile index 81d43c27cf93..ea7068f1eb6b 100644 --- a/drivers/i2c/busses/Makefile +++ b/drivers/i2c/busses/Makefile @@ -20,7 +20,6 @@ obj-$(CONFIG_I2C_I810) += i2c-i810.o obj-$(CONFIG_I2C_IBM_IIC) += i2c-ibm_iic.o obj-$(CONFIG_I2C_IOP3XX) += i2c-iop3xx.o obj-$(CONFIG_I2C_IXP2000) += i2c-ixp2000.o -obj-$(CONFIG_I2C_IXP4XX) += i2c-ixp4xx.o obj-$(CONFIG_I2C_POWERMAC) += i2c-powermac.o obj-$(CONFIG_I2C_MPC) += i2c-mpc.o obj-$(CONFIG_I2C_MV64XXX) += i2c-mv64xxx.o diff --git a/drivers/i2c/busses/i2c-ixp4xx.c b/drivers/i2c/busses/i2c-ixp4xx.c deleted file mode 100644 index 069ed7f3b395..000000000000 --- a/drivers/i2c/busses/i2c-ixp4xx.c +++ /dev/null @@ -1,178 +0,0 @@ -/* - * drivers/i2c/busses/i2c-ixp4xx.c - * - * Intel's IXP4xx XScale NPU chipsets (IXP420, 421, 422, 425) do not have - * an on board I2C controller but provide 16 GPIO pins that are often - * used to create an I2C bus. This driver provides an i2c_adapter - * interface that plugs in under algo_bit and drives the GPIO pins - * as instructed by the alogorithm driver. - * - * Author: Deepak Saxena - * - * Copyright (c) 2003-2004 MontaVista Software Inc. - * - * This file is licensed under the terms of the GNU General Public - * License version 2. This program is licensed "as is" without any - * warranty of any kind, whether express or implied. - * - * NOTE: Since different platforms will use different GPIO pins for - * I2C, this driver uses an IXP4xx-specific platform_data - * pointer to pass the GPIO numbers to the driver. This - * allows us to support all the different IXP4xx platforms - * w/o having to put #ifdefs in this driver. - * - * See arch/arm/mach-ixp4xx/ixdp425.c for an example of building a - * device list and filling in the ixp4xx_i2c_pins data structure - * that is passed as the platform_data to this driver. - */ - -#include -#include -#include -#include -#include -#include - -#include /* Pick up IXP4xx-specific bits */ - -static inline int ixp4xx_scl_pin(void *data) -{ - return ((struct ixp4xx_i2c_pins*)data)->scl_pin; -} - -static inline int ixp4xx_sda_pin(void *data) -{ - return ((struct ixp4xx_i2c_pins*)data)->sda_pin; -} - -static void ixp4xx_bit_setscl(void *data, int val) -{ - gpio_line_set(ixp4xx_scl_pin(data), 0); - gpio_line_config(ixp4xx_scl_pin(data), - val ? IXP4XX_GPIO_IN : IXP4XX_GPIO_OUT ); -} - -static void ixp4xx_bit_setsda(void *data, int val) -{ - gpio_line_set(ixp4xx_sda_pin(data), 0); - gpio_line_config(ixp4xx_sda_pin(data), - val ? IXP4XX_GPIO_IN : IXP4XX_GPIO_OUT ); -} - -static int ixp4xx_bit_getscl(void *data) -{ - int scl; - - gpio_line_config(ixp4xx_scl_pin(data), IXP4XX_GPIO_IN ); - gpio_line_get(ixp4xx_scl_pin(data), &scl); - - return scl; -} - -static int ixp4xx_bit_getsda(void *data) -{ - int sda; - - gpio_line_config(ixp4xx_sda_pin(data), IXP4XX_GPIO_IN ); - gpio_line_get(ixp4xx_sda_pin(data), &sda); - - return sda; -} - -struct ixp4xx_i2c_data { - struct ixp4xx_i2c_pins *gpio_pins; - struct i2c_adapter adapter; - struct i2c_algo_bit_data algo_data; -}; - -static int ixp4xx_i2c_remove(struct platform_device *plat_dev) -{ - struct ixp4xx_i2c_data *drv_data = platform_get_drvdata(plat_dev); - - platform_set_drvdata(plat_dev, NULL); - - i2c_del_adapter(&drv_data->adapter); - - kfree(drv_data); - - return 0; -} - -static int ixp4xx_i2c_probe(struct platform_device *plat_dev) -{ - int err; - struct ixp4xx_i2c_pins *gpio = plat_dev->dev.platform_data; - struct ixp4xx_i2c_data *drv_data = - kzalloc(sizeof(struct ixp4xx_i2c_data), GFP_KERNEL); - - if(!drv_data) - return -ENOMEM; - - drv_data->gpio_pins = gpio; - - /* - * We could make a lot of these structures static, but - * certain platforms may have multiple GPIO-based I2C - * buses for various device domains, so we need per-device - * algo_data->data. - */ - drv_data->algo_data.data = gpio; - drv_data->algo_data.setsda = ixp4xx_bit_setsda; - drv_data->algo_data.setscl = ixp4xx_bit_setscl; - drv_data->algo_data.getsda = ixp4xx_bit_getsda; - drv_data->algo_data.getscl = ixp4xx_bit_getscl; - drv_data->algo_data.udelay = 10; - drv_data->algo_data.timeout = 100; - - drv_data->adapter.id = I2C_HW_B_IXP4XX; - drv_data->adapter.class = I2C_CLASS_HWMON; - strlcpy(drv_data->adapter.name, plat_dev->dev.driver->name, - sizeof(drv_data->adapter.name)); - drv_data->adapter.algo_data = &drv_data->algo_data; - - drv_data->adapter.dev.parent = &plat_dev->dev; - - gpio_line_config(gpio->scl_pin, IXP4XX_GPIO_IN); - gpio_line_config(gpio->sda_pin, IXP4XX_GPIO_IN); - gpio_line_set(gpio->scl_pin, 0); - gpio_line_set(gpio->sda_pin, 0); - - err = i2c_bit_add_bus(&drv_data->adapter); - if (err) { - printk(KERN_ERR "ERROR: Could not install %s\n", plat_dev->dev.bus_id); - - kfree(drv_data); - return err; - } - - platform_set_drvdata(plat_dev, drv_data); - - return 0; -} - -static struct platform_driver ixp4xx_i2c_driver = { - .probe = ixp4xx_i2c_probe, - .remove = ixp4xx_i2c_remove, - .driver = { - .name = "IXP4XX-I2C", - .owner = THIS_MODULE, - }, -}; - -static int __init ixp4xx_i2c_init(void) -{ - return platform_driver_register(&ixp4xx_i2c_driver); -} - -static void __exit ixp4xx_i2c_exit(void) -{ - platform_driver_unregister(&ixp4xx_i2c_driver); -} - -module_init(ixp4xx_i2c_init); -module_exit(ixp4xx_i2c_exit); - -MODULE_DESCRIPTION("GPIO-based I2C adapter for IXP4xx systems"); -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Deepak Saxena "); - diff --git a/include/asm-arm/arch-ixp4xx/platform.h b/include/asm-arm/arch-ixp4xx/platform.h index 2a44d3d67980..2ce28e3fd325 100644 --- a/include/asm-arm/arch-ixp4xx/platform.h +++ b/include/asm-arm/arch-ixp4xx/platform.h @@ -75,17 +75,6 @@ extern unsigned long ixp4xx_exp_bus_size; #define IXP4XX_PERIPHERAL_BUS_CLOCK (66) /* 66Mhzi APB BUS */ #define IXP4XX_UART_XTAL 14745600 -/* - * The IXP4xx chips do not have an I2C unit, so GPIO lines are just - * used to - * Used as platform_data to provide GPIO pin information to the ixp42x - * I2C driver. - */ -struct ixp4xx_i2c_pins { - unsigned long sda_pin; - unsigned long scl_pin; -}; - /* * This structure provide a means for the board setup code * to give information to th pata_ixp4xx driver. It is diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h index 10c8edd7795c..6bab140a4505 100644 --- a/include/linux/i2c-id.h +++ b/include/linux/i2c-id.h @@ -191,7 +191,6 @@ #define I2C_HW_B_OMAHA 0x010014 /* Omaha I2C interface (ARM) */ #define I2C_HW_B_GUIDE 0x010015 /* Guide bit-basher */ #define I2C_HW_B_IXP2000 0x010016 /* GPIO on IXP2000 systems */ -#define I2C_HW_B_IXP4XX 0x010017 /* GPIO on IXP4XX systems */ #define I2C_HW_B_S3VIA 0x010018 /* S3Via ProSavage adapter */ #define I2C_HW_B_ZR36067 0x010019 /* Zoran-36057/36067 based boards */ #define I2C_HW_B_PCILYNX 0x01001a /* TI PCILynx I2C adapter */ -- cgit v1.2.3 From 2cdddeb8d7dd42dc1847b110228a626d25a1e468 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sun, 27 Jan 2008 18:14:47 +0100 Subject: i2c: normal_i2c can be made const (remaining drivers) Signed-off-by: Jean Delvare --- Documentation/i2c/writing-clients | 6 +++--- arch/arm/mach-pxa/akita-ioexp.c | 2 +- drivers/i2c/chips/eeprom.c | 2 +- drivers/i2c/chips/max6875.c | 2 +- drivers/i2c/chips/pcf8574.c | 8 +++++--- drivers/i2c/chips/pcf8575.c | 2 +- drivers/i2c/chips/pcf8591.c | 2 +- drivers/macintosh/therm_windtunnel.c | 2 +- drivers/usb/host/ohci-pnx4008.c | 4 ++-- drivers/video/matrox/matroxfb_maven.c | 2 +- drivers/w1/masters/ds2482.c | 2 +- sound/soc/codecs/cs4270.c | 2 +- sound/soc/s3c24xx/neo1973_wm8753.c | 2 +- 13 files changed, 20 insertions(+), 18 deletions(-) (limited to 'Documentation') diff --git a/Documentation/i2c/writing-clients b/Documentation/i2c/writing-clients index 2c170032bf37..bfb0a5520817 100644 --- a/Documentation/i2c/writing-clients +++ b/Documentation/i2c/writing-clients @@ -267,9 +267,9 @@ insmod parameter of the form force_. Fortunately, as a module writer, you just have to define the `normal_i2c' parameter. The complete declaration could look like this: - /* Scan 0x37, and 0x48 to 0x4f */ - static unsigned short normal_i2c[] = { 0x37, 0x48, 0x49, 0x4a, 0x4b, 0x4c, - 0x4d, 0x4e, 0x4f, I2C_CLIENT_END }; + /* Scan 0x4c to 0x4f */ + static const unsigned short normal_i2c[] = { 0x4c, 0x4d, 0x4e, 0x4f, + I2C_CLIENT_END }; /* Magic definition of all other variables and things */ I2C_CLIENT_INSMOD; diff --git a/arch/arm/mach-pxa/akita-ioexp.c b/arch/arm/mach-pxa/akita-ioexp.c index 12d2fe0ceff6..254892ac30cd 100644 --- a/arch/arm/mach-pxa/akita-ioexp.c +++ b/arch/arm/mach-pxa/akita-ioexp.c @@ -29,7 +29,7 @@ #define MAX7310_TIMEOUT 0x04 /* Addresses to scan */ -static unsigned short normal_i2c[] = { 0x18, I2C_CLIENT_END }; +static const unsigned short normal_i2c[] = { 0x18, I2C_CLIENT_END }; /* I2C Magic */ I2C_CLIENT_INSMOD; diff --git a/drivers/i2c/chips/eeprom.c b/drivers/i2c/chips/eeprom.c index 1a7eeebac506..fde297b21ad7 100644 --- a/drivers/i2c/chips/eeprom.c +++ b/drivers/i2c/chips/eeprom.c @@ -35,7 +35,7 @@ #include /* Addresses to scan */ -static unsigned short normal_i2c[] = { 0x50, 0x51, 0x52, 0x53, 0x54, +static const unsigned short normal_i2c[] = { 0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, I2C_CLIENT_END }; /* Insmod parameters */ diff --git a/drivers/i2c/chips/max6875.c b/drivers/i2c/chips/max6875.c index 64692f666372..fb7ea5637eca 100644 --- a/drivers/i2c/chips/max6875.c +++ b/drivers/i2c/chips/max6875.c @@ -34,7 +34,7 @@ #include /* Do not scan - the MAX6875 access method will write to some EEPROM chips */ -static unsigned short normal_i2c[] = {I2C_CLIENT_END}; +static const unsigned short normal_i2c[] = { I2C_CLIENT_END }; /* Insmod parameters */ I2C_CLIENT_INSMOD_1(max6875); diff --git a/drivers/i2c/chips/pcf8574.c b/drivers/i2c/chips/pcf8574.c index 21c6dd69193c..b3b830ccf209 100644 --- a/drivers/i2c/chips/pcf8574.c +++ b/drivers/i2c/chips/pcf8574.c @@ -41,9 +41,11 @@ #include /* Addresses to scan */ -static unsigned short normal_i2c[] = { 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, - 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f, - I2C_CLIENT_END }; +static const unsigned short normal_i2c[] = { + 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, + 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f, + I2C_CLIENT_END +}; /* Insmod parameters */ I2C_CLIENT_INSMOD_2(pcf8574, pcf8574a); diff --git a/drivers/i2c/chips/pcf8575.c b/drivers/i2c/chips/pcf8575.c index db3c87726978..3ea08ac0bfa3 100644 --- a/drivers/i2c/chips/pcf8575.c +++ b/drivers/i2c/chips/pcf8575.c @@ -33,7 +33,7 @@ #include /* sysfs_create_group() */ /* Addresses to scan */ -static unsigned short normal_i2c[] = { +static const unsigned short normal_i2c[] = { 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, I2C_CLIENT_END }; diff --git a/drivers/i2c/chips/pcf8591.c b/drivers/i2c/chips/pcf8591.c index 4dc36376eb32..865f4409c06b 100644 --- a/drivers/i2c/chips/pcf8591.c +++ b/drivers/i2c/chips/pcf8591.c @@ -27,7 +27,7 @@ #include /* Addresses to scan */ -static unsigned short normal_i2c[] = { 0x48, 0x49, 0x4a, 0x4b, 0x4c, +static const unsigned short normal_i2c[] = { 0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f, I2C_CLIENT_END }; /* Insmod parameters */ diff --git a/drivers/macintosh/therm_windtunnel.c b/drivers/macintosh/therm_windtunnel.c index 5452da1bb1a5..7c57e7a4e00f 100644 --- a/drivers/macintosh/therm_windtunnel.c +++ b/drivers/macintosh/therm_windtunnel.c @@ -52,7 +52,7 @@ static int do_probe( struct i2c_adapter *adapter, int addr, int kind); /* scan 0x48-0x4f (DS1775) and 0x2c-2x2f (ADM1030) */ -static unsigned short normal_i2c[] = { 0x48, 0x49, 0x4a, 0x4b, +static const unsigned short normal_i2c[] = { 0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f, 0x2c, 0x2d, 0x2e, 0x2f, I2C_CLIENT_END }; diff --git a/drivers/usb/host/ohci-pnx4008.c b/drivers/usb/host/ohci-pnx4008.c index ca2a6abbc117..642733d091fa 100644 --- a/drivers/usb/host/ohci-pnx4008.c +++ b/drivers/usb/host/ohci-pnx4008.c @@ -112,9 +112,9 @@ static int isp1301_detach(struct i2c_client *client); static int isp1301_command(struct i2c_client *client, unsigned int cmd, void *arg); -static unsigned short normal_i2c[] = +static const unsigned short normal_i2c[] = { ISP1301_I2C_ADDR, ISP1301_I2C_ADDR + 1, I2C_CLIENT_END }; -static unsigned short dummy_i2c_addrlist[] = { I2C_CLIENT_END }; +static const unsigned short dummy_i2c_addrlist[] = { I2C_CLIENT_END }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_i2c, diff --git a/drivers/video/matrox/matroxfb_maven.c b/drivers/video/matrox/matroxfb_maven.c index 49cd53e46c0a..0cd58f84fb46 100644 --- a/drivers/video/matrox/matroxfb_maven.c +++ b/drivers/video/matrox/matroxfb_maven.c @@ -1232,7 +1232,7 @@ static int maven_shutdown_client(struct i2c_client* clnt) { return 0; } -static unsigned short normal_i2c[] = { MAVEN_I2CID, I2C_CLIENT_END }; +static const unsigned short normal_i2c[] = { MAVEN_I2CID, I2C_CLIENT_END }; I2C_CLIENT_INSMOD; static struct i2c_driver maven_driver; diff --git a/drivers/w1/masters/ds2482.c b/drivers/w1/masters/ds2482.c index d93eb626b2f0..0fd5820d5c61 100644 --- a/drivers/w1/masters/ds2482.c +++ b/drivers/w1/masters/ds2482.c @@ -29,7 +29,7 @@ * However, the chip cannot be detected without doing an i2c write, * so use the force module parameter. */ -static unsigned short normal_i2c[] = {I2C_CLIENT_END}; +static const unsigned short normal_i2c[] = { I2C_CLIENT_END }; /** * Insmod parameters diff --git a/sound/soc/codecs/cs4270.c b/sound/soc/codecs/cs4270.c index abac62866da8..dab22cc97ead 100644 --- a/sound/soc/codecs/cs4270.c +++ b/sound/soc/codecs/cs4270.c @@ -234,7 +234,7 @@ static int cs4270_set_dai_fmt(struct snd_soc_codec_dai *codec_dai, * lower three bits are determined via the AD2, AD1, and AD0 pins * (respectively). */ -static unsigned short normal_i2c[] = { +static const unsigned short normal_i2c[] = { 0x48, 0x49, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, I2C_CLIENT_END }; I2C_CLIENT_INSMOD; diff --git a/sound/soc/s3c24xx/neo1973_wm8753.c b/sound/soc/s3c24xx/neo1973_wm8753.c index d5a8fc2cf8d6..f1f6b9478af9 100644 --- a/sound/soc/s3c24xx/neo1973_wm8753.c +++ b/sound/soc/s3c24xx/neo1973_wm8753.c @@ -573,7 +573,7 @@ static struct snd_soc_device neo1973_snd_devdata = { static struct i2c_client client_template; -static unsigned short normal_i2c[] = { 0x7C, I2C_CLIENT_END }; +static const unsigned short normal_i2c[] = { 0x7C, I2C_CLIENT_END }; /* Magic definition of all other variables and things */ I2C_CLIENT_INSMOD; -- cgit v1.2.3 From 0f79b72e45da68bf542a63a08f9c924b91b507e7 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sun, 27 Jan 2008 18:14:47 +0100 Subject: i2c: Deprecate drivers for I2C buses on video adapters The framebuffer drivers for these pieces of hardware include support for the DDC/I2C buses, so there is no need for separate drivers. Signed-off-by: Jean Delvare --- Documentation/feature-removal-schedule.txt | 7 +++++++ drivers/i2c/busses/Kconfig | 17 +++++++++++++---- 2 files changed, 20 insertions(+), 4 deletions(-) (limited to 'Documentation') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index fb51d38e8b46..878ca34844e4 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -308,3 +308,10 @@ Why: This driver has been marked obsolete for many years. Who: Stephen Hemminger --------------------------- + +What: i2c-i810, i2c-prosavage and i2c-savage4 +When: May 2008 +Why: These drivers are superseded by i810fb, intelfb and savagefb. +Who: Jean Delvare + +--------------------------- diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig index 2cef92a0bd0e..a38ba04fe36e 100644 --- a/drivers/i2c/busses/Kconfig +++ b/drivers/i2c/busses/Kconfig @@ -182,7 +182,8 @@ config I2C_I801 will be called i2c-i801. config I2C_I810 - tristate "Intel 810/815" + tristate "Intel 810/815 (DEPRECATED)" + default n depends on PCI select I2C_ALGOBIT help @@ -195,6 +196,8 @@ config I2C_I810 i815 i845G + This driver is deprecated in favor of the i810fb and intelfb drivers. + This driver can also be built as a module. If so, the module will be called i2c-i810. @@ -382,7 +385,8 @@ config I2C_PASEMI Supports the PA Semi PWRficient on-chip SMBus interfaces. config I2C_PROSAVAGE - tristate "S3/VIA (Pro)Savage" + tristate "S3/VIA (Pro)Savage (DEPRECATED)" + default n depends on PCI select I2C_ALGOBIT help @@ -393,6 +397,8 @@ config I2C_PROSAVAGE S3/VIA KM266/VT8375 aka ProSavage8 S3/VIA KM133/VT8365 aka Savage4 + This driver is deprecated in favor of the savagefb driver. + This support is also available as a module. If so, the module will be called i2c-prosavage. @@ -404,13 +410,16 @@ config I2C_S3C2410 Samsung S3C2410 based System-on-Chip devices. config I2C_SAVAGE4 - tristate "S3 Savage 4" - depends on PCI && EXPERIMENTAL + tristate "S3 Savage 4 (DEPRECATED)" + default n + depends on PCI select I2C_ALGOBIT help If you say yes to this option, support will be included for the S3 Savage 4 I2C interface. + This driver is deprecated in favor of the savagefb driver. + This driver can also be built as a module. If so, the module will be called i2c-savage4. -- cgit v1.2.3 From 87c6c22945e5d68eb96dd1e5cb26185253cd5b9d Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sun, 27 Jan 2008 18:14:48 +0100 Subject: i2c: Drop redundant i2c_adapter.list i2c_adapter.list is superfluous, this list duplicates the one maintained by the driver core. Drop it. Signed-off-by: Jean Delvare Acked-by: David Brownell --- Documentation/feature-removal-schedule.txt | 9 --------- drivers/i2c/i2c-core.c | 24 ++++++++++-------------- include/linux/i2c.h | 1 - 3 files changed, 10 insertions(+), 24 deletions(-) (limited to 'Documentation') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 878ca34844e4..25370662cc5d 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -191,15 +191,6 @@ Who: Kay Sievers --------------------------- -What: i2c_adapter.list -When: July 2007 -Why: Superfluous, this list duplicates the one maintained by the driver - core. -Who: Jean Delvare , - David Brownell - ---------------------------- - What: ACPI procfs interface When: July 2008 Why: ACPI sysfs conversion should be finished by January 2008. diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c index f75344ac4663..3495ab4035ec 100644 --- a/drivers/i2c/i2c-core.c +++ b/drivers/i2c/i2c-core.c @@ -34,11 +34,11 @@ #include #include #include +#include #include "i2c-core.h" -static LIST_HEAD(adapters); static LIST_HEAD(drivers); static DEFINE_MUTEX(core_lists); static DEFINE_IDR(i2c_adapter_idr); @@ -331,7 +331,6 @@ static int i2c_register_adapter(struct i2c_adapter *adap) INIT_LIST_HEAD(&adap->clients); mutex_lock(&core_lists); - list_add_tail(&adap->list, &adapters); /* Add the adapter to the driver core. * If the parent pointer is not set up, @@ -368,7 +367,6 @@ out_unlock: return res; out_list: - list_del(&adap->list); idr_remove(&i2c_adapter_idr, adap->nr); goto out_unlock; } @@ -473,7 +471,6 @@ EXPORT_SYMBOL_GPL(i2c_add_numbered_adapter); int i2c_del_adapter(struct i2c_adapter *adap) { struct list_head *item, *_n; - struct i2c_adapter *adap_from_list; struct i2c_driver *driver; struct i2c_client *client; int res = 0; @@ -481,11 +478,7 @@ int i2c_del_adapter(struct i2c_adapter *adap) mutex_lock(&core_lists); /* First make sure that this adapter was ever added */ - list_for_each_entry(adap_from_list, &adapters, list) { - if (adap_from_list == adap) - break; - } - if (adap_from_list != adap) { + if (idr_find(&i2c_adapter_idr, adap->nr) != adap) { pr_debug("i2c-core: attempting to delete unregistered " "adapter [%s]\n", adap->name); res = -EINVAL; @@ -529,7 +522,6 @@ int i2c_del_adapter(struct i2c_adapter *adap) /* clean up the sysfs representation */ init_completion(&adap->dev_released); device_unregister(&adap->dev); - list_del(&adap->list); /* wait for sysfs to drop all references */ wait_for_completion(&adap->dev_released); @@ -592,9 +584,12 @@ int i2c_register_driver(struct module *owner, struct i2c_driver *driver) if (driver->attach_adapter) { struct i2c_adapter *adapter; - list_for_each_entry(adapter, &adapters, list) { + down(&i2c_adapter_class.sem); + list_for_each_entry(adapter, &i2c_adapter_class.devices, + dev.node) { driver->attach_adapter(adapter); } + up(&i2c_adapter_class.sem); } mutex_unlock(&core_lists); @@ -609,7 +604,7 @@ EXPORT_SYMBOL(i2c_register_driver); */ void i2c_del_driver(struct i2c_driver *driver) { - struct list_head *item1, *item2, *_n; + struct list_head *item2, *_n; struct i2c_client *client; struct i2c_adapter *adap; @@ -623,8 +618,8 @@ void i2c_del_driver(struct i2c_driver *driver) * attached. If so, detach them to be able to kill the driver * afterwards. */ - list_for_each(item1,&adapters) { - adap = list_entry(item1, struct i2c_adapter, list); + down(&i2c_adapter_class.sem); + list_for_each_entry(adap, &i2c_adapter_class.devices, dev.node) { if (driver->detach_adapter) { if (driver->detach_adapter(adap)) { dev_err(&adap->dev, "detach_adapter failed " @@ -648,6 +643,7 @@ void i2c_del_driver(struct i2c_driver *driver) } } } + up(&i2c_adapter_class.sem); unregister: driver_unregister(&driver->driver); diff --git a/include/linux/i2c.h b/include/linux/i2c.h index 78a305bed8eb..5fa7e180501d 100644 --- a/include/linux/i2c.h +++ b/include/linux/i2c.h @@ -317,7 +317,6 @@ struct i2c_adapter { int nr; struct list_head clients; - struct list_head list; char name[48]; struct completion dev_released; }; -- cgit v1.2.3 From 6342064cad7a28d10504128d028bc4ba379d489d Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sun, 27 Jan 2008 18:14:50 +0100 Subject: i2c-i801: Implement I2C block read support I2C block read is supported since the ICH5. I couldn't get it to work using the block buffer, so it's using the old-style byte-by-byte mode for now. Note: I'm also updating the driver author... The i2c-i801 driver was really written by Mark Studebaker, even though he based his work on the i2c-piix4 driver which was written by Philip Edelbrock. Signed-off-by: Jean Delvare --- Documentation/i2c/busses/i2c-i801 | 5 +-- drivers/i2c/busses/i2c-i801.c | 79 ++++++++++++++++++++++++++------------- 2 files changed, 56 insertions(+), 28 deletions(-) (limited to 'Documentation') diff --git a/Documentation/i2c/busses/i2c-i801 b/Documentation/i2c/busses/i2c-i801 index fde4420e3f75..3bd958360159 100644 --- a/Documentation/i2c/busses/i2c-i801 +++ b/Documentation/i2c/busses/i2c-i801 @@ -17,9 +17,8 @@ Supported adapters: Datasheets: Publicly available at the Intel website Authors: - Frodo Looijaard , - Philip Edelbrock , Mark Studebaker + Jean Delvare Module Parameters @@ -62,7 +61,7 @@ Not supported. I2C Block Read Support ---------------------- -Not supported at the moment. +I2C block read is supported on the 82801EB (ICH5) and later chips. SMBus 2.0 Support diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c index 0b1b1ae5e228..aa9157913b9a 100644 --- a/drivers/i2c/busses/i2c-i801.c +++ b/drivers/i2c/busses/i2c-i801.c @@ -4,6 +4,7 @@ Copyright (c) 1998 - 2002 Frodo Looijaard , Philip Edelbrock , and Mark D. Studebaker + Copyright (C) 2007 Jean Delvare This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by @@ -46,7 +47,7 @@ Hardware PEC yes Block buffer yes Block process call transaction no - I2C block read transaction no + I2C block read transaction yes (doesn't use the block buffer) See the file Documentation/i2c/busses/i2c-i801 for details. */ @@ -102,9 +103,9 @@ #define I801_WORD_DATA 0x0C #define I801_PROC_CALL 0x10 /* unimplemented */ #define I801_BLOCK_DATA 0x14 -#define I801_I2C_BLOCK_DATA 0x18 /* unimplemented */ +#define I801_I2C_BLOCK_DATA 0x18 /* ICH5 and later */ #define I801_BLOCK_LAST 0x34 -#define I801_I2C_BLOCK_LAST 0x38 /* unimplemented */ +#define I801_I2C_BLOCK_LAST 0x38 /* ICH5 and later */ #define I801_START 0x40 #define I801_PEC_EN 0x80 /* ICH3 and later */ @@ -256,7 +257,8 @@ static int i801_block_transaction_by_block(union i2c_smbus_data *data, } static int i801_block_transaction_byte_by_byte(union i2c_smbus_data *data, - char read_write, int hwpec) + char read_write, int command, + int hwpec) { int i, len; int smbcmd; @@ -273,16 +275,24 @@ static int i801_block_transaction_byte_by_byte(union i2c_smbus_data *data, } for (i = 1; i <= len; i++) { - if (i == len && read_write == I2C_SMBUS_READ) - smbcmd = I801_BLOCK_LAST; - else - smbcmd = I801_BLOCK_DATA; + if (i == len && read_write == I2C_SMBUS_READ) { + if (command == I2C_SMBUS_I2C_BLOCK_DATA) + smbcmd = I801_I2C_BLOCK_LAST; + else + smbcmd = I801_BLOCK_LAST; + } else { + if (command == I2C_SMBUS_I2C_BLOCK_DATA + && read_write == I2C_SMBUS_READ) + smbcmd = I801_I2C_BLOCK_DATA; + else + smbcmd = I801_BLOCK_DATA; + } outb_p(smbcmd | ENABLE_INT9, SMBHSTCNT); dev_dbg(&I801_dev->dev, "Block (pre %d): CNT=%02x, CMD=%02x, " - "ADD=%02x, DAT0=%02x, BLKDAT=%02x\n", i, + "ADD=%02x, DAT0=%02x, DAT1=%02x, BLKDAT=%02x\n", i, inb_p(SMBHSTCNT), inb_p(SMBHSTCMD), inb_p(SMBHSTADD), - inb_p(SMBHSTDAT0), inb_p(SMBBLKDAT)); + inb_p(SMBHSTDAT0), inb_p(SMBHSTDAT1), inb_p(SMBBLKDAT)); /* Make sure the SMBus host is ready to start transmitting */ temp = inb_p(SMBHSTSTS); @@ -346,7 +356,8 @@ static int i801_block_transaction_byte_by_byte(union i2c_smbus_data *data, dev_dbg(&I801_dev->dev, "Error: no response!\n"); } - if (i == 1 && read_write == I2C_SMBUS_READ) { + if (i == 1 && read_write == I2C_SMBUS_READ + && command != I2C_SMBUS_I2C_BLOCK_DATA) { len = inb_p(SMBHSTDAT0); if (len < 1 || len > I2C_SMBUS_BLOCK_MAX) return -1; @@ -367,9 +378,9 @@ static int i801_block_transaction_byte_by_byte(union i2c_smbus_data *data, temp); } dev_dbg(&I801_dev->dev, "Block (post %d): CNT=%02x, CMD=%02x, " - "ADD=%02x, DAT0=%02x, BLKDAT=%02x\n", i, + "ADD=%02x, DAT0=%02x, DAT1=%02x, BLKDAT=%02x\n", i, inb_p(SMBHSTCNT), inb_p(SMBHSTCMD), inb_p(SMBHSTADD), - inb_p(SMBHSTDAT0), inb_p(SMBBLKDAT)); + inb_p(SMBHSTDAT0), inb_p(SMBHSTDAT1), inb_p(SMBBLKDAT)); if (result < 0) return result; @@ -398,34 +409,38 @@ static int i801_block_transaction(union i2c_smbus_data *data, char read_write, pci_read_config_byte(I801_dev, SMBHSTCFG, &hostc); pci_write_config_byte(I801_dev, SMBHSTCFG, hostc | SMBHSTCFG_I2C_EN); - } else { + } else if (!(i801_features & FEATURE_I2C_BLOCK_READ)) { dev_err(&I801_dev->dev, - "I2C_SMBUS_I2C_BLOCK_READ not DB!\n"); + "I2C block read is unsupported!\n"); return -1; } } - if (read_write == I2C_SMBUS_WRITE) { + if (read_write == I2C_SMBUS_WRITE + || command == I2C_SMBUS_I2C_BLOCK_DATA) { if (data->block[0] < 1) data->block[0] = 1; if (data->block[0] > I2C_SMBUS_BLOCK_MAX) data->block[0] = I2C_SMBUS_BLOCK_MAX; } else { - data->block[0] = 32; /* max for reads */ + data->block[0] = 32; /* max for SMBus block reads */ } if ((i801_features & FEATURE_BLOCK_BUFFER) + && !(command == I2C_SMBUS_I2C_BLOCK_DATA + && read_write == I2C_SMBUS_READ) && i801_set_block_buffer_mode() == 0) result = i801_block_transaction_by_block(data, read_write, hwpec); else result = i801_block_transaction_byte_by_byte(data, read_write, - hwpec); + command, hwpec); if (result == 0 && hwpec) i801_wait_hwpec(); - if (command == I2C_SMBUS_I2C_BLOCK_DATA) { + if (command == I2C_SMBUS_I2C_BLOCK_DATA + && read_write == I2C_SMBUS_WRITE) { /* restore saved configuration register value */ pci_write_config_byte(I801_dev, SMBHSTCFG, hostc); } @@ -477,12 +492,23 @@ static s32 i801_access(struct i2c_adapter * adap, u16 addr, xact = I801_WORD_DATA; break; case I2C_SMBUS_BLOCK_DATA: - case I2C_SMBUS_I2C_BLOCK_DATA: outb_p(((addr & 0x7f) << 1) | (read_write & 0x01), SMBHSTADD); outb_p(command, SMBHSTCMD); block = 1; break; + case I2C_SMBUS_I2C_BLOCK_DATA: + /* NB: page 240 of ICH5 datasheet shows that the R/#W + * bit should be cleared here, even when reading */ + outb_p((addr & 0x7f) << 1, SMBHSTADD); + if (read_write == I2C_SMBUS_READ) { + /* NB: page 240 of ICH5 datasheet also shows + * that DATA1 is the cmd field when reading */ + outb_p(command, SMBHSTDAT1); + } else + outb_p(command, SMBHSTCMD); + block = 1; + break; case I2C_SMBUS_PROC_CALL: default: dev_err(&I801_dev->dev, "Unsupported transaction %d\n", size); @@ -531,7 +557,9 @@ static u32 i801_func(struct i2c_adapter *adapter) return I2C_FUNC_SMBUS_QUICK | I2C_FUNC_SMBUS_BYTE | I2C_FUNC_SMBUS_BYTE_DATA | I2C_FUNC_SMBUS_WORD_DATA | I2C_FUNC_SMBUS_BLOCK_DATA | I2C_FUNC_SMBUS_WRITE_I2C_BLOCK | - ((i801_features & FEATURE_SMBUS_PEC) ? I2C_FUNC_SMBUS_PEC : 0); + ((i801_features & FEATURE_SMBUS_PEC) ? I2C_FUNC_SMBUS_PEC : 0) | + ((i801_features & FEATURE_I2C_BLOCK_READ) ? + I2C_FUNC_SMBUS_READ_I2C_BLOCK : 0); } static const struct i2c_algorithm smbus_algorithm = { @@ -573,7 +601,6 @@ static int __devinit i801_probe(struct pci_dev *dev, const struct pci_device_id I801_dev = dev; i801_features = 0; switch (dev->device) { - case PCI_DEVICE_ID_INTEL_82801DB_3: case PCI_DEVICE_ID_INTEL_82801EB_3: case PCI_DEVICE_ID_INTEL_ESB_4: case PCI_DEVICE_ID_INTEL_ICH6_16: @@ -581,6 +608,9 @@ static int __devinit i801_probe(struct pci_dev *dev, const struct pci_device_id case PCI_DEVICE_ID_INTEL_ESB2_17: case PCI_DEVICE_ID_INTEL_ICH8_5: case PCI_DEVICE_ID_INTEL_ICH9_6: + i801_features |= FEATURE_I2C_BLOCK_READ; + /* fall through */ + case PCI_DEVICE_ID_INTEL_82801DB_3: case PCI_DEVICE_ID_INTEL_TOLAPAI_1: i801_features |= FEATURE_SMBUS_PEC; i801_features |= FEATURE_BLOCK_BUFFER; @@ -698,9 +728,8 @@ static void __exit i2c_i801_exit(void) pci_unregister_driver(&i801_driver); } -MODULE_AUTHOR ("Frodo Looijaard , " - "Philip Edelbrock , " - "and Mark D. Studebaker "); +MODULE_AUTHOR("Mark D. Studebaker , " + "Jean Delvare "); MODULE_DESCRIPTION("I801 SMBus driver"); MODULE_LICENSE("GPL"); -- cgit v1.2.3 From 0d227a7e724460bddcd603a1feb672267bcb0d6c Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sun, 27 Jan 2008 18:14:51 +0100 Subject: i2c-viapro: Add support for the VT8237S Add support for another variant of the VT8237. I couldn't test I2C block support but I assume it is present as well. Signed-off-by: Jean Delvare --- Documentation/i2c/busses/i2c-viapro | 3 ++- drivers/i2c/busses/Kconfig | 2 +- drivers/i2c/busses/i2c-viapro.c | 6 +++++- 3 files changed, 8 insertions(+), 3 deletions(-) (limited to 'Documentation') diff --git a/Documentation/i2c/busses/i2c-viapro b/Documentation/i2c/busses/i2c-viapro index 06b4be3ef6d8..1405fb69984c 100644 --- a/Documentation/i2c/busses/i2c-viapro +++ b/Documentation/i2c/busses/i2c-viapro @@ -10,7 +10,7 @@ Supported adapters: * VIA Technologies, Inc. VT8231, VT8233, VT8233A Datasheet: available on request from VIA - * VIA Technologies, Inc. VT8235, VT8237R, VT8237A, VT8251 + * VIA Technologies, Inc. VT8235, VT8237R, VT8237A, VT8237S, VT8251 Datasheet: available on request and under NDA from VIA * VIA Technologies, Inc. CX700 @@ -46,6 +46,7 @@ Your lspci -n listing must show one of these : device 1106:3177 (VT8235) device 1106:3227 (VT8237R) device 1106:3337 (VT8237A) + device 1106:3372 (VT8237S) device 1106:3287 (VT8251) device 1106:8324 (CX700) diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig index a38ba04fe36e..8d12b26bb6c6 100644 --- a/drivers/i2c/busses/Kconfig +++ b/drivers/i2c/busses/Kconfig @@ -606,7 +606,7 @@ config I2C_VIAPRO VT8231 VT8233/A VT8235 - VT8237R/A + VT8237R/A/S VT8251 CX700 diff --git a/drivers/i2c/busses/i2c-viapro.c b/drivers/i2c/busses/i2c-viapro.c index 661ebc7a711e..77b13d027f86 100644 --- a/drivers/i2c/busses/i2c-viapro.c +++ b/drivers/i2c/busses/i2c-viapro.c @@ -4,7 +4,7 @@ Copyright (c) 1998 - 2002 Frodo Looijaard , Philip Edelbrock , Kyösti Mälkki , Mark D. Studebaker - Copyright (C) 2005 - 2007 Jean Delvare + Copyright (C) 2005 - 2008 Jean Delvare This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by @@ -35,6 +35,7 @@ VT8235 0x3177 yes VT8237R 0x3227 yes VT8237A 0x3337 yes + VT8237S 0x3372 yes VT8251 0x3287 yes CX700 0x8324 yes @@ -393,6 +394,7 @@ found: case PCI_DEVICE_ID_VIA_8251: case PCI_DEVICE_ID_VIA_8237: case PCI_DEVICE_ID_VIA_8237A: + case PCI_DEVICE_ID_VIA_8237S: case PCI_DEVICE_ID_VIA_8235: case PCI_DEVICE_ID_VIA_8233A: case PCI_DEVICE_ID_VIA_8233_0: @@ -444,6 +446,8 @@ static struct pci_device_id vt596_ids[] = { .driver_data = SMBBA3 }, { PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8237A), .driver_data = SMBBA3 }, + { PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8237S), + .driver_data = SMBBA3 }, { PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8231_4), .driver_data = SMBBA1 }, { PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8251), -- cgit v1.2.3 From 0486bc9098f4556a0aa90d57f717d08164b7647e Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Mon, 12 Nov 2007 16:17:55 -0800 Subject: kconfig: add hints/tips/tricks to Documentation/kbuild/kconfig-language.txt Add a section on kconfig hints: how to do in Kconfig files. Fix a few typos/spellos. Signed-off-by: Randy Dunlap Signed-off-by: Sam Ravnborg --- Documentation/kbuild/kconfig-language.txt | 54 +++++++++++++++++++++++++++---- 1 file changed, 48 insertions(+), 6 deletions(-) (limited to 'Documentation') diff --git a/Documentation/kbuild/kconfig-language.txt b/Documentation/kbuild/kconfig-language.txt index 616043a6da99..53ca12f7999f 100644 --- a/Documentation/kbuild/kconfig-language.txt +++ b/Documentation/kbuild/kconfig-language.txt @@ -24,7 +24,7 @@ visible if its parent entry is also visible. Menu entries ------------ -Most entries define a config option, all other entries help to organize +Most entries define a config option; all other entries help to organize them. A single configuration option is defined like this: config MODVERSIONS @@ -50,7 +50,7 @@ applicable everywhere (see syntax). - type definition: "bool"/"tristate"/"string"/"hex"/"int" Every config option must have a type. There are only two basic types: - tristate and string, the other types are based on these two. The type + tristate and string; the other types are based on these two. The type definition optionally accepts an input prompt, so these two examples are equivalent: @@ -108,7 +108,7 @@ applicable everywhere (see syntax). equal to 'y' without visiting the dependencies. So abusing select you are able to select a symbol FOO even if FOO depends on BAR that is not set. In general use select only for - non-visible symbols (no promts anywhere) and for symbols with + non-visible symbols (no prompts anywhere) and for symbols with no dependencies. That will limit the usefulness but on the other hand avoid the illegal configurations all over. kconfig should one day warn about such things. @@ -162,9 +162,9 @@ An expression can have a value of 'n', 'm' or 'y' (or 0, 1, 2 respectively for calculations). A menu entry becomes visible when it's expression evaluates to 'm' or 'y'. -There are two types of symbols: constant and nonconstant symbols. -Nonconstant symbols are the most common ones and are defined with the -'config' statement. Nonconstant symbols consist entirely of alphanumeric +There are two types of symbols: constant and non-constant symbols. +Non-constant symbols are the most common ones and are defined with the +'config' statement. Non-constant symbols consist entirely of alphanumeric characters or underscores. Constant symbols are only part of expressions. Constant symbols are always surrounded by single or double quotes. Within the quote, any @@ -301,3 +301,45 @@ mainmenu: This sets the config program's title bar if the config program chooses to use it. + + +Kconfig hints +------------- +This is a collection of Kconfig tips, most of which aren't obvious at +first glance and most of which have become idioms in several Kconfig +files. + +Build as module only +~~~~~~~~~~~~~~~~~~~~ +To restrict a component build to module-only, qualify its config symbol +with "depends on m". E.g.: + +config FOO + depends on BAR && m + +limits FOO to module (=m) or disabled (=n). + + +Build limited by a third config symbol which may be =y or =m +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +A common idiom that we see (and sometimes have problems with) is this: + +When option C in B (module or subsystem) uses interfaces from A (module +or subsystem), and both A and B are tristate (could be =y or =m if they +were independent of each other, but they aren't), then we need to limit +C such that it cannot be built statically if A is built as a loadable +module. (C already depends on B, so there is no dependency issue to +take care of here.) + +If A is linked statically into the kernel image, C can be built +statically or as loadable module(s). However, if A is built as loadable +module(s), then C must be restricted to loadable module(s) also. This +can be expressed in kconfig language as: + +config C + depends on A = y || A = B + +or for real examples, use this command in a kernel tree: + +$ find . -name Kconfig\* | xargs grep -ns "depends on.*=.*||.*=" | grep -v orig + -- cgit v1.2.3 From 93449082e905ce73d0346d617dd67c4b668b58af Mon Sep 17 00:00:00 2001 From: Roman Zippel Date: Mon, 14 Jan 2008 04:50:54 +0100 Subject: kconfig: environment symbol support Add the possibility to import a value from the environment into kconfig via the option syntax. Beside flexibility this has the advantage providing proper dependencies. Documented the options syntax. Signed-off-by: Roman Zippel Signed-off-by: Sam Ravnborg --- Documentation/kbuild/kconfig-language.txt | 21 +++++++++++++++ scripts/kconfig/expr.h | 3 ++- scripts/kconfig/lkc.h | 5 ++++ scripts/kconfig/menu.c | 5 +++- scripts/kconfig/qconf.cc | 16 ++++------- scripts/kconfig/symbol.c | 45 +++++++++++++++++++++++++++++++ scripts/kconfig/util.c | 23 ++++++++++++++-- scripts/kconfig/zconf.gperf | 1 + scripts/kconfig/zconf.hash.c_shipped | 45 +++++++++++++++++-------------- 9 files changed, 129 insertions(+), 35 deletions(-) (limited to 'Documentation') diff --git a/Documentation/kbuild/kconfig-language.txt b/Documentation/kbuild/kconfig-language.txt index 53ca12f7999f..a43fcc68e171 100644 --- a/Documentation/kbuild/kconfig-language.txt +++ b/Documentation/kbuild/kconfig-language.txt @@ -127,6 +127,27 @@ applicable everywhere (see syntax). used to help visually separate configuration logic from help within the file as an aid to developers. +- misc options: "option" [=] + Various less common options can be defined via this option syntax, + which can modify the behaviour of the menu entry and its config + symbol. These options are currently possible: + + - "defconfig_list" + This declares a list of default entries which can be used when + looking for the default configuration (which is used when the main + .config doesn't exists yet.) + + - "modules" + This declares the symbol to be used as the MODULES symbol, which + enables the third modular state for all config symbols. + + - "env"= + This imports the environment variable into Kconfig. It behaves like + a default, except that the value comes from the environment, this + also means that the behaviour when mixing it with normal defaults is + undefined at this point. The symbol is currently not exported back + to the build environment (if this is desired, it can be done via + another symbol). Menu dependencies ----------------- diff --git a/scripts/kconfig/expr.h b/scripts/kconfig/expr.h index b6f922c77e0b..9d4cba1c001d 100644 --- a/scripts/kconfig/expr.h +++ b/scripts/kconfig/expr.h @@ -106,7 +106,8 @@ struct symbol { #define SYMBOL_HASHMASK 0xff enum prop_type { - P_UNKNOWN, P_PROMPT, P_COMMENT, P_MENU, P_DEFAULT, P_CHOICE, P_SELECT, P_RANGE + P_UNKNOWN, P_PROMPT, P_COMMENT, P_MENU, P_DEFAULT, P_CHOICE, + P_SELECT, P_RANGE, P_ENV }; struct property { diff --git a/scripts/kconfig/lkc.h b/scripts/kconfig/lkc.h index 8a07ee4f6bd4..4bc68f20a73c 100644 --- a/scripts/kconfig/lkc.h +++ b/scripts/kconfig/lkc.h @@ -44,6 +44,7 @@ extern "C" { #define T_OPT_MODULES 1 #define T_OPT_DEFCONFIG_LIST 2 +#define T_OPT_ENV 3 struct kconf_id { int name; @@ -74,6 +75,7 @@ void kconfig_load(void); /* menu.c */ void menu_init(void); +void menu_warn(struct menu *menu, const char *fmt, ...); struct menu *menu_add_menu(void); void menu_end_menu(void); void menu_add_entry(struct symbol *sym); @@ -103,6 +105,8 @@ void str_printf(struct gstr *gs, const char *fmt, ...); const char *str_get(struct gstr *gs); /* symbol.c */ +extern struct expr *sym_env_list; + void sym_init(void); void sym_clear_all_valid(void); void sym_set_all_changed(void); @@ -110,6 +114,7 @@ void sym_set_changed(struct symbol *sym); struct symbol *sym_check_deps(struct symbol *sym); struct property *prop_alloc(enum prop_type type, struct symbol *sym); struct symbol *prop_get_symbol(struct property *prop); +struct property *sym_get_env_prop(struct symbol *sym); static inline tristate sym_get_tristate_value(struct symbol *sym) { diff --git a/scripts/kconfig/menu.c b/scripts/kconfig/menu.c index 3637d1057e13..e6ef171e5b14 100644 --- a/scripts/kconfig/menu.c +++ b/scripts/kconfig/menu.c @@ -15,7 +15,7 @@ static struct menu **last_entry_ptr; struct file *file_list; struct file *current_file; -static void menu_warn(struct menu *menu, const char *fmt, ...) +void menu_warn(struct menu *menu, const char *fmt, ...) { va_list ap; va_start(ap, fmt); @@ -172,6 +172,9 @@ void menu_add_option(int token, char *arg) else if (sym_defconfig_list != current_entry->sym) zconf_error("trying to redefine defconfig symbol"); break; + case T_OPT_ENV: + prop_add_env(arg); + break; } } diff --git a/scripts/kconfig/qconf.cc b/scripts/kconfig/qconf.cc index 9fe27ca8a192..5d0fd38b089b 100644 --- a/scripts/kconfig/qconf.cc +++ b/scripts/kconfig/qconf.cc @@ -1088,7 +1088,11 @@ QString ConfigInfoView::debug_info(struct symbol *sym) debug += "
"; break; case P_DEFAULT: - debug += "default: "; + case P_SELECT: + case P_RANGE: + case P_ENV: + debug += prop_get_type_name(prop->type); + debug += ": "; expr_print(prop->expr, expr_print_help, &debug, E_NONE); debug += "
"; break; @@ -1099,16 +1103,6 @@ QString ConfigInfoView::debug_info(struct symbol *sym) debug += "
"; } break; - case P_SELECT: - debug += "select: "; - expr_print(prop->expr, expr_print_help, &debug, E_NONE); - debug += "
"; - break; - case P_RANGE: - debug += "range: "; - expr_print(prop->expr, expr_print_help, &debug, E_NONE); - debug += "
"; - break; default: debug += "unknown property: "; debug += prop_get_type_name(prop->type); diff --git a/scripts/kconfig/symbol.c b/scripts/kconfig/symbol.c index a7dfc82fc858..99e3d02d1cfc 100644 --- a/scripts/kconfig/symbol.c +++ b/scripts/kconfig/symbol.c @@ -34,6 +34,8 @@ struct symbol *sym_defconfig_list; struct symbol *modules_sym; tristate modules_val; +struct expr *sym_env_list; + void sym_add_default(struct symbol *sym, const char *def) { struct property *prop = prop_alloc(P_DEFAULT, sym); @@ -117,6 +119,15 @@ struct property *sym_get_choice_prop(struct symbol *sym) return NULL; } +struct property *sym_get_env_prop(struct symbol *sym) +{ + struct property *prop; + + for_all_properties(sym, prop, P_ENV) + return prop; + return NULL; +} + struct property *sym_get_default_prop(struct symbol *sym) { struct property *prop; @@ -346,6 +357,9 @@ void sym_calc_value(struct symbol *sym) ; } + if (sym->flags & SYMBOL_AUTO) + sym->flags &= ~SYMBOL_WRITE; + sym->curr = newval; if (sym_is_choice(sym) && newval.tri == yes) sym->curr.val = sym_calc_choice(sym); @@ -860,6 +874,8 @@ const char *prop_get_type_name(enum prop_type type) switch (type) { case P_PROMPT: return "prompt"; + case P_ENV: + return "env"; case P_COMMENT: return "comment"; case P_MENU: @@ -877,3 +893,32 @@ const char *prop_get_type_name(enum prop_type type) } return "unknown"; } + +void prop_add_env(const char *env) +{ + struct symbol *sym, *sym2; + struct property *prop; + char *p; + + sym = current_entry->sym; + sym->flags |= SYMBOL_AUTO; + for_all_properties(sym, prop, P_ENV) { + sym2 = prop_get_symbol(prop); + if (strcmp(sym2->name, env)) + menu_warn(current_entry, "redefining environment symbol from %s", + sym2->name); + return; + } + + prop = prop_alloc(P_ENV, sym); + prop->expr = expr_alloc_symbol(sym_lookup(env, 1)); + + sym_env_list = expr_alloc_one(E_LIST, sym_env_list); + sym_env_list->right.sym = sym; + + p = getenv(env); + if (p) + sym_add_default(sym, p); + else + menu_warn(current_entry, "environment variable %s undefined", env); +} diff --git a/scripts/kconfig/util.c b/scripts/kconfig/util.c index e1cad924c0a4..f8e73c039dc8 100644 --- a/scripts/kconfig/util.c +++ b/scripts/kconfig/util.c @@ -29,6 +29,8 @@ struct file *file_lookup(const char *name) /* write a dependency file as used by kbuild to track dependencies */ int file_write_dep(const char *name) { + struct symbol *sym, *env_sym; + struct expr *e; struct file *file; FILE *out; @@ -45,8 +47,25 @@ int file_write_dep(const char *name) fprintf(out, "\t%s\n", file->name); } fprintf(out, "\ninclude/config/auto.conf: \\\n" - "\t$(deps_config)\n\n" - "$(deps_config): ;\n"); + "\t$(deps_config)\n\n"); + + expr_list_for_each_sym(sym_env_list, e, sym) { + struct property *prop; + const char *value; + + prop = sym_get_env_prop(sym); + env_sym = prop_get_symbol(prop); + if (!env_sym) + continue; + value = getenv(env_sym->name); + if (!value) + value = ""; + fprintf(out, "ifneq \"$(%s)\" \"%s\"\n", env_sym->name, value); + fprintf(out, "include/config/auto.conf: FORCE\n"); + fprintf(out, "endif\n"); + } + + fprintf(out, "\n$(deps_config): ;\n"); fclose(out); rename("..config.tmp", name); return 0; diff --git a/scripts/kconfig/zconf.gperf b/scripts/kconfig/zconf.gperf index 93538e567bda..4051639c0900 100644 --- a/scripts/kconfig/zconf.gperf +++ b/scripts/kconfig/zconf.gperf @@ -41,4 +41,5 @@ option, T_OPTION, TF_COMMAND on, T_ON, TF_PARAM modules, T_OPT_MODULES, TF_OPTION defconfig_list, T_OPT_DEFCONFIG_LIST,TF_OPTION +env, T_OPT_ENV, TF_OPTION %% diff --git a/scripts/kconfig/zconf.hash.c_shipped b/scripts/kconfig/zconf.hash.c_shipped index ab28b18153a7..c9d26bc1a214 100644 --- a/scripts/kconfig/zconf.hash.c_shipped +++ b/scripts/kconfig/zconf.hash.c_shipped @@ -53,10 +53,10 @@ kconf_id_hash (register const char *str, register unsigned int len) 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, - 49, 49, 49, 49, 49, 49, 49, 18, 11, 5, + 49, 49, 49, 49, 49, 49, 49, 35, 35, 5, 0, 0, 5, 49, 5, 20, 49, 49, 5, 20, - 5, 0, 30, 49, 0, 15, 0, 10, 49, 49, - 25, 49, 49, 49, 49, 49, 49, 49, 49, 49, + 5, 0, 30, 49, 0, 15, 0, 10, 0, 49, + 10, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, @@ -89,6 +89,7 @@ kconf_id_hash (register const char *str, register unsigned int len) struct kconf_id_strings_t { char kconf_id_strings_str2[sizeof("on")]; + char kconf_id_strings_str3[sizeof("env")]; char kconf_id_strings_str5[sizeof("endif")]; char kconf_id_strings_str6[sizeof("option")]; char kconf_id_strings_str7[sizeof("endmenu")]; @@ -99,30 +100,31 @@ struct kconf_id_strings_t char kconf_id_strings_str12[sizeof("default")]; char kconf_id_strings_str13[sizeof("def_bool")]; char kconf_id_strings_str14[sizeof("help")]; - char kconf_id_strings_str15[sizeof("bool")]; char kconf_id_strings_str16[sizeof("config")]; char kconf_id_strings_str17[sizeof("def_tristate")]; - char kconf_id_strings_str18[sizeof("boolean")]; + char kconf_id_strings_str18[sizeof("hex")]; char kconf_id_strings_str19[sizeof("defconfig_list")]; char kconf_id_strings_str21[sizeof("string")]; char kconf_id_strings_str22[sizeof("if")]; char kconf_id_strings_str23[sizeof("int")]; - char kconf_id_strings_str24[sizeof("enable")]; char kconf_id_strings_str26[sizeof("select")]; char kconf_id_strings_str27[sizeof("modules")]; char kconf_id_strings_str28[sizeof("tristate")]; char kconf_id_strings_str29[sizeof("menu")]; char kconf_id_strings_str31[sizeof("source")]; char kconf_id_strings_str32[sizeof("comment")]; - char kconf_id_strings_str33[sizeof("hex")]; char kconf_id_strings_str35[sizeof("menuconfig")]; char kconf_id_strings_str36[sizeof("prompt")]; char kconf_id_strings_str37[sizeof("depends")]; + char kconf_id_strings_str39[sizeof("bool")]; + char kconf_id_strings_str41[sizeof("enable")]; + char kconf_id_strings_str42[sizeof("boolean")]; char kconf_id_strings_str48[sizeof("mainmenu")]; }; static struct kconf_id_strings_t kconf_id_strings_contents = { "on", + "env", "endif", "option", "endmenu", @@ -133,25 +135,25 @@ static struct kconf_id_strings_t kconf_id_strings_contents = "default", "def_bool", "help", - "bool", "config", "def_tristate", - "boolean", + "hex", "defconfig_list", "string", "if", "int", - "enable", "select", "modules", "tristate", "menu", "source", "comment", - "hex", "menuconfig", "prompt", "depends", + "bool", + "enable", + "boolean", "mainmenu" }; #define kconf_id_strings ((const char *) &kconf_id_strings_contents) @@ -163,7 +165,7 @@ kconf_id_lookup (register const char *str, register unsigned int len) { enum { - TOTAL_KEYWORDS = 31, + TOTAL_KEYWORDS = 32, MIN_WORD_LENGTH = 2, MAX_WORD_LENGTH = 14, MIN_HASH_VALUE = 2, @@ -174,7 +176,8 @@ kconf_id_lookup (register const char *str, register unsigned int len) { {-1}, {-1}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str2, T_ON, TF_PARAM}, - {-1}, {-1}, + {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str3, T_OPT_ENV, TF_OPTION}, + {-1}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str5, T_ENDIF, TF_COMMAND}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str6, T_OPTION, TF_COMMAND}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str7, T_ENDMENU, TF_COMMAND}, @@ -185,17 +188,16 @@ kconf_id_lookup (register const char *str, register unsigned int len) {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str12, T_DEFAULT, TF_COMMAND, S_UNKNOWN}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str13, T_DEFAULT, TF_COMMAND, S_BOOLEAN}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str14, T_HELP, TF_COMMAND}, - {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str15, T_TYPE, TF_COMMAND, S_BOOLEAN}, + {-1}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str16, T_CONFIG, TF_COMMAND}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str17, T_DEFAULT, TF_COMMAND, S_TRISTATE}, - {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str18, T_TYPE, TF_COMMAND, S_BOOLEAN}, + {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str18, T_TYPE, TF_COMMAND, S_HEX}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str19, T_OPT_DEFCONFIG_LIST,TF_OPTION}, {-1}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str21, T_TYPE, TF_COMMAND, S_STRING}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str22, T_IF, TF_COMMAND|TF_PARAM}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str23, T_TYPE, TF_COMMAND, S_INT}, - {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str24, T_SELECT, TF_COMMAND}, - {-1}, + {-1}, {-1}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str26, T_SELECT, TF_COMMAND}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str27, T_OPT_MODULES, TF_OPTION}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str28, T_TYPE, TF_COMMAND, S_TRISTATE}, @@ -203,13 +205,16 @@ kconf_id_lookup (register const char *str, register unsigned int len) {-1}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str31, T_SOURCE, TF_COMMAND}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str32, T_COMMENT, TF_COMMAND}, - {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str33, T_TYPE, TF_COMMAND, S_HEX}, - {-1}, + {-1}, {-1}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str35, T_MENUCONFIG, TF_COMMAND}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str36, T_PROMPT, TF_COMMAND}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str37, T_DEPENDS, TF_COMMAND}, - {-1}, {-1}, {-1}, {-1}, {-1}, {-1}, {-1}, {-1}, {-1}, {-1}, + {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str39, T_TYPE, TF_COMMAND, S_BOOLEAN}, + {-1}, + {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str41, T_SELECT, TF_COMMAND}, + {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str42, T_TYPE, TF_COMMAND, S_BOOLEAN}, + {-1}, {-1}, {-1}, {-1}, {-1}, {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str48, T_MAINMENU, TF_COMMAND} }; -- cgit v1.2.3 From 9b3e4dad7e58405de398faa17cb4691ca85eb050 Mon Sep 17 00:00:00 2001 From: Sam Ravnborg Date: Mon, 28 Jan 2008 21:49:46 +0100 Subject: kconfig: document use of HAVE_* It has been discussed on lkml several times but we need it documented as this is new stuff. Signed-off-by: Sam Ravnborg --- Documentation/kbuild/kconfig-language.txt | 36 +++++++++++++++++++++++++++++++ 1 file changed, 36 insertions(+) (limited to 'Documentation') diff --git a/Documentation/kbuild/kconfig-language.txt b/Documentation/kbuild/kconfig-language.txt index a43fcc68e171..649cb8799890 100644 --- a/Documentation/kbuild/kconfig-language.txt +++ b/Documentation/kbuild/kconfig-language.txt @@ -330,6 +330,42 @@ This is a collection of Kconfig tips, most of which aren't obvious at first glance and most of which have become idioms in several Kconfig files. +Adding common features and make the usage configurable +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +It is a common idiom to implement a feature/functionality that are +relevant for some architectures but not all. +The recommended way to do so is to use a config variable named HAVE_* +that is defined in a common Kconfig file and selected by the relevant +architectures. +An example is the generic IOMAP functionality. + +We would in lib/Kconfig see: + +# Generic IOMAP is used to ... +config HAVE_GENERIC_IOMAP + +config GENERIC_IOMAP + depends on HAVE_GENERIC_IOMAP && FOO + +And in lib/Makefile we would see: +obj-$(CONFIG_GENERIC_IOMAP) += iomap.o + +For each architecture using the generic IOMAP functionality we would see: + +config X86 + select ... + select HAVE_GENERIC_IOMAP + select ... + +Note: we use the existing config option and avoid creating a new +config variable to select HAVE_GENERIC_IOMAP. + +Note: the use of the internal config variable HAVE_GENERIC_IOMAP, it is +introduced to overcome the limitation of select which will force a +config option to 'y' no matter the dependencies. +The dependencies are moved to the symbol GENERIC_IOMAP and we avoid the +situation where select forces a symbol equals to 'y'. + Build as module only ~~~~~~~~~~~~~~~~~~~~ To restrict a component build to module-only, qualify its config symbol -- cgit v1.2.3 From f7ab97f78a5c573e49474edbd260ea6898ddccda Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Fri, 16 Nov 2007 16:09:28 -0800 Subject: [CAN]: Add documentation This patch adds documentation for the PF_CAN protocol family. Signed-off-by: Oliver Hartkopp Signed-off-by: Urs Thuermann Signed-off-by: David S. Miller --- Documentation/networking/00-INDEX | 2 + Documentation/networking/can.txt | 629 ++++++++++++++++++++++++++++++++++++++ 2 files changed, 631 insertions(+) create mode 100644 Documentation/networking/can.txt (limited to 'Documentation') diff --git a/Documentation/networking/00-INDEX b/Documentation/networking/00-INDEX index 563e442f2d42..b4eefadb9adf 100644 --- a/Documentation/networking/00-INDEX +++ b/Documentation/networking/00-INDEX @@ -24,6 +24,8 @@ baycom.txt - info on the driver for Baycom style amateur radio modems bridge.txt - where to get user space programs for ethernet bridging with Linux. +can.txt + - documentation on CAN protocol family. cops.txt - info on the COPS LocalTalk Linux driver cs89x0.txt diff --git a/Documentation/networking/can.txt b/Documentation/networking/can.txt new file mode 100644 index 000000000000..f1b2de170929 --- /dev/null +++ b/Documentation/networking/can.txt @@ -0,0 +1,629 @@ +============================================================================ + +can.txt + +Readme file for the Controller Area Network Protocol Family (aka Socket CAN) + +This file contains + + 1 Overview / What is Socket CAN + + 2 Motivation / Why using the socket API + + 3 Socket CAN concept + 3.1 receive lists + 3.2 local loopback of sent frames + 3.3 network security issues (capabilities) + 3.4 network problem notifications + + 4 How to use Socket CAN + 4.1 RAW protocol sockets with can_filters (SOCK_RAW) + 4.1.1 RAW socket option CAN_RAW_FILTER + 4.1.2 RAW socket option CAN_RAW_ERR_FILTER + 4.1.3 RAW socket option CAN_RAW_LOOPBACK + 4.1.4 RAW socket option CAN_RAW_RECV_OWN_MSGS + 4.2 Broadcast Manager protocol sockets (SOCK_DGRAM) + 4.3 connected transport protocols (SOCK_SEQPACKET) + 4.4 unconnected transport protocols (SOCK_DGRAM) + + 5 Socket CAN core module + 5.1 can.ko module params + 5.2 procfs content + 5.3 writing own CAN protocol modules + + 6 CAN network drivers + 6.1 general settings + 6.2 local loopback of sent frames + 6.3 CAN controller hardware filters + 6.4 currently supported CAN hardware + 6.5 todo + + 7 Credits + +============================================================================ + +1. Overview / What is Socket CAN +-------------------------------- + +The socketcan package is an implementation of CAN protocols +(Controller Area Network) for Linux. CAN is a networking technology +which has widespread use in automation, embedded devices, and +automotive fields. While there have been other CAN implementations +for Linux based on character devices, Socket CAN uses the Berkeley +socket API, the Linux network stack and implements the CAN device +drivers as network interfaces. The CAN socket API has been designed +as similar as possible to the TCP/IP protocols to allow programmers, +familiar with network programming, to easily learn how to use CAN +sockets. + +2. Motivation / Why using the socket API +---------------------------------------- + +There have been CAN implementations for Linux before Socket CAN so the +question arises, why we have started another project. Most existing +implementations come as a device driver for some CAN hardware, they +are based on character devices and provide comparatively little +functionality. Usually, there is only a hardware-specific device +driver which provides a character device interface to send and +receive raw CAN frames, directly to/from the controller hardware. +Queueing of frames and higher-level transport protocols like ISO-TP +have to be implemented in user space applications. Also, most +character-device implementations support only one single process to +open the device at a time, similar to a serial interface. Exchanging +the CAN controller requires employment of another device driver and +often the need for adaption of large parts of the application to the +new driver's API. + +Socket CAN was designed to overcome all of these limitations. A new +protocol family has been implemented which provides a socket interface +to user space applications and which builds upon the Linux network +layer, so to use all of the provided queueing functionality. A device +driver for CAN controller hardware registers itself with the Linux +network layer as a network device, so that CAN frames from the +controller can be passed up to the network layer and on to the CAN +protocol family module and also vice-versa. Also, the protocol family +module provides an API for transport protocol modules to register, so +that any number of transport protocols can be loaded or unloaded +dynamically. In fact, the can core module alone does not provide any +protocol and cannot be used without loading at least one additional +protocol module. Multiple sockets can be opened at the same time, +on different or the same protocol module and they can listen/send +frames on different or the same CAN IDs. Several sockets listening on +the same interface for frames with the same CAN ID are all passed the +same received matching CAN frames. An application wishing to +communicate using a specific transport protocol, e.g. ISO-TP, just +selects that protocol when opening the socket, and then can read and +write application data byte streams, without having to deal with +CAN-IDs, frames, etc. + +Similar functionality visible from user-space could be provided by a +character device, too, but this would lead to a technically inelegant +solution for a couple of reasons: + +* Intricate usage. Instead of passing a protocol argument to + socket(2) and using bind(2) to select a CAN interface and CAN ID, an + application would have to do all these operations using ioctl(2)s. + +* Code duplication. A character device cannot make use of the Linux + network queueing code, so all that code would have to be duplicated + for CAN networking. + +* Abstraction. In most existing character-device implementations, the + hardware-specific device driver for a CAN controller directly + provides the character device for the application to work with. + This is at least very unusual in Unix systems for both, char and + block devices. For example you don't have a character device for a + certain UART of a serial interface, a certain sound chip in your + computer, a SCSI or IDE controller providing access to your hard + disk or tape streamer device. Instead, you have abstraction layers + which provide a unified character or block device interface to the + application on the one hand, and a interface for hardware-specific + device drivers on the other hand. These abstractions are provided + by subsystems like the tty layer, the audio subsystem or the SCSI + and IDE subsystems for the devices mentioned above. + + The easiest way to implement a CAN device driver is as a character + device without such a (complete) abstraction layer, as is done by most + existing drivers. The right way, however, would be to add such a + layer with all the functionality like registering for certain CAN + IDs, supporting several open file descriptors and (de)multiplexing + CAN frames between them, (sophisticated) queueing of CAN frames, and + providing an API for device drivers to register with. However, then + it would be no more difficult, or may be even easier, to use the + networking framework provided by the Linux kernel, and this is what + Socket CAN does. + + The use of the networking framework of the Linux kernel is just the + natural and most appropriate way to implement CAN for Linux. + +3. Socket CAN concept +--------------------- + + As described in chapter 2 it is the main goal of Socket CAN to + provide a socket interface to user space applications which builds + upon the Linux network layer. In contrast to the commonly known + TCP/IP and ethernet networking, the CAN bus is a broadcast-only(!) + medium that has no MAC-layer addressing like ethernet. The CAN-identifier + (can_id) is used for arbitration on the CAN-bus. Therefore the CAN-IDs + have to be chosen uniquely on the bus. When designing a CAN-ECU + network the CAN-IDs are mapped to be sent by a specific ECU. + For this reason a CAN-ID can be treated best as a kind of source address. + + 3.1 receive lists + + The network transparent access of multiple applications leads to the + problem that different applications may be interested in the same + CAN-IDs from the same CAN network interface. The Socket CAN core + module - which implements the protocol family CAN - provides several + high efficient receive lists for this reason. If e.g. a user space + application opens a CAN RAW socket, the raw protocol module itself + requests the (range of) CAN-IDs from the Socket CAN core that are + requested by the user. The subscription and unsubscription of + CAN-IDs can be done for specific CAN interfaces or for all(!) known + CAN interfaces with the can_rx_(un)register() functions provided to + CAN protocol modules by the SocketCAN core (see chapter 5). + To optimize the CPU usage at runtime the receive lists are split up + into several specific lists per device that match the requested + filter complexity for a given use-case. + + 3.2 local loopback of sent frames + + As known from other networking concepts the data exchanging + applications may run on the same or different nodes without any + change (except for the according addressing information): + + ___ ___ ___ _______ ___ + | _ | | _ | | _ | | _ _ | | _ | + ||A|| ||B|| ||C|| ||A| |B|| ||C|| + |___| |___| |___| |_______| |___| + | | | | | + -----------------(1)- CAN bus -(2)--------------- + + To ensure that application A receives the same information in the + example (2) as it would receive in example (1) there is need for + some kind of local loopback of the sent CAN frames on the appropriate + node. + + The Linux network devices (by default) just can handle the + transmission and reception of media dependent frames. Due to the + arbritration on the CAN bus the transmission of a low prio CAN-ID + may be delayed by the reception of a high prio CAN frame. To + reflect the correct* traffic on the node the loopback of the sent + data has to be performed right after a successful transmission. If + the CAN network interface is not capable of performing the loopback for + some reason the SocketCAN core can do this task as a fallback solution. + See chapter 6.2 for details (recommended). + + The loopback functionality is enabled by default to reflect standard + networking behaviour for CAN applications. Due to some requests from + the RT-SocketCAN group the loopback optionally may be disabled for each + separate socket. See sockopts from the CAN RAW sockets in chapter 4.1. + + * = you really like to have this when you're running analyser tools + like 'candump' or 'cansniffer' on the (same) node. + + 3.3 network security issues (capabilities) + + The Controller Area Network is a local field bus transmitting only + broadcast messages without any routing and security concepts. + In the majority of cases the user application has to deal with + raw CAN frames. Therefore it might be reasonable NOT to restrict + the CAN access only to the user root, as known from other networks. + Since the currently implemented CAN_RAW and CAN_BCM sockets can only + send and receive frames to/from CAN interfaces it does not affect + security of others networks to allow all users to access the CAN. + To enable non-root users to access CAN_RAW and CAN_BCM protocol + sockets the Kconfig options CAN_RAW_USER and/or CAN_BCM_USER may be + selected at kernel compile time. + + 3.4 network problem notifications + + The use of the CAN bus may lead to several problems on the physical + and media access control layer. Detecting and logging of these lower + layer problems is a vital requirement for CAN users to identify + hardware issues on the physical transceiver layer as well as + arbitration problems and error frames caused by the different + ECUs. The occurrence of detected errors are important for diagnosis + and have to be logged together with the exact timestamp. For this + reason the CAN interface driver can generate so called Error Frames + that can optionally be passed to the user application in the same + way as other CAN frames. Whenever an error on the physical layer + or the MAC layer is detected (e.g. by the CAN controller) the driver + creates an appropriate error frame. Error frames can be requested by + the user application using the common CAN filter mechanisms. Inside + this filter definition the (interested) type of errors may be + selected. The reception of error frames is disabled by default. + +4. How to use Socket CAN +------------------------ + + Like TCP/IP, you first need to open a socket for communicating over a + CAN network. Since Socket CAN implements a new protocol family, you + need to pass PF_CAN as the first argument to the socket(2) system + call. Currently, there are two CAN protocols to choose from, the raw + socket protocol and the broadcast manager (BCM). So to open a socket, + you would write + + s = socket(PF_CAN, SOCK_RAW, CAN_RAW); + + and + + s = socket(PF_CAN, SOCK_DGRAM, CAN_BCM); + + respectively. After the successful creation of the socket, you would + normally use the bind(2) system call to bind the socket to a CAN + interface (which is different from TCP/IP due to different addressing + - see chapter 3). After binding (CAN_RAW) or connecting (CAN_BCM) + the socket, you can read(2) and write(2) from/to the socket or use + send(2), sendto(2), sendmsg(2) and the recv* counterpart operations + on the socket as usual. There are also CAN specific socket options + described below. + + The basic CAN frame structure and the sockaddr structure are defined + in include/linux/can.h: + + struct can_frame { + canid_t can_id; /* 32 bit CAN_ID + EFF/RTR/ERR flags */ + __u8 can_dlc; /* data length code: 0 .. 8 */ + __u8 data[8] __attribute__((aligned(8))); + }; + + The alignment of the (linear) payload data[] to a 64bit boundary + allows the user to define own structs and unions to easily access the + CAN payload. There is no given byteorder on the CAN bus by + default. A read(2) system call on a CAN_RAW socket transfers a + struct can_frame to the user space. + + The sockaddr_can structure has an interface index like the + PF_PACKET socket, that also binds to a specific interface: + + struct sockaddr_can { + sa_family_t can_family; + int can_ifindex; + union { + struct { canid_t rx_id, tx_id; } tp16; + struct { canid_t rx_id, tx_id; } tp20; + struct { canid_t rx_id, tx_id; } mcnet; + struct { canid_t rx_id, tx_id; } isotp; + } can_addr; + }; + + To determine the interface index an appropriate ioctl() has to + be used (example for CAN_RAW sockets without error checking): + + int s; + struct sockaddr_can addr; + struct ifreq ifr; + + s = socket(PF_CAN, SOCK_RAW, CAN_RAW); + + strcpy(ifr.ifr_name, "can0" ); + ioctl(s, SIOCGIFINDEX, &ifr); + + addr.can_family = AF_CAN; + addr.can_ifindex = ifr.ifr_ifindex; + + bind(s, (struct sockaddr *)&addr, sizeof(addr)); + + (..) + + To bind a socket to all(!) CAN interfaces the interface index must + be 0 (zero). In this case the socket receives CAN frames from every + enabled CAN interface. To determine the originating CAN interface + the system call recvfrom(2) may be used instead of read(2). To send + on a socket that is bound to 'any' interface sendto(2) is needed to + specify the outgoing interface. + + Reading CAN frames from a bound CAN_RAW socket (see above) consists + of reading a struct can_frame: + + struct can_frame frame; + + nbytes = read(s, &frame, sizeof(struct can_frame)); + + if (nbytes < 0) { + perror("can raw socket read"); + return 1; + } + + /* paraniod check ... */ + if (nbytes < sizeof(struct can_frame)) { + fprintf(stderr, "read: incomplete CAN frame\n"); + return 1; + } + + /* do something with the received CAN frame */ + + Writing CAN frames can be done similarly, with the write(2) system call: + + nbytes = write(s, &frame, sizeof(struct can_frame)); + + When the CAN interface is bound to 'any' existing CAN interface + (addr.can_ifindex = 0) it is recommended to use recvfrom(2) if the + information about the originating CAN interface is needed: + + struct sockaddr_can addr; + struct ifreq ifr; + socklen_t len = sizeof(addr); + struct can_frame frame; + + nbytes = recvfrom(s, &frame, sizeof(struct can_frame), + 0, (struct sockaddr*)&addr, &len); + + /* get interface name of the received CAN frame */ + ifr.ifr_ifindex = addr.can_ifindex; + ioctl(s, SIOCGIFNAME, &ifr); + printf("Received a CAN frame from interface %s", ifr.ifr_name); + + To write CAN frames on sockets bound to 'any' CAN interface the + outgoing interface has to be defined certainly. + + strcpy(ifr.ifr_name, "can0"); + ioctl(s, SIOCGIFINDEX, &ifr); + addr.can_ifindex = ifr.ifr_ifindex; + addr.can_family = AF_CAN; + + nbytes = sendto(s, &frame, sizeof(struct can_frame), + 0, (struct sockaddr*)&addr, sizeof(addr)); + + 4.1 RAW protocol sockets with can_filters (SOCK_RAW) + + Using CAN_RAW sockets is extensively comparable to the commonly + known access to CAN character devices. To meet the new possibilities + provided by the multi user SocketCAN approach, some reasonable + defaults are set at RAW socket binding time: + + - The filters are set to exactly one filter receiving everything + - The socket only receives valid data frames (=> no error frames) + - The loopback of sent CAN frames is enabled (see chapter 3.2) + - The socket does not receive its own sent frames (in loopback mode) + + These default settings may be changed before or after binding the socket. + To use the referenced definitions of the socket options for CAN_RAW + sockets, include . + + 4.1.1 RAW socket option CAN_RAW_FILTER + + The reception of CAN frames using CAN_RAW sockets can be controlled + by defining 0 .. n filters with the CAN_RAW_FILTER socket option. + + The CAN filter structure is defined in include/linux/can.h: + + struct can_filter { + canid_t can_id; + canid_t can_mask; + }; + + A filter matches, when + + & mask == can_id & mask + + which is analogous to known CAN controllers hardware filter semantics. + The filter can be inverted in this semantic, when the CAN_INV_FILTER + bit is set in can_id element of the can_filter structure. In + contrast to CAN controller hardware filters the user may set 0 .. n + receive filters for each open socket separately: + + struct can_filter rfilter[2]; + + rfilter[0].can_id = 0x123; + rfilter[0].can_mask = CAN_SFF_MASK; + rfilter[1].can_id = 0x200; + rfilter[1].can_mask = 0x700; + + setsockopt(s, SOL_CAN_RAW, CAN_RAW_FILTER, &rfilter, sizeof(rfilter)); + + To disable the reception of CAN frames on the selected CAN_RAW socket: + + setsockopt(s, SOL_CAN_RAW, CAN_RAW_FILTER, NULL, 0); + + To set the filters to zero filters is quite obsolete as not read + data causes the raw socket to discard the received CAN frames. But + having this 'send only' use-case we may remove the receive list in the + Kernel to save a little (really a very little!) CPU usage. + + 4.1.2 RAW socket option CAN_RAW_ERR_FILTER + + As described in chapter 3.4 the CAN interface driver can generate so + called Error Frames that can optionally be passed to the user + application in the same way as other CAN frames. The possible + errors are divided into different error classes that may be filtered + using the appropriate error mask. To register for every possible + error condition CAN_ERR_MASK can be used as value for the error mask. + The values for the error mask are defined in linux/can/error.h . + + can_err_mask_t err_mask = ( CAN_ERR_TX_TIMEOUT | CAN_ERR_BUSOFF ); + + setsockopt(s, SOL_CAN_RAW, CAN_RAW_ERR_FILTER, + &err_mask, sizeof(err_mask)); + + 4.1.3 RAW socket option CAN_RAW_LOOPBACK + + To meet multi user needs the local loopback is enabled by default + (see chapter 3.2 for details). But in some embedded use-cases + (e.g. when only one application uses the CAN bus) this loopback + functionality can be disabled (separately for each socket): + + int loopback = 0; /* 0 = disabled, 1 = enabled (default) */ + + setsockopt(s, SOL_CAN_RAW, CAN_RAW_LOOPBACK, &loopback, sizeof(loopback)); + + 4.1.4 RAW socket option CAN_RAW_RECV_OWN_MSGS + + When the local loopback is enabled, all the sent CAN frames are + looped back to the open CAN sockets that registered for the CAN + frames' CAN-ID on this given interface to meet the multi user + needs. The reception of the CAN frames on the same socket that was + sending the CAN frame is assumed to be unwanted and therefore + disabled by default. This default behaviour may be changed on + demand: + + int recv_own_msgs = 1; /* 0 = disabled (default), 1 = enabled */ + + setsockopt(s, SOL_CAN_RAW, CAN_RAW_RECV_OWN_MSGS, + &recv_own_msgs, sizeof(recv_own_msgs)); + + 4.2 Broadcast Manager protocol sockets (SOCK_DGRAM) + 4.3 connected transport protocols (SOCK_SEQPACKET) + 4.4 unconnected transport protocols (SOCK_DGRAM) + + +5. Socket CAN core module +------------------------- + + The Socket CAN core module implements the protocol family + PF_CAN. CAN protocol modules are loaded by the core module at + runtime. The core module provides an interface for CAN protocol + modules to subscribe needed CAN IDs (see chapter 3.1). + + 5.1 can.ko module params + + - stats_timer: To calculate the Socket CAN core statistics + (e.g. current/maximum frames per second) this 1 second timer is + invoked at can.ko module start time by default. This timer can be + disabled by using stattimer=0 on the module comandline. + + - debug: (removed since SocketCAN SVN r546) + + 5.2 procfs content + + As described in chapter 3.1 the Socket CAN core uses several filter + lists to deliver received CAN frames to CAN protocol modules. These + receive lists, their filters and the count of filter matches can be + checked in the appropriate receive list. All entries contain the + device and a protocol module identifier: + + foo@bar:~$ cat /proc/net/can/rcvlist_all + + receive list 'rx_all': + (vcan3: no entry) + (vcan2: no entry) + (vcan1: no entry) + device can_id can_mask function userdata matches ident + vcan0 000 00000000 f88e6370 f6c6f400 0 raw + (any: no entry) + + In this example an application requests any CAN traffic from vcan0. + + rcvlist_all - list for unfiltered entries (no filter operations) + rcvlist_eff - list for single extended frame (EFF) entries + rcvlist_err - list for error frames masks + rcvlist_fil - list for mask/value filters + rcvlist_inv - list for mask/value filters (inverse semantic) + rcvlist_sff - list for single standard frame (SFF) entries + + Additional procfs files in /proc/net/can + + stats - Socket CAN core statistics (rx/tx frames, match ratios, ...) + reset_stats - manual statistic reset + version - prints the Socket CAN core version and the ABI version + + 5.3 writing own CAN protocol modules + + To implement a new protocol in the protocol family PF_CAN a new + protocol has to be defined in include/linux/can.h . + The prototypes and definitions to use the Socket CAN core can be + accessed by including include/linux/can/core.h . + In addition to functions that register the CAN protocol and the + CAN device notifier chain there are functions to subscribe CAN + frames received by CAN interfaces and to send CAN frames: + + can_rx_register - subscribe CAN frames from a specific interface + can_rx_unregister - unsubscribe CAN frames from a specific interface + can_send - transmit a CAN frame (optional with local loopback) + + For details see the kerneldoc documentation in net/can/af_can.c or + the source code of net/can/raw.c or net/can/bcm.c . + +6. CAN network drivers +---------------------- + + Writing a CAN network device driver is much easier than writing a + CAN character device driver. Similar to other known network device + drivers you mainly have to deal with: + + - TX: Put the CAN frame from the socket buffer to the CAN controller. + - RX: Put the CAN frame from the CAN controller to the socket buffer. + + See e.g. at Documentation/networking/netdevices.txt . The differences + for writing CAN network device driver are described below: + + 6.1 general settings + + dev->type = ARPHRD_CAN; /* the netdevice hardware type */ + dev->flags = IFF_NOARP; /* CAN has no arp */ + + dev->mtu = sizeof(struct can_frame); + + The struct can_frame is the payload of each socket buffer in the + protocol family PF_CAN. + + 6.2 local loopback of sent frames + + As described in chapter 3.2 the CAN network device driver should + support a local loopback functionality similar to the local echo + e.g. of tty devices. In this case the driver flag IFF_ECHO has to be + set to prevent the PF_CAN core from locally echoing sent frames + (aka loopback) as fallback solution: + + dev->flags = (IFF_NOARP | IFF_ECHO); + + 6.3 CAN controller hardware filters + + To reduce the interrupt load on deep embedded systems some CAN + controllers support the filtering of CAN IDs or ranges of CAN IDs. + These hardware filter capabilities vary from controller to + controller and have to be identified as not feasible in a multi-user + networking approach. The use of the very controller specific + hardware filters could make sense in a very dedicated use-case, as a + filter on driver level would affect all users in the multi-user + system. The high efficient filter sets inside the PF_CAN core allow + to set different multiple filters for each socket separately. + Therefore the use of hardware filters goes to the category 'handmade + tuning on deep embedded systems'. The author is running a MPC603e + @133MHz with four SJA1000 CAN controllers from 2002 under heavy bus + load without any problems ... + + 6.4 currently supported CAN hardware (September 2007) + + On the project website http://developer.berlios.de/projects/socketcan + there are different drivers available: + + vcan: Virtual CAN interface driver (if no real hardware is available) + sja1000: Philips SJA1000 CAN controller (recommended) + i82527: Intel i82527 CAN controller + mscan: Motorola/Freescale CAN controller (e.g. inside SOC MPC5200) + ccan: CCAN controller core (e.g. inside SOC h7202) + slcan: For a bunch of CAN adaptors that are attached via a + serial line ASCII protocol (for serial / USB adaptors) + + Additionally the different CAN adaptors (ISA/PCI/PCMCIA/USB/Parport) + from PEAK Systemtechnik support the CAN netdevice driver model + since Linux driver v6.0: http://www.peak-system.com/linux/index.htm + + Please check the Mailing Lists on the berlios OSS project website. + + 6.5 todo (September 2007) + + The configuration interface for CAN network drivers is still an open + issue that has not been finalized in the socketcan project. Also the + idea of having a library module (candev.ko) that holds functions + that are needed by all CAN netdevices is not ready to ship. + Your contribution is welcome. + +7. Credits +---------- + + Oliver Hartkopp (PF_CAN core, filters, drivers, bcm) + Urs Thuermann (PF_CAN core, kernel integration, socket interfaces, raw, vcan) + Jan Kizka (RT-SocketCAN core, Socket-API reconciliation) + Wolfgang Grandegger (RT-SocketCAN core & drivers, Raw Socket-API reviews) + Robert Schwebel (design reviews, PTXdist integration) + Marc Kleine-Budde (design reviews, Kernel 2.6 cleanups, drivers) + Benedikt Spranger (reviews) + Thomas Gleixner (LKML reviews, coding style, posting hints) + Andrey Volkov (kernel subtree structure, ioctls, mscan driver) + Matthias Brukner (first SJA1000 CAN netdevice implementation Q2/2003) + Klaus Hitschler (PEAK driver integration) + Uwe Koppe (CAN netdevices with PF_PACKET approach) + Michael Schulze (driver layer loopback requirement, RT CAN drivers review) -- cgit v1.2.3 From 8e8c71f1ab0ca1c4e74efad14533b991524dcb6c Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Wed, 21 Nov 2007 09:56:48 -0200 Subject: [DCCP]: Honour and make use of shutdown option set by user This extends the DCCP socket API by honouring any shutdown(2) option set by the user. The behaviour is, as much as possible, made consistent with the API for TCP's shutdown. This patch exploits the information provided by the user via the socket API to reduce processing costs: * if the read end is closed (SHUT_RD), it is not necessary to deliver to input CCID; * if the write end is closed (SHUT_WR), the same idea applies, but with a difference - as long as the TX queue has not been drained, we need to receive feedback to keep congestion-control rates up to date. Hence SHUT_WR is honoured only after the last packet (under congestion control) has been sent; * although SHUT_RDWR seems nonsensical, it is nevertheless supported in the same manner as for TCP (and agrees with test for SHUTDOWN_MASK in dccp_poll() in net/dccp/proto.c). Furthermore, most of the code already honours the sk_shutdown flags (dccp_recvmsg() for instance sets the read length to 0 if SHUT_RD had been called); CCID handling is now added to this by the present patch. There will also no longer be any delivery when the socket is in the final stages, i.e. when one of dccp_close(), dccp_fin(), or dccp_done() has been called - which is fine since at that stage the connection is its final stages. Motivation and background are on http://www.erg.abdn.ac.uk/users/gerrit/dccp/notes/shutdown A FIXME has been added to notify the other end if SHUT_RD has been set (RFC 4340, 11.7). Note: There is a comment in inet_shutdown() in net/ipv4/af_inet.c which asks to "make sure the socket is a TCP socket". This should probably be extended to mean `TCP or DCCP socket' (the code is also used by UDP and raw sockets). Signed-off-by: Gerrit Renker Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- Documentation/networking/dccp.txt | 2 ++ net/dccp/input.c | 27 ++++++++++++++++++++------- net/dccp/proto.c | 2 +- 3 files changed, 23 insertions(+), 8 deletions(-) (limited to 'Documentation') diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index afb66f9a8aff..f771034e8f27 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -72,6 +72,8 @@ DCCP_SOCKOPT_CCID_TX_INFO Returns a `struct tfrc_tx_info' in optval; the buffer for optval and optlen must be set to at least sizeof(struct tfrc_tx_info). +On unidirectional connections it is useful to close the unused half-connection +via shutdown (SHUT_WR or SHUT_RD): this will reduce per-packet processing costs. Sysctl variables ================ diff --git a/net/dccp/input.c b/net/dccp/input.c index 1ce101062824..df0fb2c149a6 100644 --- a/net/dccp/input.c +++ b/net/dccp/input.c @@ -103,6 +103,21 @@ static void dccp_event_ack_recv(struct sock *sk, struct sk_buff *skb) DCCP_SKB_CB(skb)->dccpd_ack_seq); } +static void dccp_deliver_input_to_ccids(struct sock *sk, struct sk_buff *skb) +{ + const struct dccp_sock *dp = dccp_sk(sk); + + /* Don't deliver to RX CCID when node has shut down read end. */ + if (!(sk->sk_shutdown & RCV_SHUTDOWN)) + ccid_hc_rx_packet_recv(dp->dccps_hc_rx_ccid, sk, skb); + /* + * Until the TX queue has been drained, we can not honour SHUT_WR, since + * we need received feedback as input to adjust congestion control. + */ + if (sk->sk_write_queue.qlen > 0 || !(sk->sk_shutdown & SEND_SHUTDOWN)) + ccid_hc_tx_packet_recv(dp->dccps_hc_tx_ccid, sk, skb); +} + static int dccp_check_seqno(struct sock *sk, struct sk_buff *skb) { const struct dccp_hdr *dh = dccp_hdr(skb); @@ -209,8 +224,9 @@ static int __dccp_rcv_established(struct sock *sk, struct sk_buff *skb, case DCCP_PKT_DATAACK: case DCCP_PKT_DATA: /* - * FIXME: check if sk_receive_queue is full, schedule DATA_DROPPED - * option if it is. + * FIXME: schedule DATA_DROPPED (RFC 4340, 11.7.2) if and when + * - sk_shutdown == RCV_SHUTDOWN, use Code 1, "Not Listening" + * - sk_receive_queue is full, use Code 2, "Receive Buffer" */ __skb_pull(skb, dh->dccph_doff * 4); __skb_queue_tail(&sk->sk_receive_queue, skb); @@ -300,9 +316,7 @@ int dccp_rcv_established(struct sock *sk, struct sk_buff *skb, DCCP_SKB_CB(skb)->dccpd_seq, DCCP_ACKVEC_STATE_RECEIVED)) goto discard; - - ccid_hc_rx_packet_recv(dp->dccps_hc_rx_ccid, sk, skb); - ccid_hc_tx_packet_recv(dp->dccps_hc_tx_ccid, sk, skb); + dccp_deliver_input_to_ccids(sk, skb); return __dccp_rcv_established(sk, skb, dh, len); discard: @@ -543,8 +557,7 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, DCCP_ACKVEC_STATE_RECEIVED)) goto discard; - ccid_hc_rx_packet_recv(dp->dccps_hc_rx_ccid, sk, skb); - ccid_hc_tx_packet_recv(dp->dccps_hc_tx_ccid, sk, skb); + dccp_deliver_input_to_ccids(sk, skb); } /* diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 7a3bea9c28c1..0aec73592124 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -981,7 +981,7 @@ EXPORT_SYMBOL_GPL(dccp_close); void dccp_shutdown(struct sock *sk, int how) { - dccp_pr_debug("entry\n"); + dccp_pr_debug("called shutdown(%x)\n", how); } EXPORT_SYMBOL_GPL(dccp_shutdown); -- cgit v1.2.3 From ebe6f7e73c3efec1de295205806b4550fcb468cd Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Wed, 21 Nov 2007 10:00:17 -0200 Subject: [DCCP]: Update documentation This updates the DCCP documentation, following input from Ian McDonald, clarifiying the status of DCCP, and adding a note about the test tree. Signed-off-by: Gerrit Renker Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- Documentation/networking/dccp.txt | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) (limited to 'Documentation') diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index f771034e8f27..fdc93beec057 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -19,19 +19,23 @@ for real time and multimedia traffic. It has a base protocol and pluggable congestion control IDs (CCIDs). -It is at proposed standard RFC status and the homepage for DCCP as a protocol -is at: - http://www.read.cs.ucla.edu/dccp/ +DCCP is a Proposed Standard (RFC 2026), and the homepage for DCCP as a protocol +is at http://www.ietf.org/html.charters/dccp-charter.html Missing features ================ -The DCCP implementation does not currently have all the features that are in -the RFC. +The Linux DCCP implementation does not currently support all the features that are +specified in RFCs 4340...42. The known bugs are at: http://linux-net.osdl.org/index.php/TODO#DCCP +For more up-to-date versions of the DCCP implementation, please consider using +the experimental DCCP test tree; instructions for checking this out are on: +http://linux-net.osdl.org/index.php/DCCP_Testing#Experimental_DCCP_source_tree + + Socket options ============== -- cgit v1.2.3 From e333b3edc489151afda2a4f6c798842c64cb67a4 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Wed, 21 Nov 2007 10:09:56 -0200 Subject: [DCCP]: Promote CCID2 as default CCID This patch addresses the following problems: 1. DCCP relies for its proper functioning on having at least one CCID module enabled (as in TCP plugable congestion control). Currently it is possible to disable both CCIDs and thus leave the DCCP module in a compiled, but entirely non-functional state: no sockets can be created when no CCID is available. Furthermore, the protocol is (again like TCP) not intended to be used without CCIDs. Last, a non-empty CCID list is needed for doing CCID feature negotiation. 2. Internally the default CCID that is advertised by the Linux host is set to CCID2 (DCCPF_INITIAL_CCID in include/linux/dccp.h). Disabling CCID2 in the Kconfig menu without changing the defaults leads to a failure `module not found' when trying to load the dccp module (which internally tries to load the default CCID). 3. The specification (RFC 4340, sec. 10) treats CCID2 somewhat like a `minimum common denominator'; the specification says that: * "New connections start with CCID 2 for both endpoints" * "A DCCP implementation intended for general use, such as an implementation in a general-purpose operating system kernel, SHOULD implement at least CCID 2. The intent is to make CCID 2 broadly available for interoperability [...]" Providing CCID2 as minimum-required CCID (like Reno/Cubic in TCP) thus seems reasonable. Hence this patch automatically selects CCID2 when DCCP is enabled. Documentation also added. Discussions with Ian McDonald on this subject are gratefully acknowledged. Signed-off-by: Gerrit Renker Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- Documentation/networking/dccp.txt | 11 +++++++++-- net/dccp/Kconfig | 1 + net/dccp/ccids/Kconfig | 13 ++----------- 3 files changed, 12 insertions(+), 13 deletions(-) (limited to 'Documentation') diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index fdc93beec057..ffb9ca937d65 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -14,8 +14,15 @@ Introduction ============ Datagram Congestion Control Protocol (DCCP) is an unreliable, connection -based protocol designed to solve issues present in UDP and TCP particularly -for real time and multimedia traffic. +oriented protocol designed to solve issues present in UDP and TCP, particularly +for real-time and multimedia (streaming) traffic. +It divides into a base protocol (RFC 4340) and plugable congestion control +modules called CCIDs. Like plugable TCP congestion control, at least one CCID +needs to be enabled in order for the protocol to function properly. In the Linux +implementation, this is the TCP-like CCID2 (RFC 4341). Additional CCIDs, such as +the TCP-friendly CCID3 (RFC 4342), are optional. +For a brief introduction to CCIDs and suggestions for choosing a CCID to match +given applications, see section 10 of RFC 4340. It has a base protocol and pluggable congestion control IDs (CCIDs). diff --git a/net/dccp/Kconfig b/net/dccp/Kconfig index 0549e4719b13..7aa2a7acc7ec 100644 --- a/net/dccp/Kconfig +++ b/net/dccp/Kconfig @@ -1,6 +1,7 @@ menuconfig IP_DCCP tristate "The DCCP Protocol (EXPERIMENTAL)" depends on INET && EXPERIMENTAL + select IP_DCCP_CCID2 ---help--- Datagram Congestion Control Protocol (RFC 4340) diff --git a/net/dccp/ccids/Kconfig b/net/dccp/ccids/Kconfig index 80f469887691..f37af17556ba 100644 --- a/net/dccp/ccids/Kconfig +++ b/net/dccp/ccids/Kconfig @@ -20,18 +20,9 @@ config IP_DCCP_CCID2 to the user. For example, a hypothetical application that transferred files over DCCP, using application-level retransmissions for lost packets, would prefer CCID 2 to CCID 3. On-line games may - also prefer CCID 2. + also prefer CCID 2. See RFC 4341 for further details. - CCID 2 is further described in RFC 4341, - http://www.ietf.org/rfc/rfc4341.txt - - This text was extracted from RFC 4340 (sec. 10.1), - http://www.ietf.org/rfc/rfc4340.txt - - To compile this CCID as a module, choose M here: the module will be - called dccp_ccid2. - - If in doubt, say M. + CCID2 is the default CCID used by DCCP. config IP_DCCP_CCID2_DEBUG bool "CCID2 debugging messages" -- cgit v1.2.3 From c28149016c24f4399c0a1eb0ebc15c92611223f0 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Wed, 21 Nov 2007 10:14:31 -0200 Subject: [DCCP]: Update documentation on ioctls Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- Documentation/networking/dccp.txt | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'Documentation') diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index ffb9ca937d65..d76905a5a087 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -136,6 +136,12 @@ sync_ratelimit = 125 ms sequence-invalid packets on the same socket (RFC 4340, 7.5.4). The unit of this parameter is milliseconds; a value of 0 disables rate-limiting. +IOCTLS +====== +FIONREAD + Works as in udp(7): returns in the `int' argument pointer the size of + the next pending datagram in bytes, or 0 when no datagram is pending. + Notes ===== -- cgit v1.2.3 From 003faaa17793c478ed2babc56a618396e0ef96c2 Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Sun, 27 Jan 2008 22:48:37 -0800 Subject: bcm43xx: mark as obsolete and schedule for removal Schedule bcm43xx for for removal in the 2.6.26 development window. Signed-off-by: John W. Linville Signed-off-by: David S. Miller --- Documentation/feature-removal-schedule.txt | 9 +++++++++ MAINTAINERS | 2 +- drivers/net/wireless/bcm43xx/Kconfig | 9 ++++++--- 3 files changed, 16 insertions(+), 4 deletions(-) (limited to 'Documentation') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 25370662cc5d..8d5f9f8b7a78 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -306,3 +306,12 @@ Why: These drivers are superseded by i810fb, intelfb and savagefb. Who: Jean Delvare --------------------------- + +What: bcm43xx wireless network driver +When: 2.6.26 +Files: drivers/net/wireless/bcm43xx +Why: This driver's functionality has been replaced by the + mac80211-based b43 and b43legacy drivers. +Who: John W. Linville + +--------------------------- diff --git a/MAINTAINERS b/MAINTAINERS index 6de909fe2013..2c848b46dbac 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -809,7 +809,7 @@ P: Stefano Brivio M: stefano.brivio@polimi.it L: linux-wireless@vger.kernel.org W: http://bcm43xx.berlios.de/ -S: Maintained +S: Obsolete BEFS FILE SYSTEM P: Sergey S. Kostyliov diff --git a/drivers/net/wireless/bcm43xx/Kconfig b/drivers/net/wireless/bcm43xx/Kconfig index ce397e4284f4..0159701e8456 100644 --- a/drivers/net/wireless/bcm43xx/Kconfig +++ b/drivers/net/wireless/bcm43xx/Kconfig @@ -1,12 +1,15 @@ config BCM43XX - tristate "Broadcom BCM43xx wireless support" + tristate "Broadcom BCM43xx wireless support (DEPRECATED)" depends on PCI && IEEE80211 && IEEE80211_SOFTMAC && WLAN_80211 && EXPERIMENTAL select WIRELESS_EXT select FW_LOADER select HW_RANDOM ---help--- - This is an experimental driver for the Broadcom 43xx wireless chip, - found in the Apple Airport Extreme and various other devices. + This is an experimental driver for the Broadcom 43xx wireless + chip, found in the Apple Airport Extreme and various other + devices. This driver is deprecated and will be removed + from the kernel in the near future. It has been replaced + by the b43 and b43legacy drivers. config BCM43XX_DEBUG bool "Broadcom BCM43xx debugging (RECOMMENDED)" -- cgit v1.2.3 From edae58ead57319b0bcf16e48a32365145c7bf0b7 Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Wed, 21 Nov 2007 15:24:35 -0500 Subject: softmac: mark as obsolete and schedule for removal Schedule softmac for for removal in the 2.6.26 development window. Signed-off-by: John W. Linville Signed-off-by: David S. Miller --- Documentation/feature-removal-schedule.txt | 8 ++++++++ net/ieee80211/Kconfig | 5 +++-- 2 files changed, 11 insertions(+), 2 deletions(-) (limited to 'Documentation') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 8d5f9f8b7a78..ec50a2ee7ddd 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -315,3 +315,11 @@ Why: This driver's functionality has been replaced by the Who: John W. Linville --------------------------- + +What: iee80211 softmac wireless networking component +When: 2.6.26 (or after removal of bcm43xx and port of zd1211rw to mac80211) +Files: net/ieee80211/softmac +Why: No in-kernel drivers will depend on it any longer. +Who: John W. Linville + +--------------------------- diff --git a/net/ieee80211/Kconfig b/net/ieee80211/Kconfig index 1438adedbc83..bd501046c9c0 100644 --- a/net/ieee80211/Kconfig +++ b/net/ieee80211/Kconfig @@ -1,8 +1,9 @@ config IEEE80211 - tristate "Generic IEEE 802.11 Networking Stack" + tristate "Generic IEEE 802.11 Networking Stack (DEPRECATED)" ---help--- This option enables the hardware independent IEEE 802.11 - networking stack. + networking stack. This component is deprecated in favor of the + mac80211 component. config IEEE80211_DEBUG bool "Enable full debugging output" -- cgit v1.2.3 From cb75994ec311b2cd50e5205efdcc0696abd6675d Mon Sep 17 00:00:00 2001 From: Wang Chen Date: Mon, 3 Dec 2007 22:33:28 +1100 Subject: [UDP]: Defer InDataGrams increment until recvmsg() does checksum Thanks dave, herbert, gerrit, andi and other people for your discussion about this problem. UdpInDatagrams can be confusing because it counts packets that might be dropped later. Move UdpInDatagrams into recvmsg() as allowed by the RFC. Signed-off-by: Wang Chen Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- Documentation/networking/udplite.txt | 2 +- net/ipv4/udp.c | 7 +++---- net/ipv6/udp.c | 4 +++- 3 files changed, 7 insertions(+), 6 deletions(-) (limited to 'Documentation') diff --git a/Documentation/networking/udplite.txt b/Documentation/networking/udplite.txt index b6409cab075c..3870f280280b 100644 --- a/Documentation/networking/udplite.txt +++ b/Documentation/networking/udplite.txt @@ -236,7 +236,7 @@ This displays UDP-Lite statistics variables, whose meaning is as follows. - InDatagrams: Total number of received datagrams. + InDatagrams: The total number of datagrams delivered to users. NoPorts: Number of packets received to an unknown port. These cases are counted separately (not as InErrors). diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 03c400ca14c5..3465d4ad301b 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -873,6 +873,8 @@ try_again: if (err) goto out_free; + UDP_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite); + sock_recv_timestamp(msg, sk, skb); /* Copy the address. */ @@ -966,10 +968,8 @@ int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) int ret; ret = (*up->encap_rcv)(sk, skb); - if (ret <= 0) { - UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, up->pcflag); + if (ret <= 0) return -ret; - } } /* FALLTHROUGH -- it's a UDP Packet */ @@ -1023,7 +1023,6 @@ int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) goto drop; } - UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, up->pcflag); return 0; drop: diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index ee1cc3f8599f..b0474a618bbe 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -164,6 +164,8 @@ try_again: if (err) goto out_free; + UDP6_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite); + sock_recv_timestamp(msg, sk, skb); /* Copy the address. */ @@ -292,7 +294,7 @@ int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) UDP6_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, up->pcflag); goto drop; } - UDP6_INC_STATS_BH(UDP_MIB_INDATAGRAMS, up->pcflag); + return 0; drop: UDP6_INC_STATS_BH(UDP_MIB_INERRORS, up->pcflag); -- cgit v1.2.3 From cb76c6a597350534d211ba79d92da1f9771f8226 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 4 Dec 2007 23:39:36 -0800 Subject: [NETFILTER]: ip_tables: remove obsolete SAME target Remove the ipt_SAME target as scheduled in feature-removal-schedule. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- Documentation/feature-removal-schedule.txt | 9 -- net/ipv4/netfilter/Kconfig | 9 -- net/ipv4/netfilter/Makefile | 1 - net/ipv4/netfilter/ipt_SAME.c | 174 ----------------------------- 4 files changed, 193 deletions(-) delete mode 100644 net/ipv4/netfilter/ipt_SAME.c (limited to 'Documentation') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index ec50a2ee7ddd..0927425026f1 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -249,15 +249,6 @@ Who: Tejun Heo --------------------------- -What: iptables SAME target -When: 1.1. 2008 -Files: net/ipv4/netfilter/ipt_SAME.c, include/linux/netfilter_ipv4/ipt_SAME.h -Why: Obsolete for multiple years now, NAT core provides the same behaviour. - Unfixable broken wrt. 32/64 bit cleanness. -Who: Patrick McHardy - ---------------------------- - What: The arch/ppc and include/asm-ppc directories When: Jun 2008 Why: The arch/powerpc tree is the merged architecture for ppc32 and ppc64 diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index b11231df62c9..ad26f66b53e9 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -211,15 +211,6 @@ config IP_NF_TARGET_NETMAP To compile it as a module, choose M here. If unsure, say N. -config IP_NF_TARGET_SAME - tristate "SAME target support (OBSOLETE)" - depends on NF_NAT - help - This option adds a `SAME' target, which works like the standard SNAT - target, but attempts to give clients the same IP for all connections. - - To compile it as a module, choose M here. If unsure, say N. - config NF_NAT_SNMP_BASIC tristate "Basic SNMP-ALG support (EXPERIMENTAL)" depends on EXPERIMENTAL && NF_NAT diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 2fc05619f919..fd7d4a5b436c 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -56,7 +56,6 @@ obj-$(CONFIG_IP_NF_TARGET_MASQUERADE) += ipt_MASQUERADE.o obj-$(CONFIG_IP_NF_TARGET_NETMAP) += ipt_NETMAP.o obj-$(CONFIG_IP_NF_TARGET_REDIRECT) += ipt_REDIRECT.o obj-$(CONFIG_IP_NF_TARGET_REJECT) += ipt_REJECT.o -obj-$(CONFIG_IP_NF_TARGET_SAME) += ipt_SAME.o obj-$(CONFIG_IP_NF_TARGET_TTL) += ipt_TTL.o obj-$(CONFIG_IP_NF_TARGET_ULOG) += ipt_ULOG.o diff --git a/net/ipv4/netfilter/ipt_SAME.c b/net/ipv4/netfilter/ipt_SAME.c deleted file mode 100644 index a43923dab1e6..000000000000 --- a/net/ipv4/netfilter/ipt_SAME.c +++ /dev/null @@ -1,174 +0,0 @@ -/* Same. Just like SNAT, only try to make the connections - * between client A and server B always have the same source ip. - * - * (C) 2000 Paul `Rusty' Russell - * (C) 2001 Martin Josefsson - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Martin Josefsson "); -MODULE_DESCRIPTION("iptables special SNAT module for consistent sourceip"); - -static bool -same_tg_check(const char *tablename, const void *e, - const struct xt_target *target, void *targinfo, - unsigned int hook_mask) -{ - unsigned int count, countess, rangeip, index = 0; - struct ipt_same_info *mr = targinfo; - - mr->ipnum = 0; - - if (mr->rangesize < 1) { - pr_debug("same_check: need at least one dest range.\n"); - return false; - } - if (mr->rangesize > IPT_SAME_MAX_RANGE) { - pr_debug("same_check: too many ranges specified, maximum " - "is %u ranges\n", IPT_SAME_MAX_RANGE); - return false; - } - for (count = 0; count < mr->rangesize; count++) { - if (ntohl(mr->range[count].min_ip) > - ntohl(mr->range[count].max_ip)) { - pr_debug("same_check: min_ip is larger than max_ip in " - "range `%u.%u.%u.%u-%u.%u.%u.%u'.\n", - NIPQUAD(mr->range[count].min_ip), - NIPQUAD(mr->range[count].max_ip)); - return false; - } - if (!(mr->range[count].flags & IP_NAT_RANGE_MAP_IPS)) { - pr_debug("same_check: bad MAP_IPS.\n"); - return false; - } - rangeip = (ntohl(mr->range[count].max_ip) - - ntohl(mr->range[count].min_ip) + 1); - mr->ipnum += rangeip; - - pr_debug("same_check: range %u, ipnum = %u\n", count, rangeip); - } - pr_debug("same_check: total ipaddresses = %u\n", mr->ipnum); - - mr->iparray = kmalloc((sizeof(u_int32_t) * mr->ipnum), GFP_KERNEL); - if (!mr->iparray) { - pr_debug("same_check: Couldn't allocate %Zu bytes " - "for %u ipaddresses!\n", - (sizeof(u_int32_t) * mr->ipnum), mr->ipnum); - return false; - } - pr_debug("same_check: Allocated %Zu bytes for %u ipaddresses.\n", - (sizeof(u_int32_t) * mr->ipnum), mr->ipnum); - - for (count = 0; count < mr->rangesize; count++) { - for (countess = ntohl(mr->range[count].min_ip); - countess <= ntohl(mr->range[count].max_ip); - countess++) { - mr->iparray[index] = countess; - pr_debug("same_check: Added ipaddress `%u.%u.%u.%u' " - "in index %u.\n", HIPQUAD(countess), index); - index++; - } - } - return true; -} - -static void same_tg_destroy(const struct xt_target *target, void *targinfo) -{ - struct ipt_same_info *mr = targinfo; - - kfree(mr->iparray); - - pr_debug("same_destroy: Deallocated %Zu bytes for %u ipaddresses.\n", - (sizeof(u_int32_t) * mr->ipnum), mr->ipnum); -} - -static unsigned int -same_tg(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) -{ - struct nf_conn *ct; - enum ip_conntrack_info ctinfo; - u_int32_t tmpip, aindex; - __be32 new_ip; - const struct ipt_same_info *same = targinfo; - struct nf_nat_range newrange; - const struct nf_conntrack_tuple *t; - - NF_CT_ASSERT(hooknum == NF_INET_PRE_ROUTING || - hooknum == NF_INET_POST_ROUTING); - ct = nf_ct_get(skb, &ctinfo); - - t = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; - - /* Base new source on real src ip and optionally dst ip, - giving some hope for consistency across reboots. - Here we calculate the index in same->iparray which - holds the ipaddress we should use */ - - tmpip = ntohl(t->src.u3.ip); - - if (!(same->info & IPT_SAME_NODST)) - tmpip += ntohl(t->dst.u3.ip); - aindex = tmpip % same->ipnum; - - new_ip = htonl(same->iparray[aindex]); - - pr_debug("ipt_SAME: src=%u.%u.%u.%u dst=%u.%u.%u.%u, " - "new src=%u.%u.%u.%u\n", - NIPQUAD(t->src.u3.ip), NIPQUAD(t->dst.u3.ip), NIPQUAD(new_ip)); - - /* Transfer from original range. */ - newrange = ((struct nf_nat_range) - { same->range[0].flags, new_ip, new_ip, - /* FIXME: Use ports from correct range! */ - same->range[0].min, same->range[0].max }); - - /* Hand modified range to generic setup. */ - return nf_nat_setup_info(ct, &newrange, hooknum); -} - -static struct xt_target same_tg_reg __read_mostly = { - .name = "SAME", - .family = AF_INET, - .target = same_tg, - .targetsize = sizeof(struct ipt_same_info), - .table = "nat", - .hooks = (1 << NF_INET_PRE_ROUTING) | - (1 << NF_INET_POST_ROUTING), - .checkentry = same_tg_check, - .destroy = same_tg_destroy, - .me = THIS_MODULE, -}; - -static int __init same_tg_init(void) -{ - return xt_register_target(&same_tg_reg); -} - -static void __exit same_tg_exit(void) -{ - xt_unregister_target(&same_tg_reg); -} - -module_init(same_tg_init); -module_exit(same_tg_exit); - -- cgit v1.2.3 From b8599d20708fa3bde1e414689f3474560c2d990b Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Thu, 13 Dec 2007 12:25:01 -0200 Subject: [DCCP]: Support for server holding timewait state This adds a socket option and signalling support for the case where the server holds timewait state on closing the connection, as described in RFC 4340, 8.3. Since holding timewait state at the server is the non-usual case, it is enabled via a socket option. Documentation for this socket option has been added. The setsockopt statement has been made resilient against different possible cases of expressing boolean `true' values using a suggestion by Ian McDonald. Signed-off-by: Gerrit Renker Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- Documentation/networking/dccp.txt | 6 ++++++ include/linux/dccp.h | 3 +++ net/dccp/output.c | 6 ++++-- net/dccp/proto.c | 13 ++++++++++++- 4 files changed, 25 insertions(+), 3 deletions(-) (limited to 'Documentation') diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index d76905a5a087..39131a3c78f8 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -57,6 +57,12 @@ can be set before calling bind(). DCCP_SOCKOPT_GET_CUR_MPS is read-only and retrieves the current maximum packet size (application payload size) in bytes, see RFC 4340, section 14. +DCCP_SOCKOPT_SERVER_TIMEWAIT enables the server (listening socket) to hold +timewait state when closing the connection (RFC 4340, 8.3). The usual case is +that the closing server sends a CloseReq, whereupon the client holds timewait +state. When this boolean socket option is on, the server sends a Close instead +and will enter TIMEWAIT. This option must be set after accept() returns. + DCCP_SOCKOPT_SEND_CSCOV and DCCP_SOCKOPT_RECV_CSCOV are used for setting the partial checksum coverage (RFC 4340, sec. 9.2). The default is that checksums always cover the entire packet and that only fully covered application data is diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 312b989c7edb..c676021603f5 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -205,6 +205,7 @@ struct dccp_so_feat { #define DCCP_SOCKOPT_CHANGE_L 3 #define DCCP_SOCKOPT_CHANGE_R 4 #define DCCP_SOCKOPT_GET_CUR_MPS 5 +#define DCCP_SOCKOPT_SERVER_TIMEWAIT 6 #define DCCP_SOCKOPT_SEND_CSCOV 10 #define DCCP_SOCKOPT_RECV_CSCOV 11 #define DCCP_SOCKOPT_CCID_RX_INFO 128 @@ -492,6 +493,7 @@ struct dccp_ackvec; * @dccps_role - role of this sock, one of %dccp_role * @dccps_hc_rx_insert_options - receiver wants to add options when acking * @dccps_hc_tx_insert_options - sender wants to add options when sending + * @dccps_server_timewait - server holds timewait state on close (RFC 4340, 8.3) * @dccps_xmit_timer - timer for when CCID is not ready to send * @dccps_syn_rtt - RTT sample from Request/Response exchange (in usecs) */ @@ -528,6 +530,7 @@ struct dccp_sock { enum dccp_role dccps_role:2; __u8 dccps_hc_rx_insert_options:1; __u8 dccps_hc_tx_insert_options:1; + __u8 dccps_server_timewait:1; struct timer_list dccps_xmit_timer; }; diff --git a/net/dccp/output.c b/net/dccp/output.c index e97584aa4898..b2e17910930d 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -567,8 +567,10 @@ void dccp_send_close(struct sock *sk, const int active) /* Reserve space for headers and prepare control bits. */ skb_reserve(skb, sk->sk_prot->max_header); - DCCP_SKB_CB(skb)->dccpd_type = dp->dccps_role == DCCP_ROLE_CLIENT ? - DCCP_PKT_CLOSE : DCCP_PKT_CLOSEREQ; + if (dp->dccps_role == DCCP_ROLE_SERVER && !dp->dccps_server_timewait) + DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_CLOSEREQ; + else + DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_CLOSE; if (active) { dccp_write_xmit(sk, 1); diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 8a73c8f98d76..cc87c500bfb8 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -551,6 +551,12 @@ static int do_dccp_setsockopt(struct sock *sk, int level, int optname, (struct dccp_so_feat __user *) optval); break; + case DCCP_SOCKOPT_SERVER_TIMEWAIT: + if (dp->dccps_role != DCCP_ROLE_SERVER) + err = -EOPNOTSUPP; + else + dp->dccps_server_timewait = (val != 0); + break; case DCCP_SOCKOPT_SEND_CSCOV: /* sender side, RFC 4340, sec. 9.2 */ if (val < 0 || val > 15) err = -EINVAL; @@ -653,6 +659,10 @@ static int do_dccp_getsockopt(struct sock *sk, int level, int optname, val = dp->dccps_mss_cache; len = sizeof(val); break; + case DCCP_SOCKOPT_SERVER_TIMEWAIT: + val = dp->dccps_server_timewait; + len = sizeof(val); + break; case DCCP_SOCKOPT_SEND_CSCOV: val = dp->dccps_pcslen; len = sizeof(val); @@ -918,7 +928,8 @@ static void dccp_terminate_connection(struct sock *sk) case DCCP_OPEN: dccp_send_close(sk, 1); - if (dccp_sk(sk)->dccps_role == DCCP_ROLE_SERVER) + if (dccp_sk(sk)->dccps_role == DCCP_ROLE_SERVER && + !dccp_sk(sk)->dccps_server_timewait) next_state = DCCP_ACTIVE_CLOSEREQ; else next_state = DCCP_CLOSING; -- cgit v1.2.3 From 558f82ef6e0d25e87f7468c07b6db1fbbf95a855 Mon Sep 17 00:00:00 2001 From: Masahide NAKAMURA Date: Thu, 20 Dec 2007 20:42:57 -0800 Subject: [XFRM]: Define packet dropping statistics. This statistics is shown factor dropped by transformation at /proc/net/xfrm_stat for developer. It is a counter designed from current transformation source code and defined as linux private MIB. See Documentation/networking/xfrm_proc.txt for the detail. Signed-off-by: Masahide NAKAMURA Signed-off-by: David S. Miller --- Documentation/networking/xfrm_proc.txt | 71 +++++++++++++++++++++++++ include/linux/snmp.h | 31 +++++++++++ include/net/snmp.h | 5 ++ include/net/xfrm.h | 18 +++++++ net/xfrm/Makefile | 1 + net/xfrm/xfrm_policy.c | 24 +++++++++ net/xfrm/xfrm_proc.c | 96 ++++++++++++++++++++++++++++++++++ 7 files changed, 246 insertions(+) create mode 100644 Documentation/networking/xfrm_proc.txt create mode 100644 net/xfrm/xfrm_proc.c (limited to 'Documentation') diff --git a/Documentation/networking/xfrm_proc.txt b/Documentation/networking/xfrm_proc.txt new file mode 100644 index 000000000000..ec9045b5c34a --- /dev/null +++ b/Documentation/networking/xfrm_proc.txt @@ -0,0 +1,71 @@ +XFRM proc - /proc/net/xfrm_* files +================================== +Masahide NAKAMURA + + +Transformation Statistics +------------------------- +xfrm_proc is a statistics shown factor dropped by transformation +for developer. +It is a counter designed from current transformation source code +and defined like linux private MIB. + +Inbound statistics +~~~~~~~~~~~~~~~~~~ +XfrmInError: + All errors which is not matched others +XfrmInBufferError: + No buffer is left +XfrmInHdrError: + Header error +XfrmInNoStates: + No state is found + i.e. Either inbound SPI, address, or IPsec protocol at SA is wrong +XfrmInStateProtoError: + Transformation protocol specific error + e.g. SA key is wrong +XfrmInStateModeError: + Transformation mode specific error +XfrmInSeqOutOfWindow: + Sequence out of window +XfrmInStateExpired: + State is expired +XfrmInStateMismatch: + State has mismatch option + e.g. UDP encapsulation type is mismatch +XfrmInStateInvalid: + State is invalid +XfrmInTmplMismatch: + No matching template for states + e.g. Inbound SAs are correct but SP rule is wrong +XfrmInNoPols: + No policy is found for states + e.g. Inbound SAs are correct but no SP is found +XfrmInPolBlock: + Policy discards +XfrmInPolError: + Policy error + +Outbound errors +~~~~~~~~~~~~~~~ +XfrmOutError: + All errors which is not matched others +XfrmOutBundleGenError: + Bundle generation error +XfrmOutBundleCheckError: + Bundle check error +XfrmOutNoStates: + No state is found +XfrmOutStateProtoError: + Transformation protocol specific error +XfrmOutStateModeError: + Transformation mode specific error + e.g. Outer header space is not enough +XfrmOutStateExpired: + State is expired +XfrmOutPolBlock: + Policy discards +XfrmOutPolDead: + Policy is dead +XfrmOutPolError: + Policy error diff --git a/include/linux/snmp.h b/include/linux/snmp.h index 89f0c2b5f405..86d3effb2836 100644 --- a/include/linux/snmp.h +++ b/include/linux/snmp.h @@ -217,4 +217,35 @@ enum __LINUX_MIB_MAX }; +/* linux Xfrm mib definitions */ +enum +{ + LINUX_MIB_XFRMNUM = 0, + LINUX_MIB_XFRMINERROR, /* XfrmInError */ + LINUX_MIB_XFRMINBUFFERERROR, /* XfrmInBufferError */ + LINUX_MIB_XFRMINHDRERROR, /* XfrmInHdrError */ + LINUX_MIB_XFRMINNOSTATES, /* XfrmInNoStates */ + LINUX_MIB_XFRMINSTATEPROTOERROR, /* XfrmInStateProtoError */ + LINUX_MIB_XFRMINSTATEMODEERROR, /* XfrmInStateModeError */ + LINUX_MIB_XFRMINSEQOUTOFWINDOW, /* XfrmInSeqOutOfWindow */ + LINUX_MIB_XFRMINSTATEEXPIRED, /* XfrmInStateExpired */ + LINUX_MIB_XFRMINSTATEMISMATCH, /* XfrmInStateMismatch */ + LINUX_MIB_XFRMINSTATEINVALID, /* XfrmInStateInvalid */ + LINUX_MIB_XFRMINTMPLMISMATCH, /* XfrmInTmplMismatch */ + LINUX_MIB_XFRMINNOPOLS, /* XfrmInNoPols */ + LINUX_MIB_XFRMINPOLBLOCK, /* XfrmInPolBlock */ + LINUX_MIB_XFRMINPOLERROR, /* XfrmInPolError */ + LINUX_MIB_XFRMOUTERROR, /* XfrmOutError */ + LINUX_MIB_XFRMOUTBUNDLEGENERROR, /* XfrmOutBundleGenError */ + LINUX_MIB_XFRMOUTBUNDLECHECKERROR, /* XfrmOutBundleCheckError */ + LINUX_MIB_XFRMOUTNOSTATES, /* XfrmOutNoStates */ + LINUX_MIB_XFRMOUTSTATEPROTOERROR, /* XfrmOutStateProtoError */ + LINUX_MIB_XFRMOUTSTATEMODEERROR, /* XfrmOutStateModeError */ + LINUX_MIB_XFRMOUTSTATEEXPIRED, /* XfrmOutStateExpired */ + LINUX_MIB_XFRMOUTPOLBLOCK, /* XfrmOutPolBlock */ + LINUX_MIB_XFRMOUTPOLDEAD, /* XfrmOutPolDead */ + LINUX_MIB_XFRMOUTPOLERROR, /* XfrmOutPolError */ + __LINUX_MIB_XFRMMAX +}; + #endif /* _LINUX_SNMP_H */ diff --git a/include/net/snmp.h b/include/net/snmp.h index fbb66663a42c..ce2f48507510 100644 --- a/include/net/snmp.h +++ b/include/net/snmp.h @@ -118,6 +118,11 @@ struct linux_mib { unsigned long mibs[LINUX_MIB_MAX]; }; +/* Linux Xfrm */ +#define LINUX_MIB_XFRMMAX __LINUX_MIB_XFRMMAX +struct linux_xfrm_mib { + unsigned long mibs[LINUX_MIB_XFRMMAX]; +}; /* * FIXME: On x86 and some other CPUs the split into user and softirq parts diff --git a/include/net/xfrm.h b/include/net/xfrm.h index eea1c327c93e..a79702bcdcd0 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -19,6 +19,9 @@ #include #include #include +#ifdef CONFIG_XFRM_STATISTICS +#include +#endif #define XFRM_PROTO_ESP 50 #define XFRM_PROTO_AH 51 @@ -34,6 +37,17 @@ #define MODULE_ALIAS_XFRM_TYPE(family, proto) \ MODULE_ALIAS("xfrm-type-" __stringify(family) "-" __stringify(proto)) +#ifdef CONFIG_XFRM_STATISTICS +DECLARE_SNMP_STAT(struct linux_xfrm_mib, xfrm_statistics); +#define XFRM_INC_STATS(field) SNMP_INC_STATS(xfrm_statistics, field) +#define XFRM_INC_STATS_BH(field) SNMP_INC_STATS_BH(xfrm_statistics, field) +#define XFRM_INC_STATS_USER(field) SNMP_INC_STATS_USER(xfrm_statistics, field) +#else +#define XFRM_INC_STATS(field) +#define XFRM_INC_STATS_BH(field) +#define XFRM_INC_STATS_USER(field) +#endif + extern struct sock *xfrm_nl; extern u32 sysctl_xfrm_aevent_etime; extern u32 sysctl_xfrm_aevent_rseqth; @@ -1139,6 +1153,10 @@ static inline void xfrm6_fini(void) } #endif +#ifdef CONFIG_XFRM_STATISTICS +extern int xfrm_proc_init(void); +#endif + extern int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*), void *); extern struct xfrm_state *xfrm_state_alloc(void); extern struct xfrm_state *xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, diff --git a/net/xfrm/Makefile b/net/xfrm/Makefile index 45744a3d3a51..332cfb0ff566 100644 --- a/net/xfrm/Makefile +++ b/net/xfrm/Makefile @@ -4,5 +4,6 @@ obj-$(CONFIG_XFRM) := xfrm_policy.o xfrm_state.o xfrm_hash.o \ xfrm_input.o xfrm_output.o xfrm_algo.o +obj-$(CONFIG_XFRM_STATISTICS) += xfrm_proc.o obj-$(CONFIG_XFRM_USER) += xfrm_user.o diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 521cb6e12561..32ddb7b12e7f 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -27,11 +27,19 @@ #include #include #include +#ifdef CONFIG_XFRM_STATISTICS +#include +#endif #include "xfrm_hash.h" int sysctl_xfrm_larval_drop __read_mostly; +#ifdef CONFIG_XFRM_STATISTICS +DEFINE_SNMP_STAT(struct linux_xfrm_mib, xfrm_statistics) __read_mostly; +EXPORT_SYMBOL(xfrm_statistics); +#endif + DEFINE_MUTEX(xfrm_cfg_mutex); EXPORT_SYMBOL(xfrm_cfg_mutex); @@ -2258,6 +2266,16 @@ static struct notifier_block xfrm_dev_notifier = { 0 }; +#ifdef CONFIG_XFRM_STATISTICS +static int __init xfrm_statistics_init(void) +{ + if (snmp_mib_init((void **)xfrm_statistics, + sizeof(struct linux_xfrm_mib)) < 0) + return -ENOMEM; + return 0; +} +#endif + static void __init xfrm_policy_init(void) { unsigned int hmask, sz; @@ -2294,9 +2312,15 @@ static void __init xfrm_policy_init(void) void __init xfrm_init(void) { +#ifdef CONFIG_XFRM_STATISTICS + xfrm_statistics_init(); +#endif xfrm_state_init(); xfrm_policy_init(); xfrm_input_init(); +#ifdef CONFIG_XFRM_STATISTICS + xfrm_proc_init(); +#endif } #ifdef CONFIG_AUDITSYSCALL diff --git a/net/xfrm/xfrm_proc.c b/net/xfrm/xfrm_proc.c new file mode 100644 index 000000000000..31d035415ecd --- /dev/null +++ b/net/xfrm/xfrm_proc.c @@ -0,0 +1,96 @@ +/* + * xfrm_proc.c + * + * Copyright (C)2006-2007 USAGI/WIDE Project + * + * Authors: Masahide NAKAMURA + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#include +#include +#include +#include + +static struct snmp_mib xfrm_mib_list[] = { + SNMP_MIB_ITEM("XfrmInError", LINUX_MIB_XFRMINERROR), + SNMP_MIB_ITEM("XfrmInBufferError", LINUX_MIB_XFRMINBUFFERERROR), + SNMP_MIB_ITEM("XfrmInHdrError", LINUX_MIB_XFRMINHDRERROR), + SNMP_MIB_ITEM("XfrmInNoStates", LINUX_MIB_XFRMINNOSTATES), + SNMP_MIB_ITEM("XfrmInStateProtoError", LINUX_MIB_XFRMINSTATEPROTOERROR), + SNMP_MIB_ITEM("XfrmInStateModeError", LINUX_MIB_XFRMINSTATEMODEERROR), + SNMP_MIB_ITEM("XfrmInSeqOutOfWindow", LINUX_MIB_XFRMINSEQOUTOFWINDOW), + SNMP_MIB_ITEM("XfrmInStateExpired", LINUX_MIB_XFRMINSTATEEXPIRED), + SNMP_MIB_ITEM("XfrmInStateMismatch", LINUX_MIB_XFRMINSTATEMISMATCH), + SNMP_MIB_ITEM("XfrmInStateInvalid", LINUX_MIB_XFRMINSTATEINVALID), + SNMP_MIB_ITEM("XfrmInTmplMismatch", LINUX_MIB_XFRMINTMPLMISMATCH), + SNMP_MIB_ITEM("XfrmInNoPols", LINUX_MIB_XFRMINNOPOLS), + SNMP_MIB_ITEM("XfrmInPolBlock", LINUX_MIB_XFRMINPOLBLOCK), + SNMP_MIB_ITEM("XfrmInPolError", LINUX_MIB_XFRMINPOLERROR), + SNMP_MIB_ITEM("XfrmOutError", LINUX_MIB_XFRMOUTERROR), + SNMP_MIB_ITEM("XfrmOutBundleGenError", LINUX_MIB_XFRMOUTBUNDLEGENERROR), + SNMP_MIB_ITEM("XfrmOutBundleCheckError", LINUX_MIB_XFRMOUTBUNDLECHECKERROR), + SNMP_MIB_ITEM("XfrmOutNoStates", LINUX_MIB_XFRMOUTNOSTATES), + SNMP_MIB_ITEM("XfrmOutStateProtoError", LINUX_MIB_XFRMOUTSTATEPROTOERROR), + SNMP_MIB_ITEM("XfrmOutStateModeError", LINUX_MIB_XFRMOUTSTATEMODEERROR), + SNMP_MIB_ITEM("XfrmOutStateExpired", LINUX_MIB_XFRMOUTSTATEEXPIRED), + SNMP_MIB_ITEM("XfrmOutPolBlock", LINUX_MIB_XFRMOUTPOLBLOCK), + SNMP_MIB_ITEM("XfrmOutPolDead", LINUX_MIB_XFRMOUTPOLDEAD), + SNMP_MIB_ITEM("XfrmOutPolError", LINUX_MIB_XFRMOUTPOLERROR), + SNMP_MIB_SENTINEL +}; + +static unsigned long +fold_field(void *mib[], int offt) +{ + unsigned long res = 0; + int i; + + for_each_possible_cpu(i) { + res += *(((unsigned long *)per_cpu_ptr(mib[0], i)) + offt); + res += *(((unsigned long *)per_cpu_ptr(mib[1], i)) + offt); + } + return res; +} + +static int xfrm_statistics_seq_show(struct seq_file *seq, void *v) +{ + int i; + for (i=0; xfrm_mib_list[i].name; i++) + seq_printf(seq, "%-24s\t%lu\n", xfrm_mib_list[i].name, + fold_field((void **)xfrm_statistics, + xfrm_mib_list[i].entry)); + return 0; +} + +static int xfrm_statistics_seq_open(struct inode *inode, struct file *file) +{ + return single_open(file, xfrm_statistics_seq_show, NULL); +} + +static struct file_operations xfrm_statistics_seq_fops = { + .owner = THIS_MODULE, + .open = xfrm_statistics_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +int __init xfrm_proc_init(void) +{ + int rc = 0; + + if (!proc_net_fops_create(&init_net, "xfrm_stat", S_IRUGO, + &xfrm_statistics_seq_fops)) + goto stat_fail; + + out: + return rc; + + stat_fail: + rc = -ENOMEM; + goto out; +} -- cgit v1.2.3 From c21b39aca4f8f4975784e54cd3a1b80bab80dcc0 Mon Sep 17 00:00:00 2001 From: Stefano Brivio Date: Wed, 19 Dec 2007 01:26:16 +0100 Subject: mac80211: make PID rate control algorithm the default This makes the new PID TX rate control algorithm the default instead of the rc80211_simple rate control algorithm. The simple algorithm was flawed in several ways: it wasn't responsive at all and didn't age the information it was relying on properly. The PID algorithm allows us to tune characteristics such as responsiveness by adjusting parameters and was found to generally behave better. The default algorithm can be overridden to select simple instead. Which ever algorithm is the default is included as part of the mac80211 module automatically. The other algorithm (simple vs. pid) can be selected for inclusion as well. If EMBEDDED is selected then the choice is available to have no default specified and neither algorithm included in mac80211. The default algorithm can be set through a modparam. While at it, mark rc80211-simple as deprecated, and schedule it for removal. Signed-off-by: Stefano Brivio Signed-off-by: John W. Linville Signed-off-by: David S. Miller --- Documentation/feature-removal-schedule.txt | 8 ++++ net/mac80211/Kconfig | 72 ++++++++++++++++++++++++------ net/mac80211/Makefile | 4 +- net/mac80211/ieee80211.c | 12 ++--- net/mac80211/ieee80211_rate.c | 24 +++++++--- 5 files changed, 93 insertions(+), 27 deletions(-) (limited to 'Documentation') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 0927425026f1..a6cf8f6fc586 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -314,3 +314,11 @@ Why: No in-kernel drivers will depend on it any longer. Who: John W. Linville --------------------------- + +What: rc80211-simple rate control algorithm for mac80211 +When: 2.6.26 +Files: net/mac80211/rc80211-simple.c +Why: This algorithm was provided for reference but always exhibited bad + responsiveness and performance and has some serious flaws. It has been + replaced by rc80211-pid. +Who: Stefano Brivio diff --git a/net/mac80211/Kconfig b/net/mac80211/Kconfig index 9f8663b412d9..297f4d967e2d 100644 --- a/net/mac80211/Kconfig +++ b/net/mac80211/Kconfig @@ -13,29 +13,75 @@ config MAC80211 This option enables the hardware independent IEEE 802.11 networking stack. -config MAC80211_RCSIMPLE - bool "'simple' rate control algorithm" if EMBEDDED +config MAC80211_RC_DEFAULT_CHOICE + bool "Choose default rate control algorithm" if EMBEDDED default y depends on MAC80211 - help - This option allows you to turn off the 'simple' rate - control algorithm in mac80211. If you do turn it off, - you absolutely need another rate control algorithm. + ---help--- + This options enables selection of a default rate control + algorithm to be built into the mac80211 module. Alternate + rate control algorithms might be built into the mac80211 + module as well. + +choice + prompt "Default rate control algorithm" + default MAC80211_RC_DEFAULT_PID + depends on MAC80211 && MAC80211_RC_DEFAULT_CHOICE + ---help--- + This option selects the default rate control algorithm + mac80211 will use. Note that this default can still be + overriden through the ieee80211_default_rc_algo module + parameter. + +config MAC80211_RC_DEFAULT_PID + bool "PID controller based rate control algorithm" + select MAC80211_RC_PID + ---help--- + Select the PID controller based rate control as the + default rate control algorithm. You should choose + this unless you know what you are doing. + +config MAC80211_RC_DEFAULT_SIMPLE + bool "Simple rate control algorithm" + select MAC80211_RC_SIMPLE + ---help--- + Select the simple rate control as the default rate + control algorithm. Note that this is a non-responsive, + dumb algorithm. You should choose the PID rate control + instead. + +endchoice - Say Y unless you know you will have another algorithm - available. +config MAC80211_RC_DEFAULT + string + depends on MAC80211 + default "pid" if MAC80211_RC_DEFAULT_PID + default "simple" if MAC80211_RC_DEFAULT_SIMPLE + default "" -config MAC80211_RCPID - bool "'PID' rate control algorithm" if EMBEDDED +config MAC80211_RC_PID + bool "PID controller based rate control algorithm" default y depends on MAC80211 - help + ---help--- This option enables a TX rate control algorithm for mac80211 that uses a PID controller to select the TX rate. - Say Y unless you're sure you want to use a different - rate control algorithm. + Say Y or M unless you're sure you want to use a + different rate control algorithm. + +config MAC80211_RC_SIMPLE + bool "Simple rate control algorithm (DEPRECATED)" + default n + depends on MAC80211 + ---help--- + This option enables a very simple, non-responsive TX + rate control algorithm. This algorithm is deprecated + and will be removed from the kernel in near future. + It has been replaced by the PID algorithm. + + Say N unless you know what you are doing. config MAC80211_LEDS bool "Enable LED triggers" diff --git a/net/mac80211/Makefile b/net/mac80211/Makefile index 62c01ca099e5..a375f0477da0 100644 --- a/net/mac80211/Makefile +++ b/net/mac80211/Makefile @@ -3,8 +3,8 @@ obj-$(CONFIG_MAC80211) += mac80211.o mac80211-objs-$(CONFIG_MAC80211_LEDS) += ieee80211_led.o mac80211-objs-$(CONFIG_MAC80211_DEBUGFS) += debugfs.o debugfs_sta.o debugfs_netdev.o debugfs_key.o mac80211-objs-$(CONFIG_NET_SCHED) += wme.o -mac80211-objs-$(CONFIG_MAC80211_RCSIMPLE) += rc80211_simple.o -mac80211-objs-$(CONFIG_MAC80211_RCPID) += rc80211_pid.o +mac80211-objs-$(CONFIG_MAC80211_RC_SIMPLE) += rc80211_simple.o +mac80211-objs-$(CONFIG_MAC80211_RC_PID) += rc80211_pid.o mac80211-objs := \ ieee80211.o \ diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c index 3d7b4341914e..c0dbf77547f7 100644 --- a/net/mac80211/ieee80211.c +++ b/net/mac80211/ieee80211.c @@ -1312,13 +1312,13 @@ static int __init ieee80211_init(void) BUILD_BUG_ON(sizeof(struct ieee80211_tx_packet_data) > sizeof(skb->cb)); -#ifdef CONFIG_MAC80211_RCSIMPLE +#ifdef CONFIG_MAC80211_RC_SIMPLE ret = ieee80211_rate_control_register(&mac80211_rcsimple); if (ret) goto fail; #endif -#ifdef CONFIG_MAC80211_RCPID +#ifdef CONFIG_MAC80211_RC_PID ret = ieee80211_rate_control_register(&mac80211_rcpid); if (ret) goto fail; @@ -1338,10 +1338,10 @@ static int __init ieee80211_init(void) fail: -#ifdef CONFIG_MAC80211_RCSIMPLE +#ifdef CONFIG_MAC80211_RC_SIMPLE ieee80211_rate_control_unregister(&mac80211_rcsimple); #endif -#ifdef CONFIG_MAC80211_RCPID +#ifdef CONFIG_MAC80211_RC_PID ieee80211_rate_control_unregister(&mac80211_rcpid); #endif @@ -1350,10 +1350,10 @@ fail: static void __exit ieee80211_exit(void) { -#ifdef CONFIG_MAC80211_RCSIMPLE +#ifdef CONFIG_MAC80211_RC_SIMPLE ieee80211_rate_control_unregister(&mac80211_rcsimple); #endif -#ifdef CONFIG_MAC80211_RCPID +#ifdef CONFIG_MAC80211_RC_PID ieee80211_rate_control_unregister(&mac80211_rcpid); #endif diff --git a/net/mac80211/ieee80211_rate.c b/net/mac80211/ieee80211_rate.c index e495b0998b4d..65fc9ad615e9 100644 --- a/net/mac80211/ieee80211_rate.c +++ b/net/mac80211/ieee80211_rate.c @@ -21,6 +21,11 @@ struct rate_control_alg { static LIST_HEAD(rate_ctrl_algs); static DEFINE_MUTEX(rate_ctrl_mutex); +static char *ieee80211_default_rc_algo = CONFIG_MAC80211_RC_DEFAULT; +module_param(ieee80211_default_rc_algo, charp, 0644); +MODULE_PARM_DESC(ieee80211_default_rc_algo, + "Default rate control algorithm for mac80211 to use"); + int ieee80211_rate_control_register(struct rate_control_ops *ops) { struct rate_control_alg *alg; @@ -89,21 +94,27 @@ ieee80211_try_rate_control_ops_get(const char *name) return ops; } -/* Get the rate control algorithm. If `name' is NULL, get the first - * available algorithm. */ +/* Get the rate control algorithm. */ static struct rate_control_ops * ieee80211_rate_control_ops_get(const char *name) { struct rate_control_ops *ops; + const char *alg_name; if (!name) - name = "simple"; + alg_name = ieee80211_default_rc_algo; + else + alg_name = name; - ops = ieee80211_try_rate_control_ops_get(name); + ops = ieee80211_try_rate_control_ops_get(alg_name); if (!ops) { - request_module("rc80211_%s", name); - ops = ieee80211_try_rate_control_ops_get(name); + request_module("rc80211_%s", alg_name); + ops = ieee80211_try_rate_control_ops_get(alg_name); } + if (!ops && name) + /* try default if specific alg requested but not found */ + ops = ieee80211_try_rate_control_ops_get(ieee80211_default_rc_algo); + return ops; } @@ -244,3 +255,4 @@ void rate_control_deinitialize(struct ieee80211_local *local) local->rate_ctrl = NULL; rate_control_put(ref); } + -- cgit v1.2.3 From a1464ab61e66c96f9cffea335755de850fe8bdbd Mon Sep 17 00:00:00 2001 From: Stefano Brivio Date: Wed, 19 Dec 2007 01:46:53 +0100 Subject: doc: fix typo in feature-removal-schedule Signed-off-by: Stefano Brivio Signed-off-by: John W. Linville Signed-off-by: David S. Miller --- Documentation/feature-removal-schedule.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index a6cf8f6fc586..2c3ae6c02a47 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -307,7 +307,7 @@ Who: John W. Linville --------------------------- -What: iee80211 softmac wireless networking component +What: ieee80211 softmac wireless networking component When: 2.6.26 (or after removal of bcm43xx and port of zd1211rw to mac80211) Files: net/ieee80211/softmac Why: No in-kernel drivers will depend on it any longer. -- cgit v1.2.3 From d9727bb2d516bc16bafee4216eec91cd2be5f30a Mon Sep 17 00:00:00 2001 From: Masahide NAKAMURA Date: Tue, 25 Dec 2007 20:56:26 -0800 Subject: [XFRM] Documentaion: Fix error example at XFRMOUTSTATEMODEERROR. Signed-off-by: Masahide NAKAMURA Signed-off-by: David S. Miller --- Documentation/networking/xfrm_proc.txt | 1 - 1 file changed, 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/networking/xfrm_proc.txt b/Documentation/networking/xfrm_proc.txt index ec9045b5c34a..53c1a58b02f1 100644 --- a/Documentation/networking/xfrm_proc.txt +++ b/Documentation/networking/xfrm_proc.txt @@ -60,7 +60,6 @@ XfrmOutStateProtoError: Transformation protocol specific error XfrmOutStateModeError: Transformation mode specific error - e.g. Outer header space is not enough XfrmOutStateExpired: State is expired XfrmOutPolBlock: -- cgit v1.2.3 From 95766fff6b9a78d11fc2d3812dd035381690b55d Mon Sep 17 00:00:00 2001 From: Hideo Aoki Date: Mon, 31 Dec 2007 00:29:24 -0800 Subject: [UDP]: Add memory accounting. Signed-off-by: Takahiro Yasui Signed-off-by: Hideo Aoki Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 27 ++++++++++++++++ include/net/udp.h | 9 ++++++ net/ipv4/af_inet.c | 5 +++ net/ipv4/proc.c | 3 +- net/ipv4/sysctl_net_ipv4.c | 31 ++++++++++++++++++ net/ipv4/udp.c | 57 ++++++++++++++++++++++++++++++++-- net/ipv6/udp.c | 32 ++++++++++++++++--- 7 files changed, 157 insertions(+), 7 deletions(-) (limited to 'Documentation') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index 6f7872ba1def..17a6e46fbd43 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -446,6 +446,33 @@ tcp_dma_copybreak - INTEGER and CONFIG_NET_DMA is enabled. Default: 4096 +UDP variables: + +udp_mem - vector of 3 INTEGERs: min, pressure, max + Number of pages allowed for queueing by all UDP sockets. + + min: Below this number of pages UDP is not bothered about its + memory appetite. When amount of memory allocated by UDP exceeds + this number, UDP starts to moderate memory usage. + + pressure: This value was introduced to follow format of tcp_mem. + + max: Number of pages allowed for queueing by all UDP sockets. + + Default is calculated at boot time from amount of available memory. + +udp_rmem_min - INTEGER + Minimal size of receive buffer used by UDP sockets in moderation. + Each UDP socket is able to use the size for receiving data, even if + total pages of UDP sockets exceed udp_mem pressure. The unit is byte. + Default: 4096 + +udp_wmem_min - INTEGER + Minimal size of send buffer used by UDP sockets in moderation. + Each UDP socket is able to use the size for sending data, even if + total pages of UDP sockets exceed udp_mem pressure. The unit is byte. + Default: 4096 + CIPSOv4 Variables: cipso_cache_enable - BOOLEAN diff --git a/include/net/udp.h b/include/net/udp.h index 98cb09ca3a27..93796beac8ff 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -65,6 +65,13 @@ extern rwlock_t udp_hash_lock; extern struct proto udp_prot; +extern atomic_t udp_memory_allocated; + +/* sysctl variables for udp */ +extern int sysctl_udp_mem[3]; +extern int sysctl_udp_rmem_min; +extern int sysctl_udp_wmem_min; + struct sk_buff; /* @@ -198,4 +205,6 @@ extern void udp_proc_unregister(struct udp_seq_afinfo *afinfo); extern int udp4_proc_init(void); extern void udp4_proc_exit(void); #endif + +extern void udp_init(void); #endif /* _UDP_H */ diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 03633b7b9b4a..0e12cf646071 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -139,6 +139,8 @@ void inet_sock_destruct(struct sock *sk) __skb_queue_purge(&sk->sk_receive_queue); __skb_queue_purge(&sk->sk_error_queue); + sk_mem_reclaim(sk); + if (sk->sk_type == SOCK_STREAM && sk->sk_state != TCP_CLOSE) { printk("Attempt to release TCP socket in state %d %p\n", sk->sk_state, sk); @@ -1417,6 +1419,9 @@ static int __init inet_init(void) /* Setup TCP slab cache for open requests. */ tcp_init(); + /* Setup UDP memory threshold */ + udp_init(); + /* Add UDP-Lite (RFC 3828) */ udplite4_register(); diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index 41734db677be..53bc010beefd 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -56,7 +56,8 @@ static int sockstat_seq_show(struct seq_file *seq, void *v) sock_prot_inuse(&tcp_prot), atomic_read(&tcp_orphan_count), tcp_death_row.tw_count, atomic_read(&tcp_sockets_allocated), atomic_read(&tcp_memory_allocated)); - seq_printf(seq, "UDP: inuse %d\n", sock_prot_inuse(&udp_prot)); + seq_printf(seq, "UDP: inuse %d mem %d\n", sock_prot_inuse(&udp_prot), + atomic_read(&udp_memory_allocated)); seq_printf(seq, "UDPLITE: inuse %d\n", sock_prot_inuse(&udplite_prot)); seq_printf(seq, "RAW: inuse %d\n", sock_prot_inuse(&raw_prot)); seq_printf(seq, "FRAG: inuse %d memory %d\n", diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 844f26fab06f..a5a9f8e3bb25 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -812,6 +813,36 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = &proc_dointvec, }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "udp_mem", + .data = &sysctl_udp_mem, + .maxlen = sizeof(sysctl_udp_mem), + .mode = 0644, + .proc_handler = &proc_dointvec_minmax, + .strategy = &sysctl_intvec, + .extra1 = &zero + }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "udp_rmem_min", + .data = &sysctl_udp_rmem_min, + .maxlen = sizeof(sysctl_udp_rmem_min), + .mode = 0644, + .proc_handler = &proc_dointvec_minmax, + .strategy = &sysctl_intvec, + .extra1 = &zero + }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "udp_wmem_min", + .data = &sysctl_udp_wmem_min, + .maxlen = sizeof(sysctl_udp_wmem_min), + .mode = 0644, + .proc_handler = &proc_dointvec_minmax, + .strategy = &sysctl_intvec, + .extra1 = &zero + }, { .ctl_name = 0 } }; diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 1ce6b60b7f93..353284360751 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -82,6 +82,7 @@ #include #include #include +#include #include #include #include @@ -118,6 +119,17 @@ EXPORT_SYMBOL(udp_stats_in6); struct hlist_head udp_hash[UDP_HTABLE_SIZE]; DEFINE_RWLOCK(udp_hash_lock); +int sysctl_udp_mem[3] __read_mostly; +int sysctl_udp_rmem_min __read_mostly; +int sysctl_udp_wmem_min __read_mostly; + +EXPORT_SYMBOL(sysctl_udp_mem); +EXPORT_SYMBOL(sysctl_udp_rmem_min); +EXPORT_SYMBOL(sysctl_udp_wmem_min); + +atomic_t udp_memory_allocated; +EXPORT_SYMBOL(udp_memory_allocated); + static inline int __udp_lib_lport_inuse(__u16 num, const struct hlist_head udptable[]) { @@ -901,13 +913,17 @@ try_again: err = ulen; out_free: + lock_sock(sk); skb_free_datagram(sk, skb); + release_sock(sk); out: return err; csum_copy_err: + lock_sock(sk); if (!skb_kill_datagram(sk, skb, flags)) UDP_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite); + release_sock(sk); if (noblock) return -EAGAIN; @@ -1072,7 +1088,15 @@ static int __udp4_lib_mcast_deliver(struct sk_buff *skb, skb1 = skb_clone(skb, GFP_ATOMIC); if (skb1) { - int ret = udp_queue_rcv_skb(sk, skb1); + int ret = 0; + + bh_lock_sock_nested(sk); + if (!sock_owned_by_user(sk)) + ret = udp_queue_rcv_skb(sk, skb1); + else + sk_add_backlog(sk, skb1); + bh_unlock_sock(sk); + if (ret > 0) /* we should probably re-process instead * of dropping packets here. */ @@ -1165,7 +1189,13 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], inet_iif(skb), udptable); if (sk != NULL) { - int ret = udp_queue_rcv_skb(sk, skb); + int ret = 0; + bh_lock_sock_nested(sk); + if (!sock_owned_by_user(sk)) + ret = udp_queue_rcv_skb(sk, skb); + else + sk_add_backlog(sk, skb); + bh_unlock_sock(sk); sock_put(sk); /* a return value > 0 means to resubmit the input, but @@ -1460,6 +1490,10 @@ struct proto udp_prot = { .hash = udp_lib_hash, .unhash = udp_lib_unhash, .get_port = udp_v4_get_port, + .memory_allocated = &udp_memory_allocated, + .sysctl_mem = sysctl_udp_mem, + .sysctl_wmem = &sysctl_udp_wmem_min, + .sysctl_rmem = &sysctl_udp_rmem_min, .obj_size = sizeof(struct udp_sock), #ifdef CONFIG_COMPAT .compat_setsockopt = compat_udp_setsockopt, @@ -1655,6 +1689,25 @@ void udp4_proc_exit(void) } #endif /* CONFIG_PROC_FS */ +void __init udp_init(void) +{ + unsigned long limit; + + /* Set the pressure threshold up by the same strategy of TCP. It is a + * fraction of global memory that is up to 1/2 at 256 MB, decreasing + * toward zero with the amount of memory, with a floor of 128 pages. + */ + limit = min(nr_all_pages, 1UL<<(28-PAGE_SHIFT)) >> (20-PAGE_SHIFT); + limit = (limit * (nr_all_pages >> (20-PAGE_SHIFT))) >> (PAGE_SHIFT-11); + limit = max(limit, 128UL); + sysctl_udp_mem[0] = limit / 4 * 3; + sysctl_udp_mem[1] = limit; + sysctl_udp_mem[2] = sysctl_udp_mem[0] * 2; + + sysctl_udp_rmem_min = SK_MEM_QUANTUM; + sysctl_udp_wmem_min = SK_MEM_QUANTUM; +} + EXPORT_SYMBOL(udp_disconnect); EXPORT_SYMBOL(udp_hash); EXPORT_SYMBOL(udp_hash_lock); diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index c9a97b405511..bf58acab2064 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -204,13 +204,17 @@ try_again: err = ulen; out_free: + lock_sock(sk); skb_free_datagram(sk, skb); + release_sock(sk); out: return err; csum_copy_err: + lock_sock(sk); if (!skb_kill_datagram(sk, skb, flags)) UDP6_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite); + release_sock(sk); if (flags & MSG_DONTWAIT) return -EAGAIN; @@ -366,10 +370,21 @@ static int __udp6_lib_mcast_deliver(struct sk_buff *skb, struct in6_addr *saddr, while ((sk2 = udp_v6_mcast_next(sk_next(sk2), uh->dest, daddr, uh->source, saddr, dif))) { struct sk_buff *buff = skb_clone(skb, GFP_ATOMIC); - if (buff) - udpv6_queue_rcv_skb(sk2, buff); + if (buff) { + bh_lock_sock_nested(sk2); + if (!sock_owned_by_user(sk2)) + udpv6_queue_rcv_skb(sk2, buff); + else + sk_add_backlog(sk2, buff); + bh_unlock_sock(sk2); + } } - udpv6_queue_rcv_skb(sk, skb); + bh_lock_sock_nested(sk); + if (!sock_owned_by_user(sk)) + udpv6_queue_rcv_skb(sk, skb); + else + sk_add_backlog(sk, skb); + bh_unlock_sock(sk); out: read_unlock(&udp_hash_lock); return 0; @@ -482,7 +497,12 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], /* deliver */ - udpv6_queue_rcv_skb(sk, skb); + bh_lock_sock_nested(sk); + if (!sock_owned_by_user(sk)) + udpv6_queue_rcv_skb(sk, skb); + else + sk_add_backlog(sk, skb); + bh_unlock_sock(sk); sock_put(sk); return 0; @@ -994,6 +1014,10 @@ struct proto udpv6_prot = { .hash = udp_lib_hash, .unhash = udp_lib_unhash, .get_port = udp_v6_get_port, + .memory_allocated = &udp_memory_allocated, + .sysctl_mem = sysctl_udp_mem, + .sysctl_wmem = &sysctl_udp_wmem_min, + .sysctl_rmem = &sysctl_udp_rmem_min, .obj_size = sizeof(struct udp6_sock), #ifdef CONFIG_COMPAT .compat_setsockopt = compat_udpv6_setsockopt, -- cgit v1.2.3 From f1862b0ae2294f6970f695abf02392d025e02dbe Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Sun, 27 Jan 2008 23:04:43 -0800 Subject: [SHAPER]: The scheduled shaper removal. This patch contains the scheduled removal of the shaper driver. Signed-off-by: Adrian Bunk Acked-by: Alan Cox Signed-off-by: David S. Miller --- Documentation/feature-removal-schedule.txt | 9 - Documentation/networking/00-INDEX | 2 - Documentation/networking/shaper.txt | 48 --- drivers/net/Kconfig | 17 - drivers/net/Makefile | 1 - drivers/net/shaper.c | 603 ----------------------------- include/linux/Kbuild | 1 - include/linux/if_shaper.h | 51 --- 8 files changed, 732 deletions(-) delete mode 100644 Documentation/networking/shaper.txt delete mode 100644 drivers/net/shaper.c delete mode 100644 include/linux/if_shaper.h (limited to 'Documentation') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 2c3ae6c02a47..cc02e67239ec 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -280,15 +280,6 @@ Who: Thomas Gleixner --------------------------- -What: shaper network driver -When: January 2008 -Files: drivers/net/shaper.c, include/linux/if_shaper.h -Why: This driver has been marked obsolete for many years. - It was only designed to work on lower speed links and has design - flaws that lead to machine crashes. The qdisc infrastructure in - 2.4 or later kernels, provides richer features and is more robust. -Who: Stephen Hemminger - --------------------------- What: i2c-i810, i2c-prosavage and i2c-savage4 diff --git a/Documentation/networking/00-INDEX b/Documentation/networking/00-INDEX index b4eefadb9adf..02e56d447a8f 100644 --- a/Documentation/networking/00-INDEX +++ b/Documentation/networking/00-INDEX @@ -84,8 +84,6 @@ policy-routing.txt - IP policy-based routing ray_cs.txt - Raylink Wireless LAN card driver info. -shaper.txt - - info on the module that can shape/limit transmitted traffic. sk98lin.txt - Marvell Yukon Chipset / SysKonnect SK-98xx compliant Gigabit Ethernet Adapter family driver info diff --git a/Documentation/networking/shaper.txt b/Documentation/networking/shaper.txt deleted file mode 100644 index 6c4ebb66a906..000000000000 --- a/Documentation/networking/shaper.txt +++ /dev/null @@ -1,48 +0,0 @@ -Traffic Shaper For Linux - -This is the current BETA release of the traffic shaper for Linux. It works -within the following limits: - -o Minimum shaping speed is currently about 9600 baud (it can only -shape down to 1 byte per clock tick) - -o Maximum is about 256K, it will go above this but get a bit blocky. - -o If you ifconfig the master device that a shaper is attached to down -then your machine will follow. - -o The shaper must be a module. - - -Setup: - - A shaper device is configured using the shapeconfig program. -Typically you will do something like this - -shapecfg attach shaper0 eth1 -shapecfg speed shaper0 64000 -ifconfig shaper0 myhost netmask 255.255.255.240 broadcast 1.2.3.4.255 up -route add -net some.network netmask a.b.c.d dev shaper0 - -The shaper should have the same IP address as the device it is attached to -for normal use. - -Gotchas: - - The shaper shapes transmitted traffic. It's rather impossible to -shape received traffic except at the end (or a router) transmitting it. - - Gated/routed/rwhod/mrouted all see the shaper as an additional device -and will treat it as such unless patched. Note that for mrouted you can run -mrouted tunnels via a traffic shaper to control bandwidth usage. - - The shaper is device/route based. This makes it very easy to use -with any setup BUT less flexible. You may need to use iproute2 to set up -multiple route tables to get the flexibility. - - There is no "borrowing" or "sharing" scheme. This is a simple -traffic limiter. We implement Van Jacobson and Sally Floyd's CBQ -architecture into Linux 2.2. This is the preferred solution. Shaper is -for simple or back compatible setups. - -Alan diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index a6728661c416..777bb81d9ada 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -3015,23 +3015,6 @@ config NET_FC adaptor below. You also should have said Y to "SCSI support" and "SCSI generic support". -config SHAPER - tristate "Traffic Shaper (OBSOLETE)" - depends on EXPERIMENTAL - ---help--- - The traffic shaper is a virtual network device that allows you to - limit the rate of outgoing data flow over some other network device. - The traffic that you want to slow down can then be routed through - these virtual devices. See - for more information. - - An alternative to this traffic shaper are traffic schedulers which - you'll get if you say Y to "QoS and/or fair queuing" in - "Networking options". - - To compile this driver as a module, choose M here: the module - will be called shaper. If unsure, say N. - config NETCONSOLE tristate "Network console logging support (EXPERIMENTAL)" depends on EXPERIMENTAL diff --git a/drivers/net/Makefile b/drivers/net/Makefile index 17a2ffd73d7d..a2f2662c244b 100644 --- a/drivers/net/Makefile +++ b/drivers/net/Makefile @@ -93,7 +93,6 @@ obj-$(CONFIG_NET_SB1000) += sb1000.o obj-$(CONFIG_MAC8390) += mac8390.o obj-$(CONFIG_APNE) += apne.o 8390.o obj-$(CONFIG_PCMCIA_PCNET) += 8390.o -obj-$(CONFIG_SHAPER) += shaper.o obj-$(CONFIG_HP100) += hp100.o obj-$(CONFIG_SMC9194) += smc9194.o obj-$(CONFIG_FEC) += fec.o diff --git a/drivers/net/shaper.c b/drivers/net/shaper.c deleted file mode 100644 index 228f650250f6..000000000000 --- a/drivers/net/shaper.c +++ /dev/null @@ -1,603 +0,0 @@ -/* - * Simple traffic shaper for Linux NET3. - * - * (c) Copyright 1996 Alan Cox , All Rights Reserved. - * http://www.redhat.com - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * Neither Alan Cox nor CymruNet Ltd. admit liability nor provide - * warranty for any of this software. This material is provided - * "AS-IS" and at no charge. - * - * - * Algorithm: - * - * Queue Frame: - * Compute time length of frame at regulated speed - * Add frame to queue at appropriate point - * Adjust time length computation for followup frames - * Any frame that falls outside of its boundaries is freed - * - * We work to the following constants - * - * SHAPER_QLEN Maximum queued frames - * SHAPER_LATENCY Bounding latency on a frame. Leaving this latency - * window drops the frame. This stops us queueing - * frames for a long time and confusing a remote - * host. - * SHAPER_MAXSLIP Maximum time a priority frame may jump forward. - * That bounds the penalty we will inflict on low - * priority traffic. - * SHAPER_BURST Time range we call "now" in order to reduce - * system load. The more we make this the burstier - * the behaviour, the better local performance you - * get through packet clustering on routers and the - * worse the remote end gets to judge rtts. - * - * This is designed to handle lower speed links ( < 200K/second or so). We - * run off a 100-150Hz base clock typically. This gives us a resolution at - * 200Kbit/second of about 2Kbit or 256 bytes. Above that our timer - * resolution may start to cause much more burstiness in the traffic. We - * could avoid a lot of that by calling kick_shaper() at the end of the - * tied device transmissions. If you run above about 100K second you - * may need to tune the supposed speed rate for the right values. - * - * BUGS: - * Downing the interface under the shaper before the shaper - * will render your machine defunct. Don't for now shape over - * PPP or SLIP therefore! - * This will be fixed in BETA4 - * - * Update History : - * - * bh_atomic() SMP races fixes and rewritten the locking code to - * be SMP safe and irq-mask friendly. - * NOTE: we can't use start_bh_atomic() in kick_shaper() - * because it's going to be recalled from an irq handler, - * and synchronize_bh() is a nono if called from irq context. - * 1999 Andrea Arcangeli - * - * Device statistics (tx_pakets, tx_bytes, - * tx_drops: queue_over_time and collisions: max_queue_exceded) - * 1999/06/18 Jordi Murgo - * - * Use skb->cb for private data. - * 2000/03 Andi Kleen - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include - -struct shaper_cb { - unsigned long shapeclock; /* Time it should go out */ - unsigned long shapestamp; /* Stamp for shaper */ - __u32 shapelatency; /* Latency on frame */ - __u32 shapelen; /* Frame length in clocks */ - __u16 shapepend; /* Pending */ -}; -#define SHAPERCB(skb) ((struct shaper_cb *) ((skb)->cb)) - -static int sh_debug; /* Debug flag */ - -#define SHAPER_BANNER "CymruNet Traffic Shaper BETA 0.04 for Linux 2.1\n" - -static void shaper_kick(struct shaper *sh); - -/* - * Compute clocks on a buffer - */ - -static int shaper_clocks(struct shaper *shaper, struct sk_buff *skb) -{ - int t=skb->len/shaper->bytespertick; - return t; -} - -/* - * Set the speed of a shaper. We compute this in bytes per tick since - * thats how the machine wants to run. Quoted input is in bits per second - * as is traditional (note not BAUD). We assume 8 bit bytes. - */ - -static void shaper_setspeed(struct shaper *shaper, int bitspersec) -{ - shaper->bitspersec=bitspersec; - shaper->bytespertick=(bitspersec/HZ)/8; - if(!shaper->bytespertick) - shaper->bytespertick++; -} - -/* - * Throw a frame at a shaper. - */ - - -static int shaper_start_xmit(struct sk_buff *skb, struct net_device *dev) -{ - struct shaper *shaper = dev->priv; - struct sk_buff *ptr; - - spin_lock(&shaper->lock); - ptr=shaper->sendq.prev; - - /* - * Set up our packet details - */ - - SHAPERCB(skb)->shapelatency=0; - SHAPERCB(skb)->shapeclock=shaper->recovery; - if(time_before(SHAPERCB(skb)->shapeclock, jiffies)) - SHAPERCB(skb)->shapeclock=jiffies; - skb->priority=0; /* short term bug fix */ - SHAPERCB(skb)->shapestamp=jiffies; - - /* - * Time slots for this packet. - */ - - SHAPERCB(skb)->shapelen= shaper_clocks(shaper,skb); - - { - struct sk_buff *tmp; - /* - * Up our shape clock by the time pending on the queue - * (Should keep this in the shaper as a variable..) - */ - for(tmp=skb_peek(&shaper->sendq); tmp!=NULL && - tmp!=(struct sk_buff *)&shaper->sendq; tmp=tmp->next) - SHAPERCB(skb)->shapeclock+=SHAPERCB(tmp)->shapelen; - /* - * Queue over time. Spill packet. - */ - if(time_after(SHAPERCB(skb)->shapeclock,jiffies + SHAPER_LATENCY)) { - dev_kfree_skb(skb); - dev->stats.tx_dropped++; - } else - skb_queue_tail(&shaper->sendq, skb); - } - - if(sh_debug) - printk("Frame queued.\n"); - if(skb_queue_len(&shaper->sendq)>SHAPER_QLEN) - { - ptr=skb_dequeue(&shaper->sendq); - dev_kfree_skb(ptr); - dev->stats.collisions++; - } - shaper_kick(shaper); - spin_unlock(&shaper->lock); - return 0; -} - -/* - * Transmit from a shaper - */ - -static void shaper_queue_xmit(struct shaper *shaper, struct sk_buff *skb) -{ - struct sk_buff *newskb=skb_clone(skb, GFP_ATOMIC); - if(sh_debug) - printk("Kick frame on %p\n",newskb); - if(newskb) - { - newskb->dev=shaper->dev; - newskb->priority=2; - if(sh_debug) - printk("Kick new frame to %s, %d\n", - shaper->dev->name,newskb->priority); - dev_queue_xmit(newskb); - - shaper->dev->stats.tx_bytes += skb->len; - shaper->dev->stats.tx_packets++; - - if(sh_debug) - printk("Kicked new frame out.\n"); - dev_kfree_skb(skb); - } -} - -/* - * Timer handler for shaping clock - */ - -static void shaper_timer(unsigned long data) -{ - struct shaper *shaper = (struct shaper *)data; - - spin_lock(&shaper->lock); - shaper_kick(shaper); - spin_unlock(&shaper->lock); -} - -/* - * Kick a shaper queue and try and do something sensible with the - * queue. - */ - -static void shaper_kick(struct shaper *shaper) -{ - struct sk_buff *skb; - - /* - * Walk the list (may be empty) - */ - - while((skb=skb_peek(&shaper->sendq))!=NULL) - { - /* - * Each packet due to go out by now (within an error - * of SHAPER_BURST) gets kicked onto the link - */ - - if(sh_debug) - printk("Clock = %ld, jiffies = %ld\n", SHAPERCB(skb)->shapeclock, jiffies); - if(time_before_eq(SHAPERCB(skb)->shapeclock, jiffies + SHAPER_BURST)) - { - /* - * Pull the frame and get interrupts back on. - */ - - skb_unlink(skb, &shaper->sendq); - if (shaper->recovery < - SHAPERCB(skb)->shapeclock + SHAPERCB(skb)->shapelen) - shaper->recovery = SHAPERCB(skb)->shapeclock + SHAPERCB(skb)->shapelen; - /* - * Pass on to the physical target device via - * our low level packet thrower. - */ - - SHAPERCB(skb)->shapepend=0; - shaper_queue_xmit(shaper, skb); /* Fire */ - } - else - break; - } - - /* - * Next kick. - */ - - if(skb!=NULL) - mod_timer(&shaper->timer, SHAPERCB(skb)->shapeclock); -} - - -/* - * Bring the interface up. We just disallow this until a - * bind. - */ - -static int shaper_open(struct net_device *dev) -{ - struct shaper *shaper=dev->priv; - - /* - * Can't open until attached. - * Also can't open until speed is set, or we'll get - * a division by zero. - */ - - if(shaper->dev==NULL) - return -ENODEV; - if(shaper->bitspersec==0) - return -EINVAL; - return 0; -} - -/* - * Closing a shaper flushes the queues. - */ - -static int shaper_close(struct net_device *dev) -{ - struct shaper *shaper=dev->priv; - struct sk_buff *skb; - - while ((skb = skb_dequeue(&shaper->sendq)) != NULL) - dev_kfree_skb(skb); - - spin_lock_bh(&shaper->lock); - shaper_kick(shaper); - spin_unlock_bh(&shaper->lock); - - del_timer_sync(&shaper->timer); - return 0; -} - -/* - * Revectored calls. We alter the parameters and call the functions - * for our attached device. This enables us to bandwidth allocate after - * ARP and other resolutions and not before. - */ - -static int shaper_header(struct sk_buff *skb, struct net_device *dev, - unsigned short type, - const void *daddr, const void *saddr, unsigned len) -{ - struct shaper *sh=dev->priv; - int v; - if(sh_debug) - printk("Shaper header\n"); - skb->dev = sh->dev; - v = dev_hard_header(skb, sh->dev, type, daddr, saddr, len); - skb->dev = dev; - return v; -} - -static int shaper_rebuild_header(struct sk_buff *skb) -{ - struct shaper *sh=skb->dev->priv; - struct net_device *dev=skb->dev; - int v; - if(sh_debug) - printk("Shaper rebuild header\n"); - skb->dev=sh->dev; - v = sh->dev->header_ops->rebuild(skb); - skb->dev=dev; - return v; -} - -#if 0 -static int shaper_cache(struct neighbour *neigh, struct hh_cache *hh) -{ - struct shaper *sh=neigh->dev->priv; - struct net_device *tmp; - int ret; - if(sh_debug) - printk("Shaper header cache bind\n"); - tmp=neigh->dev; - neigh->dev=sh->dev; - ret=sh->hard_header_cache(neigh,hh); - neigh->dev=tmp; - return ret; -} - -static void shaper_cache_update(struct hh_cache *hh, struct net_device *dev, - unsigned char *haddr) -{ - struct shaper *sh=dev->priv; - if(sh_debug) - printk("Shaper cache update\n"); - sh->header_cache_update(hh, sh->dev, haddr); -} -#endif - -#ifdef CONFIG_INET - -static int shaper_neigh_setup(struct neighbour *n) -{ -#ifdef CONFIG_INET - if (n->nud_state == NUD_NONE) { - n->ops = &arp_broken_ops; - n->output = n->ops->output; - } -#endif - return 0; -} - -static int shaper_neigh_setup_dev(struct net_device *dev, struct neigh_parms *p) -{ -#ifdef CONFIG_INET - if (p->tbl->family == AF_INET) { - p->neigh_setup = shaper_neigh_setup; - p->ucast_probes = 0; - p->mcast_probes = 0; - } -#endif - return 0; -} - -#else /* !(CONFIG_INET) */ - -static int shaper_neigh_setup_dev(struct net_device *dev, struct neigh_parms *p) -{ - return 0; -} - -#endif - -static const struct header_ops shaper_ops = { - .create = shaper_header, - .rebuild = shaper_rebuild_header, -}; - -static int shaper_attach(struct net_device *shdev, struct shaper *sh, struct net_device *dev) -{ - sh->dev = dev; - sh->get_stats=dev->get_stats; - - shdev->neigh_setup = shaper_neigh_setup_dev; - shdev->hard_header_len=dev->hard_header_len; - shdev->type=dev->type; - shdev->addr_len=dev->addr_len; - shdev->mtu=dev->mtu; - sh->bitspersec=0; - return 0; -} - -static int shaper_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) -{ - struct shaperconf *ss= (struct shaperconf *)&ifr->ifr_ifru; - struct shaper *sh=dev->priv; - - if(ss->ss_cmd == SHAPER_SET_DEV || ss->ss_cmd == SHAPER_SET_SPEED) - { - if(!capable(CAP_NET_ADMIN)) - return -EPERM; - } - - switch(ss->ss_cmd) - { - case SHAPER_SET_DEV: - { - struct net_device *them=__dev_get_by_name(&init_net, ss->ss_name); - if(them==NULL) - return -ENODEV; - if(sh->dev) - return -EBUSY; - return shaper_attach(dev,dev->priv, them); - } - case SHAPER_GET_DEV: - if(sh->dev==NULL) - return -ENODEV; - strcpy(ss->ss_name, sh->dev->name); - return 0; - case SHAPER_SET_SPEED: - shaper_setspeed(sh,ss->ss_speed); - return 0; - case SHAPER_GET_SPEED: - ss->ss_speed=sh->bitspersec; - return 0; - default: - return -EINVAL; - } -} - -static void shaper_init_priv(struct net_device *dev) -{ - struct shaper *sh = dev->priv; - - skb_queue_head_init(&sh->sendq); - init_timer(&sh->timer); - sh->timer.function=shaper_timer; - sh->timer.data=(unsigned long)sh; - spin_lock_init(&sh->lock); -} - -/* - * Add a shaper device to the system - */ - -static void __init shaper_setup(struct net_device *dev) -{ - /* - * Set up the shaper. - */ - - shaper_init_priv(dev); - - dev->open = shaper_open; - dev->stop = shaper_close; - dev->hard_start_xmit = shaper_start_xmit; - dev->set_multicast_list = NULL; - - /* - * Intialise the packet queues - */ - - /* - * Handlers for when we attach to a device. - */ - - dev->neigh_setup = shaper_neigh_setup_dev; - dev->do_ioctl = shaper_ioctl; - dev->hard_header_len = 0; - dev->type = ARPHRD_ETHER; /* initially */ - dev->set_mac_address = NULL; - dev->mtu = 1500; - dev->addr_len = 0; - dev->tx_queue_len = 10; - dev->flags = 0; -} - -static int shapers = 1; -#ifdef MODULE - -module_param(shapers, int, 0); -MODULE_PARM_DESC(shapers, "Traffic shaper: maximum number of shapers"); - -#else /* MODULE */ - -static int __init set_num_shapers(char *str) -{ - shapers = simple_strtol(str, NULL, 0); - return 1; -} - -__setup("shapers=", set_num_shapers); - -#endif /* MODULE */ - -static struct net_device **devs; - -static unsigned int shapers_registered = 0; - -static int __init shaper_init(void) -{ - int i; - size_t alloc_size; - struct net_device *dev; - char name[IFNAMSIZ]; - - if (shapers < 1) - return -ENODEV; - - alloc_size = sizeof(*dev) * shapers; - devs = kzalloc(alloc_size, GFP_KERNEL); - if (!devs) - return -ENOMEM; - - for (i = 0; i < shapers; i++) { - - snprintf(name, IFNAMSIZ, "shaper%d", i); - dev = alloc_netdev(sizeof(struct shaper), name, - shaper_setup); - if (!dev) - break; - - if (register_netdev(dev)) { - free_netdev(dev); - break; - } - - devs[i] = dev; - shapers_registered++; - } - - if (!shapers_registered) { - kfree(devs); - devs = NULL; - } - - return (shapers_registered ? 0 : -ENODEV); -} - -static void __exit shaper_exit (void) -{ - int i; - - for (i = 0; i < shapers_registered; i++) { - if (devs[i]) { - unregister_netdev(devs[i]); - free_netdev(devs[i]); - } - } - - kfree(devs); - devs = NULL; -} - -module_init(shaper_init); -module_exit(shaper_exit); -MODULE_LICENSE("GPL"); - diff --git a/include/linux/Kbuild b/include/linux/Kbuild index a85e87fd60d7..bc33a5c87d64 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -231,7 +231,6 @@ unifdef-y += if_ltalk.h unifdef-y += if_link.h unifdef-y += if_pppol2tp.h unifdef-y += if_pppox.h -unifdef-y += if_shaper.h unifdef-y += if_tr.h unifdef-y += if_tun.h unifdef-y += if_vlan.h diff --git a/include/linux/if_shaper.h b/include/linux/if_shaper.h deleted file mode 100644 index 3b1b7ba19825..000000000000 --- a/include/linux/if_shaper.h +++ /dev/null @@ -1,51 +0,0 @@ -#ifndef __LINUX_SHAPER_H -#define __LINUX_SHAPER_H - -#ifdef __KERNEL__ - -#define SHAPER_QLEN 10 -/* - * This is a bit speed dependent (read it shouldn't be a constant!) - * - * 5 is about right for 28.8 upwards. Below that double for every - * halving of speed or so. - ie about 20 for 9600 baud. - */ -#define SHAPER_LATENCY (5*HZ) -#define SHAPER_MAXSLIP 2 -#define SHAPER_BURST (HZ/50) /* Good for >128K then */ - -struct shaper -{ - struct sk_buff_head sendq; - __u32 bytespertick; - __u32 bitspersec; - __u32 shapelatency; - __u32 shapeclock; - unsigned long recovery; /* Time we can next clock a packet out on - an empty queue */ - spinlock_t lock; - struct net_device *dev; - struct net_device_stats* (*get_stats)(struct net_device *dev); - struct timer_list timer; -}; - -#endif - -#define SHAPER_SET_DEV 0x0001 -#define SHAPER_SET_SPEED 0x0002 -#define SHAPER_GET_DEV 0x0003 -#define SHAPER_GET_SPEED 0x0004 - -struct shaperconf -{ - __u16 ss_cmd; - union - { - char ssu_name[14]; - __u32 ssu_speed; - } ss_u; -#define ss_speed ss_u.ssu_speed -#define ss_name ss_u.ssu_name -}; - -#endif -- cgit v1.2.3 From f9ef8a23c37677929e95ad4026a2fa0d449c1d3e Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 14 Jan 2008 23:43:34 -0800 Subject: [NETFILTER]: Update feature-removal-schedule.txt With all the newly introduced features, there is a lot to remove later on after a compatibility grace period of 2 years. Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- Documentation/feature-removal-schedule.txt | 32 ++++++++++++++++++++++++++++++ 1 file changed, 32 insertions(+) (limited to 'Documentation') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index cc02e67239ec..166915df8d6a 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -313,3 +313,35 @@ Why: This algorithm was provided for reference but always exhibited bad responsiveness and performance and has some serious flaws. It has been replaced by rc80211-pid. Who: Stefano Brivio + +--------------------------- + +What (Why): + - include/linux/netfilter_ipv4/ipt_TOS.h ipt_tos.h header files + (superseded by xt_TOS/xt_tos target & match) + + - "forwarding" header files like ipt_mac.h in + include/linux/netfilter_ipv4/ and include/linux/netfilter_ipv6/ + + - xt_CONNMARK match revision 0 + (superseded by xt_CONNMARK match revision 1) + + - xt_MARK target revisions 0 and 1 + (superseded by xt_MARK match revision 2) + + - xt_connmark match revision 0 + (superseded by xt_connmark match revision 1) + + - xt_conntrack match revision 0 + (superseded by xt_conntrack match revision 1) + + - xt_iprange match revision 0, + include/linux/netfilter_ipv4/ipt_iprange.h + (superseded by xt_iprange match revision 1) + + - xt_mark match revision 0 + (superseded by xt_mark match revision 1) + +When: January 2009 or Linux 2.7.0, whichever comes first +Why: Superseded by newer revisions or modules +Who: Jan Engelhardt -- cgit v1.2.3 From 9a6c686799346b6c95405c9e051f5023873504fa Mon Sep 17 00:00:00 2001 From: Jay Vosburgh Date: Tue, 13 Nov 2007 20:25:48 -0800 Subject: [BONDING]: Documentation update Update the bonding documentation: more discussion on initialization and configuration, changes to discussion of packet reordering in balance-rr, update some out of date information. Based in part on input from Rick Jones and Andy Gospodarek . Signed-off-by: Jay Vosburgh Acked-by: Andy Gospodarek Signed-off-by: David S. Miller --- Documentation/networking/bonding.txt | 204 ++++++++++++++++++++++++----------- 1 file changed, 143 insertions(+), 61 deletions(-) (limited to 'Documentation') diff --git a/Documentation/networking/bonding.txt b/Documentation/networking/bonding.txt index 6cc30e0d5795..a0cda062bc33 100644 --- a/Documentation/networking/bonding.txt +++ b/Documentation/networking/bonding.txt @@ -1,7 +1,7 @@ Linux Ethernet Bonding Driver HOWTO - Latest update: 24 April 2006 + Latest update: 12 November 2007 Initial release : Thomas Davis Corrections, HA extensions : 2000/10/03-15 : @@ -166,12 +166,17 @@ to use ifenslave. 2. Bonding Driver Options ========================= - Options for the bonding driver are supplied as parameters to -the bonding module at load time. They may be given as command line -arguments to the insmod or modprobe command, but are usually specified -in either the /etc/modules.conf or /etc/modprobe.conf configuration -file, or in a distro-specific configuration file (some of which are -detailed in the next section). + Options for the bonding driver are supplied as parameters to the +bonding module at load time, or are specified via sysfs. + + Module options may be given as command line arguments to the +insmod or modprobe command, but are usually specified in either the +/etc/modules.conf or /etc/modprobe.conf configuration file, or in a +distro-specific configuration file (some of which are detailed in the next +section). + + Details on bonding support for sysfs is provided in the +"Configuring Bonding Manually via Sysfs" section, below. The available bonding driver parameters are listed below. If a parameter is not specified the default value is used. When initially @@ -812,11 +817,13 @@ the system /etc/modules.conf or /etc/modprobe.conf configuration file. 3.2 Configuration with Initscripts Support ------------------------------------------ - This section applies to distros using a version of initscripts -with bonding support, for example, Red Hat Linux 9 or Red Hat -Enterprise Linux version 3 or 4. On these systems, the network -initialization scripts have some knowledge of bonding, and can be -configured to control bonding devices. + This section applies to distros using a recent version of +initscripts with bonding support, for example, Red Hat Enterprise Linux +version 3 or later, Fedora, etc. On these systems, the network +initialization scripts have knowledge of bonding, and can be configured to +control bonding devices. Note that older versions of the initscripts +package have lower levels of support for bonding; this will be noted where +applicable. These distros will not automatically load the network adapter driver unless the ethX device is configured with an IP address. @@ -864,11 +871,31 @@ USERCTL=no Be sure to change the networking specific lines (IPADDR, NETMASK, NETWORK and BROADCAST) to match your network configuration. - Finally, it is necessary to edit /etc/modules.conf (or -/etc/modprobe.conf, depending upon your distro) to load the bonding -module with your desired options when the bond0 interface is brought -up. The following lines in /etc/modules.conf (or modprobe.conf) will -load the bonding module, and select its options: + For later versions of initscripts, such as that found with Fedora +7 and Red Hat Enterprise Linux version 5 (or later), it is possible, and, +indeed, preferable, to specify the bonding options in the ifcfg-bond0 +file, e.g. a line of the format: + +BONDING_OPTS="mode=active-backup arp_interval=60 arp_ip_target=+192.168.1.254" + + will configure the bond with the specified options. The options +specified in BONDING_OPTS are identical to the bonding module parameters +except for the arp_ip_target field. Each target should be included as a +separate option and should be preceded by a '+' to indicate it should be +added to the list of queried targets, e.g., + + arp_ip_target=+192.168.1.1 arp_ip_target=+192.168.1.2 + + is the proper syntax to specify multiple targets. When specifying +options via BONDING_OPTS, it is not necessary to edit /etc/modules.conf or +/etc/modprobe.conf. + + For older versions of initscripts that do not support +BONDING_OPTS, it is necessary to edit /etc/modules.conf (or +/etc/modprobe.conf, depending upon your distro) to load the bonding module +with your desired options when the bond0 interface is brought up. The +following lines in /etc/modules.conf (or modprobe.conf) will load the +bonding module, and select its options: alias bond0 bonding options bond0 mode=balance-alb miimon=100 @@ -883,9 +910,10 @@ up and running. 3.2.1 Using DHCP with Initscripts --------------------------------- - Recent versions of initscripts (the version supplied with -Fedora Core 3 and Red Hat Enterprise Linux 4 is reported to work) do -have support for assigning IP information to bonding devices via DHCP. + Recent versions of initscripts (the versions supplied with Fedora +Core 3 and Red Hat Enterprise Linux 4, or later versions, are reported to +work) have support for assigning IP information to bonding devices via +DHCP. To configure bonding for DHCP, configure it as described above, except replace the line "BOOTPROTO=none" with "BOOTPROTO=dhcp" @@ -895,18 +923,14 @@ is case sensitive. 3.2.2 Configuring Multiple Bonds with Initscripts ------------------------------------------------- - At this writing, the initscripts package does not directly -support loading the bonding driver multiple times, so the process for -doing so is the same as described in the "Configuring Multiple Bonds -Manually" section, below. - - NOTE: It has been observed that some Red Hat supplied kernels -are apparently unable to rename modules at load time (the "-o bond1" -part). Attempts to pass that option to modprobe will produce an -"Operation not permitted" error. This has been reported on some -Fedora Core kernels, and has been seen on RHEL 4 as well. On kernels -exhibiting this problem, it will be impossible to configure multiple -bonds with differing parameters. + Initscripts packages that are included with Fedora 7 and Red Hat +Enterprise Linux 5 support multiple bonding interfaces by simply +specifying the appropriate BONDING_OPTS= in ifcfg-bondX where X is the +number of the bond. This support requires sysfs support in the kernel, +and a bonding driver of version 3.0.0 or later. Other configurations may +not support this method for specifying multiple bonding interfaces; for +those instances, see the "Configuring Multiple Bonds Manually" section, +below. 3.3 Configuring Bonding Manually with Ifenslave ----------------------------------------------- @@ -977,15 +1001,58 @@ initialization scripts lack support for configuring multiple bonds. options, you may wish to use the "max_bonds" module parameter, documented above. - To create multiple bonding devices with differing options, it -is necessary to use bonding parameters exported by sysfs, documented -in the section below. + To create multiple bonding devices with differing options, it is +preferrable to use bonding parameters exported by sysfs, documented in the +section below. + + For versions of bonding without sysfs support, the only means to +provide multiple instances of bonding with differing options is to load +the bonding driver multiple times. Note that current versions of the +sysconfig network initialization scripts handle this automatically; if +your distro uses these scripts, no special action is needed. See the +section Configuring Bonding Devices, above, if you're not sure about your +network initialization scripts. + + To load multiple instances of the module, it is necessary to +specify a different name for each instance (the module loading system +requires that every loaded module, even multiple instances of the same +module, have a unique name). This is accomplished by supplying multiple +sets of bonding options in /etc/modprobe.conf, for example: + +alias bond0 bonding +options bond0 -o bond0 mode=balance-rr miimon=100 + +alias bond1 bonding +options bond1 -o bond1 mode=balance-alb miimon=50 + + will load the bonding module two times. The first instance is +named "bond0" and creates the bond0 device in balance-rr mode with an +miimon of 100. The second instance is named "bond1" and creates the +bond1 device in balance-alb mode with an miimon of 50. + + In some circumstances (typically with older distributions), +the above does not work, and the second bonding instance never sees +its options. In that case, the second options line can be substituted +as follows: + +install bond1 /sbin/modprobe --ignore-install bonding -o bond1 \ + mode=balance-alb miimon=50 + This may be repeated any number of times, specifying a new and +unique name in place of bond1 for each subsequent instance. + + It has been observed that some Red Hat supplied kernels are unable +to rename modules at load time (the "-o bond1" part). Attempts to pass +that option to modprobe will produce an "Operation not permitted" error. +This has been reported on some Fedora Core kernels, and has been seen on +RHEL 4 as well. On kernels exhibiting this problem, it will be impossible +to configure multiple bonds with differing parameters (as they are older +kernels, and also lack sysfs support). 3.4 Configuring Bonding Manually via Sysfs ------------------------------------------ - Starting with version 3.0, Channel Bonding may be configured + Starting with version 3.0.0, Channel Bonding may be configured via the sysfs interface. This interface allows dynamic configuration of all bonds in the system without unloading the module. It also allows for adding and removing bonds at runtime. Ifenslave is no @@ -1030,9 +1097,6 @@ To enslave interface eth0 to bond bond0: To free slave eth0 from bond bond0: # echo -eth0 > /sys/class/net/bond0/bonding/slaves - NOTE: The bond must be up before slaves can be added. All -slaves are freed when the interface is brought down. - When an interface is enslaved to a bond, symlinks between the two are created in the sysfs filesystem. In this case, you would get /sys/class/net/bond0/slave_eth0 pointing to /sys/class/net/eth0, and @@ -1622,6 +1686,15 @@ one for each switch in the network). This will insure that, regardless of which switch is active, the ARP monitor has a suitable target to query. + Note, also, that of late many switches now support a functionality +generally referred to as "trunk failover." This is a feature of the +switch that causes the link state of a particular switch port to be set +down (or up) when the state of another switch port goes down (or up). +It's purpose is to propogate link failures from logically "exterior" ports +to the logically "interior" ports that bonding is able to monitor via +miimon. Availability and configuration for trunk failover varies by +switch, but this can be a viable alternative to the ARP monitor when using +suitable switches. 12. Configuring Bonding for Maximum Throughput ============================================== @@ -1709,7 +1782,7 @@ balance-rr: This mode is the only mode that will permit a single interfaces. It is therefore the only mode that will allow a single TCP/IP stream to utilize more than one interface's worth of throughput. This comes at a cost, however: the - striping often results in peer systems receiving packets out + striping generally results in peer systems receiving packets out of order, causing TCP/IP's congestion control system to kick in, often by retransmitting segments. @@ -1721,22 +1794,20 @@ balance-rr: This mode is the only mode that will permit a single interface's worth of throughput, even after adjusting tcp_reordering. - Note that this out of order delivery occurs when both the - sending and receiving systems are utilizing a multiple - interface bond. Consider a configuration in which a - balance-rr bond feeds into a single higher capacity network - channel (e.g., multiple 100Mb/sec ethernets feeding a single - gigabit ethernet via an etherchannel capable switch). In this - configuration, traffic sent from the multiple 100Mb devices to - a destination connected to the gigabit device will not see - packets out of order. However, traffic sent from the gigabit - device to the multiple 100Mb devices may or may not see - traffic out of order, depending upon the balance policy of the - switch. Many switches do not support any modes that stripe - traffic (instead choosing a port based upon IP or MAC level - addresses); for those devices, traffic flowing from the - gigabit device to the many 100Mb devices will only utilize one - interface. + Note that the fraction of packets that will be delivered out of + order is highly variable, and is unlikely to be zero. The level + of reordering depends upon a variety of factors, including the + networking interfaces, the switch, and the topology of the + configuration. Speaking in general terms, higher speed network + cards produce more reordering (due to factors such as packet + coalescing), and a "many to many" topology will reorder at a + higher rate than a "many slow to one fast" configuration. + + Many switches do not support any modes that stripe traffic + (instead choosing a port based upon IP or MAC level addresses); + for those devices, traffic for a particular connection flowing + through the switch to a balance-rr bond will not utilize greater + than one interface's worth of bandwidth. If you are utilizing protocols other than TCP/IP, UDP for example, and your application can tolerate out of order @@ -1936,6 +2007,10 @@ Failover may be delayed via the downdelay bonding module option. 13.2 Duplicated Incoming Packets -------------------------------- + NOTE: Starting with version 3.0.2, the bonding driver has logic to +suppress duplicate packets, which should largely eliminate this problem. +The following description is kept for reference. + It is not uncommon to observe a short burst of duplicated traffic when the bonding device is first used, or after it has been idle for some period of time. This is most easily observed by issuing @@ -2096,6 +2171,9 @@ The new driver was designed to be SMP safe from the start. EtherExpress PRO/100 and a 3com 3c905b, for example). For most modes, devices need not be of the same speed. + Starting with version 3.2.1, bonding also supports Infiniband +slaves in active-backup mode. + 3. How many bonding devices can I have? There is no limit. @@ -2154,11 +2232,15 @@ switches currently available support 802.3ad. 8. Where does a bonding device get its MAC address from? - If not explicitly configured (with ifconfig or ip link), the -MAC address of the bonding device is taken from its first slave -device. This MAC address is then passed to all following slaves and -remains persistent (even if the first slave is removed) until the -bonding device is brought down or reconfigured. + When using slave devices that have fixed MAC addresses, or when +the fail_over_mac option is enabled, the bonding device's MAC address is +the MAC address of the active slave. + + For other configurations, if not explicitly configured (with +ifconfig or ip link), the MAC address of the bonding device is taken from +its first slave device. This MAC address is then passed to all following +slaves and remains persistent (even if the first slave is removed) until +the bonding device is brought down or reconfigured. If you wish to change the MAC address, you can set it with ifconfig or ip link: -- cgit v1.2.3 From eb189d8bc9824bcb2187ffdab27d77ab469264c3 Mon Sep 17 00:00:00 2001 From: Michael Buesch Date: Mon, 28 Jan 2008 14:47:41 -0800 Subject: b43: Add support for new firmware This patch adds support for new firmware. Old firmware is still supported until July 2008. To get new firmware, go to ftp://ftp.linksys.com/opensourcecode/wrt150nv11/1.51.3/ and download the tarball. We don't have a smaller tarball, yet. That will be fixed later. You can extract firmware out of the "wl_ap.o" file contained in this tarball using latest fwcutter. You must pass the option --unsupported to fwcutter. Fwcutter-010 with official support for a new firmware image will be released soon. Signed-off-by: Michael Buesch Signed-off-by: John W. Linville Signed-off-by: David S. Miller --- Documentation/feature-removal-schedule.txt | 9 ++ drivers/net/wireless/b43/dma.c | 22 ++-- drivers/net/wireless/b43/main.c | 50 ++++++-- drivers/net/wireless/b43/phy.h | 2 + drivers/net/wireless/b43/xmit.c | 144 +++++++++++++-------- drivers/net/wireless/b43/xmit.h | 200 +++++++++++++++++++++-------- 6 files changed, 298 insertions(+), 129 deletions(-) (limited to 'Documentation') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 166915df8d6a..181bff005167 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -345,3 +345,12 @@ What (Why): When: January 2009 or Linux 2.7.0, whichever comes first Why: Superseded by newer revisions or modules Who: Jan Engelhardt + +--------------------------- + +What: b43 support for firmware revision < 410 +When: July 2008 +Why: The support code for the old firmware hurts code readability/maintainability + and slightly hurts runtime performance. Bugfixes for the old firmware + are not provided by Broadcom anymore. +Who: Michael Buesch diff --git a/drivers/net/wireless/b43/dma.c b/drivers/net/wireless/b43/dma.c index cf92853a2180..3e73d2a523aa 100644 --- a/drivers/net/wireless/b43/dma.c +++ b/drivers/net/wireless/b43/dma.c @@ -807,7 +807,7 @@ struct b43_dmaring *b43_setup_dmaring(struct b43_wldev *dev, goto err_kfree_ring; if (for_tx) { ring->txhdr_cache = kcalloc(nr_slots, - sizeof(struct b43_txhdr_fw4), + b43_txhdr_size(dev), GFP_KERNEL); if (!ring->txhdr_cache) goto err_kfree_meta; @@ -815,22 +815,21 @@ struct b43_dmaring *b43_setup_dmaring(struct b43_wldev *dev, /* test for ability to dma to txhdr_cache */ dma_test = dma_map_single(dev->dev->dev, ring->txhdr_cache, - sizeof(struct b43_txhdr_fw4), + b43_txhdr_size(dev), DMA_TO_DEVICE); if (dma_mapping_error(dma_test)) { /* ugh realloc */ kfree(ring->txhdr_cache); ring->txhdr_cache = kcalloc(nr_slots, - sizeof(struct - b43_txhdr_fw4), + b43_txhdr_size(dev), GFP_KERNEL | GFP_DMA); if (!ring->txhdr_cache) goto err_kfree_meta; dma_test = dma_map_single(dev->dev->dev, ring->txhdr_cache, - sizeof(struct b43_txhdr_fw4), + b43_txhdr_size(dev), DMA_TO_DEVICE); if (dma_mapping_error(dma_test)) @@ -838,7 +837,7 @@ struct b43_dmaring *b43_setup_dmaring(struct b43_wldev *dev, } dma_unmap_single(dev->dev->dev, - dma_test, sizeof(struct b43_txhdr_fw4), + dma_test, b43_txhdr_size(dev), DMA_TO_DEVICE); } @@ -1122,6 +1121,7 @@ static int dma_tx_fragment(struct b43_dmaring *ring, struct b43_dmadesc_meta *meta_hdr; struct sk_buff *bounce_skb; u16 cookie; + size_t hdrsize = b43_txhdr_size(ring->dev); #define SLOTS_PER_PACKET 2 B43_WARN_ON(skb_shinfo(skb)->nr_frags); @@ -1131,17 +1131,17 @@ static int dma_tx_fragment(struct b43_dmaring *ring, desc = ops->idx2desc(ring, slot, &meta_hdr); memset(meta_hdr, 0, sizeof(*meta_hdr)); - header = &(ring->txhdr_cache[slot * sizeof(struct b43_txhdr_fw4)]); + header = &(ring->txhdr_cache[slot * hdrsize]); cookie = generate_cookie(ring, slot); b43_generate_txhdr(ring->dev, header, skb->data, skb->len, ctl, cookie); meta_hdr->dmaaddr = map_descbuffer(ring, (unsigned char *)header, - sizeof(struct b43_txhdr_fw4), 1); + hdrsize, 1); if (dma_mapping_error(meta_hdr->dmaaddr)) return -EIO; ops->fill_descriptor(ring, desc, meta_hdr->dmaaddr, - sizeof(struct b43_txhdr_fw4), 1, 0, 0); + hdrsize, 1, 0, 0); /* Get a slot for the payload. */ slot = request_slot(ring); @@ -1189,7 +1189,7 @@ out_free_bounce: dev_kfree_skb_any(skb); out_unmap_hdr: unmap_descbuffer(ring, meta_hdr->dmaaddr, - sizeof(struct b43_txhdr_fw4), 1); + hdrsize, 1); return err; } @@ -1298,7 +1298,7 @@ void b43_dma_handle_txstatus(struct b43_wldev *dev, 1); else unmap_descbuffer(ring, meta->dmaaddr, - sizeof(struct b43_txhdr_fw4), 1); + b43_txhdr_size(dev), 1); if (meta->is_last_fragment) { B43_WARN_ON(!meta->skb); diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c index 481bc8238e7c..560d1421e679 100644 --- a/drivers/net/wireless/b43/main.c +++ b/drivers/net/wireless/b43/main.c @@ -1569,11 +1569,17 @@ static void b43_release_firmware(struct b43_wldev *dev) dev->fw.initvals_band = NULL; } -static void b43_print_fw_helptext(struct b43_wl *wl) +static void b43_print_fw_helptext(struct b43_wl *wl, bool error) { - b43err(wl, "You must go to " + const char *text; + + text = "You must go to " "http://linuxwireless.org/en/users/Drivers/b43#devicefirmware " - "and download the correct firmware (version 4).\n"); + "and download the latest firmware (version 4).\n"; + if (error) + b43err(wl, text); + else + b43warn(wl, text); } static int do_request_fw(struct b43_wldev *dev, @@ -1725,7 +1731,7 @@ static int b43_request_firmware(struct b43_wldev *dev) return 0; err_load: - b43_print_fw_helptext(dev->wl); + b43_print_fw_helptext(dev->wl, 1); goto error; err_no_ucode: @@ -1795,7 +1801,7 @@ static int b43_upload_microcode(struct b43_wldev *dev) i++; if (i >= 50) { b43err(dev->wl, "Microcode not responding\n"); - b43_print_fw_helptext(dev->wl); + b43_print_fw_helptext(dev->wl, 1); err = -ENODEV; goto out; } @@ -1813,7 +1819,7 @@ static int b43_upload_microcode(struct b43_wldev *dev) b43err(dev->wl, "YOUR FIRMWARE IS TOO OLD. Firmware from " "binary drivers older than version 4.x is unsupported. " "You must upgrade your firmware files.\n"); - b43_print_fw_helptext(dev->wl); + b43_print_fw_helptext(dev->wl, 1); b43_write32(dev, B43_MMIO_MACCTL, 0); err = -EOPNOTSUPP; goto out; @@ -1827,7 +1833,13 @@ static int b43_upload_microcode(struct b43_wldev *dev) dev->fw.rev = fwrev; dev->fw.patch = fwpatch; - out: + if (b43_is_old_txhdr_format(dev)) { + b43warn(dev->wl, "You are using an old firmware image. " + "Support for old firmware will be removed in July 2008.\n"); + b43_print_fw_helptext(dev->wl, 0); + } + +out: return err; } @@ -1887,7 +1899,7 @@ static int b43_write_initvals(struct b43_wldev *dev, err_format: b43err(dev->wl, "Initial Values Firmware file-format error.\n"); - b43_print_fw_helptext(dev->wl); + b43_print_fw_helptext(dev->wl, 1); return -EPROTO; } @@ -2149,13 +2161,19 @@ static void b43_mgmtframe_txantenna(struct b43_wldev *dev, int antenna) switch (antenna) { case B43_ANTENNA0: - ant |= B43_TX4_PHY_ANT0; + ant |= B43_TXH_PHY_ANT0; break; case B43_ANTENNA1: - ant |= B43_TX4_PHY_ANT1; + ant |= B43_TXH_PHY_ANT1; + break; + case B43_ANTENNA2: + ant |= B43_TXH_PHY_ANT2; + break; + case B43_ANTENNA3: + ant |= B43_TXH_PHY_ANT3; break; case B43_ANTENNA_AUTO: - ant |= B43_TX4_PHY_ANTLAST; + ant |= B43_TXH_PHY_ANT01AUTO; break; default: B43_WARN_ON(1); @@ -2165,15 +2183,15 @@ static void b43_mgmtframe_txantenna(struct b43_wldev *dev, int antenna) /* For Beacons */ tmp = b43_shm_read16(dev, B43_SHM_SHARED, B43_SHM_SH_BEACPHYCTL); - tmp = (tmp & ~B43_TX4_PHY_ANT) | ant; + tmp = (tmp & ~B43_TXH_PHY_ANT) | ant; b43_shm_write16(dev, B43_SHM_SHARED, B43_SHM_SH_BEACPHYCTL, tmp); /* For ACK/CTS */ tmp = b43_shm_read16(dev, B43_SHM_SHARED, B43_SHM_SH_ACKCTSPHYCTL); - tmp = (tmp & ~B43_TX4_PHY_ANT) | ant; + tmp = (tmp & ~B43_TXH_PHY_ANT) | ant; b43_shm_write16(dev, B43_SHM_SHARED, B43_SHM_SH_ACKCTSPHYCTL, tmp); /* For Probe Resposes */ tmp = b43_shm_read16(dev, B43_SHM_SHARED, B43_SHM_SH_PRPHYCTL); - tmp = (tmp & ~B43_TX4_PHY_ANT) | ant; + tmp = (tmp & ~B43_TXH_PHY_ANT) | ant; b43_shm_write16(dev, B43_SHM_SHARED, B43_SHM_SH_PRPHYCTL, tmp); } @@ -2738,6 +2756,10 @@ static int b43_antenna_from_ieee80211(struct b43_wldev *dev, u8 antenna) return B43_ANTENNA0; case 2: /* Antenna 1 */ return B43_ANTENNA1; + case 3: /* Antenna 2 */ + return B43_ANTENNA2; + case 4: /* Antenna 3 */ + return B43_ANTENNA3; default: return B43_ANTENNA_DEFAULT; } diff --git a/drivers/net/wireless/b43/phy.h b/drivers/net/wireless/b43/phy.h index 31bd4d87b404..ab1e7f097022 100644 --- a/drivers/net/wireless/b43/phy.h +++ b/drivers/net/wireless/b43/phy.h @@ -180,6 +180,8 @@ enum { B43_ANTENNA1, /* Antenna 0 */ B43_ANTENNA_AUTO1, /* Automatic, starting with antenna 1 */ B43_ANTENNA_AUTO0, /* Automatic, starting with antenna 0 */ + B43_ANTENNA2, + B43_ANTENNA3 = 8, B43_ANTENNA_AUTO = B43_ANTENNA_AUTO0, B43_ANTENNA_DEFAULT = B43_ANTENNA_AUTO, diff --git a/drivers/net/wireless/b43/xmit.c b/drivers/net/wireless/b43/xmit.c index 5014213b7752..3fc53e8b4416 100644 --- a/drivers/net/wireless/b43/xmit.c +++ b/drivers/net/wireless/b43/xmit.c @@ -177,13 +177,15 @@ static u8 b43_calc_fallback_rate(u8 bitrate) return 0; } -static void generate_txhdr_fw4(struct b43_wldev *dev, - struct b43_txhdr_fw4 *txhdr, - const unsigned char *fragment_data, - unsigned int fragment_len, - const struct ieee80211_tx_control *txctl, - u16 cookie) +/* Generate a TX data header. */ +void b43_generate_txhdr(struct b43_wldev *dev, + u8 *_txhdr, + const unsigned char *fragment_data, + unsigned int fragment_len, + const struct ieee80211_tx_control *txctl, + u16 cookie) { + struct b43_txhdr *txhdr = (struct b43_txhdr *)_txhdr; const struct b43_phy *phy = &dev->phy; const struct ieee80211_hdr *wlhdr = (const struct ieee80211_hdr *)fragment_data; @@ -241,23 +243,30 @@ static void generate_txhdr_fw4(struct b43_wldev *dev, plcp_fragment_len += txctl->icv_len; key_idx = b43_kidx_to_fw(dev, key_idx); - mac_ctl |= (key_idx << B43_TX4_MAC_KEYIDX_SHIFT) & - B43_TX4_MAC_KEYIDX; - mac_ctl |= (key->algorithm << B43_TX4_MAC_KEYALG_SHIFT) & - B43_TX4_MAC_KEYALG; + mac_ctl |= (key_idx << B43_TXH_MAC_KEYIDX_SHIFT) & + B43_TXH_MAC_KEYIDX; + mac_ctl |= (key->algorithm << B43_TXH_MAC_KEYALG_SHIFT) & + B43_TXH_MAC_KEYALG; wlhdr_len = ieee80211_get_hdrlen(fctl); iv_len = min((size_t) txctl->iv_len, ARRAY_SIZE(txhdr->iv)); memcpy(txhdr->iv, ((u8 *) wlhdr) + wlhdr_len, iv_len); } - b43_generate_plcp_hdr((struct b43_plcp_hdr4 *)(&txhdr->plcp), - plcp_fragment_len, rate); + if (b43_is_old_txhdr_format(dev)) { + b43_generate_plcp_hdr((struct b43_plcp_hdr4 *)(&txhdr->old_format.plcp), + plcp_fragment_len, rate); + } else { + b43_generate_plcp_hdr((struct b43_plcp_hdr4 *)(&txhdr->new_format.plcp), + plcp_fragment_len, rate); + } b43_generate_plcp_hdr((struct b43_plcp_hdr4 *)(&txhdr->plcp_fb), plcp_fragment_len, rate_fb); /* Extra Frame Types */ if (rate_fb_ofdm) - extra_ft |= B43_TX4_EFT_FBOFDM; + extra_ft |= B43_TXH_EFT_FB_OFDM; + else + extra_ft |= B43_TXH_EFT_FB_CCK; /* Set channel radio code. Note that the micrcode ORs 0x100 to * this value before comparing it to the value in SHM, if this @@ -267,19 +276,27 @@ static void generate_txhdr_fw4(struct b43_wldev *dev, /* PHY TX Control word */ if (rate_ofdm) - phy_ctl |= B43_TX4_PHY_OFDM; + phy_ctl |= B43_TXH_PHY_ENC_OFDM; + else + phy_ctl |= B43_TXH_PHY_ENC_CCK; if (dev->short_preamble) - phy_ctl |= B43_TX4_PHY_SHORTPRMBL; + phy_ctl |= B43_TXH_PHY_SHORTPRMBL; switch (b43_ieee80211_antenna_sanitize(dev, txctl->antenna_sel_tx)) { case 0: /* Default */ - phy_ctl |= B43_TX4_PHY_ANTLAST; + phy_ctl |= B43_TXH_PHY_ANT01AUTO; break; case 1: /* Antenna 0 */ - phy_ctl |= B43_TX4_PHY_ANT0; + phy_ctl |= B43_TXH_PHY_ANT0; break; case 2: /* Antenna 1 */ - phy_ctl |= B43_TX4_PHY_ANT1; + phy_ctl |= B43_TXH_PHY_ANT1; + break; + case 3: /* Antenna 2 */ + phy_ctl |= B43_TXH_PHY_ANT2; + break; + case 4: /* Antenna 3 */ + phy_ctl |= B43_TXH_PHY_ANT3; break; default: B43_WARN_ON(1); @@ -287,16 +304,16 @@ static void generate_txhdr_fw4(struct b43_wldev *dev, /* MAC control */ if (!(txctl->flags & IEEE80211_TXCTL_NO_ACK)) - mac_ctl |= B43_TX4_MAC_ACK; + mac_ctl |= B43_TXH_MAC_ACK; if (!(((fctl & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_CTL) && ((fctl & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_PSPOLL))) - mac_ctl |= B43_TX4_MAC_HWSEQ; + mac_ctl |= B43_TXH_MAC_HWSEQ; if (txctl->flags & IEEE80211_TXCTL_FIRST_FRAGMENT) - mac_ctl |= B43_TX4_MAC_STMSDU; + mac_ctl |= B43_TXH_MAC_STMSDU; if (phy->type == B43_PHYTYPE_A) - mac_ctl |= B43_TX4_MAC_5GHZ; + mac_ctl |= B43_TXH_MAC_5GHZ; if (txctl->flags & IEEE80211_TXCTL_LONG_RETRY_LIMIT) - mac_ctl |= B43_TX4_MAC_LONGFRAME; + mac_ctl |= B43_TXH_MAC_LONGFRAME; /* Generate the RTS or CTS-to-self frame */ if ((txctl->flags & IEEE80211_TXCTL_USE_RTS_CTS) || @@ -305,6 +322,7 @@ static void generate_txhdr_fw4(struct b43_wldev *dev, struct ieee80211_hdr *hdr; int rts_rate, rts_rate_fb; int rts_rate_ofdm, rts_rate_fb_ofdm; + struct b43_plcp_hdr6 *plcp; rts_rate = txctl->rts_cts_rate; rts_rate_ofdm = b43_is_ofdm_rate(rts_rate); @@ -312,58 +330,84 @@ static void generate_txhdr_fw4(struct b43_wldev *dev, rts_rate_fb_ofdm = b43_is_ofdm_rate(rts_rate_fb); if (txctl->flags & IEEE80211_TXCTL_USE_CTS_PROTECT) { + struct ieee80211_cts *cts; + + if (b43_is_old_txhdr_format(dev)) { + cts = (struct ieee80211_cts *) + (txhdr->old_format.rts_frame); + } else { + cts = (struct ieee80211_cts *) + (txhdr->new_format.rts_frame); + } ieee80211_ctstoself_get(dev->wl->hw, txctl->vif, fragment_data, fragment_len, - txctl, - (struct ieee80211_cts *)(txhdr-> - rts_frame)); - mac_ctl |= B43_TX4_MAC_SENDCTS; + txctl, cts); + mac_ctl |= B43_TXH_MAC_SENDCTS; len = sizeof(struct ieee80211_cts); } else { + struct ieee80211_rts *rts; + + if (b43_is_old_txhdr_format(dev)) { + rts = (struct ieee80211_rts *) + (txhdr->old_format.rts_frame); + } else { + rts = (struct ieee80211_rts *) + (txhdr->new_format.rts_frame); + } ieee80211_rts_get(dev->wl->hw, txctl->vif, - fragment_data, fragment_len, txctl, - (struct ieee80211_rts *)(txhdr-> - rts_frame)); - mac_ctl |= B43_TX4_MAC_SENDRTS; + fragment_data, fragment_len, + txctl, rts); + mac_ctl |= B43_TXH_MAC_SENDRTS; len = sizeof(struct ieee80211_rts); } len += FCS_LEN; - b43_generate_plcp_hdr((struct b43_plcp_hdr4 *)(&txhdr-> - rts_plcp), len, - rts_rate); - b43_generate_plcp_hdr((struct b43_plcp_hdr4 *)(&txhdr-> - rts_plcp_fb), + + /* Generate the PLCP headers for the RTS/CTS frame */ + if (b43_is_old_txhdr_format(dev)) + plcp = &txhdr->old_format.rts_plcp; + else + plcp = &txhdr->new_format.rts_plcp; + b43_generate_plcp_hdr((struct b43_plcp_hdr4 *)plcp, + len, rts_rate); + plcp = &txhdr->rts_plcp_fb; + b43_generate_plcp_hdr((struct b43_plcp_hdr4 *)plcp, len, rts_rate_fb); - hdr = (struct ieee80211_hdr *)(&txhdr->rts_frame); + + if (b43_is_old_txhdr_format(dev)) { + hdr = (struct ieee80211_hdr *) + (&txhdr->old_format.rts_frame); + } else { + hdr = (struct ieee80211_hdr *) + (&txhdr->new_format.rts_frame); + } txhdr->rts_dur_fb = hdr->duration_id; + if (rts_rate_ofdm) { - extra_ft |= B43_TX4_EFT_RTSOFDM; + extra_ft |= B43_TXH_EFT_RTS_OFDM; txhdr->phy_rate_rts = b43_plcp_get_ratecode_ofdm(rts_rate); - } else + } else { + extra_ft |= B43_TXH_EFT_RTS_CCK; txhdr->phy_rate_rts = b43_plcp_get_ratecode_cck(rts_rate); + } if (rts_rate_fb_ofdm) - extra_ft |= B43_TX4_EFT_RTSFBOFDM; + extra_ft |= B43_TXH_EFT_RTSFB_OFDM; + else + extra_ft |= B43_TXH_EFT_RTSFB_CCK; } /* Magic cookie */ - txhdr->cookie = cpu_to_le16(cookie); + if (b43_is_old_txhdr_format(dev)) + txhdr->old_format.cookie = cpu_to_le16(cookie); + else + txhdr->new_format.cookie = cpu_to_le16(cookie); /* Apply the bitfields */ txhdr->mac_ctl = cpu_to_le32(mac_ctl); txhdr->phy_ctl = cpu_to_le16(phy_ctl); txhdr->extra_ft = extra_ft; -} -void b43_generate_txhdr(struct b43_wldev *dev, - u8 * txhdr, - const unsigned char *fragment_data, - unsigned int fragment_len, - const struct ieee80211_tx_control *txctl, u16 cookie) -{ - generate_txhdr_fw4(dev, (struct b43_txhdr_fw4 *)txhdr, - fragment_data, fragment_len, txctl, cookie); } static s8 b43_rssi_postprocess(struct b43_wldev *dev, diff --git a/drivers/net/wireless/b43/xmit.h b/drivers/net/wireless/b43/xmit.h index 6dc079382f7f..ca2a2ab8654c 100644 --- a/drivers/net/wireless/b43/xmit.h +++ b/drivers/net/wireless/b43/xmit.h @@ -19,68 +19,160 @@ _b43_declare_plcp_hdr(6); #undef _b43_declare_plcp_hdr /* TX header for v4 firmware */ -struct b43_txhdr_fw4 { - __le32 mac_ctl; /* MAC TX control */ - __le16 mac_frame_ctl; /* Copy of the FrameControl field */ +struct b43_txhdr { + __le32 mac_ctl; /* MAC TX control */ + __le16 mac_frame_ctl; /* Copy of the FrameControl field */ __le16 tx_fes_time_norm; /* TX FES Time Normal */ - __le16 phy_ctl; /* PHY TX control */ - __le16 phy_ctl_0; /* Unused */ - __le16 phy_ctl_1; /* Unused */ - __le16 phy_ctl_rts_0; /* Unused */ - __le16 phy_ctl_rts_1; /* Unused */ - __u8 phy_rate; /* PHY rate */ - __u8 phy_rate_rts; /* PHY rate for RTS/CTS */ - __u8 extra_ft; /* Extra Frame Types */ - __u8 chan_radio_code; /* Channel Radio Code */ - __u8 iv[16]; /* Encryption IV */ - __u8 tx_receiver[6]; /* TX Frame Receiver address */ - __le16 tx_fes_time_fb; /* TX FES Time Fallback */ - struct b43_plcp_hdr6 rts_plcp_fb; /* RTS fallback PLCP */ - __le16 rts_dur_fb; /* RTS fallback duration */ - struct b43_plcp_hdr6 plcp_fb; /* Fallback PLCP */ - __le16 dur_fb; /* Fallback duration */ - __le16 mm_dur_time; /* Unused */ - __le16 mm_dur_time_fb; /* Unused */ - __le32 time_stamp; /* Timestamp */ - PAD_BYTES(2); - __le16 cookie; /* TX frame cookie */ - __le16 tx_status; /* TX status */ - struct b43_plcp_hdr6 rts_plcp; /* RTS PLCP */ - __u8 rts_frame[16]; /* The RTS frame (if used) */ - PAD_BYTES(2); - struct b43_plcp_hdr6 plcp; /* Main PLCP */ + __le16 phy_ctl; /* PHY TX control */ + __le16 phy_ctl1; /* PHY TX control word 1 */ + __le16 phy_ctl1_fb; /* PHY TX control word 1 for fallback rates */ + __le16 phy_ctl1_rts; /* PHY TX control word 1 RTS */ + __le16 phy_ctl1_rts_fb; /* PHY TX control word 1 RTS for fallback rates */ + __u8 phy_rate; /* PHY rate */ + __u8 phy_rate_rts; /* PHY rate for RTS/CTS */ + __u8 extra_ft; /* Extra Frame Types */ + __u8 chan_radio_code; /* Channel Radio Code */ + __u8 iv[16]; /* Encryption IV */ + __u8 tx_receiver[6]; /* TX Frame Receiver address */ + __le16 tx_fes_time_fb; /* TX FES Time Fallback */ + struct b43_plcp_hdr6 rts_plcp_fb; /* RTS fallback PLCP header */ + __le16 rts_dur_fb; /* RTS fallback duration */ + struct b43_plcp_hdr6 plcp_fb; /* Fallback PLCP header */ + __le16 dur_fb; /* Fallback duration */ + __le16 mimo_modelen; /* MIMO mode length */ + __le16 mimo_ratelen_fb; /* MIMO fallback rate length */ + __le32 timeout; /* Timeout */ + + union { + /* The new r410 format. */ + struct { + __le16 mimo_antenna; /* MIMO antenna select */ + __le16 preload_size; /* Preload size */ + PAD_BYTES(2); + __le16 cookie; /* TX frame cookie */ + __le16 tx_status; /* TX status */ + struct b43_plcp_hdr6 rts_plcp; /* RTS PLCP header */ + __u8 rts_frame[16]; /* The RTS frame (if used) */ + PAD_BYTES(2); + struct b43_plcp_hdr6 plcp; /* Main PLCP header */ + } new_format __attribute__ ((__packed__)); + + /* The old r351 format. */ + struct { + PAD_BYTES(2); + __le16 cookie; /* TX frame cookie */ + __le16 tx_status; /* TX status */ + struct b43_plcp_hdr6 rts_plcp; /* RTS PLCP header */ + __u8 rts_frame[16]; /* The RTS frame (if used) */ + PAD_BYTES(2); + struct b43_plcp_hdr6 plcp; /* Main PLCP header */ + } old_format __attribute__ ((__packed__)); + + } __attribute__ ((__packed__)); } __attribute__ ((__packed__)); /* MAC TX control */ -#define B43_TX4_MAC_KEYIDX 0x0FF00000 /* Security key index */ -#define B43_TX4_MAC_KEYIDX_SHIFT 20 -#define B43_TX4_MAC_KEYALG 0x00070000 /* Security key algorithm */ -#define B43_TX4_MAC_KEYALG_SHIFT 16 -#define B43_TX4_MAC_LIFETIME 0x00001000 -#define B43_TX4_MAC_FRAMEBURST 0x00000800 -#define B43_TX4_MAC_SENDCTS 0x00000400 -#define B43_TX4_MAC_AMPDU 0x00000300 -#define B43_TX4_MAC_AMPDU_SHIFT 8 -#define B43_TX4_MAC_5GHZ 0x00000080 -#define B43_TX4_MAC_IGNPMQ 0x00000020 -#define B43_TX4_MAC_HWSEQ 0x00000010 /* Use Hardware Sequence Number */ -#define B43_TX4_MAC_STMSDU 0x00000008 /* Start MSDU */ -#define B43_TX4_MAC_SENDRTS 0x00000004 -#define B43_TX4_MAC_LONGFRAME 0x00000002 -#define B43_TX4_MAC_ACK 0x00000001 +#define B43_TXH_MAC_USEFBR 0x10000000 /* Use fallback rate for this AMPDU */ +#define B43_TXH_MAC_KEYIDX 0x0FF00000 /* Security key index */ +#define B43_TXH_MAC_KEYIDX_SHIFT 20 +#define B43_TXH_MAC_KEYALG 0x00070000 /* Security key algorithm */ +#define B43_TXH_MAC_KEYALG_SHIFT 16 +#define B43_TXH_MAC_AMIC 0x00008000 /* AMIC */ +#define B43_TXH_MAC_RIFS 0x00004000 /* Use RIFS */ +#define B43_TXH_MAC_LIFETIME 0x00002000 /* Lifetime */ +#define B43_TXH_MAC_FRAMEBURST 0x00001000 /* Frameburst */ +#define B43_TXH_MAC_SENDCTS 0x00000800 /* Send CTS-to-self */ +#define B43_TXH_MAC_AMPDU 0x00000600 /* AMPDU status */ +#define B43_TXH_MAC_AMPDU_MPDU 0x00000000 /* Regular MPDU, not an AMPDU */ +#define B43_TXH_MAC_AMPDU_FIRST 0x00000200 /* First MPDU or AMPDU */ +#define B43_TXH_MAC_AMPDU_INTER 0x00000400 /* Intermediate MPDU or AMPDU */ +#define B43_TXH_MAC_AMPDU_LAST 0x00000600 /* Last (or only) MPDU of AMPDU */ +#define B43_TXH_MAC_40MHZ 0x00000100 /* Use 40 MHz bandwidth */ +#define B43_TXH_MAC_5GHZ 0x00000080 /* 5GHz band */ +#define B43_TXH_MAC_DFCS 0x00000040 /* DFCS */ +#define B43_TXH_MAC_IGNPMQ 0x00000020 /* Ignore PMQ */ +#define B43_TXH_MAC_HWSEQ 0x00000010 /* Use Hardware Sequence Number */ +#define B43_TXH_MAC_STMSDU 0x00000008 /* Start MSDU */ +#define B43_TXH_MAC_SENDRTS 0x00000004 /* Send RTS */ +#define B43_TXH_MAC_LONGFRAME 0x00000002 /* Long frame */ +#define B43_TXH_MAC_ACK 0x00000001 /* Immediate ACK */ /* Extra Frame Types */ -#define B43_TX4_EFT_FBOFDM 0x0001 /* Data frame fallback rate type */ -#define B43_TX4_EFT_RTSOFDM 0x0004 /* RTS/CTS rate type */ -#define B43_TX4_EFT_RTSFBOFDM 0x0010 /* RTS/CTS fallback rate type */ +#define B43_TXH_EFT_FB 0x03 /* Data frame fallback encoding */ +#define B43_TXH_EFT_FB_CCK 0x00 /* CCK */ +#define B43_TXH_EFT_FB_OFDM 0x01 /* OFDM */ +#define B43_TXH_EFT_FB_EWC 0x02 /* EWC */ +#define B43_TXH_EFT_FB_N 0x03 /* N */ +#define B43_TXH_EFT_RTS 0x0C /* RTS/CTS encoding */ +#define B43_TXH_EFT_RTS_CCK 0x00 /* CCK */ +#define B43_TXH_EFT_RTS_OFDM 0x04 /* OFDM */ +#define B43_TXH_EFT_RTS_EWC 0x08 /* EWC */ +#define B43_TXH_EFT_RTS_N 0x0C /* N */ +#define B43_TXH_EFT_RTSFB 0x30 /* RTS/CTS fallback encoding */ +#define B43_TXH_EFT_RTSFB_CCK 0x00 /* CCK */ +#define B43_TXH_EFT_RTSFB_OFDM 0x10 /* OFDM */ +#define B43_TXH_EFT_RTSFB_EWC 0x20 /* EWC */ +#define B43_TXH_EFT_RTSFB_N 0x30 /* N */ /* PHY TX control word */ -#define B43_TX4_PHY_OFDM 0x0001 /* Data frame rate type */ -#define B43_TX4_PHY_SHORTPRMBL 0x0010 /* Use short preamble */ -#define B43_TX4_PHY_ANT 0x03C0 /* Antenna selection */ -#define B43_TX4_PHY_ANT0 0x0000 /* Use antenna 0 */ -#define B43_TX4_PHY_ANT1 0x0100 /* Use antenna 1 */ -#define B43_TX4_PHY_ANTLAST 0x0300 /* Use last used antenna */ +#define B43_TXH_PHY_ENC 0x0003 /* Data frame encoding */ +#define B43_TXH_PHY_ENC_CCK 0x0000 /* CCK */ +#define B43_TXH_PHY_ENC_OFDM 0x0001 /* OFDM */ +#define B43_TXH_PHY_ENC_EWC 0x0002 /* EWC */ +#define B43_TXH_PHY_ENC_N 0x0003 /* N */ +#define B43_TXH_PHY_SHORTPRMBL 0x0010 /* Use short preamble */ +#define B43_TXH_PHY_ANT 0x03C0 /* Antenna selection */ +#define B43_TXH_PHY_ANT0 0x0000 /* Use antenna 0 */ +#define B43_TXH_PHY_ANT1 0x0040 /* Use antenna 1 */ +#define B43_TXH_PHY_ANT01AUTO 0x00C0 /* Use antenna 0/1 auto */ +#define B43_TXH_PHY_ANT2 0x0100 /* Use antenna 2 */ +#define B43_TXH_PHY_ANT3 0x0200 /* Use antenna 3 */ +#define B43_TXH_PHY_TXPWR 0xFC00 /* TX power */ +#define B43_TXH_PHY_TXPWR_SHIFT 10 + +/* PHY TX control word 1 */ +#define B43_TXH_PHY1_BW 0x0007 /* Bandwidth */ +#define B43_TXH_PHY1_BW_10 0x0000 /* 10 MHz */ +#define B43_TXH_PHY1_BW_10U 0x0001 /* 10 MHz upper */ +#define B43_TXH_PHY1_BW_20 0x0002 /* 20 MHz */ +#define B43_TXH_PHY1_BW_20U 0x0003 /* 20 MHz upper */ +#define B43_TXH_PHY1_BW_40 0x0004 /* 40 MHz */ +#define B43_TXH_PHY1_BW_40DUP 0x0005 /* 50 MHz duplicate */ +#define B43_TXH_PHY1_MODE 0x0038 /* Mode */ +#define B43_TXH_PHY1_MODE_SISO 0x0000 /* SISO */ +#define B43_TXH_PHY1_MODE_CDD 0x0008 /* CDD */ +#define B43_TXH_PHY1_MODE_STBC 0x0010 /* STBC */ +#define B43_TXH_PHY1_MODE_SDM 0x0018 /* SDM */ +#define B43_TXH_PHY1_CRATE 0x0700 /* Coding rate */ +#define B43_TXH_PHY1_CRATE_1_2 0x0000 /* 1/2 */ +#define B43_TXH_PHY1_CRATE_2_3 0x0100 /* 2/3 */ +#define B43_TXH_PHY1_CRATE_3_4 0x0200 /* 3/4 */ +#define B43_TXH_PHY1_CRATE_4_5 0x0300 /* 4/5 */ +#define B43_TXH_PHY1_CRATE_5_6 0x0400 /* 5/6 */ +#define B43_TXH_PHY1_CRATE_7_8 0x0600 /* 7/8 */ +#define B43_TXH_PHY1_MODUL 0x3800 /* Modulation scheme */ +#define B43_TXH_PHY1_MODUL_BPSK 0x0000 /* BPSK */ +#define B43_TXH_PHY1_MODUL_QPSK 0x0800 /* QPSK */ +#define B43_TXH_PHY1_MODUL_QAM16 0x1000 /* QAM16 */ +#define B43_TXH_PHY1_MODUL_QAM64 0x1800 /* QAM64 */ +#define B43_TXH_PHY1_MODUL_QAM256 0x2000 /* QAM256 */ + + +/* r351 firmware compatibility stuff. */ +static inline +bool b43_is_old_txhdr_format(struct b43_wldev *dev) +{ + return (dev->fw.rev <= 351); +} + +static inline +size_t b43_txhdr_size(struct b43_wldev *dev) +{ + if (b43_is_old_txhdr_format(dev)) + return 100 + sizeof(struct b43_plcp_hdr6); + return 104 + sizeof(struct b43_plcp_hdr6); +} + void b43_generate_txhdr(struct b43_wldev *dev, u8 * txhdr, -- cgit v1.2.3 From 818d276ceb83aa9fdebb5e0a53188290312de987 Mon Sep 17 00:00:00 2001 From: Girish Shilamkar Date: Mon, 28 Jan 2008 23:58:27 -0500 Subject: ext4: Add the journal checksum feature The journal checksum feature adds two new flags i.e JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT and JBD2_FEATURE_COMPAT_CHECKSUM. JBD2_FEATURE_CHECKSUM flag indicates that the commit block contains the checksum for the blocks described by the descriptor blocks. Due to checksums, writing of the commit record no longer needs to be synchronous. Now commit record can be sent to disk without waiting for descriptor blocks to be written to disk. This behavior is controlled using JBD2_FEATURE_ASYNC_COMMIT flag. Older kernels/e2fsck should not be able to recover the journal with _ASYNC_COMMIT hence it is made incompat. The commit header has been extended to hold the checksum along with the type of the checksum. For recovery in pass scan checksums are verified to ensure the sanity and completeness(in case of _ASYNC_COMMIT) of every transaction. Signed-off-by: Andreas Dilger Signed-off-by: Girish Shilamkar Signed-off-by: Dave Kleikamp Signed-off-by: Mingming Cao --- Documentation/filesystems/ext4.txt | 10 ++ fs/Kconfig | 1 + fs/ext4/super.c | 25 +++++ fs/jbd2/commit.c | 198 ++++++++++++++++++++++++++++--------- fs/jbd2/journal.c | 26 +++++ fs/jbd2/recovery.c | 151 ++++++++++++++++++++++++++-- include/linux/ext4_fs.h | 3 +- include/linux/jbd2.h | 36 ++++++- 8 files changed, 388 insertions(+), 62 deletions(-) (limited to 'Documentation') diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt index 6a4adcae9f9a..4f329afe20ec 100644 --- a/Documentation/filesystems/ext4.txt +++ b/Documentation/filesystems/ext4.txt @@ -89,6 +89,16 @@ When mounting an ext4 filesystem, the following option are accepted: extents ext4 will use extents to address file data. The file system will no longer be mountable by ext3. +journal_checksum Enable checksumming of the journal transactions. + This will allow the recovery code in e2fsck and the + kernel to detect corruption in the kernel. It is a + compatible change and will be ignored by older kernels. + +journal_async_commit Commit block can be written to disk without waiting + for descriptor blocks. If enabled older kernels cannot + mount the device. This will enable 'journal_checksum' + internally. + journal=update Update the ext4 file system's journal to the current format. diff --git a/fs/Kconfig b/fs/Kconfig index 9656139d2e99..219ec06a8c7e 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -236,6 +236,7 @@ config JBD_DEBUG config JBD2 tristate + select CRC32 help This is a generic journaling layer for block devices that support both 32-bit and 64-bit block numbers. It is currently used by diff --git a/fs/ext4/super.c b/fs/ext4/super.c index c7305443e100..f7479d30735e 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -869,6 +869,7 @@ enum { Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl, Opt_reservation, Opt_noreservation, Opt_noload, Opt_nobh, Opt_bh, Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_journal_dev, + Opt_journal_checksum, Opt_journal_async_commit, Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback, Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota, Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota, @@ -908,6 +909,8 @@ static match_table_t tokens = { {Opt_journal_update, "journal=update"}, {Opt_journal_inum, "journal=%u"}, {Opt_journal_dev, "journal_dev=%u"}, + {Opt_journal_checksum, "journal_checksum"}, + {Opt_journal_async_commit, "journal_async_commit"}, {Opt_abort, "abort"}, {Opt_data_journal, "data=journal"}, {Opt_data_ordered, "data=ordered"}, @@ -1095,6 +1098,13 @@ static int parse_options (char *options, struct super_block *sb, return 0; *journal_devnum = option; break; + case Opt_journal_checksum: + set_opt(sbi->s_mount_opt, JOURNAL_CHECKSUM); + break; + case Opt_journal_async_commit: + set_opt(sbi->s_mount_opt, JOURNAL_ASYNC_COMMIT); + set_opt(sbi->s_mount_opt, JOURNAL_CHECKSUM); + break; case Opt_noload: set_opt (sbi->s_mount_opt, NOLOAD); break; @@ -2114,6 +2124,21 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) goto failed_mount4; } + if (test_opt(sb, JOURNAL_ASYNC_COMMIT)) { + jbd2_journal_set_features(sbi->s_journal, + JBD2_FEATURE_COMPAT_CHECKSUM, 0, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT); + } else if (test_opt(sb, JOURNAL_CHECKSUM)) { + jbd2_journal_set_features(sbi->s_journal, + JBD2_FEATURE_COMPAT_CHECKSUM, 0, 0); + jbd2_journal_clear_features(sbi->s_journal, 0, 0, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT); + } else { + jbd2_journal_clear_features(sbi->s_journal, + JBD2_FEATURE_COMPAT_CHECKSUM, 0, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT); + } + /* We have now updated the journal if required, so we can * validate the data journaling mode. */ switch (test_opt(sb, DATA_FLAGS)) { diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 8749a86f4175..da8d0eb3b7b9 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -21,6 +21,7 @@ #include #include #include +#include /* * Default IO end handler for temporary BJ_IO buffer_heads. @@ -93,19 +94,23 @@ static int inverted_lock(journal_t *journal, struct buffer_head *bh) return 1; } -/* Done it all: now write the commit record. We should have +/* + * Done it all: now submit the commit record. We should have * cleaned up our previous buffers by now, so if we are in abort * mode we can now just skip the rest of the journal write * entirely. * * Returns 1 if the journal needs to be aborted or 0 on success */ -static int journal_write_commit_record(journal_t *journal, - transaction_t *commit_transaction) +static int journal_submit_commit_record(journal_t *journal, + transaction_t *commit_transaction, + struct buffer_head **cbh, + __u32 crc32_sum) { struct journal_head *descriptor; + struct commit_header *tmp; struct buffer_head *bh; - int i, ret; + int ret; int barrier_done = 0; if (is_journal_aborted(journal)) @@ -117,21 +122,33 @@ static int journal_write_commit_record(journal_t *journal, bh = jh2bh(descriptor); - /* AKPM: buglet - add `i' to tmp! */ - for (i = 0; i < bh->b_size; i += 512) { - journal_header_t *tmp = (journal_header_t*)bh->b_data; - tmp->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); - tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK); - tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid); + tmp = (struct commit_header *)bh->b_data; + tmp->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); + tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK); + tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid); + + if (JBD2_HAS_COMPAT_FEATURE(journal, + JBD2_FEATURE_COMPAT_CHECKSUM)) { + tmp->h_chksum_type = JBD2_CRC32_CHKSUM; + tmp->h_chksum_size = JBD2_CRC32_CHKSUM_SIZE; + tmp->h_chksum[0] = cpu_to_be32(crc32_sum); } - JBUFFER_TRACE(descriptor, "write commit block"); + JBUFFER_TRACE(descriptor, "submit commit block"); + lock_buffer(bh); + set_buffer_dirty(bh); - if (journal->j_flags & JBD2_BARRIER) { + set_buffer_uptodate(bh); + bh->b_end_io = journal_end_buffer_io_sync; + + if (journal->j_flags & JBD2_BARRIER && + !JBD2_HAS_COMPAT_FEATURE(journal, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) { set_buffer_ordered(bh); barrier_done = 1; } - ret = sync_dirty_buffer(bh); + ret = submit_bh(WRITE, bh); + /* is it possible for another commit to fail at roughly * the same time as this one? If so, we don't want to * trust the barrier flag in the super, but instead want @@ -152,14 +169,72 @@ static int journal_write_commit_record(journal_t *journal, clear_buffer_ordered(bh); set_buffer_uptodate(bh); set_buffer_dirty(bh); - ret = sync_dirty_buffer(bh); + ret = submit_bh(WRITE, bh); } - put_bh(bh); /* One for getblk() */ - jbd2_journal_put_journal_head(descriptor); + *cbh = bh; + return ret; +} + +/* + * This function along with journal_submit_commit_record + * allows to write the commit record asynchronously. + */ +static int journal_wait_on_commit_record(struct buffer_head *bh) +{ + int ret = 0; + + clear_buffer_dirty(bh); + wait_on_buffer(bh); - return (ret == -EIO); + if (unlikely(!buffer_uptodate(bh))) + ret = -EIO; + put_bh(bh); /* One for getblk() */ + jbd2_journal_put_journal_head(bh2jh(bh)); + + return ret; } +/* + * Wait for all submitted IO to complete. + */ +static int journal_wait_on_locked_list(journal_t *journal, + transaction_t *commit_transaction) +{ + int ret = 0; + struct journal_head *jh; + + while (commit_transaction->t_locked_list) { + struct buffer_head *bh; + + jh = commit_transaction->t_locked_list->b_tprev; + bh = jh2bh(jh); + get_bh(bh); + if (buffer_locked(bh)) { + spin_unlock(&journal->j_list_lock); + wait_on_buffer(bh); + if (unlikely(!buffer_uptodate(bh))) + ret = -EIO; + spin_lock(&journal->j_list_lock); + } + if (!inverted_lock(journal, bh)) { + put_bh(bh); + spin_lock(&journal->j_list_lock); + continue; + } + if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) { + __jbd2_journal_unfile_buffer(jh); + jbd_unlock_bh_state(bh); + jbd2_journal_remove_journal_head(bh); + put_bh(bh); + } else { + jbd_unlock_bh_state(bh); + } + put_bh(bh); + cond_resched_lock(&journal->j_list_lock); + } + return ret; + } + static void journal_do_submit_data(struct buffer_head **wbuf, int bufs) { int i; @@ -275,7 +350,21 @@ write_out_data: journal_do_submit_data(wbuf, bufs); } -static inline void write_tag_block(int tag_bytes, journal_block_tag_t *tag, +static __u32 jbd2_checksum_data(__u32 crc32_sum, struct buffer_head *bh) +{ + struct page *page = bh->b_page; + char *addr; + __u32 checksum; + + addr = kmap_atomic(page, KM_USER0); + checksum = crc32_be(crc32_sum, + (void *)(addr + offset_in_page(bh->b_data)), bh->b_size); + kunmap_atomic(addr, KM_USER0); + + return checksum; +} + +static void write_tag_block(int tag_bytes, journal_block_tag_t *tag, unsigned long long block) { tag->t_blocknr = cpu_to_be32(block & (u32)~0); @@ -307,6 +396,8 @@ void jbd2_journal_commit_transaction(journal_t *journal) int tag_flag; int i; int tag_bytes = journal_tag_bytes(journal); + struct buffer_head *cbh = NULL; /* For transactional checksums */ + __u32 crc32_sum = ~0; /* * First job: lock down the current transaction and wait for @@ -451,38 +542,15 @@ void jbd2_journal_commit_transaction(journal_t *journal) journal_submit_data_buffers(journal, commit_transaction); /* - * Wait for all previously submitted IO to complete. + * Wait for all previously submitted IO to complete if commit + * record is to be written synchronously. */ spin_lock(&journal->j_list_lock); - while (commit_transaction->t_locked_list) { - struct buffer_head *bh; + if (!JBD2_HAS_INCOMPAT_FEATURE(journal, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) + err = journal_wait_on_locked_list(journal, + commit_transaction); - jh = commit_transaction->t_locked_list->b_tprev; - bh = jh2bh(jh); - get_bh(bh); - if (buffer_locked(bh)) { - spin_unlock(&journal->j_list_lock); - wait_on_buffer(bh); - if (unlikely(!buffer_uptodate(bh))) - err = -EIO; - spin_lock(&journal->j_list_lock); - } - if (!inverted_lock(journal, bh)) { - put_bh(bh); - spin_lock(&journal->j_list_lock); - continue; - } - if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) { - __jbd2_journal_unfile_buffer(jh); - jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); - put_bh(bh); - } else { - jbd_unlock_bh_state(bh); - } - put_bh(bh); - cond_resched_lock(&journal->j_list_lock); - } spin_unlock(&journal->j_list_lock); if (err) @@ -656,6 +724,15 @@ void jbd2_journal_commit_transaction(journal_t *journal) start_journal_io: for (i = 0; i < bufs; i++) { struct buffer_head *bh = wbuf[i]; + /* + * Compute checksum. + */ + if (JBD2_HAS_COMPAT_FEATURE(journal, + JBD2_FEATURE_COMPAT_CHECKSUM)) { + crc32_sum = + jbd2_checksum_data(crc32_sum, bh); + } + lock_buffer(bh); clear_buffer_dirty(bh); set_buffer_uptodate(bh); @@ -672,6 +749,23 @@ start_journal_io: } } + /* Done it all: now write the commit record asynchronously. */ + + if (JBD2_HAS_INCOMPAT_FEATURE(journal, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) { + err = journal_submit_commit_record(journal, commit_transaction, + &cbh, crc32_sum); + if (err) + __jbd2_journal_abort_hard(journal); + + spin_lock(&journal->j_list_lock); + err = journal_wait_on_locked_list(journal, + commit_transaction); + spin_unlock(&journal->j_list_lock); + if (err) + __jbd2_journal_abort_hard(journal); + } + /* Lo and behold: we have just managed to send a transaction to the log. Before we can commit it, wait for the IO so far to complete. Control buffers being written are on the @@ -771,8 +865,14 @@ wait_for_iobuf: jbd_debug(3, "JBD: commit phase 6\n"); - if (journal_write_commit_record(journal, commit_transaction)) - err = -EIO; + if (!JBD2_HAS_INCOMPAT_FEATURE(journal, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) { + err = journal_submit_commit_record(journal, commit_transaction, + &cbh, crc32_sum); + if (err) + __jbd2_journal_abort_hard(journal); + } + err = journal_wait_on_commit_record(cbh); if (err) jbd2_journal_abort(journal, err); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 3667c91bc786..59ba2494dcaf 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1578,6 +1578,32 @@ int jbd2_journal_set_features (journal_t *journal, unsigned long compat, return 1; } +/* + * jbd2_journal_clear_features () - Clear a given journal feature in the + * superblock + * @journal: Journal to act on. + * @compat: bitmask of compatible features + * @ro: bitmask of features that force read-only mount + * @incompat: bitmask of incompatible features + * + * Clear a given journal feature as present on the + * superblock. + */ +void jbd2_journal_clear_features(journal_t *journal, unsigned long compat, + unsigned long ro, unsigned long incompat) +{ + journal_superblock_t *sb; + + jbd_debug(1, "Clear features 0x%lx/0x%lx/0x%lx\n", + compat, ro, incompat); + + sb = journal->j_superblock; + + sb->s_feature_compat &= ~cpu_to_be32(compat); + sb->s_feature_ro_compat &= ~cpu_to_be32(ro); + sb->s_feature_incompat &= ~cpu_to_be32(incompat); +} +EXPORT_SYMBOL(jbd2_journal_clear_features); /** * int jbd2_journal_update_format () - Update on-disk journal structure. diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index d0ce627539ef..921680663fa2 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -21,6 +21,7 @@ #include #include #include +#include #endif /* @@ -316,6 +317,37 @@ static inline unsigned long long read_tag_block(int tag_bytes, journal_block_tag return block; } +/* + * calc_chksums calculates the checksums for the blocks described in the + * descriptor block. + */ +static int calc_chksums(journal_t *journal, struct buffer_head *bh, + unsigned long *next_log_block, __u32 *crc32_sum) +{ + int i, num_blks, err; + unsigned long io_block; + struct buffer_head *obh; + + num_blks = count_tags(journal, bh); + /* Calculate checksum of the descriptor block. */ + *crc32_sum = crc32_be(*crc32_sum, (void *)bh->b_data, bh->b_size); + + for (i = 0; i < num_blks; i++) { + io_block = (*next_log_block)++; + wrap(journal, *next_log_block); + err = jread(&obh, journal, io_block); + if (err) { + printk(KERN_ERR "JBD: IO error %d recovering block " + "%lu in log\n", err, io_block); + return 1; + } else { + *crc32_sum = crc32_be(*crc32_sum, (void *)obh->b_data, + obh->b_size); + } + } + return 0; +} + static int do_one_pass(journal_t *journal, struct recovery_info *info, enum passtype pass) { @@ -328,6 +360,7 @@ static int do_one_pass(journal_t *journal, unsigned int sequence; int blocktype; int tag_bytes = journal_tag_bytes(journal); + __u32 crc32_sum = ~0; /* Transactional Checksums */ /* Precompute the maximum metadata descriptors in a descriptor block */ int MAX_BLOCKS_PER_DESC; @@ -419,12 +452,26 @@ static int do_one_pass(journal_t *journal, switch(blocktype) { case JBD2_DESCRIPTOR_BLOCK: /* If it is a valid descriptor block, replay it - * in pass REPLAY; otherwise, just skip over the - * blocks it describes. */ + * in pass REPLAY; if journal_checksums enabled, then + * calculate checksums in PASS_SCAN, otherwise, + * just skip over the blocks it describes. */ if (pass != PASS_REPLAY) { + if (pass == PASS_SCAN && + JBD2_HAS_COMPAT_FEATURE(journal, + JBD2_FEATURE_COMPAT_CHECKSUM) && + !info->end_transaction) { + if (calc_chksums(journal, bh, + &next_log_block, + &crc32_sum)) { + put_bh(bh); + break; + } + put_bh(bh); + continue; + } next_log_block += count_tags(journal, bh); wrap(journal, next_log_block); - brelse(bh); + put_bh(bh); continue; } @@ -516,9 +563,96 @@ static int do_one_pass(journal_t *journal, continue; case JBD2_COMMIT_BLOCK: - /* Found an expected commit block: not much to - * do other than move on to the next sequence + /* How to differentiate between interrupted commit + * and journal corruption ? + * + * {nth transaction} + * Checksum Verification Failed + * | + * ____________________ + * | | + * async_commit sync_commit + * | | + * | GO TO NEXT "Journal Corruption" + * | TRANSACTION + * | + * {(n+1)th transanction} + * | + * _______|______________ + * | | + * Commit block found Commit block not found + * | | + * "Journal Corruption" | + * _____________|_________ + * | | + * nth trans corrupt OR nth trans + * and (n+1)th interrupted interrupted + * before commit block + * could reach the disk. + * (Cannot find the difference in above + * mentioned conditions. Hence assume + * "Interrupted Commit".) + */ + + /* Found an expected commit block: if checksums + * are present verify them in PASS_SCAN; else not + * much to do other than move on to the next sequence * number. */ + if (pass == PASS_SCAN && + JBD2_HAS_COMPAT_FEATURE(journal, + JBD2_FEATURE_COMPAT_CHECKSUM)) { + int chksum_err, chksum_seen; + struct commit_header *cbh = + (struct commit_header *)bh->b_data; + unsigned found_chksum = + be32_to_cpu(cbh->h_chksum[0]); + + chksum_err = chksum_seen = 0; + + if (info->end_transaction) { + printk(KERN_ERR "JBD: Transaction %u " + "found to be corrupt.\n", + next_commit_ID - 1); + brelse(bh); + break; + } + + if (crc32_sum == found_chksum && + cbh->h_chksum_type == JBD2_CRC32_CHKSUM && + cbh->h_chksum_size == + JBD2_CRC32_CHKSUM_SIZE) + chksum_seen = 1; + else if (!(cbh->h_chksum_type == 0 && + cbh->h_chksum_size == 0 && + found_chksum == 0 && + !chksum_seen)) + /* + * If fs is mounted using an old kernel and then + * kernel with journal_chksum is used then we + * get a situation where the journal flag has + * checksum flag set but checksums are not + * present i.e chksum = 0, in the individual + * commit blocks. + * Hence to avoid checksum failures, in this + * situation, this extra check is added. + */ + chksum_err = 1; + + if (chksum_err) { + info->end_transaction = next_commit_ID; + + if (!JBD2_HAS_COMPAT_FEATURE(journal, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)){ + printk(KERN_ERR + "JBD: Transaction %u " + "found to be corrupt.\n", + next_commit_ID); + brelse(bh); + break; + } + } + crc32_sum = ~0; + } brelse(bh); next_commit_ID++; continue; @@ -554,9 +688,10 @@ static int do_one_pass(journal_t *journal, * transaction marks the end of the valid log. */ - if (pass == PASS_SCAN) - info->end_transaction = next_commit_ID; - else { + if (pass == PASS_SCAN) { + if (!info->end_transaction) + info->end_transaction = next_commit_ID; + } else { /* It's really bad news if different passes end up at * different places (but possible due to IO errors). */ if (info->end_transaction != next_commit_ID) { diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h index 300cc5a5adb9..cd406dba0e64 100644 --- a/include/linux/ext4_fs.h +++ b/include/linux/ext4_fs.h @@ -467,7 +467,8 @@ do { \ #define EXT4_MOUNT_USRQUOTA 0x100000 /* "old" user quota */ #define EXT4_MOUNT_GRPQUOTA 0x200000 /* "old" group quota */ #define EXT4_MOUNT_EXTENTS 0x400000 /* Extents support */ - +#define EXT4_MOUNT_JOURNAL_CHECKSUM 0x800000 /* Journal checksums */ +#define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT 0x1000000 /* Journal Async Commit */ /* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */ #ifndef _LINUX_EXT2_FS_H #define clear_opt(o, opt) o &= ~EXT4_MOUNT_##opt diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 685640036e81..98a2bc5d3e3f 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -149,6 +149,28 @@ typedef struct journal_header_s __be32 h_sequence; } journal_header_t; +/* + * Checksum types. + */ +#define JBD2_CRC32_CHKSUM 1 +#define JBD2_MD5_CHKSUM 2 +#define JBD2_SHA1_CHKSUM 3 + +#define JBD2_CRC32_CHKSUM_SIZE 4 + +#define JBD2_CHECKSUM_BYTES (32 / sizeof(u32)) +/* + * Commit block header for storing transactional checksums: + */ +struct commit_header { + __be32 h_magic; + __be32 h_blocktype; + __be32 h_sequence; + unsigned char h_chksum_type; + unsigned char h_chksum_size; + unsigned char h_padding[2]; + __be32 h_chksum[JBD2_CHECKSUM_BYTES]; +}; /* * The block tag: used to describe a single buffer in the journal. @@ -242,14 +264,18 @@ typedef struct journal_superblock_s ((j)->j_format_version >= 2 && \ ((j)->j_superblock->s_feature_incompat & cpu_to_be32((mask)))) -#define JBD2_FEATURE_INCOMPAT_REVOKE 0x00000001 -#define JBD2_FEATURE_INCOMPAT_64BIT 0x00000002 +#define JBD2_FEATURE_COMPAT_CHECKSUM 0x00000001 + +#define JBD2_FEATURE_INCOMPAT_REVOKE 0x00000001 +#define JBD2_FEATURE_INCOMPAT_64BIT 0x00000002 +#define JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT 0x00000004 /* Features known to this kernel version: */ -#define JBD2_KNOWN_COMPAT_FEATURES 0 +#define JBD2_KNOWN_COMPAT_FEATURES JBD2_FEATURE_COMPAT_CHECKSUM #define JBD2_KNOWN_ROCOMPAT_FEATURES 0 #define JBD2_KNOWN_INCOMPAT_FEATURES (JBD2_FEATURE_INCOMPAT_REVOKE | \ - JBD2_FEATURE_INCOMPAT_64BIT) + JBD2_FEATURE_INCOMPAT_64BIT | \ + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT) #ifdef __KERNEL__ @@ -997,6 +1023,8 @@ extern int jbd2_journal_check_available_features (journal_t *, unsigned long, unsigned long, unsigned long); extern int jbd2_journal_set_features (journal_t *, unsigned long, unsigned long, unsigned long); +extern void jbd2_journal_clear_features + (journal_t *, unsigned long, unsigned long, unsigned long); extern int jbd2_journal_create (journal_t *); extern int jbd2_journal_load (journal_t *journal); extern void jbd2_journal_destroy (journal_t *); -- cgit v1.2.3 From c9de560ded61faa5b754137b7753da252391c55a Mon Sep 17 00:00:00 2001 From: Alex Tomas Date: Tue, 29 Jan 2008 00:19:52 -0500 Subject: ext4: Add multi block allocator for ext4 Signed-off-by: Alex Tomas Signed-off-by: Andreas Dilger Signed-off-by: Aneesh Kumar K.V Signed-off-by: Eric Sandeen Signed-off-by: "Theodore Ts'o" --- Documentation/filesystems/ext4.txt | 10 +- Documentation/filesystems/proc.txt | 39 + fs/ext4/Makefile | 2 +- fs/ext4/balloc.c | 67 +- fs/ext4/extents.c | 45 +- fs/ext4/inode.c | 15 +- fs/ext4/mballoc.c | 4552 ++++++++++++++++++++++++++++++++++++ fs/ext4/migrate.c | 10 +- fs/ext4/super.c | 62 +- fs/ext4/xattr.c | 4 +- include/linux/ext4_fs.h | 76 +- include/linux/ext4_fs_i.h | 4 + include/linux/ext4_fs_sb.h | 52 + 13 files changed, 4900 insertions(+), 38 deletions(-) create mode 100644 fs/ext4/mballoc.c (limited to 'Documentation') diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt index 4f329afe20ec..560f88dc7090 100644 --- a/Documentation/filesystems/ext4.txt +++ b/Documentation/filesystems/ext4.txt @@ -86,9 +86,11 @@ Alex is working on a new set of patches right now. When mounting an ext4 filesystem, the following option are accepted: (*) == default -extents ext4 will use extents to address file data. The +extents (*) ext4 will use extents to address file data. The file system will no longer be mountable by ext3. +noextents ext4 will not use extents for newly created files + journal_checksum Enable checksumming of the journal transactions. This will allow the recovery code in e2fsck and the kernel to detect corruption in the kernel. It is a @@ -206,6 +208,12 @@ nobh (a) cache disk block mapping information "nobh" option tries to avoid associating buffer heads (supported only for "writeback" mode). +mballoc (*) Use the multiple block allocator for block allocation +nomballoc disabled multiple block allocator for block allocation. +stripe=n Number of filesystem blocks that mballoc will try + to use for allocation size and alignment. For RAID5/6 + systems this should be the number of data + disks * RAID chunk size in file system blocks. Data Mode --------- diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index dec99455321f..4413a2d4646f 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -857,6 +857,45 @@ CPUs. The "procs_blocked" line gives the number of processes currently blocked, waiting for I/O to complete. +1.9 Ext4 file system parameters +------------------------------ +Ext4 file system have one directory per partition under /proc/fs/ext4/ +# ls /proc/fs/ext4/hdc/ +group_prealloc max_to_scan mb_groups mb_history min_to_scan order2_req +stats stream_req + +mb_groups: +This file gives the details of mutiblock allocator buddy cache of free blocks + +mb_history: +Multiblock allocation history. + +stats: +This file indicate whether the multiblock allocator should start collecting +statistics. The statistics are shown during unmount + +group_prealloc: +The multiblock allocator normalize the block allocation request to +group_prealloc filesystem blocks if we don't have strip value set. +The stripe value can be specified at mount time or during mke2fs. + +max_to_scan: +How long multiblock allocator can look for a best extent (in found extents) + +min_to_scan: +How long multiblock allocator must look for a best extent + +order2_req: +Multiblock allocator use 2^N search using buddies only for requests greater +than or equal to order2_req. The request size is specfied in file system +blocks. A value of 2 indicate only if the requests are greater than or equal +to 4 blocks. + +stream_req: +Files smaller than stream_req are served by the stream allocator, whose +purpose is to pack requests as close each to other as possible to +produce smooth I/O traffic. Avalue of 16 indicate that file smaller than 16 +filesystem block size will use group based preallocation. ------------------------------------------------------------------------------ Summary diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile index d5fd80bc0d04..ac6fa8ca0a2f 100644 --- a/fs/ext4/Makefile +++ b/fs/ext4/Makefile @@ -6,7 +6,7 @@ obj-$(CONFIG_EXT4DEV_FS) += ext4dev.o ext4dev-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o \ - ext4_jbd2.o migrate.o + ext4_jbd2.o migrate.o mballoc.o ext4dev-$(CONFIG_EXT4DEV_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o ext4dev-$(CONFIG_EXT4DEV_FS_POSIX_ACL) += acl.o diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 80a4616c8244..ac75ea953d83 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -577,6 +577,8 @@ void ext4_discard_reservation(struct inode *inode) struct ext4_reserve_window_node *rsv; spinlock_t *rsv_lock = &EXT4_SB(inode->i_sb)->s_rsv_window_lock; + ext4_mb_discard_inode_preallocations(inode); + if (!block_i) return; @@ -785,19 +787,29 @@ error_return: * @inode: inode * @block: start physical block to free * @count: number of blocks to count + * @metadata: Are these metadata blocks */ void ext4_free_blocks(handle_t *handle, struct inode *inode, - ext4_fsblk_t block, unsigned long count) + ext4_fsblk_t block, unsigned long count, + int metadata) { struct super_block * sb; unsigned long dquot_freed_blocks; + /* this isn't the right place to decide whether block is metadata + * inode.c/extents.c knows better, but for safety ... */ + if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode) || + ext4_should_journal_data(inode)) + metadata = 1; + sb = inode->i_sb; - if (!sb) { - printk ("ext4_free_blocks: nonexistent device"); - return; - } - ext4_free_blocks_sb(handle, sb, block, count, &dquot_freed_blocks); + + if (!test_opt(sb, MBALLOC) || !EXT4_SB(sb)->s_group_info) + ext4_free_blocks_sb(handle, sb, block, count, + &dquot_freed_blocks); + else + ext4_mb_free_blocks(handle, inode, block, count, + metadata, &dquot_freed_blocks); if (dquot_freed_blocks) DQUOT_FREE_BLOCK(inode, dquot_freed_blocks); return; @@ -1576,7 +1588,7 @@ int ext4_should_retry_alloc(struct super_block *sb, int *retries) } /** - * ext4_new_blocks() -- core block(s) allocation function + * ext4_new_blocks_old() -- core block(s) allocation function * @handle: handle to this transaction * @inode: file inode * @goal: given target block(filesystem wide) @@ -1589,7 +1601,7 @@ int ext4_should_retry_alloc(struct super_block *sb, int *retries) * any specific goal block. * */ -ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, +ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode, ext4_fsblk_t goal, unsigned long *count, int *errp) { struct buffer_head *bitmap_bh = NULL; @@ -1849,13 +1861,46 @@ out: } ext4_fsblk_t ext4_new_block(handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, int *errp) + ext4_fsblk_t goal, int *errp) +{ + struct ext4_allocation_request ar; + ext4_fsblk_t ret; + + if (!test_opt(inode->i_sb, MBALLOC)) { + unsigned long count = 1; + ret = ext4_new_blocks_old(handle, inode, goal, &count, errp); + return ret; + } + + memset(&ar, 0, sizeof(ar)); + ar.inode = inode; + ar.goal = goal; + ar.len = 1; + ret = ext4_mb_new_blocks(handle, &ar, errp); + return ret; +} + +ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, unsigned long *count, int *errp) { - unsigned long count = 1; + struct ext4_allocation_request ar; + ext4_fsblk_t ret; - return ext4_new_blocks(handle, inode, goal, &count, errp); + if (!test_opt(inode->i_sb, MBALLOC)) { + ret = ext4_new_blocks_old(handle, inode, goal, count, errp); + return ret; + } + + memset(&ar, 0, sizeof(ar)); + ar.inode = inode; + ar.goal = goal; + ar.len = *count; + ret = ext4_mb_new_blocks(handle, &ar, errp); + *count = ar.len; + return ret; } + /** * ext4_count_free_blocks() -- count filesystem free blocks * @sb: superblock diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index f5cf2a94b6fc..0cffb59fff46 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -853,7 +853,7 @@ cleanup: for (i = 0; i < depth; i++) { if (!ablocks[i]) continue; - ext4_free_blocks(handle, inode, ablocks[i], 1); + ext4_free_blocks(handle, inode, ablocks[i], 1, 1); } } kfree(ablocks); @@ -1698,7 +1698,7 @@ static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, ext_debug("index is empty, remove it, free block %llu\n", leaf); bh = sb_find_get_block(inode->i_sb, leaf); ext4_forget(handle, 1, inode, bh, leaf); - ext4_free_blocks(handle, inode, leaf, 1); + ext4_free_blocks(handle, inode, leaf, 1, 1); return err; } @@ -1759,8 +1759,10 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode, { struct buffer_head *bh; unsigned short ee_len = ext4_ext_get_actual_len(ex); - int i; + int i, metadata = 0; + if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) + metadata = 1; #ifdef EXTENTS_STATS { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); @@ -1789,7 +1791,7 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode, bh = sb_find_get_block(inode->i_sb, start + i); ext4_forget(handle, 0, inode, bh, start + i); } - ext4_free_blocks(handle, inode, start, num); + ext4_free_blocks(handle, inode, start, num, metadata); } else if (from == le32_to_cpu(ex->ee_block) && to <= le32_to_cpu(ex->ee_block) + ee_len - 1) { printk(KERN_INFO "strange request: removal %u-%u from %u:%u\n", @@ -2287,6 +2289,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, ext4_fsblk_t goal, newblock; int err = 0, depth, ret; unsigned long allocated = 0; + struct ext4_allocation_request ar; __clear_bit(BH_New, &bh_result->b_state); ext_debug("blocks %u/%lu requested for inode %u\n", @@ -2397,8 +2400,15 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, if (S_ISREG(inode->i_mode) && (!EXT4_I(inode)->i_block_alloc_info)) ext4_init_block_alloc_info(inode); - /* allocate new block */ - goal = ext4_ext_find_goal(inode, path, iblock); + /* find neighbour allocated blocks */ + ar.lleft = iblock; + err = ext4_ext_search_left(inode, path, &ar.lleft, &ar.pleft); + if (err) + goto out2; + ar.lright = iblock; + err = ext4_ext_search_right(inode, path, &ar.lright, &ar.pright); + if (err) + goto out2; /* * See if request is beyond maximum number of blocks we can have in @@ -2421,7 +2431,18 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, allocated = le16_to_cpu(newex.ee_len); else allocated = max_blocks; - newblock = ext4_new_blocks(handle, inode, goal, &allocated, &err); + + /* allocate new block */ + ar.inode = inode; + ar.goal = ext4_ext_find_goal(inode, path, iblock); + ar.logical = iblock; + ar.len = allocated; + if (S_ISREG(inode->i_mode)) + ar.flags = EXT4_MB_HINT_DATA; + else + /* disable in-core preallocation for non-regular files */ + ar.flags = 0; + newblock = ext4_mb_new_blocks(handle, &ar, &err); if (!newblock) goto out2; ext_debug("allocate new block: goal %llu, found %llu/%lu\n", @@ -2429,14 +2450,17 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, /* try to insert new extent into found leaf and return */ ext4_ext_store_pblock(&newex, newblock); - newex.ee_len = cpu_to_le16(allocated); + newex.ee_len = cpu_to_le16(ar.len); if (create == EXT4_CREATE_UNINITIALIZED_EXT) /* Mark uninitialized */ ext4_ext_mark_uninitialized(&newex); err = ext4_ext_insert_extent(handle, inode, path, &newex); if (err) { /* free data blocks we just allocated */ + /* not a good idea to call discard here directly, + * but otherwise we'd need to call it every free() */ + ext4_mb_discard_inode_preallocations(inode); ext4_free_blocks(handle, inode, ext_pblock(&newex), - le16_to_cpu(newex.ee_len)); + le16_to_cpu(newex.ee_len), 0); goto out2; } @@ -2445,6 +2469,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, /* previous routine could use block we allocated */ newblock = ext_pblock(&newex); + allocated = le16_to_cpu(newex.ee_len); outnew: __set_bit(BH_New, &bh_result->b_state); @@ -2496,6 +2521,8 @@ void ext4_ext_truncate(struct inode * inode, struct page *page) down_write(&EXT4_I(inode)->i_data_sem); ext4_ext_invalidate_cache(inode); + ext4_mb_discard_inode_preallocations(inode); + /* * TODO: optimization is possible here. * Probably we need not scan at all, diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index a06a3b7cfc34..bb717cbb749c 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -551,7 +551,7 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, return ret; failed_out: for (i = 0; i + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public Licens + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111- + */ + + +/* + * mballoc.c contains the multiblocks allocation routines + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "group.h" + +/* + * MUSTDO: + * - test ext4_ext_search_left() and ext4_ext_search_right() + * - search for metadata in few groups + * + * TODO v4: + * - normalization should take into account whether file is still open + * - discard preallocations if no free space left (policy?) + * - don't normalize tails + * - quota + * - reservation for superuser + * + * TODO v3: + * - bitmap read-ahead (proposed by Oleg Drokin aka green) + * - track min/max extents in each group for better group selection + * - mb_mark_used() may allocate chunk right after splitting buddy + * - tree of groups sorted by number of free blocks + * - error handling + */ + +/* + * The allocation request involve request for multiple number of blocks + * near to the goal(block) value specified. + * + * During initialization phase of the allocator we decide to use the group + * preallocation or inode preallocation depending on the size file. The + * size of the file could be the resulting file size we would have after + * allocation or the current file size which ever is larger. If the size is + * less that sbi->s_mb_stream_request we select the group + * preallocation. The default value of s_mb_stream_request is 16 + * blocks. This can also be tuned via + * /proc/fs/ext4//stream_req. The value is represented in terms + * of number of blocks. + * + * The main motivation for having small file use group preallocation is to + * ensure that we have small file closer in the disk. + * + * First stage the allocator looks at the inode prealloc list + * ext4_inode_info->i_prealloc_list contain list of prealloc spaces for + * this particular inode. The inode prealloc space is represented as: + * + * pa_lstart -> the logical start block for this prealloc space + * pa_pstart -> the physical start block for this prealloc space + * pa_len -> lenght for this prealloc space + * pa_free -> free space available in this prealloc space + * + * The inode preallocation space is used looking at the _logical_ start + * block. If only the logical file block falls within the range of prealloc + * space we will consume the particular prealloc space. This make sure that + * that the we have contiguous physical blocks representing the file blocks + * + * The important thing to be noted in case of inode prealloc space is that + * we don't modify the values associated to inode prealloc space except + * pa_free. + * + * If we are not able to find blocks in the inode prealloc space and if we + * have the group allocation flag set then we look at the locality group + * prealloc space. These are per CPU prealloc list repreasented as + * + * ext4_sb_info.s_locality_groups[smp_processor_id()] + * + * The reason for having a per cpu locality group is to reduce the contention + * between CPUs. It is possible to get scheduled at this point. + * + * The locality group prealloc space is used looking at whether we have + * enough free space (pa_free) withing the prealloc space. + * + * If we can't allocate blocks via inode prealloc or/and locality group + * prealloc then we look at the buddy cache. The buddy cache is represented + * by ext4_sb_info.s_buddy_cache (struct inode) whose file offset gets + * mapped to the buddy and bitmap information regarding different + * groups. The buddy information is attached to buddy cache inode so that + * we can access them through the page cache. The information regarding + * each group is loaded via ext4_mb_load_buddy. The information involve + * block bitmap and buddy information. The information are stored in the + * inode as: + * + * { page } + * [ group 0 buddy][ group 0 bitmap] [group 1][ group 1]... + * + * + * one block each for bitmap and buddy information. So for each group we + * take up 2 blocks. A page can contain blocks_per_page (PAGE_CACHE_SIZE / + * blocksize) blocks. So it can have information regarding groups_per_page + * which is blocks_per_page/2 + * + * The buddy cache inode is not stored on disk. The inode is thrown + * away when the filesystem is unmounted. + * + * We look for count number of blocks in the buddy cache. If we were able + * to locate that many free blocks we return with additional information + * regarding rest of the contiguous physical block available + * + * Before allocating blocks via buddy cache we normalize the request + * blocks. This ensure we ask for more blocks that we needed. The extra + * blocks that we get after allocation is added to the respective prealloc + * list. In case of inode preallocation we follow a list of heuristics + * based on file size. This can be found in ext4_mb_normalize_request. If + * we are doing a group prealloc we try to normalize the request to + * sbi->s_mb_group_prealloc. Default value of s_mb_group_prealloc is set to + * 512 blocks. This can be tuned via + * /proc/fs/ext4/ option the group prealloc request is normalized to the + * stripe value (sbi->s_stripe) + * + * The regular allocator(using the buddy cache) support few tunables. + * + * /proc/fs/ext4//min_to_scan + * /proc/fs/ext4//max_to_scan + * /proc/fs/ext4//order2_req + * + * The regular allocator use buddy scan only if the request len is power of + * 2 blocks and the order of allocation is >= sbi->s_mb_order2_reqs. The + * value of s_mb_order2_reqs can be tuned via + * /proc/fs/ext4//order2_req. If the request len is equal to + * stripe size (sbi->s_stripe), we try to search for contigous block in + * stripe size. This should result in better allocation on RAID setup. If + * not we search in the specific group using bitmap for best extents. The + * tunable min_to_scan and max_to_scan controll the behaviour here. + * min_to_scan indicate how long the mballoc __must__ look for a best + * extent and max_to_scanindicate how long the mballoc __can__ look for a + * best extent in the found extents. Searching for the blocks starts with + * the group specified as the goal value in allocation context via + * ac_g_ex. Each group is first checked based on the criteria whether it + * can used for allocation. ext4_mb_good_group explains how the groups are + * checked. + * + * Both the prealloc space are getting populated as above. So for the first + * request we will hit the buddy cache which will result in this prealloc + * space getting filled. The prealloc space is then later used for the + * subsequent request. + */ + +/* + * mballoc operates on the following data: + * - on-disk bitmap + * - in-core buddy (actually includes buddy and bitmap) + * - preallocation descriptors (PAs) + * + * there are two types of preallocations: + * - inode + * assiged to specific inode and can be used for this inode only. + * it describes part of inode's space preallocated to specific + * physical blocks. any block from that preallocated can be used + * independent. the descriptor just tracks number of blocks left + * unused. so, before taking some block from descriptor, one must + * make sure corresponded logical block isn't allocated yet. this + * also means that freeing any block within descriptor's range + * must discard all preallocated blocks. + * - locality group + * assigned to specific locality group which does not translate to + * permanent set of inodes: inode can join and leave group. space + * from this type of preallocation can be used for any inode. thus + * it's consumed from the beginning to the end. + * + * relation between them can be expressed as: + * in-core buddy = on-disk bitmap + preallocation descriptors + * + * this mean blocks mballoc considers used are: + * - allocated blocks (persistent) + * - preallocated blocks (non-persistent) + * + * consistency in mballoc world means that at any time a block is either + * free or used in ALL structures. notice: "any time" should not be read + * literally -- time is discrete and delimited by locks. + * + * to keep it simple, we don't use block numbers, instead we count number of + * blocks: how many blocks marked used/free in on-disk bitmap, buddy and PA. + * + * all operations can be expressed as: + * - init buddy: buddy = on-disk + PAs + * - new PA: buddy += N; PA = N + * - use inode PA: on-disk += N; PA -= N + * - discard inode PA buddy -= on-disk - PA; PA = 0 + * - use locality group PA on-disk += N; PA -= N + * - discard locality group PA buddy -= PA; PA = 0 + * note: 'buddy -= on-disk - PA' is used to show that on-disk bitmap + * is used in real operation because we can't know actual used + * bits from PA, only from on-disk bitmap + * + * if we follow this strict logic, then all operations above should be atomic. + * given some of them can block, we'd have to use something like semaphores + * killing performance on high-end SMP hardware. let's try to relax it using + * the following knowledge: + * 1) if buddy is referenced, it's already initialized + * 2) while block is used in buddy and the buddy is referenced, + * nobody can re-allocate that block + * 3) we work on bitmaps and '+' actually means 'set bits'. if on-disk has + * bit set and PA claims same block, it's OK. IOW, one can set bit in + * on-disk bitmap if buddy has same bit set or/and PA covers corresponded + * block + * + * so, now we're building a concurrency table: + * - init buddy vs. + * - new PA + * blocks for PA are allocated in the buddy, buddy must be referenced + * until PA is linked to allocation group to avoid concurrent buddy init + * - use inode PA + * we need to make sure that either on-disk bitmap or PA has uptodate data + * given (3) we care that PA-=N operation doesn't interfere with init + * - discard inode PA + * the simplest way would be to have buddy initialized by the discard + * - use locality group PA + * again PA-=N must be serialized with init + * - discard locality group PA + * the simplest way would be to have buddy initialized by the discard + * - new PA vs. + * - use inode PA + * i_data_sem serializes them + * - discard inode PA + * discard process must wait until PA isn't used by another process + * - use locality group PA + * some mutex should serialize them + * - discard locality group PA + * discard process must wait until PA isn't used by another process + * - use inode PA + * - use inode PA + * i_data_sem or another mutex should serializes them + * - discard inode PA + * discard process must wait until PA isn't used by another process + * - use locality group PA + * nothing wrong here -- they're different PAs covering different blocks + * - discard locality group PA + * discard process must wait until PA isn't used by another process + * + * now we're ready to make few consequences: + * - PA is referenced and while it is no discard is possible + * - PA is referenced until block isn't marked in on-disk bitmap + * - PA changes only after on-disk bitmap + * - discard must not compete with init. either init is done before + * any discard or they're serialized somehow + * - buddy init as sum of on-disk bitmap and PAs is done atomically + * + * a special case when we've used PA to emptiness. no need to modify buddy + * in this case, but we should care about concurrent init + * + */ + + /* + * Logic in few words: + * + * - allocation: + * load group + * find blocks + * mark bits in on-disk bitmap + * release group + * + * - use preallocation: + * find proper PA (per-inode or group) + * load group + * mark bits in on-disk bitmap + * release group + * release PA + * + * - free: + * load group + * mark bits in on-disk bitmap + * release group + * + * - discard preallocations in group: + * mark PAs deleted + * move them onto local list + * load on-disk bitmap + * load group + * remove PA from object (inode or locality group) + * mark free blocks in-core + * + * - discard inode's preallocations: + */ + +/* + * Locking rules + * + * Locks: + * - bitlock on a group (group) + * - object (inode/locality) (object) + * - per-pa lock (pa) + * + * Paths: + * - new pa + * object + * group + * + * - find and use pa: + * pa + * + * - release consumed pa: + * pa + * group + * object + * + * - generate in-core bitmap: + * group + * pa + * + * - discard all for given object (inode, locality group): + * object + * pa + * group + * + * - discard all for given group: + * group + * pa + * group + * object + * + */ + +/* + * with AGGRESSIVE_CHECK allocator runs consistency checks over + * structures. these checks slow things down a lot + */ +#define AGGRESSIVE_CHECK__ + +/* + * with DOUBLE_CHECK defined mballoc creates persistent in-core + * bitmaps, maintains and uses them to check for double allocations + */ +#define DOUBLE_CHECK__ + +/* + */ +#define MB_DEBUG__ +#ifdef MB_DEBUG +#define mb_debug(fmt, a...) printk(fmt, ##a) +#else +#define mb_debug(fmt, a...) +#endif + +/* + * with EXT4_MB_HISTORY mballoc stores last N allocations in memory + * and you can monitor it in /proc/fs/ext4//mb_history + */ +#define EXT4_MB_HISTORY +#define EXT4_MB_HISTORY_ALLOC 1 /* allocation */ +#define EXT4_MB_HISTORY_PREALLOC 2 /* preallocated blocks used */ +#define EXT4_MB_HISTORY_DISCARD 4 /* preallocation discarded */ +#define EXT4_MB_HISTORY_FREE 8 /* free */ + +#define EXT4_MB_HISTORY_DEFAULT (EXT4_MB_HISTORY_ALLOC | \ + EXT4_MB_HISTORY_PREALLOC) + +/* + * How long mballoc can look for a best extent (in found extents) + */ +#define MB_DEFAULT_MAX_TO_SCAN 200 + +/* + * How long mballoc must look for a best extent + */ +#define MB_DEFAULT_MIN_TO_SCAN 10 + +/* + * How many groups mballoc will scan looking for the best chunk + */ +#define MB_DEFAULT_MAX_GROUPS_TO_SCAN 5 + +/* + * with 'ext4_mb_stats' allocator will collect stats that will be + * shown at umount. The collecting costs though! + */ +#define MB_DEFAULT_STATS 1 + +/* + * files smaller than MB_DEFAULT_STREAM_THRESHOLD are served + * by the stream allocator, which purpose is to pack requests + * as close each to other as possible to produce smooth I/O traffic + * We use locality group prealloc space for stream request. + * We can tune the same via /proc/fs/ext4//stream_req + */ +#define MB_DEFAULT_STREAM_THRESHOLD 16 /* 64K */ + +/* + * for which requests use 2^N search using buddies + */ +#define MB_DEFAULT_ORDER2_REQS 2 + +/* + * default group prealloc size 512 blocks + */ +#define MB_DEFAULT_GROUP_PREALLOC 512 + +static struct kmem_cache *ext4_pspace_cachep; + +#ifdef EXT4_BB_MAX_BLOCKS +#undef EXT4_BB_MAX_BLOCKS +#endif +#define EXT4_BB_MAX_BLOCKS 30 + +struct ext4_free_metadata { + ext4_group_t group; + unsigned short num; + ext4_grpblk_t blocks[EXT4_BB_MAX_BLOCKS]; + struct list_head list; +}; + +struct ext4_group_info { + unsigned long bb_state; + unsigned long bb_tid; + struct ext4_free_metadata *bb_md_cur; + unsigned short bb_first_free; + unsigned short bb_free; + unsigned short bb_fragments; + struct list_head bb_prealloc_list; +#ifdef DOUBLE_CHECK + void *bb_bitmap; +#endif + unsigned short bb_counters[]; +}; + +#define EXT4_GROUP_INFO_NEED_INIT_BIT 0 +#define EXT4_GROUP_INFO_LOCKED_BIT 1 + +#define EXT4_MB_GRP_NEED_INIT(grp) \ + (test_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &((grp)->bb_state))) + + +struct ext4_prealloc_space { + struct list_head pa_inode_list; + struct list_head pa_group_list; + union { + struct list_head pa_tmp_list; + struct rcu_head pa_rcu; + } u; + spinlock_t pa_lock; + atomic_t pa_count; + unsigned pa_deleted; + ext4_fsblk_t pa_pstart; /* phys. block */ + ext4_lblk_t pa_lstart; /* log. block */ + unsigned short pa_len; /* len of preallocated chunk */ + unsigned short pa_free; /* how many blocks are free */ + unsigned short pa_linear; /* consumed in one direction + * strictly, for grp prealloc */ + spinlock_t *pa_obj_lock; + struct inode *pa_inode; /* hack, for history only */ +}; + + +struct ext4_free_extent { + ext4_lblk_t fe_logical; + ext4_grpblk_t fe_start; + ext4_group_t fe_group; + int fe_len; +}; + +/* + * Locality group: + * we try to group all related changes together + * so that writeback can flush/allocate them together as well + */ +struct ext4_locality_group { + /* for allocator */ + struct mutex lg_mutex; /* to serialize allocates */ + struct list_head lg_prealloc_list;/* list of preallocations */ + spinlock_t lg_prealloc_lock; +}; + +struct ext4_allocation_context { + struct inode *ac_inode; + struct super_block *ac_sb; + + /* original request */ + struct ext4_free_extent ac_o_ex; + + /* goal request (after normalization) */ + struct ext4_free_extent ac_g_ex; + + /* the best found extent */ + struct ext4_free_extent ac_b_ex; + + /* copy of the bext found extent taken before preallocation efforts */ + struct ext4_free_extent ac_f_ex; + + /* number of iterations done. we have to track to limit searching */ + unsigned long ac_ex_scanned; + __u16 ac_groups_scanned; + __u16 ac_found; + __u16 ac_tail; + __u16 ac_buddy; + __u16 ac_flags; /* allocation hints */ + __u8 ac_status; + __u8 ac_criteria; + __u8 ac_repeats; + __u8 ac_2order; /* if request is to allocate 2^N blocks and + * N > 0, the field stores N, otherwise 0 */ + __u8 ac_op; /* operation, for history only */ + struct page *ac_bitmap_page; + struct page *ac_buddy_page; + struct ext4_prealloc_space *ac_pa; + struct ext4_locality_group *ac_lg; +}; + +#define AC_STATUS_CONTINUE 1 +#define AC_STATUS_FOUND 2 +#define AC_STATUS_BREAK 3 + +struct ext4_mb_history { + struct ext4_free_extent orig; /* orig allocation */ + struct ext4_free_extent goal; /* goal allocation */ + struct ext4_free_extent result; /* result allocation */ + unsigned pid; + unsigned ino; + __u16 found; /* how many extents have been found */ + __u16 groups; /* how many groups have been scanned */ + __u16 tail; /* what tail broke some buddy */ + __u16 buddy; /* buddy the tail ^^^ broke */ + __u16 flags; + __u8 cr:3; /* which phase the result extent was found at */ + __u8 op:4; + __u8 merged:1; +}; + +struct ext4_buddy { + struct page *bd_buddy_page; + void *bd_buddy; + struct page *bd_bitmap_page; + void *bd_bitmap; + struct ext4_group_info *bd_info; + struct super_block *bd_sb; + __u16 bd_blkbits; + ext4_group_t bd_group; +}; +#define EXT4_MB_BITMAP(e4b) ((e4b)->bd_bitmap) +#define EXT4_MB_BUDDY(e4b) ((e4b)->bd_buddy) + +#ifndef EXT4_MB_HISTORY +static inline void ext4_mb_store_history(struct ext4_allocation_context *ac) +{ + return; +} +#else +static void ext4_mb_store_history(struct ext4_allocation_context *ac); +#endif + +#define in_range(b, first, len) ((b) >= (first) && (b) <= (first) + (len) - 1) + +static struct proc_dir_entry *proc_root_ext4; +struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t); +ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, unsigned long *count, int *errp); + +static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap, + ext4_group_t group); +static void ext4_mb_poll_new_transaction(struct super_block *, handle_t *); +static void ext4_mb_free_committed_blocks(struct super_block *); +static void ext4_mb_return_to_preallocation(struct inode *inode, + struct ext4_buddy *e4b, sector_t block, + int count); +static void ext4_mb_put_pa(struct ext4_allocation_context *, + struct super_block *, struct ext4_prealloc_space *pa); +static int ext4_mb_init_per_dev_proc(struct super_block *sb); +static int ext4_mb_destroy_per_dev_proc(struct super_block *sb); + + +static inline void ext4_lock_group(struct super_block *sb, ext4_group_t group) +{ + struct ext4_group_info *grinfo = ext4_get_group_info(sb, group); + + bit_spin_lock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state)); +} + +static inline void ext4_unlock_group(struct super_block *sb, + ext4_group_t group) +{ + struct ext4_group_info *grinfo = ext4_get_group_info(sb, group); + + bit_spin_unlock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state)); +} + +static inline int ext4_is_group_locked(struct super_block *sb, + ext4_group_t group) +{ + struct ext4_group_info *grinfo = ext4_get_group_info(sb, group); + + return bit_spin_is_locked(EXT4_GROUP_INFO_LOCKED_BIT, + &(grinfo->bb_state)); +} + +static ext4_fsblk_t ext4_grp_offs_to_block(struct super_block *sb, + struct ext4_free_extent *fex) +{ + ext4_fsblk_t block; + + block = (ext4_fsblk_t) fex->fe_group * EXT4_BLOCKS_PER_GROUP(sb) + + fex->fe_start + + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); + return block; +} + +#if BITS_PER_LONG == 64 +#define mb_correct_addr_and_bit(bit, addr) \ +{ \ + bit += ((unsigned long) addr & 7UL) << 3; \ + addr = (void *) ((unsigned long) addr & ~7UL); \ +} +#elif BITS_PER_LONG == 32 +#define mb_correct_addr_and_bit(bit, addr) \ +{ \ + bit += ((unsigned long) addr & 3UL) << 3; \ + addr = (void *) ((unsigned long) addr & ~3UL); \ +} +#else +#error "how many bits you are?!" +#endif + +static inline int mb_test_bit(int bit, void *addr) +{ + /* + * ext4_test_bit on architecture like powerpc + * needs unsigned long aligned address + */ + mb_correct_addr_and_bit(bit, addr); + return ext4_test_bit(bit, addr); +} + +static inline void mb_set_bit(int bit, void *addr) +{ + mb_correct_addr_and_bit(bit, addr); + ext4_set_bit(bit, addr); +} + +static inline void mb_set_bit_atomic(spinlock_t *lock, int bit, void *addr) +{ + mb_correct_addr_and_bit(bit, addr); + ext4_set_bit_atomic(lock, bit, addr); +} + +static inline void mb_clear_bit(int bit, void *addr) +{ + mb_correct_addr_and_bit(bit, addr); + ext4_clear_bit(bit, addr); +} + +static inline void mb_clear_bit_atomic(spinlock_t *lock, int bit, void *addr) +{ + mb_correct_addr_and_bit(bit, addr); + ext4_clear_bit_atomic(lock, bit, addr); +} + +static void *mb_find_buddy(struct ext4_buddy *e4b, int order, int *max) +{ + char *bb; + + /* FIXME!! is this needed */ + BUG_ON(EXT4_MB_BITMAP(e4b) == EXT4_MB_BUDDY(e4b)); + BUG_ON(max == NULL); + + if (order > e4b->bd_blkbits + 1) { + *max = 0; + return NULL; + } + + /* at order 0 we see each particular block */ + *max = 1 << (e4b->bd_blkbits + 3); + if (order == 0) + return EXT4_MB_BITMAP(e4b); + + bb = EXT4_MB_BUDDY(e4b) + EXT4_SB(e4b->bd_sb)->s_mb_offsets[order]; + *max = EXT4_SB(e4b->bd_sb)->s_mb_maxs[order]; + + return bb; +} + +#ifdef DOUBLE_CHECK +static void mb_free_blocks_double(struct inode *inode, struct ext4_buddy *e4b, + int first, int count) +{ + int i; + struct super_block *sb = e4b->bd_sb; + + if (unlikely(e4b->bd_info->bb_bitmap == NULL)) + return; + BUG_ON(!ext4_is_group_locked(sb, e4b->bd_group)); + for (i = 0; i < count; i++) { + if (!mb_test_bit(first + i, e4b->bd_info->bb_bitmap)) { + ext4_fsblk_t blocknr; + blocknr = e4b->bd_group * EXT4_BLOCKS_PER_GROUP(sb); + blocknr += first + i; + blocknr += + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); + + ext4_error(sb, __FUNCTION__, "double-free of inode" + " %lu's block %llu(bit %u in group %lu)\n", + inode ? inode->i_ino : 0, blocknr, + first + i, e4b->bd_group); + } + mb_clear_bit(first + i, e4b->bd_info->bb_bitmap); + } +} + +static void mb_mark_used_double(struct ext4_buddy *e4b, int first, int count) +{ + int i; + + if (unlikely(e4b->bd_info->bb_bitmap == NULL)) + return; + BUG_ON(!ext4_is_group_locked(e4b->bd_sb, e4b->bd_group)); + for (i = 0; i < count; i++) { + BUG_ON(mb_test_bit(first + i, e4b->bd_info->bb_bitmap)); + mb_set_bit(first + i, e4b->bd_info->bb_bitmap); + } +} + +static void mb_cmp_bitmaps(struct ext4_buddy *e4b, void *bitmap) +{ + if (memcmp(e4b->bd_info->bb_bitmap, bitmap, e4b->bd_sb->s_blocksize)) { + unsigned char *b1, *b2; + int i; + b1 = (unsigned char *) e4b->bd_info->bb_bitmap; + b2 = (unsigned char *) bitmap; + for (i = 0; i < e4b->bd_sb->s_blocksize; i++) { + if (b1[i] != b2[i]) { + printk("corruption in group %lu at byte %u(%u):" + " %x in copy != %x on disk/prealloc\n", + e4b->bd_group, i, i * 8, b1[i], b2[i]); + BUG(); + } + } + } +} + +#else +static inline void mb_free_blocks_double(struct inode *inode, + struct ext4_buddy *e4b, int first, int count) +{ + return; +} +static inline void mb_mark_used_double(struct ext4_buddy *e4b, + int first, int count) +{ + return; +} +static inline void mb_cmp_bitmaps(struct ext4_buddy *e4b, void *bitmap) +{ + return; +} +#endif + +#ifdef AGGRESSIVE_CHECK + +#define MB_CHECK_ASSERT(assert) \ +do { \ + if (!(assert)) { \ + printk(KERN_EMERG \ + "Assertion failure in %s() at %s:%d: \"%s\"\n", \ + function, file, line, # assert); \ + BUG(); \ + } \ +} while (0) + +static int __mb_check_buddy(struct ext4_buddy *e4b, char *file, + const char *function, int line) +{ + struct super_block *sb = e4b->bd_sb; + int order = e4b->bd_blkbits + 1; + int max; + int max2; + int i; + int j; + int k; + int count; + struct ext4_group_info *grp; + int fragments = 0; + int fstart; + struct list_head *cur; + void *buddy; + void *buddy2; + + if (!test_opt(sb, MBALLOC)) + return 0; + + { + static int mb_check_counter; + if (mb_check_counter++ % 100 != 0) + return 0; + } + + while (order > 1) { + buddy = mb_find_buddy(e4b, order, &max); + MB_CHECK_ASSERT(buddy); + buddy2 = mb_find_buddy(e4b, order - 1, &max2); + MB_CHECK_ASSERT(buddy2); + MB_CHECK_ASSERT(buddy != buddy2); + MB_CHECK_ASSERT(max * 2 == max2); + + count = 0; + for (i = 0; i < max; i++) { + + if (mb_test_bit(i, buddy)) { + /* only single bit in buddy2 may be 1 */ + if (!mb_test_bit(i << 1, buddy2)) { + MB_CHECK_ASSERT( + mb_test_bit((i<<1)+1, buddy2)); + } else if (!mb_test_bit((i << 1) + 1, buddy2)) { + MB_CHECK_ASSERT( + mb_test_bit(i << 1, buddy2)); + } + continue; + } + + /* both bits in buddy2 must be 0 */ + MB_CHECK_ASSERT(mb_test_bit(i << 1, buddy2)); + MB_CHECK_ASSERT(mb_test_bit((i << 1) + 1, buddy2)); + + for (j = 0; j < (1 << order); j++) { + k = (i * (1 << order)) + j; + MB_CHECK_ASSERT( + !mb_test_bit(k, EXT4_MB_BITMAP(e4b))); + } + count++; + } + MB_CHECK_ASSERT(e4b->bd_info->bb_counters[order] == count); + order--; + } + + fstart = -1; + buddy = mb_find_buddy(e4b, 0, &max); + for (i = 0; i < max; i++) { + if (!mb_test_bit(i, buddy)) { + MB_CHECK_ASSERT(i >= e4b->bd_info->bb_first_free); + if (fstart == -1) { + fragments++; + fstart = i; + } + continue; + } + fstart = -1; + /* check used bits only */ + for (j = 0; j < e4b->bd_blkbits + 1; j++) { + buddy2 = mb_find_buddy(e4b, j, &max2); + k = i >> j; + MB_CHECK_ASSERT(k < max2); + MB_CHECK_ASSERT(mb_test_bit(k, buddy2)); + } + } + MB_CHECK_ASSERT(!EXT4_MB_GRP_NEED_INIT(e4b->bd_info)); + MB_CHECK_ASSERT(e4b->bd_info->bb_fragments == fragments); + + grp = ext4_get_group_info(sb, e4b->bd_group); + buddy = mb_find_buddy(e4b, 0, &max); + list_for_each(cur, &grp->bb_prealloc_list) { + ext4_group_t groupnr; + struct ext4_prealloc_space *pa; + pa = list_entry(cur, struct ext4_prealloc_space, group_list); + ext4_get_group_no_and_offset(sb, pa->pstart, &groupnr, &k); + MB_CHECK_ASSERT(groupnr == e4b->bd_group); + for (i = 0; i < pa->len; i++) + MB_CHECK_ASSERT(mb_test_bit(k + i, buddy)); + } + return 0; +} +#undef MB_CHECK_ASSERT +#define mb_check_buddy(e4b) __mb_check_buddy(e4b, \ + __FILE__, __FUNCTION__, __LINE__) +#else +#define mb_check_buddy(e4b) +#endif + +/* FIXME!! need more doc */ +static void ext4_mb_mark_free_simple(struct super_block *sb, + void *buddy, unsigned first, int len, + struct ext4_group_info *grp) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + unsigned short min; + unsigned short max; + unsigned short chunk; + unsigned short border; + + BUG_ON(len >= EXT4_BLOCKS_PER_GROUP(sb)); + + border = 2 << sb->s_blocksize_bits; + + while (len > 0) { + /* find how many blocks can be covered since this position */ + max = ffs(first | border) - 1; + + /* find how many blocks of power 2 we need to mark */ + min = fls(len) - 1; + + if (max < min) + min = max; + chunk = 1 << min; + + /* mark multiblock chunks only */ + grp->bb_counters[min]++; + if (min > 0) + mb_clear_bit(first >> min, + buddy + sbi->s_mb_offsets[min]); + + len -= chunk; + first += chunk; + } +} + +static void ext4_mb_generate_buddy(struct super_block *sb, + void *buddy, void *bitmap, ext4_group_t group) +{ + struct ext4_group_info *grp = ext4_get_group_info(sb, group); + unsigned short max = EXT4_BLOCKS_PER_GROUP(sb); + unsigned short i = 0; + unsigned short first; + unsigned short len; + unsigned free = 0; + unsigned fragments = 0; + unsigned long long period = get_cycles(); + + /* initialize buddy from bitmap which is aggregation + * of on-disk bitmap and preallocations */ + i = ext4_find_next_zero_bit(bitmap, max, 0); + grp->bb_first_free = i; + while (i < max) { + fragments++; + first = i; + i = ext4_find_next_bit(bitmap, max, i); + len = i - first; + free += len; + if (len > 1) + ext4_mb_mark_free_simple(sb, buddy, first, len, grp); + else + grp->bb_counters[0]++; + if (i < max) + i = ext4_find_next_zero_bit(bitmap, max, i); + } + grp->bb_fragments = fragments; + + if (free != grp->bb_free) { + printk(KERN_DEBUG + "EXT4-fs: group %lu: %u blocks in bitmap, %u in gd\n", + group, free, grp->bb_free); + grp->bb_free = free; + } + + clear_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &(grp->bb_state)); + + period = get_cycles() - period; + spin_lock(&EXT4_SB(sb)->s_bal_lock); + EXT4_SB(sb)->s_mb_buddies_generated++; + EXT4_SB(sb)->s_mb_generation_time += period; + spin_unlock(&EXT4_SB(sb)->s_bal_lock); +} + +/* The buddy information is attached the buddy cache inode + * for convenience. The information regarding each group + * is loaded via ext4_mb_load_buddy. The information involve + * block bitmap and buddy information. The information are + * stored in the inode as + * + * { page } + * [ group 0 buddy][ group 0 bitmap] [group 1][ group 1]... + * + * + * one block each for bitmap and buddy information. + * So for each group we take up 2 blocks. A page can + * contain blocks_per_page (PAGE_CACHE_SIZE / blocksize) blocks. + * So it can have information regarding groups_per_page which + * is blocks_per_page/2 + */ + +static int ext4_mb_init_cache(struct page *page, char *incore) +{ + int blocksize; + int blocks_per_page; + int groups_per_page; + int err = 0; + int i; + ext4_group_t first_group; + int first_block; + struct super_block *sb; + struct buffer_head *bhs; + struct buffer_head **bh; + struct inode *inode; + char *data; + char *bitmap; + + mb_debug("init page %lu\n", page->index); + + inode = page->mapping->host; + sb = inode->i_sb; + blocksize = 1 << inode->i_blkbits; + blocks_per_page = PAGE_CACHE_SIZE / blocksize; + + groups_per_page = blocks_per_page >> 1; + if (groups_per_page == 0) + groups_per_page = 1; + + /* allocate buffer_heads to read bitmaps */ + if (groups_per_page > 1) { + err = -ENOMEM; + i = sizeof(struct buffer_head *) * groups_per_page; + bh = kzalloc(i, GFP_NOFS); + if (bh == NULL) + goto out; + } else + bh = &bhs; + + first_group = page->index * blocks_per_page / 2; + + /* read all groups the page covers into the cache */ + for (i = 0; i < groups_per_page; i++) { + struct ext4_group_desc *desc; + + if (first_group + i >= EXT4_SB(sb)->s_groups_count) + break; + + err = -EIO; + desc = ext4_get_group_desc(sb, first_group + i, NULL); + if (desc == NULL) + goto out; + + err = -ENOMEM; + bh[i] = sb_getblk(sb, ext4_block_bitmap(sb, desc)); + if (bh[i] == NULL) + goto out; + + if (bh_uptodate_or_lock(bh[i])) + continue; + + if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { + ext4_init_block_bitmap(sb, bh[i], + first_group + i, desc); + set_buffer_uptodate(bh[i]); + unlock_buffer(bh[i]); + continue; + } + get_bh(bh[i]); + bh[i]->b_end_io = end_buffer_read_sync; + submit_bh(READ, bh[i]); + mb_debug("read bitmap for group %lu\n", first_group + i); + } + + /* wait for I/O completion */ + for (i = 0; i < groups_per_page && bh[i]; i++) + wait_on_buffer(bh[i]); + + err = -EIO; + for (i = 0; i < groups_per_page && bh[i]; i++) + if (!buffer_uptodate(bh[i])) + goto out; + + first_block = page->index * blocks_per_page; + for (i = 0; i < blocks_per_page; i++) { + int group; + struct ext4_group_info *grinfo; + + group = (first_block + i) >> 1; + if (group >= EXT4_SB(sb)->s_groups_count) + break; + + /* + * data carry information regarding this + * particular group in the format specified + * above + * + */ + data = page_address(page) + (i * blocksize); + bitmap = bh[group - first_group]->b_data; + + /* + * We place the buddy block and bitmap block + * close together + */ + if ((first_block + i) & 1) { + /* this is block of buddy */ + BUG_ON(incore == NULL); + mb_debug("put buddy for group %u in page %lu/%x\n", + group, page->index, i * blocksize); + memset(data, 0xff, blocksize); + grinfo = ext4_get_group_info(sb, group); + grinfo->bb_fragments = 0; + memset(grinfo->bb_counters, 0, + sizeof(unsigned short)*(sb->s_blocksize_bits+2)); + /* + * incore got set to the group block bitmap below + */ + ext4_mb_generate_buddy(sb, data, incore, group); + incore = NULL; + } else { + /* this is block of bitmap */ + BUG_ON(incore != NULL); + mb_debug("put bitmap for group %u in page %lu/%x\n", + group, page->index, i * blocksize); + + /* see comments in ext4_mb_put_pa() */ + ext4_lock_group(sb, group); + memcpy(data, bitmap, blocksize); + + /* mark all preallocated blks used in in-core bitmap */ + ext4_mb_generate_from_pa(sb, data, group); + ext4_unlock_group(sb, group); + + /* set incore so that the buddy information can be + * generated using this + */ + incore = data; + } + } + SetPageUptodate(page); + +out: + if (bh) { + for (i = 0; i < groups_per_page && bh[i]; i++) + brelse(bh[i]); + if (bh != &bhs) + kfree(bh); + } + return err; +} + +static int ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group, + struct ext4_buddy *e4b) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct inode *inode = sbi->s_buddy_cache; + int blocks_per_page; + int block; + int pnum; + int poff; + struct page *page; + + mb_debug("load group %lu\n", group); + + blocks_per_page = PAGE_CACHE_SIZE / sb->s_blocksize; + + e4b->bd_blkbits = sb->s_blocksize_bits; + e4b->bd_info = ext4_get_group_info(sb, group); + e4b->bd_sb = sb; + e4b->bd_group = group; + e4b->bd_buddy_page = NULL; + e4b->bd_bitmap_page = NULL; + + /* + * the buddy cache inode stores the block bitmap + * and buddy information in consecutive blocks. + * So for each group we need two blocks. + */ + block = group * 2; + pnum = block / blocks_per_page; + poff = block % blocks_per_page; + + /* we could use find_or_create_page(), but it locks page + * what we'd like to avoid in fast path ... */ + page = find_get_page(inode->i_mapping, pnum); + if (page == NULL || !PageUptodate(page)) { + if (page) + page_cache_release(page); + page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS); + if (page) { + BUG_ON(page->mapping != inode->i_mapping); + if (!PageUptodate(page)) { + ext4_mb_init_cache(page, NULL); + mb_cmp_bitmaps(e4b, page_address(page) + + (poff * sb->s_blocksize)); + } + unlock_page(page); + } + } + if (page == NULL || !PageUptodate(page)) + goto err; + e4b->bd_bitmap_page = page; + e4b->bd_bitmap = page_address(page) + (poff * sb->s_blocksize); + mark_page_accessed(page); + + block++; + pnum = block / blocks_per_page; + poff = block % blocks_per_page; + + page = find_get_page(inode->i_mapping, pnum); + if (page == NULL || !PageUptodate(page)) { + if (page) + page_cache_release(page); + page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS); + if (page) { + BUG_ON(page->mapping != inode->i_mapping); + if (!PageUptodate(page)) + ext4_mb_init_cache(page, e4b->bd_bitmap); + + unlock_page(page); + } + } + if (page == NULL || !PageUptodate(page)) + goto err; + e4b->bd_buddy_page = page; + e4b->bd_buddy = page_address(page) + (poff * sb->s_blocksize); + mark_page_accessed(page); + + BUG_ON(e4b->bd_bitmap_page == NULL); + BUG_ON(e4b->bd_buddy_page == NULL); + + return 0; + +err: + if (e4b->bd_bitmap_page) + page_cache_release(e4b->bd_bitmap_page); + if (e4b->bd_buddy_page) + page_cache_release(e4b->bd_buddy_page); + e4b->bd_buddy = NULL; + e4b->bd_bitmap = NULL; + return -EIO; +} + +static void ext4_mb_release_desc(struct ext4_buddy *e4b) +{ + if (e4b->bd_bitmap_page) + page_cache_release(e4b->bd_bitmap_page); + if (e4b->bd_buddy_page) + page_cache_release(e4b->bd_buddy_page); +} + + +static int mb_find_order_for_block(struct ext4_buddy *e4b, int block) +{ + int order = 1; + void *bb; + + BUG_ON(EXT4_MB_BITMAP(e4b) == EXT4_MB_BUDDY(e4b)); + BUG_ON(block >= (1 << (e4b->bd_blkbits + 3))); + + bb = EXT4_MB_BUDDY(e4b); + while (order <= e4b->bd_blkbits + 1) { + block = block >> 1; + if (!mb_test_bit(block, bb)) { + /* this block is part of buddy of order 'order' */ + return order; + } + bb += 1 << (e4b->bd_blkbits - order); + order++; + } + return 0; +} + +static void mb_clear_bits(spinlock_t *lock, void *bm, int cur, int len) +{ + __u32 *addr; + + len = cur + len; + while (cur < len) { + if ((cur & 31) == 0 && (len - cur) >= 32) { + /* fast path: clear whole word at once */ + addr = bm + (cur >> 3); + *addr = 0; + cur += 32; + continue; + } + mb_clear_bit_atomic(lock, cur, bm); + cur++; + } +} + +static void mb_set_bits(spinlock_t *lock, void *bm, int cur, int len) +{ + __u32 *addr; + + len = cur + len; + while (cur < len) { + if ((cur & 31) == 0 && (len - cur) >= 32) { + /* fast path: set whole word at once */ + addr = bm + (cur >> 3); + *addr = 0xffffffff; + cur += 32; + continue; + } + mb_set_bit_atomic(lock, cur, bm); + cur++; + } +} + +static int mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b, + int first, int count) +{ + int block = 0; + int max = 0; + int order; + void *buddy; + void *buddy2; + struct super_block *sb = e4b->bd_sb; + + BUG_ON(first + count > (sb->s_blocksize << 3)); + BUG_ON(!ext4_is_group_locked(sb, e4b->bd_group)); + mb_check_buddy(e4b); + mb_free_blocks_double(inode, e4b, first, count); + + e4b->bd_info->bb_free += count; + if (first < e4b->bd_info->bb_first_free) + e4b->bd_info->bb_first_free = first; + + /* let's maintain fragments counter */ + if (first != 0) + block = !mb_test_bit(first - 1, EXT4_MB_BITMAP(e4b)); + if (first + count < EXT4_SB(sb)->s_mb_maxs[0]) + max = !mb_test_bit(first + count, EXT4_MB_BITMAP(e4b)); + if (block && max) + e4b->bd_info->bb_fragments--; + else if (!block && !max) + e4b->bd_info->bb_fragments++; + + /* let's maintain buddy itself */ + while (count-- > 0) { + block = first++; + order = 0; + + if (!mb_test_bit(block, EXT4_MB_BITMAP(e4b))) { + ext4_fsblk_t blocknr; + blocknr = e4b->bd_group * EXT4_BLOCKS_PER_GROUP(sb); + blocknr += block; + blocknr += + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); + + ext4_error(sb, __FUNCTION__, "double-free of inode" + " %lu's block %llu(bit %u in group %lu)\n", + inode ? inode->i_ino : 0, blocknr, block, + e4b->bd_group); + } + mb_clear_bit(block, EXT4_MB_BITMAP(e4b)); + e4b->bd_info->bb_counters[order]++; + + /* start of the buddy */ + buddy = mb_find_buddy(e4b, order, &max); + + do { + block &= ~1UL; + if (mb_test_bit(block, buddy) || + mb_test_bit(block + 1, buddy)) + break; + + /* both the buddies are free, try to coalesce them */ + buddy2 = mb_find_buddy(e4b, order + 1, &max); + + if (!buddy2) + break; + + if (order > 0) { + /* for special purposes, we don't set + * free bits in bitmap */ + mb_set_bit(block, buddy); + mb_set_bit(block + 1, buddy); + } + e4b->bd_info->bb_counters[order]--; + e4b->bd_info->bb_counters[order]--; + + block = block >> 1; + order++; + e4b->bd_info->bb_counters[order]++; + + mb_clear_bit(block, buddy2); + buddy = buddy2; + } while (1); + } + mb_check_buddy(e4b); + + return 0; +} + +static int mb_find_extent(struct ext4_buddy *e4b, int order, int block, + int needed, struct ext4_free_extent *ex) +{ + int next = block; + int max; + int ord; + void *buddy; + + BUG_ON(!ext4_is_group_locked(e4b->bd_sb, e4b->bd_group)); + BUG_ON(ex == NULL); + + buddy = mb_find_buddy(e4b, order, &max); + BUG_ON(buddy == NULL); + BUG_ON(block >= max); + if (mb_test_bit(block, buddy)) { + ex->fe_len = 0; + ex->fe_start = 0; + ex->fe_group = 0; + return 0; + } + + /* FIXME dorp order completely ? */ + if (likely(order == 0)) { + /* find actual order */ + order = mb_find_order_for_block(e4b, block); + block = block >> order; + } + + ex->fe_len = 1 << order; + ex->fe_start = block << order; + ex->fe_group = e4b->bd_group; + + /* calc difference from given start */ + next = next - ex->fe_start; + ex->fe_len -= next; + ex->fe_start += next; + + while (needed > ex->fe_len && + (buddy = mb_find_buddy(e4b, order, &max))) { + + if (block + 1 >= max) + break; + + next = (block + 1) * (1 << order); + if (mb_test_bit(next, EXT4_MB_BITMAP(e4b))) + break; + + ord = mb_find_order_for_block(e4b, next); + + order = ord; + block = next >> order; + ex->fe_len += 1 << order; + } + + BUG_ON(ex->fe_start + ex->fe_len > (1 << (e4b->bd_blkbits + 3))); + return ex->fe_len; +} + +static int mb_mark_used(struct ext4_buddy *e4b, struct ext4_free_extent *ex) +{ + int ord; + int mlen = 0; + int max = 0; + int cur; + int start = ex->fe_start; + int len = ex->fe_len; + unsigned ret = 0; + int len0 = len; + void *buddy; + + BUG_ON(start + len > (e4b->bd_sb->s_blocksize << 3)); + BUG_ON(e4b->bd_group != ex->fe_group); + BUG_ON(!ext4_is_group_locked(e4b->bd_sb, e4b->bd_group)); + mb_check_buddy(e4b); + mb_mark_used_double(e4b, start, len); + + e4b->bd_info->bb_free -= len; + if (e4b->bd_info->bb_first_free == start) + e4b->bd_info->bb_first_free += len; + + /* let's maintain fragments counter */ + if (start != 0) + mlen = !mb_test_bit(start - 1, EXT4_MB_BITMAP(e4b)); + if (start + len < EXT4_SB(e4b->bd_sb)->s_mb_maxs[0]) + max = !mb_test_bit(start + len, EXT4_MB_BITMAP(e4b)); + if (mlen && max) + e4b->bd_info->bb_fragments++; + else if (!mlen && !max) + e4b->bd_info->bb_fragments--; + + /* let's maintain buddy itself */ + while (len) { + ord = mb_find_order_for_block(e4b, start); + + if (((start >> ord) << ord) == start && len >= (1 << ord)) { + /* the whole chunk may be allocated at once! */ + mlen = 1 << ord; + buddy = mb_find_buddy(e4b, ord, &max); + BUG_ON((start >> ord) >= max); + mb_set_bit(start >> ord, buddy); + e4b->bd_info->bb_counters[ord]--; + start += mlen; + len -= mlen; + BUG_ON(len < 0); + continue; + } + + /* store for history */ + if (ret == 0) + ret = len | (ord << 16); + + /* we have to split large buddy */ + BUG_ON(ord <= 0); + buddy = mb_find_buddy(e4b, ord, &max); + mb_set_bit(start >> ord, buddy); + e4b->bd_info->bb_counters[ord]--; + + ord--; + cur = (start >> ord) & ~1U; + buddy = mb_find_buddy(e4b, ord, &max); + mb_clear_bit(cur, buddy); + mb_clear_bit(cur + 1, buddy); + e4b->bd_info->bb_counters[ord]++; + e4b->bd_info->bb_counters[ord]++; + } + + mb_set_bits(sb_bgl_lock(EXT4_SB(e4b->bd_sb), ex->fe_group), + EXT4_MB_BITMAP(e4b), ex->fe_start, len0); + mb_check_buddy(e4b); + + return ret; +} + +/* + * Must be called under group lock! + */ +static void ext4_mb_use_best_found(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b) +{ + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + int ret; + + BUG_ON(ac->ac_b_ex.fe_group != e4b->bd_group); + BUG_ON(ac->ac_status == AC_STATUS_FOUND); + + ac->ac_b_ex.fe_len = min(ac->ac_b_ex.fe_len, ac->ac_g_ex.fe_len); + ac->ac_b_ex.fe_logical = ac->ac_g_ex.fe_logical; + ret = mb_mark_used(e4b, &ac->ac_b_ex); + + /* preallocation can change ac_b_ex, thus we store actually + * allocated blocks for history */ + ac->ac_f_ex = ac->ac_b_ex; + + ac->ac_status = AC_STATUS_FOUND; + ac->ac_tail = ret & 0xffff; + ac->ac_buddy = ret >> 16; + + /* XXXXXXX: SUCH A HORRIBLE **CK */ + /*FIXME!! Why ? */ + ac->ac_bitmap_page = e4b->bd_bitmap_page; + get_page(ac->ac_bitmap_page); + ac->ac_buddy_page = e4b->bd_buddy_page; + get_page(ac->ac_buddy_page); + + /* store last allocated for subsequent stream allocation */ + if ((ac->ac_flags & EXT4_MB_HINT_DATA)) { + spin_lock(&sbi->s_md_lock); + sbi->s_mb_last_group = ac->ac_f_ex.fe_group; + sbi->s_mb_last_start = ac->ac_f_ex.fe_start; + spin_unlock(&sbi->s_md_lock); + } +} + +/* + * regular allocator, for general purposes allocation + */ + +static void ext4_mb_check_limits(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b, + int finish_group) +{ + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + struct ext4_free_extent *bex = &ac->ac_b_ex; + struct ext4_free_extent *gex = &ac->ac_g_ex; + struct ext4_free_extent ex; + int max; + + /* + * We don't want to scan for a whole year + */ + if (ac->ac_found > sbi->s_mb_max_to_scan && + !(ac->ac_flags & EXT4_MB_HINT_FIRST)) { + ac->ac_status = AC_STATUS_BREAK; + return; + } + + /* + * Haven't found good chunk so far, let's continue + */ + if (bex->fe_len < gex->fe_len) + return; + + if ((finish_group || ac->ac_found > sbi->s_mb_min_to_scan) + && bex->fe_group == e4b->bd_group) { + /* recheck chunk's availability - we don't know + * when it was found (within this lock-unlock + * period or not) */ + max = mb_find_extent(e4b, 0, bex->fe_start, gex->fe_len, &ex); + if (max >= gex->fe_len) { + ext4_mb_use_best_found(ac, e4b); + return; + } + } +} + +/* + * The routine checks whether found extent is good enough. If it is, + * then the extent gets marked used and flag is set to the context + * to stop scanning. Otherwise, the extent is compared with the + * previous found extent and if new one is better, then it's stored + * in the context. Later, the best found extent will be used, if + * mballoc can't find good enough extent. + * + * FIXME: real allocation policy is to be designed yet! + */ +static void ext4_mb_measure_extent(struct ext4_allocation_context *ac, + struct ext4_free_extent *ex, + struct ext4_buddy *e4b) +{ + struct ext4_free_extent *bex = &ac->ac_b_ex; + struct ext4_free_extent *gex = &ac->ac_g_ex; + + BUG_ON(ex->fe_len <= 0); + BUG_ON(ex->fe_len >= EXT4_BLOCKS_PER_GROUP(ac->ac_sb)); + BUG_ON(ex->fe_start >= EXT4_BLOCKS_PER_GROUP(ac->ac_sb)); + BUG_ON(ac->ac_status != AC_STATUS_CONTINUE); + + ac->ac_found++; + + /* + * The special case - take what you catch first + */ + if (unlikely(ac->ac_flags & EXT4_MB_HINT_FIRST)) { + *bex = *ex; + ext4_mb_use_best_found(ac, e4b); + return; + } + + /* + * Let's check whether the chuck is good enough + */ + if (ex->fe_len == gex->fe_len) { + *bex = *ex; + ext4_mb_use_best_found(ac, e4b); + return; + } + + /* + * If this is first found extent, just store it in the context + */ + if (bex->fe_len == 0) { + *bex = *ex; + return; + } + + /* + * If new found extent is better, store it in the context + */ + if (bex->fe_len < gex->fe_len) { + /* if the request isn't satisfied, any found extent + * larger than previous best one is better */ + if (ex->fe_len > bex->fe_len) + *bex = *ex; + } else if (ex->fe_len > gex->fe_len) { + /* if the request is satisfied, then we try to find + * an extent that still satisfy the request, but is + * smaller than previous one */ + if (ex->fe_len < bex->fe_len) + *bex = *ex; + } + + ext4_mb_check_limits(ac, e4b, 0); +} + +static int ext4_mb_try_best_found(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b) +{ + struct ext4_free_extent ex = ac->ac_b_ex; + ext4_group_t group = ex.fe_group; + int max; + int err; + + BUG_ON(ex.fe_len <= 0); + err = ext4_mb_load_buddy(ac->ac_sb, group, e4b); + if (err) + return err; + + ext4_lock_group(ac->ac_sb, group); + max = mb_find_extent(e4b, 0, ex.fe_start, ex.fe_len, &ex); + + if (max > 0) { + ac->ac_b_ex = ex; + ext4_mb_use_best_found(ac, e4b); + } + + ext4_unlock_group(ac->ac_sb, group); + ext4_mb_release_desc(e4b); + + return 0; +} + +static int ext4_mb_find_by_goal(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b) +{ + ext4_group_t group = ac->ac_g_ex.fe_group; + int max; + int err; + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + struct ext4_super_block *es = sbi->s_es; + struct ext4_free_extent ex; + + if (!(ac->ac_flags & EXT4_MB_HINT_TRY_GOAL)) + return 0; + + err = ext4_mb_load_buddy(ac->ac_sb, group, e4b); + if (err) + return err; + + ext4_lock_group(ac->ac_sb, group); + max = mb_find_extent(e4b, 0, ac->ac_g_ex.fe_start, + ac->ac_g_ex.fe_len, &ex); + + if (max >= ac->ac_g_ex.fe_len && ac->ac_g_ex.fe_len == sbi->s_stripe) { + ext4_fsblk_t start; + + start = (e4b->bd_group * EXT4_BLOCKS_PER_GROUP(ac->ac_sb)) + + ex.fe_start + le32_to_cpu(es->s_first_data_block); + /* use do_div to get remainder (would be 64-bit modulo) */ + if (do_div(start, sbi->s_stripe) == 0) { + ac->ac_found++; + ac->ac_b_ex = ex; + ext4_mb_use_best_found(ac, e4b); + } + } else if (max >= ac->ac_g_ex.fe_len) { + BUG_ON(ex.fe_len <= 0); + BUG_ON(ex.fe_group != ac->ac_g_ex.fe_group); + BUG_ON(ex.fe_start != ac->ac_g_ex.fe_start); + ac->ac_found++; + ac->ac_b_ex = ex; + ext4_mb_use_best_found(ac, e4b); + } else if (max > 0 && (ac->ac_flags & EXT4_MB_HINT_MERGE)) { + /* Sometimes, caller may want to merge even small + * number of blocks to an existing extent */ + BUG_ON(ex.fe_len <= 0); + BUG_ON(ex.fe_group != ac->ac_g_ex.fe_group); + BUG_ON(ex.fe_start != ac->ac_g_ex.fe_start); + ac->ac_found++; + ac->ac_b_ex = ex; + ext4_mb_use_best_found(ac, e4b); + } + ext4_unlock_group(ac->ac_sb, group); + ext4_mb_release_desc(e4b); + + return 0; +} + +/* + * The routine scans buddy structures (not bitmap!) from given order + * to max order and tries to find big enough chunk to satisfy the req + */ +static void ext4_mb_simple_scan_group(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b) +{ + struct super_block *sb = ac->ac_sb; + struct ext4_group_info *grp = e4b->bd_info; + void *buddy; + int i; + int k; + int max; + + BUG_ON(ac->ac_2order <= 0); + for (i = ac->ac_2order; i <= sb->s_blocksize_bits + 1; i++) { + if (grp->bb_counters[i] == 0) + continue; + + buddy = mb_find_buddy(e4b, i, &max); + BUG_ON(buddy == NULL); + + k = ext4_find_next_zero_bit(buddy, max, 0); + BUG_ON(k >= max); + + ac->ac_found++; + + ac->ac_b_ex.fe_len = 1 << i; + ac->ac_b_ex.fe_start = k << i; + ac->ac_b_ex.fe_group = e4b->bd_group; + + ext4_mb_use_best_found(ac, e4b); + + BUG_ON(ac->ac_b_ex.fe_len != ac->ac_g_ex.fe_len); + + if (EXT4_SB(sb)->s_mb_stats) + atomic_inc(&EXT4_SB(sb)->s_bal_2orders); + + break; + } +} + +/* + * The routine scans the group and measures all found extents. + * In order to optimize scanning, caller must pass number of + * free blocks in the group, so the routine can know upper limit. + */ +static void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b) +{ + struct super_block *sb = ac->ac_sb; + void *bitmap = EXT4_MB_BITMAP(e4b); + struct ext4_free_extent ex; + int i; + int free; + + free = e4b->bd_info->bb_free; + BUG_ON(free <= 0); + + i = e4b->bd_info->bb_first_free; + + while (free && ac->ac_status == AC_STATUS_CONTINUE) { + i = ext4_find_next_zero_bit(bitmap, + EXT4_BLOCKS_PER_GROUP(sb), i); + if (i >= EXT4_BLOCKS_PER_GROUP(sb)) { + BUG_ON(free != 0); + break; + } + + mb_find_extent(e4b, 0, i, ac->ac_g_ex.fe_len, &ex); + BUG_ON(ex.fe_len <= 0); + BUG_ON(free < ex.fe_len); + + ext4_mb_measure_extent(ac, &ex, e4b); + + i += ex.fe_len; + free -= ex.fe_len; + } + + ext4_mb_check_limits(ac, e4b, 1); +} + +/* + * This is a special case for storages like raid5 + * we try to find stripe-aligned chunks for stripe-size requests + * XXX should do so at least for multiples of stripe size as well + */ +static void ext4_mb_scan_aligned(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b) +{ + struct super_block *sb = ac->ac_sb; + struct ext4_sb_info *sbi = EXT4_SB(sb); + void *bitmap = EXT4_MB_BITMAP(e4b); + struct ext4_free_extent ex; + ext4_fsblk_t first_group_block; + ext4_fsblk_t a; + ext4_grpblk_t i; + int max; + + BUG_ON(sbi->s_stripe == 0); + + /* find first stripe-aligned block in group */ + first_group_block = e4b->bd_group * EXT4_BLOCKS_PER_GROUP(sb) + + le32_to_cpu(sbi->s_es->s_first_data_block); + a = first_group_block + sbi->s_stripe - 1; + do_div(a, sbi->s_stripe); + i = (a * sbi->s_stripe) - first_group_block; + + while (i < EXT4_BLOCKS_PER_GROUP(sb)) { + if (!mb_test_bit(i, bitmap)) { + max = mb_find_extent(e4b, 0, i, sbi->s_stripe, &ex); + if (max >= sbi->s_stripe) { + ac->ac_found++; + ac->ac_b_ex = ex; + ext4_mb_use_best_found(ac, e4b); + break; + } + } + i += sbi->s_stripe; + } +} + +static int ext4_mb_good_group(struct ext4_allocation_context *ac, + ext4_group_t group, int cr) +{ + unsigned free, fragments; + unsigned i, bits; + struct ext4_group_desc *desc; + struct ext4_group_info *grp = ext4_get_group_info(ac->ac_sb, group); + + BUG_ON(cr < 0 || cr >= 4); + BUG_ON(EXT4_MB_GRP_NEED_INIT(grp)); + + free = grp->bb_free; + fragments = grp->bb_fragments; + if (free == 0) + return 0; + if (fragments == 0) + return 0; + + switch (cr) { + case 0: + BUG_ON(ac->ac_2order == 0); + /* If this group is uninitialized, skip it initially */ + desc = ext4_get_group_desc(ac->ac_sb, group, NULL); + if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) + return 0; + + bits = ac->ac_sb->s_blocksize_bits + 1; + for (i = ac->ac_2order; i <= bits; i++) + if (grp->bb_counters[i] > 0) + return 1; + break; + case 1: + if ((free / fragments) >= ac->ac_g_ex.fe_len) + return 1; + break; + case 2: + if (free >= ac->ac_g_ex.fe_len) + return 1; + break; + case 3: + return 1; + default: + BUG(); + } + + return 0; +} + +static int ext4_mb_regular_allocator(struct ext4_allocation_context *ac) +{ + ext4_group_t group; + ext4_group_t i; + int cr; + int err = 0; + int bsbits; + struct ext4_sb_info *sbi; + struct super_block *sb; + struct ext4_buddy e4b; + loff_t size, isize; + + sb = ac->ac_sb; + sbi = EXT4_SB(sb); + BUG_ON(ac->ac_status == AC_STATUS_FOUND); + + /* first, try the goal */ + err = ext4_mb_find_by_goal(ac, &e4b); + if (err || ac->ac_status == AC_STATUS_FOUND) + goto out; + + if (unlikely(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY)) + goto out; + + /* + * ac->ac2_order is set only if the fe_len is a power of 2 + * if ac2_order is set we also set criteria to 0 so that we + * try exact allocation using buddy. + */ + i = fls(ac->ac_g_ex.fe_len); + ac->ac_2order = 0; + /* + * We search using buddy data only if the order of the request + * is greater than equal to the sbi_s_mb_order2_reqs + * You can tune it via /proc/fs/ext4//order2_req + */ + if (i >= sbi->s_mb_order2_reqs) { + /* + * This should tell if fe_len is exactly power of 2 + */ + if ((ac->ac_g_ex.fe_len & (~(1 << (i - 1)))) == 0) + ac->ac_2order = i - 1; + } + + bsbits = ac->ac_sb->s_blocksize_bits; + /* if stream allocation is enabled, use global goal */ + size = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len; + isize = i_size_read(ac->ac_inode) >> bsbits; + if (size < isize) + size = isize; + + if (size < sbi->s_mb_stream_request && + (ac->ac_flags & EXT4_MB_HINT_DATA)) { + /* TBD: may be hot point */ + spin_lock(&sbi->s_md_lock); + ac->ac_g_ex.fe_group = sbi->s_mb_last_group; + ac->ac_g_ex.fe_start = sbi->s_mb_last_start; + spin_unlock(&sbi->s_md_lock); + } + + /* searching for the right group start from the goal value specified */ + group = ac->ac_g_ex.fe_group; + + /* Let's just scan groups to find more-less suitable blocks */ + cr = ac->ac_2order ? 0 : 1; + /* + * cr == 0 try to get exact allocation, + * cr == 3 try to get anything + */ +repeat: + for (; cr < 4 && ac->ac_status == AC_STATUS_CONTINUE; cr++) { + ac->ac_criteria = cr; + for (i = 0; i < EXT4_SB(sb)->s_groups_count; group++, i++) { + struct ext4_group_info *grp; + struct ext4_group_desc *desc; + + if (group == EXT4_SB(sb)->s_groups_count) + group = 0; + + /* quick check to skip empty groups */ + grp = ext4_get_group_info(ac->ac_sb, group); + if (grp->bb_free == 0) + continue; + + /* + * if the group is already init we check whether it is + * a good group and if not we don't load the buddy + */ + if (EXT4_MB_GRP_NEED_INIT(grp)) { + /* + * we need full data about the group + * to make a good selection + */ + err = ext4_mb_load_buddy(sb, group, &e4b); + if (err) + goto out; + ext4_mb_release_desc(&e4b); + } + + /* + * If the particular group doesn't satisfy our + * criteria we continue with the next group + */ + if (!ext4_mb_good_group(ac, group, cr)) + continue; + + err = ext4_mb_load_buddy(sb, group, &e4b); + if (err) + goto out; + + ext4_lock_group(sb, group); + if (!ext4_mb_good_group(ac, group, cr)) { + /* someone did allocation from this group */ + ext4_unlock_group(sb, group); + ext4_mb_release_desc(&e4b); + continue; + } + + ac->ac_groups_scanned++; + desc = ext4_get_group_desc(sb, group, NULL); + if (cr == 0 || (desc->bg_flags & + cpu_to_le16(EXT4_BG_BLOCK_UNINIT) && + ac->ac_2order != 0)) + ext4_mb_simple_scan_group(ac, &e4b); + else if (cr == 1 && + ac->ac_g_ex.fe_len == sbi->s_stripe) + ext4_mb_scan_aligned(ac, &e4b); + else + ext4_mb_complex_scan_group(ac, &e4b); + + ext4_unlock_group(sb, group); + ext4_mb_release_desc(&e4b); + + if (ac->ac_status != AC_STATUS_CONTINUE) + break; + } + } + + if (ac->ac_b_ex.fe_len > 0 && ac->ac_status != AC_STATUS_FOUND && + !(ac->ac_flags & EXT4_MB_HINT_FIRST)) { + /* + * We've been searching too long. Let's try to allocate + * the best chunk we've found so far + */ + + ext4_mb_try_best_found(ac, &e4b); + if (ac->ac_status != AC_STATUS_FOUND) { + /* + * Someone more lucky has already allocated it. + * The only thing we can do is just take first + * found block(s) + printk(KERN_DEBUG "EXT4-fs: someone won our chunk\n"); + */ + ac->ac_b_ex.fe_group = 0; + ac->ac_b_ex.fe_start = 0; + ac->ac_b_ex.fe_len = 0; + ac->ac_status = AC_STATUS_CONTINUE; + ac->ac_flags |= EXT4_MB_HINT_FIRST; + cr = 3; + atomic_inc(&sbi->s_mb_lost_chunks); + goto repeat; + } + } +out: + return err; +} + +#ifdef EXT4_MB_HISTORY +struct ext4_mb_proc_session { + struct ext4_mb_history *history; + struct super_block *sb; + int start; + int max; +}; + +static void *ext4_mb_history_skip_empty(struct ext4_mb_proc_session *s, + struct ext4_mb_history *hs, + int first) +{ + if (hs == s->history + s->max) + hs = s->history; + if (!first && hs == s->history + s->start) + return NULL; + while (hs->orig.fe_len == 0) { + hs++; + if (hs == s->history + s->max) + hs = s->history; + if (hs == s->history + s->start) + return NULL; + } + return hs; +} + +static void *ext4_mb_seq_history_start(struct seq_file *seq, loff_t *pos) +{ + struct ext4_mb_proc_session *s = seq->private; + struct ext4_mb_history *hs; + int l = *pos; + + if (l == 0) + return SEQ_START_TOKEN; + hs = ext4_mb_history_skip_empty(s, s->history + s->start, 1); + if (!hs) + return NULL; + while (--l && (hs = ext4_mb_history_skip_empty(s, ++hs, 0)) != NULL); + return hs; +} + +static void *ext4_mb_seq_history_next(struct seq_file *seq, void *v, + loff_t *pos) +{ + struct ext4_mb_proc_session *s = seq->private; + struct ext4_mb_history *hs = v; + + ++*pos; + if (v == SEQ_START_TOKEN) + return ext4_mb_history_skip_empty(s, s->history + s->start, 1); + else + return ext4_mb_history_skip_empty(s, ++hs, 0); +} + +static int ext4_mb_seq_history_show(struct seq_file *seq, void *v) +{ + char buf[25], buf2[25], buf3[25], *fmt; + struct ext4_mb_history *hs = v; + + if (v == SEQ_START_TOKEN) { + seq_printf(seq, "%-5s %-8s %-23s %-23s %-23s %-5s " + "%-5s %-2s %-5s %-5s %-5s %-6s\n", + "pid", "inode", "original", "goal", "result", "found", + "grps", "cr", "flags", "merge", "tail", "broken"); + return 0; + } + + if (hs->op == EXT4_MB_HISTORY_ALLOC) { + fmt = "%-5u %-8u %-23s %-23s %-23s %-5u %-5u %-2u " + "%-5u %-5s %-5u %-6u\n"; + sprintf(buf2, "%lu/%d/%u@%u", hs->result.fe_group, + hs->result.fe_start, hs->result.fe_len, + hs->result.fe_logical); + sprintf(buf, "%lu/%d/%u@%u", hs->orig.fe_group, + hs->orig.fe_start, hs->orig.fe_len, + hs->orig.fe_logical); + sprintf(buf3, "%lu/%d/%u@%u", hs->goal.fe_group, + hs->goal.fe_start, hs->goal.fe_len, + hs->goal.fe_logical); + seq_printf(seq, fmt, hs->pid, hs->ino, buf, buf3, buf2, + hs->found, hs->groups, hs->cr, hs->flags, + hs->merged ? "M" : "", hs->tail, + hs->buddy ? 1 << hs->buddy : 0); + } else if (hs->op == EXT4_MB_HISTORY_PREALLOC) { + fmt = "%-5u %-8u %-23s %-23s %-23s\n"; + sprintf(buf2, "%lu/%d/%u@%u", hs->result.fe_group, + hs->result.fe_start, hs->result.fe_len, + hs->result.fe_logical); + sprintf(buf, "%lu/%d/%u@%u", hs->orig.fe_group, + hs->orig.fe_start, hs->orig.fe_len, + hs->orig.fe_logical); + seq_printf(seq, fmt, hs->pid, hs->ino, buf, "", buf2); + } else if (hs->op == EXT4_MB_HISTORY_DISCARD) { + sprintf(buf2, "%lu/%d/%u", hs->result.fe_group, + hs->result.fe_start, hs->result.fe_len); + seq_printf(seq, "%-5u %-8u %-23s discard\n", + hs->pid, hs->ino, buf2); + } else if (hs->op == EXT4_MB_HISTORY_FREE) { + sprintf(buf2, "%lu/%d/%u", hs->result.fe_group, + hs->result.fe_start, hs->result.fe_len); + seq_printf(seq, "%-5u %-8u %-23s free\n", + hs->pid, hs->ino, buf2); + } + return 0; +} + +static void ext4_mb_seq_history_stop(struct seq_file *seq, void *v) +{ +} + +static struct seq_operations ext4_mb_seq_history_ops = { + .start = ext4_mb_seq_history_start, + .next = ext4_mb_seq_history_next, + .stop = ext4_mb_seq_history_stop, + .show = ext4_mb_seq_history_show, +}; + +static int ext4_mb_seq_history_open(struct inode *inode, struct file *file) +{ + struct super_block *sb = PDE(inode)->data; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_mb_proc_session *s; + int rc; + int size; + + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (s == NULL) + return -ENOMEM; + s->sb = sb; + size = sizeof(struct ext4_mb_history) * sbi->s_mb_history_max; + s->history = kmalloc(size, GFP_KERNEL); + if (s->history == NULL) { + kfree(s); + return -ENOMEM; + } + + spin_lock(&sbi->s_mb_history_lock); + memcpy(s->history, sbi->s_mb_history, size); + s->max = sbi->s_mb_history_max; + s->start = sbi->s_mb_history_cur % s->max; + spin_unlock(&sbi->s_mb_history_lock); + + rc = seq_open(file, &ext4_mb_seq_history_ops); + if (rc == 0) { + struct seq_file *m = (struct seq_file *)file->private_data; + m->private = s; + } else { + kfree(s->history); + kfree(s); + } + return rc; + +} + +static int ext4_mb_seq_history_release(struct inode *inode, struct file *file) +{ + struct seq_file *seq = (struct seq_file *)file->private_data; + struct ext4_mb_proc_session *s = seq->private; + kfree(s->history); + kfree(s); + return seq_release(inode, file); +} + +static ssize_t ext4_mb_seq_history_write(struct file *file, + const char __user *buffer, + size_t count, loff_t *ppos) +{ + struct seq_file *seq = (struct seq_file *)file->private_data; + struct ext4_mb_proc_session *s = seq->private; + struct super_block *sb = s->sb; + char str[32]; + int value; + + if (count >= sizeof(str)) { + printk(KERN_ERR "EXT4-fs: %s string too long, max %u bytes\n", + "mb_history", (int)sizeof(str)); + return -EOVERFLOW; + } + + if (copy_from_user(str, buffer, count)) + return -EFAULT; + + value = simple_strtol(str, NULL, 0); + if (value < 0) + return -ERANGE; + EXT4_SB(sb)->s_mb_history_filter = value; + + return count; +} + +static struct file_operations ext4_mb_seq_history_fops = { + .owner = THIS_MODULE, + .open = ext4_mb_seq_history_open, + .read = seq_read, + .write = ext4_mb_seq_history_write, + .llseek = seq_lseek, + .release = ext4_mb_seq_history_release, +}; + +static void *ext4_mb_seq_groups_start(struct seq_file *seq, loff_t *pos) +{ + struct super_block *sb = seq->private; + struct ext4_sb_info *sbi = EXT4_SB(sb); + ext4_group_t group; + + if (*pos < 0 || *pos >= sbi->s_groups_count) + return NULL; + + group = *pos + 1; + return (void *) group; +} + +static void *ext4_mb_seq_groups_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct super_block *sb = seq->private; + struct ext4_sb_info *sbi = EXT4_SB(sb); + ext4_group_t group; + + ++*pos; + if (*pos < 0 || *pos >= sbi->s_groups_count) + return NULL; + group = *pos + 1; + return (void *) group;; +} + +static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v) +{ + struct super_block *sb = seq->private; + long group = (long) v; + int i; + int err; + struct ext4_buddy e4b; + struct sg { + struct ext4_group_info info; + unsigned short counters[16]; + } sg; + + group--; + if (group == 0) + seq_printf(seq, "#%-5s: %-5s %-5s %-5s " + "[ %-5s %-5s %-5s %-5s %-5s %-5s %-5s " + "%-5s %-5s %-5s %-5s %-5s %-5s %-5s ]\n", + "group", "free", "frags", "first", + "2^0", "2^1", "2^2", "2^3", "2^4", "2^5", "2^6", + "2^7", "2^8", "2^9", "2^10", "2^11", "2^12", "2^13"); + + i = (sb->s_blocksize_bits + 2) * sizeof(sg.info.bb_counters[0]) + + sizeof(struct ext4_group_info); + err = ext4_mb_load_buddy(sb, group, &e4b); + if (err) { + seq_printf(seq, "#%-5lu: I/O error\n", group); + return 0; + } + ext4_lock_group(sb, group); + memcpy(&sg, ext4_get_group_info(sb, group), i); + ext4_unlock_group(sb, group); + ext4_mb_release_desc(&e4b); + + seq_printf(seq, "#%-5lu: %-5u %-5u %-5u [", group, sg.info.bb_free, + sg.info.bb_fragments, sg.info.bb_first_free); + for (i = 0; i <= 13; i++) + seq_printf(seq, " %-5u", i <= sb->s_blocksize_bits + 1 ? + sg.info.bb_counters[i] : 0); + seq_printf(seq, " ]\n"); + + return 0; +} + +static void ext4_mb_seq_groups_stop(struct seq_file *seq, void *v) +{ +} + +static struct seq_operations ext4_mb_seq_groups_ops = { + .start = ext4_mb_seq_groups_start, + .next = ext4_mb_seq_groups_next, + .stop = ext4_mb_seq_groups_stop, + .show = ext4_mb_seq_groups_show, +}; + +static int ext4_mb_seq_groups_open(struct inode *inode, struct file *file) +{ + struct super_block *sb = PDE(inode)->data; + int rc; + + rc = seq_open(file, &ext4_mb_seq_groups_ops); + if (rc == 0) { + struct seq_file *m = (struct seq_file *)file->private_data; + m->private = sb; + } + return rc; + +} + +static struct file_operations ext4_mb_seq_groups_fops = { + .owner = THIS_MODULE, + .open = ext4_mb_seq_groups_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static void ext4_mb_history_release(struct super_block *sb) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + + remove_proc_entry("mb_groups", sbi->s_mb_proc); + remove_proc_entry("mb_history", sbi->s_mb_proc); + + kfree(sbi->s_mb_history); +} + +static void ext4_mb_history_init(struct super_block *sb) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + int i; + + if (sbi->s_mb_proc != NULL) { + struct proc_dir_entry *p; + p = create_proc_entry("mb_history", S_IRUGO, sbi->s_mb_proc); + if (p) { + p->proc_fops = &ext4_mb_seq_history_fops; + p->data = sb; + } + p = create_proc_entry("mb_groups", S_IRUGO, sbi->s_mb_proc); + if (p) { + p->proc_fops = &ext4_mb_seq_groups_fops; + p->data = sb; + } + } + + sbi->s_mb_history_max = 1000; + sbi->s_mb_history_cur = 0; + spin_lock_init(&sbi->s_mb_history_lock); + i = sbi->s_mb_history_max * sizeof(struct ext4_mb_history); + sbi->s_mb_history = kmalloc(i, GFP_KERNEL); + if (likely(sbi->s_mb_history != NULL)) + memset(sbi->s_mb_history, 0, i); + /* if we can't allocate history, then we simple won't use it */ +} + +static void ext4_mb_store_history(struct ext4_allocation_context *ac) +{ + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + struct ext4_mb_history h; + + if (unlikely(sbi->s_mb_history == NULL)) + return; + + if (!(ac->ac_op & sbi->s_mb_history_filter)) + return; + + h.op = ac->ac_op; + h.pid = current->pid; + h.ino = ac->ac_inode ? ac->ac_inode->i_ino : 0; + h.orig = ac->ac_o_ex; + h.result = ac->ac_b_ex; + h.flags = ac->ac_flags; + h.found = ac->ac_found; + h.groups = ac->ac_groups_scanned; + h.cr = ac->ac_criteria; + h.tail = ac->ac_tail; + h.buddy = ac->ac_buddy; + h.merged = 0; + if (ac->ac_op == EXT4_MB_HISTORY_ALLOC) { + if (ac->ac_g_ex.fe_start == ac->ac_b_ex.fe_start && + ac->ac_g_ex.fe_group == ac->ac_b_ex.fe_group) + h.merged = 1; + h.goal = ac->ac_g_ex; + h.result = ac->ac_f_ex; + } + + spin_lock(&sbi->s_mb_history_lock); + memcpy(sbi->s_mb_history + sbi->s_mb_history_cur, &h, sizeof(h)); + if (++sbi->s_mb_history_cur >= sbi->s_mb_history_max) + sbi->s_mb_history_cur = 0; + spin_unlock(&sbi->s_mb_history_lock); +} + +#else +#define ext4_mb_history_release(sb) +#define ext4_mb_history_init(sb) +#endif + +static int ext4_mb_init_backend(struct super_block *sb) +{ + ext4_group_t i; + int j, len, metalen; + struct ext4_sb_info *sbi = EXT4_SB(sb); + int num_meta_group_infos = + (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) >> + EXT4_DESC_PER_BLOCK_BITS(sb); + struct ext4_group_info **meta_group_info; + + /* An 8TB filesystem with 64-bit pointers requires a 4096 byte + * kmalloc. A 128kb malloc should suffice for a 256TB filesystem. + * So a two level scheme suffices for now. */ + sbi->s_group_info = kmalloc(sizeof(*sbi->s_group_info) * + num_meta_group_infos, GFP_KERNEL); + if (sbi->s_group_info == NULL) { + printk(KERN_ERR "EXT4-fs: can't allocate buddy meta group\n"); + return -ENOMEM; + } + sbi->s_buddy_cache = new_inode(sb); + if (sbi->s_buddy_cache == NULL) { + printk(KERN_ERR "EXT4-fs: can't get new inode\n"); + goto err_freesgi; + } + EXT4_I(sbi->s_buddy_cache)->i_disksize = 0; + + metalen = sizeof(*meta_group_info) << EXT4_DESC_PER_BLOCK_BITS(sb); + for (i = 0; i < num_meta_group_infos; i++) { + if ((i + 1) == num_meta_group_infos) + metalen = sizeof(*meta_group_info) * + (sbi->s_groups_count - + (i << EXT4_DESC_PER_BLOCK_BITS(sb))); + meta_group_info = kmalloc(metalen, GFP_KERNEL); + if (meta_group_info == NULL) { + printk(KERN_ERR "EXT4-fs: can't allocate mem for a " + "buddy group\n"); + goto err_freemeta; + } + sbi->s_group_info[i] = meta_group_info; + } + + /* + * calculate needed size. if change bb_counters size, + * don't forget about ext4_mb_generate_buddy() + */ + len = sizeof(struct ext4_group_info); + len += sizeof(unsigned short) * (sb->s_blocksize_bits + 2); + for (i = 0; i < sbi->s_groups_count; i++) { + struct ext4_group_desc *desc; + + meta_group_info = + sbi->s_group_info[i >> EXT4_DESC_PER_BLOCK_BITS(sb)]; + j = i & (EXT4_DESC_PER_BLOCK(sb) - 1); + + meta_group_info[j] = kzalloc(len, GFP_KERNEL); + if (meta_group_info[j] == NULL) { + printk(KERN_ERR "EXT4-fs: can't allocate buddy mem\n"); + i--; + goto err_freebuddy; + } + desc = ext4_get_group_desc(sb, i, NULL); + if (desc == NULL) { + printk(KERN_ERR + "EXT4-fs: can't read descriptor %lu\n", i); + goto err_freebuddy; + } + memset(meta_group_info[j], 0, len); + set_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, + &(meta_group_info[j]->bb_state)); + + /* + * initialize bb_free to be able to skip + * empty groups without initialization + */ + if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { + meta_group_info[j]->bb_free = + ext4_free_blocks_after_init(sb, i, desc); + } else { + meta_group_info[j]->bb_free = + le16_to_cpu(desc->bg_free_blocks_count); + } + + INIT_LIST_HEAD(&meta_group_info[j]->bb_prealloc_list); + +#ifdef DOUBLE_CHECK + { + struct buffer_head *bh; + meta_group_info[j]->bb_bitmap = + kmalloc(sb->s_blocksize, GFP_KERNEL); + BUG_ON(meta_group_info[j]->bb_bitmap == NULL); + bh = read_block_bitmap(sb, i); + BUG_ON(bh == NULL); + memcpy(meta_group_info[j]->bb_bitmap, bh->b_data, + sb->s_blocksize); + put_bh(bh); + } +#endif + + } + + return 0; + +err_freebuddy: + while (i >= 0) { + kfree(ext4_get_group_info(sb, i)); + i--; + } + i = num_meta_group_infos; +err_freemeta: + while (--i >= 0) + kfree(sbi->s_group_info[i]); + iput(sbi->s_buddy_cache); +err_freesgi: + kfree(sbi->s_group_info); + return -ENOMEM; +} + +int ext4_mb_init(struct super_block *sb, int needs_recovery) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + unsigned i; + unsigned offset; + unsigned max; + + if (!test_opt(sb, MBALLOC)) + return 0; + + i = (sb->s_blocksize_bits + 2) * sizeof(unsigned short); + + sbi->s_mb_offsets = kmalloc(i, GFP_KERNEL); + if (sbi->s_mb_offsets == NULL) { + clear_opt(sbi->s_mount_opt, MBALLOC); + return -ENOMEM; + } + sbi->s_mb_maxs = kmalloc(i, GFP_KERNEL); + if (sbi->s_mb_maxs == NULL) { + clear_opt(sbi->s_mount_opt, MBALLOC); + kfree(sbi->s_mb_maxs); + return -ENOMEM; + } + + /* order 0 is regular bitmap */ + sbi->s_mb_maxs[0] = sb->s_blocksize << 3; + sbi->s_mb_offsets[0] = 0; + + i = 1; + offset = 0; + max = sb->s_blocksize << 2; + do { + sbi->s_mb_offsets[i] = offset; + sbi->s_mb_maxs[i] = max; + offset += 1 << (sb->s_blocksize_bits - i); + max = max >> 1; + i++; + } while (i <= sb->s_blocksize_bits + 1); + + /* init file for buddy data */ + i = ext4_mb_init_backend(sb); + if (i) { + clear_opt(sbi->s_mount_opt, MBALLOC); + kfree(sbi->s_mb_offsets); + kfree(sbi->s_mb_maxs); + return i; + } + + spin_lock_init(&sbi->s_md_lock); + INIT_LIST_HEAD(&sbi->s_active_transaction); + INIT_LIST_HEAD(&sbi->s_closed_transaction); + INIT_LIST_HEAD(&sbi->s_committed_transaction); + spin_lock_init(&sbi->s_bal_lock); + + sbi->s_mb_max_to_scan = MB_DEFAULT_MAX_TO_SCAN; + sbi->s_mb_min_to_scan = MB_DEFAULT_MIN_TO_SCAN; + sbi->s_mb_stats = MB_DEFAULT_STATS; + sbi->s_mb_stream_request = MB_DEFAULT_STREAM_THRESHOLD; + sbi->s_mb_order2_reqs = MB_DEFAULT_ORDER2_REQS; + sbi->s_mb_history_filter = EXT4_MB_HISTORY_DEFAULT; + sbi->s_mb_group_prealloc = MB_DEFAULT_GROUP_PREALLOC; + + i = sizeof(struct ext4_locality_group) * NR_CPUS; + sbi->s_locality_groups = kmalloc(i, GFP_KERNEL); + if (sbi->s_locality_groups == NULL) { + clear_opt(sbi->s_mount_opt, MBALLOC); + kfree(sbi->s_mb_offsets); + kfree(sbi->s_mb_maxs); + return -ENOMEM; + } + for (i = 0; i < NR_CPUS; i++) { + struct ext4_locality_group *lg; + lg = &sbi->s_locality_groups[i]; + mutex_init(&lg->lg_mutex); + INIT_LIST_HEAD(&lg->lg_prealloc_list); + spin_lock_init(&lg->lg_prealloc_lock); + } + + ext4_mb_init_per_dev_proc(sb); + ext4_mb_history_init(sb); + + printk("EXT4-fs: mballoc enabled\n"); + return 0; +} + +/* need to called with ext4 group lock (ext4_lock_group) */ +static void ext4_mb_cleanup_pa(struct ext4_group_info *grp) +{ + struct ext4_prealloc_space *pa; + struct list_head *cur, *tmp; + int count = 0; + + list_for_each_safe(cur, tmp, &grp->bb_prealloc_list) { + pa = list_entry(cur, struct ext4_prealloc_space, pa_group_list); + list_del(&pa->pa_group_list); + count++; + kfree(pa); + } + if (count) + mb_debug("mballoc: %u PAs left\n", count); + +} + +int ext4_mb_release(struct super_block *sb) +{ + ext4_group_t i; + int num_meta_group_infos; + struct ext4_group_info *grinfo; + struct ext4_sb_info *sbi = EXT4_SB(sb); + + if (!test_opt(sb, MBALLOC)) + return 0; + + /* release freed, non-committed blocks */ + spin_lock(&sbi->s_md_lock); + list_splice_init(&sbi->s_closed_transaction, + &sbi->s_committed_transaction); + list_splice_init(&sbi->s_active_transaction, + &sbi->s_committed_transaction); + spin_unlock(&sbi->s_md_lock); + ext4_mb_free_committed_blocks(sb); + + if (sbi->s_group_info) { + for (i = 0; i < sbi->s_groups_count; i++) { + grinfo = ext4_get_group_info(sb, i); +#ifdef DOUBLE_CHECK + kfree(grinfo->bb_bitmap); +#endif + ext4_lock_group(sb, i); + ext4_mb_cleanup_pa(grinfo); + ext4_unlock_group(sb, i); + kfree(grinfo); + } + num_meta_group_infos = (sbi->s_groups_count + + EXT4_DESC_PER_BLOCK(sb) - 1) >> + EXT4_DESC_PER_BLOCK_BITS(sb); + for (i = 0; i < num_meta_group_infos; i++) + kfree(sbi->s_group_info[i]); + kfree(sbi->s_group_info); + } + kfree(sbi->s_mb_offsets); + kfree(sbi->s_mb_maxs); + if (sbi->s_buddy_cache) + iput(sbi->s_buddy_cache); + if (sbi->s_mb_stats) { + printk(KERN_INFO + "EXT4-fs: mballoc: %u blocks %u reqs (%u success)\n", + atomic_read(&sbi->s_bal_allocated), + atomic_read(&sbi->s_bal_reqs), + atomic_read(&sbi->s_bal_success)); + printk(KERN_INFO + "EXT4-fs: mballoc: %u extents scanned, %u goal hits, " + "%u 2^N hits, %u breaks, %u lost\n", + atomic_read(&sbi->s_bal_ex_scanned), + atomic_read(&sbi->s_bal_goals), + atomic_read(&sbi->s_bal_2orders), + atomic_read(&sbi->s_bal_breaks), + atomic_read(&sbi->s_mb_lost_chunks)); + printk(KERN_INFO + "EXT4-fs: mballoc: %lu generated and it took %Lu\n", + sbi->s_mb_buddies_generated++, + sbi->s_mb_generation_time); + printk(KERN_INFO + "EXT4-fs: mballoc: %u preallocated, %u discarded\n", + atomic_read(&sbi->s_mb_preallocated), + atomic_read(&sbi->s_mb_discarded)); + } + + kfree(sbi->s_locality_groups); + + ext4_mb_history_release(sb); + ext4_mb_destroy_per_dev_proc(sb); + + return 0; +} + +static void ext4_mb_free_committed_blocks(struct super_block *sb) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + int err; + int i; + int count = 0; + int count2 = 0; + struct ext4_free_metadata *md; + struct ext4_buddy e4b; + + if (list_empty(&sbi->s_committed_transaction)) + return; + + /* there is committed blocks to be freed yet */ + do { + /* get next array of blocks */ + md = NULL; + spin_lock(&sbi->s_md_lock); + if (!list_empty(&sbi->s_committed_transaction)) { + md = list_entry(sbi->s_committed_transaction.next, + struct ext4_free_metadata, list); + list_del(&md->list); + } + spin_unlock(&sbi->s_md_lock); + + if (md == NULL) + break; + + mb_debug("gonna free %u blocks in group %lu (0x%p):", + md->num, md->group, md); + + err = ext4_mb_load_buddy(sb, md->group, &e4b); + /* we expect to find existing buddy because it's pinned */ + BUG_ON(err != 0); + + /* there are blocks to put in buddy to make them really free */ + count += md->num; + count2++; + ext4_lock_group(sb, md->group); + for (i = 0; i < md->num; i++) { + mb_debug(" %u", md->blocks[i]); + err = mb_free_blocks(NULL, &e4b, md->blocks[i], 1); + BUG_ON(err != 0); + } + mb_debug("\n"); + ext4_unlock_group(sb, md->group); + + /* balance refcounts from ext4_mb_free_metadata() */ + page_cache_release(e4b.bd_buddy_page); + page_cache_release(e4b.bd_bitmap_page); + + kfree(md); + ext4_mb_release_desc(&e4b); + + } while (md); + + mb_debug("freed %u blocks in %u structures\n", count, count2); +} + +#define EXT4_ROOT "ext4" +#define EXT4_MB_STATS_NAME "stats" +#define EXT4_MB_MAX_TO_SCAN_NAME "max_to_scan" +#define EXT4_MB_MIN_TO_SCAN_NAME "min_to_scan" +#define EXT4_MB_ORDER2_REQ "order2_req" +#define EXT4_MB_STREAM_REQ "stream_req" +#define EXT4_MB_GROUP_PREALLOC "group_prealloc" + + + +#define MB_PROC_VALUE_READ(name) \ +static int ext4_mb_read_##name(char *page, char **start, \ + off_t off, int count, int *eof, void *data) \ +{ \ + struct ext4_sb_info *sbi = data; \ + int len; \ + *eof = 1; \ + if (off != 0) \ + return 0; \ + len = sprintf(page, "%ld\n", sbi->s_mb_##name); \ + *start = page; \ + return len; \ +} + +#define MB_PROC_VALUE_WRITE(name) \ +static int ext4_mb_write_##name(struct file *file, \ + const char __user *buf, unsigned long cnt, void *data) \ +{ \ + struct ext4_sb_info *sbi = data; \ + char str[32]; \ + long value; \ + if (cnt >= sizeof(str)) \ + return -EINVAL; \ + if (copy_from_user(str, buf, cnt)) \ + return -EFAULT; \ + value = simple_strtol(str, NULL, 0); \ + if (value <= 0) \ + return -ERANGE; \ + sbi->s_mb_##name = value; \ + return cnt; \ +} + +MB_PROC_VALUE_READ(stats); +MB_PROC_VALUE_WRITE(stats); +MB_PROC_VALUE_READ(max_to_scan); +MB_PROC_VALUE_WRITE(max_to_scan); +MB_PROC_VALUE_READ(min_to_scan); +MB_PROC_VALUE_WRITE(min_to_scan); +MB_PROC_VALUE_READ(order2_reqs); +MB_PROC_VALUE_WRITE(order2_reqs); +MB_PROC_VALUE_READ(stream_request); +MB_PROC_VALUE_WRITE(stream_request); +MB_PROC_VALUE_READ(group_prealloc); +MB_PROC_VALUE_WRITE(group_prealloc); + +#define MB_PROC_HANDLER(name, var) \ +do { \ + proc = create_proc_entry(name, mode, sbi->s_mb_proc); \ + if (proc == NULL) { \ + printk(KERN_ERR "EXT4-fs: can't to create %s\n", name); \ + goto err_out; \ + } \ + proc->data = sbi; \ + proc->read_proc = ext4_mb_read_##var ; \ + proc->write_proc = ext4_mb_write_##var; \ +} while (0) + +static int ext4_mb_init_per_dev_proc(struct super_block *sb) +{ + mode_t mode = S_IFREG | S_IRUGO | S_IWUSR; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct proc_dir_entry *proc; + char devname[64]; + + snprintf(devname, sizeof(devname) - 1, "%s", + bdevname(sb->s_bdev, devname)); + sbi->s_mb_proc = proc_mkdir(devname, proc_root_ext4); + + MB_PROC_HANDLER(EXT4_MB_STATS_NAME, stats); + MB_PROC_HANDLER(EXT4_MB_MAX_TO_SCAN_NAME, max_to_scan); + MB_PROC_HANDLER(EXT4_MB_MIN_TO_SCAN_NAME, min_to_scan); + MB_PROC_HANDLER(EXT4_MB_ORDER2_REQ, order2_reqs); + MB_PROC_HANDLER(EXT4_MB_STREAM_REQ, stream_request); + MB_PROC_HANDLER(EXT4_MB_GROUP_PREALLOC, group_prealloc); + + return 0; + +err_out: + printk(KERN_ERR "EXT4-fs: Unable to create %s\n", devname); + remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_mb_proc); + remove_proc_entry(devname, proc_root_ext4); + sbi->s_mb_proc = NULL; + + return -ENOMEM; +} + +static int ext4_mb_destroy_per_dev_proc(struct super_block *sb) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + char devname[64]; + + if (sbi->s_mb_proc == NULL) + return -EINVAL; + + snprintf(devname, sizeof(devname) - 1, "%s", + bdevname(sb->s_bdev, devname)); + remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_mb_proc); + remove_proc_entry(devname, proc_root_ext4); + + return 0; +} + +int __init init_ext4_mballoc(void) +{ + ext4_pspace_cachep = + kmem_cache_create("ext4_prealloc_space", + sizeof(struct ext4_prealloc_space), + 0, SLAB_RECLAIM_ACCOUNT, NULL); + if (ext4_pspace_cachep == NULL) + return -ENOMEM; + +#ifdef CONFIG_PROC_FS + proc_root_ext4 = proc_mkdir(EXT4_ROOT, proc_root_fs); + if (proc_root_ext4 == NULL) + printk(KERN_ERR "EXT4-fs: Unable to create %s\n", EXT4_ROOT); +#endif + + return 0; +} + +void exit_ext4_mballoc(void) +{ + /* XXX: synchronize_rcu(); */ + kmem_cache_destroy(ext4_pspace_cachep); +#ifdef CONFIG_PROC_FS + remove_proc_entry(EXT4_ROOT, proc_root_fs); +#endif +} + + +/* + * Check quota and mark choosed space (ac->ac_b_ex) non-free in bitmaps + * Returns 0 if success or error code + */ +static int ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, + handle_t *handle) +{ + struct buffer_head *bitmap_bh = NULL; + struct ext4_super_block *es; + struct ext4_group_desc *gdp; + struct buffer_head *gdp_bh; + struct ext4_sb_info *sbi; + struct super_block *sb; + ext4_fsblk_t block; + int err; + + BUG_ON(ac->ac_status != AC_STATUS_FOUND); + BUG_ON(ac->ac_b_ex.fe_len <= 0); + + sb = ac->ac_sb; + sbi = EXT4_SB(sb); + es = sbi->s_es; + + ext4_debug("using block group %lu(%d)\n", ac->ac_b_ex.fe_group, + gdp->bg_free_blocks_count); + + err = -EIO; + bitmap_bh = read_block_bitmap(sb, ac->ac_b_ex.fe_group); + if (!bitmap_bh) + goto out_err; + + err = ext4_journal_get_write_access(handle, bitmap_bh); + if (err) + goto out_err; + + err = -EIO; + gdp = ext4_get_group_desc(sb, ac->ac_b_ex.fe_group, &gdp_bh); + if (!gdp) + goto out_err; + + err = ext4_journal_get_write_access(handle, gdp_bh); + if (err) + goto out_err; + + block = ac->ac_b_ex.fe_group * EXT4_BLOCKS_PER_GROUP(sb) + + ac->ac_b_ex.fe_start + + le32_to_cpu(es->s_first_data_block); + + if (block == ext4_block_bitmap(sb, gdp) || + block == ext4_inode_bitmap(sb, gdp) || + in_range(block, ext4_inode_table(sb, gdp), + EXT4_SB(sb)->s_itb_per_group)) { + + ext4_error(sb, __FUNCTION__, + "Allocating block in system zone - block = %llu", + block); + } +#ifdef AGGRESSIVE_CHECK + { + int i; + for (i = 0; i < ac->ac_b_ex.fe_len; i++) { + BUG_ON(mb_test_bit(ac->ac_b_ex.fe_start + i, + bitmap_bh->b_data)); + } + } +#endif + mb_set_bits(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group), bitmap_bh->b_data, + ac->ac_b_ex.fe_start, ac->ac_b_ex.fe_len); + + spin_lock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group)); + if (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { + gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT); + gdp->bg_free_blocks_count = + cpu_to_le16(ext4_free_blocks_after_init(sb, + ac->ac_b_ex.fe_group, + gdp)); + } + gdp->bg_free_blocks_count = + cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) + - ac->ac_b_ex.fe_len); + gdp->bg_checksum = ext4_group_desc_csum(sbi, ac->ac_b_ex.fe_group, gdp); + spin_unlock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group)); + percpu_counter_sub(&sbi->s_freeblocks_counter, ac->ac_b_ex.fe_len); + + err = ext4_journal_dirty_metadata(handle, bitmap_bh); + if (err) + goto out_err; + err = ext4_journal_dirty_metadata(handle, gdp_bh); + +out_err: + sb->s_dirt = 1; + put_bh(bitmap_bh); + return err; +} + +/* + * here we normalize request for locality group + * Group request are normalized to s_strip size if we set the same via mount + * option. If not we set it to s_mb_group_prealloc which can be configured via + * /proc/fs/ext4//group_prealloc + * + * XXX: should we try to preallocate more than the group has now? + */ +static void ext4_mb_normalize_group_request(struct ext4_allocation_context *ac) +{ + struct super_block *sb = ac->ac_sb; + struct ext4_locality_group *lg = ac->ac_lg; + + BUG_ON(lg == NULL); + if (EXT4_SB(sb)->s_stripe) + ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_stripe; + else + ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_mb_group_prealloc; + mb_debug("#%u: goal %lu blocks for locality group\n", + current->pid, ac->ac_g_ex.fe_len); +} + +/* + * Normalization means making request better in terms of + * size and alignment + */ +static void ext4_mb_normalize_request(struct ext4_allocation_context *ac, + struct ext4_allocation_request *ar) +{ + int bsbits, max; + ext4_lblk_t end; + struct list_head *cur; + loff_t size, orig_size, start_off; + ext4_lblk_t start, orig_start; + struct ext4_inode_info *ei = EXT4_I(ac->ac_inode); + + /* do normalize only data requests, metadata requests + do not need preallocation */ + if (!(ac->ac_flags & EXT4_MB_HINT_DATA)) + return; + + /* sometime caller may want exact blocks */ + if (unlikely(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY)) + return; + + /* caller may indicate that preallocation isn't + * required (it's a tail, for example) */ + if (ac->ac_flags & EXT4_MB_HINT_NOPREALLOC) + return; + + if (ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC) { + ext4_mb_normalize_group_request(ac); + return ; + } + + bsbits = ac->ac_sb->s_blocksize_bits; + + /* first, let's learn actual file size + * given current request is allocated */ + size = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len; + size = size << bsbits; + if (size < i_size_read(ac->ac_inode)) + size = i_size_read(ac->ac_inode); + + /* max available blocks in a free group */ + max = EXT4_BLOCKS_PER_GROUP(ac->ac_sb) - 1 - 1 - + EXT4_SB(ac->ac_sb)->s_itb_per_group; + +#define NRL_CHECK_SIZE(req, size, max,bits) \ + (req <= (size) || max <= ((size) >> bits)) + + /* first, try to predict filesize */ + /* XXX: should this table be tunable? */ + start_off = 0; + if (size <= 16 * 1024) { + size = 16 * 1024; + } else if (size <= 32 * 1024) { + size = 32 * 1024; + } else if (size <= 64 * 1024) { + size = 64 * 1024; + } else if (size <= 128 * 1024) { + size = 128 * 1024; + } else if (size <= 256 * 1024) { + size = 256 * 1024; + } else if (size <= 512 * 1024) { + size = 512 * 1024; + } else if (size <= 1024 * 1024) { + size = 1024 * 1024; + } else if (NRL_CHECK_SIZE(size, 4 * 1024 * 1024, max, bsbits)) { + start_off = ((loff_t)ac->ac_o_ex.fe_logical >> + (20 - bsbits)) << 20; + size = 1024 * 1024; + } else if (NRL_CHECK_SIZE(size, 8 * 1024 * 1024, max, bsbits)) { + start_off = ((loff_t)ac->ac_o_ex.fe_logical >> + (22 - bsbits)) << 22; + size = 4 * 1024 * 1024; + } else if (NRL_CHECK_SIZE(ac->ac_o_ex.fe_len, + (8<<20)>>bsbits, max, bsbits)) { + start_off = ((loff_t)ac->ac_o_ex.fe_logical >> + (23 - bsbits)) << 23; + size = 8 * 1024 * 1024; + } else { + start_off = (loff_t)ac->ac_o_ex.fe_logical << bsbits; + size = ac->ac_o_ex.fe_len << bsbits; + } + orig_size = size = size >> bsbits; + orig_start = start = start_off >> bsbits; + + /* don't cover already allocated blocks in selected range */ + if (ar->pleft && start <= ar->lleft) { + size -= ar->lleft + 1 - start; + start = ar->lleft + 1; + } + if (ar->pright && start + size - 1 >= ar->lright) + size -= start + size - ar->lright; + + end = start + size; + + /* check we don't cross already preallocated blocks */ + rcu_read_lock(); + list_for_each_rcu(cur, &ei->i_prealloc_list) { + struct ext4_prealloc_space *pa; + unsigned long pa_end; + + pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list); + + if (pa->pa_deleted) + continue; + spin_lock(&pa->pa_lock); + if (pa->pa_deleted) { + spin_unlock(&pa->pa_lock); + continue; + } + + pa_end = pa->pa_lstart + pa->pa_len; + + /* PA must not overlap original request */ + BUG_ON(!(ac->ac_o_ex.fe_logical >= pa_end || + ac->ac_o_ex.fe_logical < pa->pa_lstart)); + + /* skip PA normalized request doesn't overlap with */ + if (pa->pa_lstart >= end) { + spin_unlock(&pa->pa_lock); + continue; + } + if (pa_end <= start) { + spin_unlock(&pa->pa_lock); + continue; + } + BUG_ON(pa->pa_lstart <= start && pa_end >= end); + + if (pa_end <= ac->ac_o_ex.fe_logical) { + BUG_ON(pa_end < start); + start = pa_end; + } + + if (pa->pa_lstart > ac->ac_o_ex.fe_logical) { + BUG_ON(pa->pa_lstart > end); + end = pa->pa_lstart; + } + spin_unlock(&pa->pa_lock); + } + rcu_read_unlock(); + size = end - start; + + /* XXX: extra loop to check we really don't overlap preallocations */ + rcu_read_lock(); + list_for_each_rcu(cur, &ei->i_prealloc_list) { + struct ext4_prealloc_space *pa; + unsigned long pa_end; + pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list); + spin_lock(&pa->pa_lock); + if (pa->pa_deleted == 0) { + pa_end = pa->pa_lstart + pa->pa_len; + BUG_ON(!(start >= pa_end || end <= pa->pa_lstart)); + } + spin_unlock(&pa->pa_lock); + } + rcu_read_unlock(); + + if (start + size <= ac->ac_o_ex.fe_logical && + start > ac->ac_o_ex.fe_logical) { + printk(KERN_ERR "start %lu, size %lu, fe_logical %lu\n", + (unsigned long) start, (unsigned long) size, + (unsigned long) ac->ac_o_ex.fe_logical); + } + BUG_ON(start + size <= ac->ac_o_ex.fe_logical && + start > ac->ac_o_ex.fe_logical); + BUG_ON(size <= 0 || size >= EXT4_BLOCKS_PER_GROUP(ac->ac_sb)); + + /* now prepare goal request */ + + /* XXX: is it better to align blocks WRT to logical + * placement or satisfy big request as is */ + ac->ac_g_ex.fe_logical = start; + ac->ac_g_ex.fe_len = size; + + /* define goal start in order to merge */ + if (ar->pright && (ar->lright == (start + size))) { + /* merge to the right */ + ext4_get_group_no_and_offset(ac->ac_sb, ar->pright - size, + &ac->ac_f_ex.fe_group, + &ac->ac_f_ex.fe_start); + ac->ac_flags |= EXT4_MB_HINT_TRY_GOAL; + } + if (ar->pleft && (ar->lleft + 1 == start)) { + /* merge to the left */ + ext4_get_group_no_and_offset(ac->ac_sb, ar->pleft + 1, + &ac->ac_f_ex.fe_group, + &ac->ac_f_ex.fe_start); + ac->ac_flags |= EXT4_MB_HINT_TRY_GOAL; + } + + mb_debug("goal: %u(was %u) blocks at %u\n", (unsigned) size, + (unsigned) orig_size, (unsigned) start); +} + +static void ext4_mb_collect_stats(struct ext4_allocation_context *ac) +{ + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + + if (sbi->s_mb_stats && ac->ac_g_ex.fe_len > 1) { + atomic_inc(&sbi->s_bal_reqs); + atomic_add(ac->ac_b_ex.fe_len, &sbi->s_bal_allocated); + if (ac->ac_o_ex.fe_len >= ac->ac_g_ex.fe_len) + atomic_inc(&sbi->s_bal_success); + atomic_add(ac->ac_found, &sbi->s_bal_ex_scanned); + if (ac->ac_g_ex.fe_start == ac->ac_b_ex.fe_start && + ac->ac_g_ex.fe_group == ac->ac_b_ex.fe_group) + atomic_inc(&sbi->s_bal_goals); + if (ac->ac_found > sbi->s_mb_max_to_scan) + atomic_inc(&sbi->s_bal_breaks); + } + + ext4_mb_store_history(ac); +} + +/* + * use blocks preallocated to inode + */ +static void ext4_mb_use_inode_pa(struct ext4_allocation_context *ac, + struct ext4_prealloc_space *pa) +{ + ext4_fsblk_t start; + ext4_fsblk_t end; + int len; + + /* found preallocated blocks, use them */ + start = pa->pa_pstart + (ac->ac_o_ex.fe_logical - pa->pa_lstart); + end = min(pa->pa_pstart + pa->pa_len, start + ac->ac_o_ex.fe_len); + len = end - start; + ext4_get_group_no_and_offset(ac->ac_sb, start, &ac->ac_b_ex.fe_group, + &ac->ac_b_ex.fe_start); + ac->ac_b_ex.fe_len = len; + ac->ac_status = AC_STATUS_FOUND; + ac->ac_pa = pa; + + BUG_ON(start < pa->pa_pstart); + BUG_ON(start + len > pa->pa_pstart + pa->pa_len); + BUG_ON(pa->pa_free < len); + pa->pa_free -= len; + + mb_debug("use %llu/%lu from inode pa %p\n", start, len, pa); +} + +/* + * use blocks preallocated to locality group + */ +static void ext4_mb_use_group_pa(struct ext4_allocation_context *ac, + struct ext4_prealloc_space *pa) +{ + unsigned len = ac->ac_o_ex.fe_len; + + ext4_get_group_no_and_offset(ac->ac_sb, pa->pa_pstart, + &ac->ac_b_ex.fe_group, + &ac->ac_b_ex.fe_start); + ac->ac_b_ex.fe_len = len; + ac->ac_status = AC_STATUS_FOUND; + ac->ac_pa = pa; + + /* we don't correct pa_pstart or pa_plen here to avoid + * possible race when tte group is being loaded concurrently + * instead we correct pa later, after blocks are marked + * in on-disk bitmap -- see ext4_mb_release_context() */ + /* + * FIXME!! but the other CPUs can look at this particular + * pa and think that it have enought free blocks if we + * don't update pa_free here right ? + */ + mb_debug("use %u/%u from group pa %p\n", pa->pa_lstart-len, len, pa); +} + +/* + * search goal blocks in preallocated space + */ +static int ext4_mb_use_preallocated(struct ext4_allocation_context *ac) +{ + struct ext4_inode_info *ei = EXT4_I(ac->ac_inode); + struct ext4_locality_group *lg; + struct ext4_prealloc_space *pa; + struct list_head *cur; + + /* only data can be preallocated */ + if (!(ac->ac_flags & EXT4_MB_HINT_DATA)) + return 0; + + /* first, try per-file preallocation */ + rcu_read_lock(); + list_for_each_rcu(cur, &ei->i_prealloc_list) { + pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list); + + /* all fields in this condition don't change, + * so we can skip locking for them */ + if (ac->ac_o_ex.fe_logical < pa->pa_lstart || + ac->ac_o_ex.fe_logical >= pa->pa_lstart + pa->pa_len) + continue; + + /* found preallocated blocks, use them */ + spin_lock(&pa->pa_lock); + if (pa->pa_deleted == 0 && pa->pa_free) { + atomic_inc(&pa->pa_count); + ext4_mb_use_inode_pa(ac, pa); + spin_unlock(&pa->pa_lock); + ac->ac_criteria = 10; + rcu_read_unlock(); + return 1; + } + spin_unlock(&pa->pa_lock); + } + rcu_read_unlock(); + + /* can we use group allocation? */ + if (!(ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC)) + return 0; + + /* inode may have no locality group for some reason */ + lg = ac->ac_lg; + if (lg == NULL) + return 0; + + rcu_read_lock(); + list_for_each_rcu(cur, &lg->lg_prealloc_list) { + pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list); + spin_lock(&pa->pa_lock); + if (pa->pa_deleted == 0 && pa->pa_free >= ac->ac_o_ex.fe_len) { + atomic_inc(&pa->pa_count); + ext4_mb_use_group_pa(ac, pa); + spin_unlock(&pa->pa_lock); + ac->ac_criteria = 20; + rcu_read_unlock(); + return 1; + } + spin_unlock(&pa->pa_lock); + } + rcu_read_unlock(); + + return 0; +} + +/* + * the function goes through all preallocation in this group and marks them + * used in in-core bitmap. buddy must be generated from this bitmap + * Need to be called with ext4 group lock (ext4_lock_group) + */ +static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap, + ext4_group_t group) +{ + struct ext4_group_info *grp = ext4_get_group_info(sb, group); + struct ext4_prealloc_space *pa; + struct list_head *cur; + ext4_group_t groupnr; + ext4_grpblk_t start; + int preallocated = 0; + int count = 0; + int len; + + /* all form of preallocation discards first load group, + * so the only competing code is preallocation use. + * we don't need any locking here + * notice we do NOT ignore preallocations with pa_deleted + * otherwise we could leave used blocks available for + * allocation in buddy when concurrent ext4_mb_put_pa() + * is dropping preallocation + */ + list_for_each(cur, &grp->bb_prealloc_list) { + pa = list_entry(cur, struct ext4_prealloc_space, pa_group_list); + spin_lock(&pa->pa_lock); + ext4_get_group_no_and_offset(sb, pa->pa_pstart, + &groupnr, &start); + len = pa->pa_len; + spin_unlock(&pa->pa_lock); + if (unlikely(len == 0)) + continue; + BUG_ON(groupnr != group); + mb_set_bits(sb_bgl_lock(EXT4_SB(sb), group), + bitmap, start, len); + preallocated += len; + count++; + } + mb_debug("prellocated %u for group %lu\n", preallocated, group); +} + +static void ext4_mb_pa_callback(struct rcu_head *head) +{ + struct ext4_prealloc_space *pa; + pa = container_of(head, struct ext4_prealloc_space, u.pa_rcu); + kmem_cache_free(ext4_pspace_cachep, pa); +} + +/* + * drops a reference to preallocated space descriptor + * if this was the last reference and the space is consumed + */ +static void ext4_mb_put_pa(struct ext4_allocation_context *ac, + struct super_block *sb, struct ext4_prealloc_space *pa) +{ + unsigned long grp; + + if (!atomic_dec_and_test(&pa->pa_count) || pa->pa_free != 0) + return; + + /* in this short window concurrent discard can set pa_deleted */ + spin_lock(&pa->pa_lock); + if (pa->pa_deleted == 1) { + spin_unlock(&pa->pa_lock); + return; + } + + pa->pa_deleted = 1; + spin_unlock(&pa->pa_lock); + + /* -1 is to protect from crossing allocation group */ + ext4_get_group_no_and_offset(sb, pa->pa_pstart - 1, &grp, NULL); + + /* + * possible race: + * + * P1 (buddy init) P2 (regular allocation) + * find block B in PA + * copy on-disk bitmap to buddy + * mark B in on-disk bitmap + * drop PA from group + * mark all PAs in buddy + * + * thus, P1 initializes buddy with B available. to prevent this + * we make "copy" and "mark all PAs" atomic and serialize "drop PA" + * against that pair + */ + ext4_lock_group(sb, grp); + list_del(&pa->pa_group_list); + ext4_unlock_group(sb, grp); + + spin_lock(pa->pa_obj_lock); + list_del_rcu(&pa->pa_inode_list); + spin_unlock(pa->pa_obj_lock); + + call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback); +} + +/* + * creates new preallocated space for given inode + */ +static int ext4_mb_new_inode_pa(struct ext4_allocation_context *ac) +{ + struct super_block *sb = ac->ac_sb; + struct ext4_prealloc_space *pa; + struct ext4_group_info *grp; + struct ext4_inode_info *ei; + + /* preallocate only when found space is larger then requested */ + BUG_ON(ac->ac_o_ex.fe_len >= ac->ac_b_ex.fe_len); + BUG_ON(ac->ac_status != AC_STATUS_FOUND); + BUG_ON(!S_ISREG(ac->ac_inode->i_mode)); + + pa = kmem_cache_alloc(ext4_pspace_cachep, GFP_NOFS); + if (pa == NULL) + return -ENOMEM; + + if (ac->ac_b_ex.fe_len < ac->ac_g_ex.fe_len) { + int winl; + int wins; + int win; + int offs; + + /* we can't allocate as much as normalizer wants. + * so, found space must get proper lstart + * to cover original request */ + BUG_ON(ac->ac_g_ex.fe_logical > ac->ac_o_ex.fe_logical); + BUG_ON(ac->ac_g_ex.fe_len < ac->ac_o_ex.fe_len); + + /* we're limited by original request in that + * logical block must be covered any way + * winl is window we can move our chunk within */ + winl = ac->ac_o_ex.fe_logical - ac->ac_g_ex.fe_logical; + + /* also, we should cover whole original request */ + wins = ac->ac_b_ex.fe_len - ac->ac_o_ex.fe_len; + + /* the smallest one defines real window */ + win = min(winl, wins); + + offs = ac->ac_o_ex.fe_logical % ac->ac_b_ex.fe_len; + if (offs && offs < win) + win = offs; + + ac->ac_b_ex.fe_logical = ac->ac_o_ex.fe_logical - win; + BUG_ON(ac->ac_o_ex.fe_logical < ac->ac_b_ex.fe_logical); + BUG_ON(ac->ac_o_ex.fe_len > ac->ac_b_ex.fe_len); + } + + /* preallocation can change ac_b_ex, thus we store actually + * allocated blocks for history */ + ac->ac_f_ex = ac->ac_b_ex; + + pa->pa_lstart = ac->ac_b_ex.fe_logical; + pa->pa_pstart = ext4_grp_offs_to_block(sb, &ac->ac_b_ex); + pa->pa_len = ac->ac_b_ex.fe_len; + pa->pa_free = pa->pa_len; + atomic_set(&pa->pa_count, 1); + spin_lock_init(&pa->pa_lock); + pa->pa_deleted = 0; + pa->pa_linear = 0; + + mb_debug("new inode pa %p: %llu/%u for %u\n", pa, + pa->pa_pstart, pa->pa_len, pa->pa_lstart); + + ext4_mb_use_inode_pa(ac, pa); + atomic_add(pa->pa_free, &EXT4_SB(sb)->s_mb_preallocated); + + ei = EXT4_I(ac->ac_inode); + grp = ext4_get_group_info(sb, ac->ac_b_ex.fe_group); + + pa->pa_obj_lock = &ei->i_prealloc_lock; + pa->pa_inode = ac->ac_inode; + + ext4_lock_group(sb, ac->ac_b_ex.fe_group); + list_add(&pa->pa_group_list, &grp->bb_prealloc_list); + ext4_unlock_group(sb, ac->ac_b_ex.fe_group); + + spin_lock(pa->pa_obj_lock); + list_add_rcu(&pa->pa_inode_list, &ei->i_prealloc_list); + spin_unlock(pa->pa_obj_lock); + + return 0; +} + +/* + * creates new preallocated space for locality group inodes belongs to + */ +static int ext4_mb_new_group_pa(struct ext4_allocation_context *ac) +{ + struct super_block *sb = ac->ac_sb; + struct ext4_locality_group *lg; + struct ext4_prealloc_space *pa; + struct ext4_group_info *grp; + + /* preallocate only when found space is larger then requested */ + BUG_ON(ac->ac_o_ex.fe_len >= ac->ac_b_ex.fe_len); + BUG_ON(ac->ac_status != AC_STATUS_FOUND); + BUG_ON(!S_ISREG(ac->ac_inode->i_mode)); + + BUG_ON(ext4_pspace_cachep == NULL); + pa = kmem_cache_alloc(ext4_pspace_cachep, GFP_NOFS); + if (pa == NULL) + return -ENOMEM; + + /* preallocation can change ac_b_ex, thus we store actually + * allocated blocks for history */ + ac->ac_f_ex = ac->ac_b_ex; + + pa->pa_pstart = ext4_grp_offs_to_block(sb, &ac->ac_b_ex); + pa->pa_lstart = pa->pa_pstart; + pa->pa_len = ac->ac_b_ex.fe_len; + pa->pa_free = pa->pa_len; + atomic_set(&pa->pa_count, 1); + spin_lock_init(&pa->pa_lock); + pa->pa_deleted = 0; + pa->pa_linear = 1; + + mb_debug("new group pa %p: %llu/%u for %u\n", pa, + pa->pa_pstart, pa->pa_len, pa->pa_lstart); + + ext4_mb_use_group_pa(ac, pa); + atomic_add(pa->pa_free, &EXT4_SB(sb)->s_mb_preallocated); + + grp = ext4_get_group_info(sb, ac->ac_b_ex.fe_group); + lg = ac->ac_lg; + BUG_ON(lg == NULL); + + pa->pa_obj_lock = &lg->lg_prealloc_lock; + pa->pa_inode = NULL; + + ext4_lock_group(sb, ac->ac_b_ex.fe_group); + list_add(&pa->pa_group_list, &grp->bb_prealloc_list); + ext4_unlock_group(sb, ac->ac_b_ex.fe_group); + + spin_lock(pa->pa_obj_lock); + list_add_tail_rcu(&pa->pa_inode_list, &lg->lg_prealloc_list); + spin_unlock(pa->pa_obj_lock); + + return 0; +} + +static int ext4_mb_new_preallocation(struct ext4_allocation_context *ac) +{ + int err; + + if (ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC) + err = ext4_mb_new_group_pa(ac); + else + err = ext4_mb_new_inode_pa(ac); + return err; +} + +/* + * finds all unused blocks in on-disk bitmap, frees them in + * in-core bitmap and buddy. + * @pa must be unlinked from inode and group lists, so that + * nobody else can find/use it. + * the caller MUST hold group/inode locks. + * TODO: optimize the case when there are no in-core structures yet + */ +static int ext4_mb_release_inode_pa(struct ext4_buddy *e4b, + struct buffer_head *bitmap_bh, + struct ext4_prealloc_space *pa) +{ + struct ext4_allocation_context ac; + struct super_block *sb = e4b->bd_sb; + struct ext4_sb_info *sbi = EXT4_SB(sb); + unsigned long end; + unsigned long next; + ext4_group_t group; + ext4_grpblk_t bit; + sector_t start; + int err = 0; + int free = 0; + + BUG_ON(pa->pa_deleted == 0); + ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, &bit); + BUG_ON(group != e4b->bd_group && pa->pa_len != 0); + end = bit + pa->pa_len; + + ac.ac_sb = sb; + ac.ac_inode = pa->pa_inode; + ac.ac_op = EXT4_MB_HISTORY_DISCARD; + + while (bit < end) { + bit = ext4_find_next_zero_bit(bitmap_bh->b_data, end, bit); + if (bit >= end) + break; + next = ext4_find_next_bit(bitmap_bh->b_data, end, bit); + if (next > end) + next = end; + start = group * EXT4_BLOCKS_PER_GROUP(sb) + bit + + le32_to_cpu(sbi->s_es->s_first_data_block); + mb_debug(" free preallocated %u/%u in group %u\n", + (unsigned) start, (unsigned) next - bit, + (unsigned) group); + free += next - bit; + + ac.ac_b_ex.fe_group = group; + ac.ac_b_ex.fe_start = bit; + ac.ac_b_ex.fe_len = next - bit; + ac.ac_b_ex.fe_logical = 0; + ext4_mb_store_history(&ac); + + mb_free_blocks(pa->pa_inode, e4b, bit, next - bit); + bit = next + 1; + } + if (free != pa->pa_free) { + printk(KERN_ERR "pa %p: logic %lu, phys. %lu, len %lu\n", + pa, (unsigned long) pa->pa_lstart, + (unsigned long) pa->pa_pstart, + (unsigned long) pa->pa_len); + printk(KERN_ERR "free %u, pa_free %u\n", free, pa->pa_free); + } + BUG_ON(free != pa->pa_free); + atomic_add(free, &sbi->s_mb_discarded); + + return err; +} + +static int ext4_mb_release_group_pa(struct ext4_buddy *e4b, + struct ext4_prealloc_space *pa) +{ + struct ext4_allocation_context ac; + struct super_block *sb = e4b->bd_sb; + ext4_group_t group; + ext4_grpblk_t bit; + + ac.ac_op = EXT4_MB_HISTORY_DISCARD; + + BUG_ON(pa->pa_deleted == 0); + ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, &bit); + BUG_ON(group != e4b->bd_group && pa->pa_len != 0); + mb_free_blocks(pa->pa_inode, e4b, bit, pa->pa_len); + atomic_add(pa->pa_len, &EXT4_SB(sb)->s_mb_discarded); + + ac.ac_sb = sb; + ac.ac_inode = NULL; + ac.ac_b_ex.fe_group = group; + ac.ac_b_ex.fe_start = bit; + ac.ac_b_ex.fe_len = pa->pa_len; + ac.ac_b_ex.fe_logical = 0; + ext4_mb_store_history(&ac); + + return 0; +} + +/* + * releases all preallocations in given group + * + * first, we need to decide discard policy: + * - when do we discard + * 1) ENOSPC + * - how many do we discard + * 1) how many requested + */ +static int ext4_mb_discard_group_preallocations(struct super_block *sb, + ext4_group_t group, int needed) +{ + struct ext4_group_info *grp = ext4_get_group_info(sb, group); + struct buffer_head *bitmap_bh = NULL; + struct ext4_prealloc_space *pa, *tmp; + struct list_head list; + struct ext4_buddy e4b; + int err; + int busy = 0; + int free = 0; + + mb_debug("discard preallocation for group %lu\n", group); + + if (list_empty(&grp->bb_prealloc_list)) + return 0; + + bitmap_bh = read_block_bitmap(sb, group); + if (bitmap_bh == NULL) { + /* error handling here */ + ext4_mb_release_desc(&e4b); + BUG_ON(bitmap_bh == NULL); + } + + err = ext4_mb_load_buddy(sb, group, &e4b); + BUG_ON(err != 0); /* error handling here */ + + if (needed == 0) + needed = EXT4_BLOCKS_PER_GROUP(sb) + 1; + + grp = ext4_get_group_info(sb, group); + INIT_LIST_HEAD(&list); + +repeat: + ext4_lock_group(sb, group); + list_for_each_entry_safe(pa, tmp, + &grp->bb_prealloc_list, pa_group_list) { + spin_lock(&pa->pa_lock); + if (atomic_read(&pa->pa_count)) { + spin_unlock(&pa->pa_lock); + busy = 1; + continue; + } + if (pa->pa_deleted) { + spin_unlock(&pa->pa_lock); + continue; + } + + /* seems this one can be freed ... */ + pa->pa_deleted = 1; + + /* we can trust pa_free ... */ + free += pa->pa_free; + + spin_unlock(&pa->pa_lock); + + list_del(&pa->pa_group_list); + list_add(&pa->u.pa_tmp_list, &list); + } + + /* if we still need more blocks and some PAs were used, try again */ + if (free < needed && busy) { + busy = 0; + ext4_unlock_group(sb, group); + /* + * Yield the CPU here so that we don't get soft lockup + * in non preempt case. + */ + yield(); + goto repeat; + } + + /* found anything to free? */ + if (list_empty(&list)) { + BUG_ON(free != 0); + goto out; + } + + /* now free all selected PAs */ + list_for_each_entry_safe(pa, tmp, &list, u.pa_tmp_list) { + + /* remove from object (inode or locality group) */ + spin_lock(pa->pa_obj_lock); + list_del_rcu(&pa->pa_inode_list); + spin_unlock(pa->pa_obj_lock); + + if (pa->pa_linear) + ext4_mb_release_group_pa(&e4b, pa); + else + ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa); + + list_del(&pa->u.pa_tmp_list); + call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback); + } + +out: + ext4_unlock_group(sb, group); + ext4_mb_release_desc(&e4b); + put_bh(bitmap_bh); + return free; +} + +/* + * releases all non-used preallocated blocks for given inode + * + * It's important to discard preallocations under i_data_sem + * We don't want another block to be served from the prealloc + * space when we are discarding the inode prealloc space. + * + * FIXME!! Make sure it is valid at all the call sites + */ +void ext4_mb_discard_inode_preallocations(struct inode *inode) +{ + struct ext4_inode_info *ei = EXT4_I(inode); + struct super_block *sb = inode->i_sb; + struct buffer_head *bitmap_bh = NULL; + struct ext4_prealloc_space *pa, *tmp; + ext4_group_t group = 0; + struct list_head list; + struct ext4_buddy e4b; + int err; + + if (!test_opt(sb, MBALLOC) || !S_ISREG(inode->i_mode)) { + /*BUG_ON(!list_empty(&ei->i_prealloc_list));*/ + return; + } + + mb_debug("discard preallocation for inode %lu\n", inode->i_ino); + + INIT_LIST_HEAD(&list); + +repeat: + /* first, collect all pa's in the inode */ + spin_lock(&ei->i_prealloc_lock); + while (!list_empty(&ei->i_prealloc_list)) { + pa = list_entry(ei->i_prealloc_list.next, + struct ext4_prealloc_space, pa_inode_list); + BUG_ON(pa->pa_obj_lock != &ei->i_prealloc_lock); + spin_lock(&pa->pa_lock); + if (atomic_read(&pa->pa_count)) { + /* this shouldn't happen often - nobody should + * use preallocation while we're discarding it */ + spin_unlock(&pa->pa_lock); + spin_unlock(&ei->i_prealloc_lock); + printk(KERN_ERR "uh-oh! used pa while discarding\n"); + WARN_ON(1); + schedule_timeout_uninterruptible(HZ); + goto repeat; + + } + if (pa->pa_deleted == 0) { + pa->pa_deleted = 1; + spin_unlock(&pa->pa_lock); + list_del_rcu(&pa->pa_inode_list); + list_add(&pa->u.pa_tmp_list, &list); + continue; + } + + /* someone is deleting pa right now */ + spin_unlock(&pa->pa_lock); + spin_unlock(&ei->i_prealloc_lock); + + /* we have to wait here because pa_deleted + * doesn't mean pa is already unlinked from + * the list. as we might be called from + * ->clear_inode() the inode will get freed + * and concurrent thread which is unlinking + * pa from inode's list may access already + * freed memory, bad-bad-bad */ + + /* XXX: if this happens too often, we can + * add a flag to force wait only in case + * of ->clear_inode(), but not in case of + * regular truncate */ + schedule_timeout_uninterruptible(HZ); + goto repeat; + } + spin_unlock(&ei->i_prealloc_lock); + + list_for_each_entry_safe(pa, tmp, &list, u.pa_tmp_list) { + BUG_ON(pa->pa_linear != 0); + ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, NULL); + + err = ext4_mb_load_buddy(sb, group, &e4b); + BUG_ON(err != 0); /* error handling here */ + + bitmap_bh = read_block_bitmap(sb, group); + if (bitmap_bh == NULL) { + /* error handling here */ + ext4_mb_release_desc(&e4b); + BUG_ON(bitmap_bh == NULL); + } + + ext4_lock_group(sb, group); + list_del(&pa->pa_group_list); + ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa); + ext4_unlock_group(sb, group); + + ext4_mb_release_desc(&e4b); + put_bh(bitmap_bh); + + list_del(&pa->u.pa_tmp_list); + call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback); + } +} + +/* + * finds all preallocated spaces and return blocks being freed to them + * if preallocated space becomes full (no block is used from the space) + * then the function frees space in buddy + * XXX: at the moment, truncate (which is the only way to free blocks) + * discards all preallocations + */ +static void ext4_mb_return_to_preallocation(struct inode *inode, + struct ext4_buddy *e4b, + sector_t block, int count) +{ + BUG_ON(!list_empty(&EXT4_I(inode)->i_prealloc_list)); +} +#ifdef MB_DEBUG +static void ext4_mb_show_ac(struct ext4_allocation_context *ac) +{ + struct super_block *sb = ac->ac_sb; + ext4_group_t i; + + printk(KERN_ERR "EXT4-fs: Can't allocate:" + " Allocation context details:\n"); + printk(KERN_ERR "EXT4-fs: status %d flags %d\n", + ac->ac_status, ac->ac_flags); + printk(KERN_ERR "EXT4-fs: orig %lu/%lu/%lu@%lu, goal %lu/%lu/%lu@%lu, " + "best %lu/%lu/%lu@%lu cr %d\n", + (unsigned long)ac->ac_o_ex.fe_group, + (unsigned long)ac->ac_o_ex.fe_start, + (unsigned long)ac->ac_o_ex.fe_len, + (unsigned long)ac->ac_o_ex.fe_logical, + (unsigned long)ac->ac_g_ex.fe_group, + (unsigned long)ac->ac_g_ex.fe_start, + (unsigned long)ac->ac_g_ex.fe_len, + (unsigned long)ac->ac_g_ex.fe_logical, + (unsigned long)ac->ac_b_ex.fe_group, + (unsigned long)ac->ac_b_ex.fe_start, + (unsigned long)ac->ac_b_ex.fe_len, + (unsigned long)ac->ac_b_ex.fe_logical, + (int)ac->ac_criteria); + printk(KERN_ERR "EXT4-fs: %lu scanned, %d found\n", ac->ac_ex_scanned, + ac->ac_found); + printk(KERN_ERR "EXT4-fs: groups: \n"); + for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { + struct ext4_group_info *grp = ext4_get_group_info(sb, i); + struct ext4_prealloc_space *pa; + ext4_grpblk_t start; + struct list_head *cur; + ext4_lock_group(sb, i); + list_for_each(cur, &grp->bb_prealloc_list) { + pa = list_entry(cur, struct ext4_prealloc_space, + pa_group_list); + spin_lock(&pa->pa_lock); + ext4_get_group_no_and_offset(sb, pa->pa_pstart, + NULL, &start); + spin_unlock(&pa->pa_lock); + printk(KERN_ERR "PA:%lu:%d:%u \n", i, + start, pa->pa_len); + } + ext4_lock_group(sb, i); + + if (grp->bb_free == 0) + continue; + printk(KERN_ERR "%lu: %d/%d \n", + i, grp->bb_free, grp->bb_fragments); + } + printk(KERN_ERR "\n"); +} +#else +static inline void ext4_mb_show_ac(struct ext4_allocation_context *ac) +{ + return; +} +#endif + +/* + * We use locality group preallocation for small size file. The size of the + * file is determined by the current size or the resulting size after + * allocation which ever is larger + * + * One can tune this size via /proc/fs/ext4//stream_req + */ +static void ext4_mb_group_or_file(struct ext4_allocation_context *ac) +{ + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + int bsbits = ac->ac_sb->s_blocksize_bits; + loff_t size, isize; + + if (!(ac->ac_flags & EXT4_MB_HINT_DATA)) + return; + + size = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len; + isize = i_size_read(ac->ac_inode) >> bsbits; + size = max(size, isize); + + /* don't use group allocation for large files */ + if (size >= sbi->s_mb_stream_request) + return; + + if (unlikely(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY)) + return; + + BUG_ON(ac->ac_lg != NULL); + /* + * locality group prealloc space are per cpu. The reason for having + * per cpu locality group is to reduce the contention between block + * request from multiple CPUs. + */ + ac->ac_lg = &sbi->s_locality_groups[get_cpu()]; + put_cpu(); + + /* we're going to use group allocation */ + ac->ac_flags |= EXT4_MB_HINT_GROUP_ALLOC; + + /* serialize all allocations in the group */ + mutex_lock(&ac->ac_lg->lg_mutex); +} + +static int ext4_mb_initialize_context(struct ext4_allocation_context *ac, + struct ext4_allocation_request *ar) +{ + struct super_block *sb = ar->inode->i_sb; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + ext4_group_t group; + unsigned long len; + unsigned long goal; + ext4_grpblk_t block; + + /* we can't allocate > group size */ + len = ar->len; + + /* just a dirty hack to filter too big requests */ + if (len >= EXT4_BLOCKS_PER_GROUP(sb) - 10) + len = EXT4_BLOCKS_PER_GROUP(sb) - 10; + + /* start searching from the goal */ + goal = ar->goal; + if (goal < le32_to_cpu(es->s_first_data_block) || + goal >= ext4_blocks_count(es)) + goal = le32_to_cpu(es->s_first_data_block); + ext4_get_group_no_and_offset(sb, goal, &group, &block); + + /* set up allocation goals */ + ac->ac_b_ex.fe_logical = ar->logical; + ac->ac_b_ex.fe_group = 0; + ac->ac_b_ex.fe_start = 0; + ac->ac_b_ex.fe_len = 0; + ac->ac_status = AC_STATUS_CONTINUE; + ac->ac_groups_scanned = 0; + ac->ac_ex_scanned = 0; + ac->ac_found = 0; + ac->ac_sb = sb; + ac->ac_inode = ar->inode; + ac->ac_o_ex.fe_logical = ar->logical; + ac->ac_o_ex.fe_group = group; + ac->ac_o_ex.fe_start = block; + ac->ac_o_ex.fe_len = len; + ac->ac_g_ex.fe_logical = ar->logical; + ac->ac_g_ex.fe_group = group; + ac->ac_g_ex.fe_start = block; + ac->ac_g_ex.fe_len = len; + ac->ac_f_ex.fe_len = 0; + ac->ac_flags = ar->flags; + ac->ac_2order = 0; + ac->ac_criteria = 0; + ac->ac_pa = NULL; + ac->ac_bitmap_page = NULL; + ac->ac_buddy_page = NULL; + ac->ac_lg = NULL; + + /* we have to define context: we'll we work with a file or + * locality group. this is a policy, actually */ + ext4_mb_group_or_file(ac); + + mb_debug("init ac: %u blocks @ %u, goal %u, flags %x, 2^%d, " + "left: %u/%u, right %u/%u to %swritable\n", + (unsigned) ar->len, (unsigned) ar->logical, + (unsigned) ar->goal, ac->ac_flags, ac->ac_2order, + (unsigned) ar->lleft, (unsigned) ar->pleft, + (unsigned) ar->lright, (unsigned) ar->pright, + atomic_read(&ar->inode->i_writecount) ? "" : "non-"); + return 0; + +} + +/* + * release all resource we used in allocation + */ +static int ext4_mb_release_context(struct ext4_allocation_context *ac) +{ + if (ac->ac_pa) { + if (ac->ac_pa->pa_linear) { + /* see comment in ext4_mb_use_group_pa() */ + spin_lock(&ac->ac_pa->pa_lock); + ac->ac_pa->pa_pstart += ac->ac_b_ex.fe_len; + ac->ac_pa->pa_lstart += ac->ac_b_ex.fe_len; + ac->ac_pa->pa_free -= ac->ac_b_ex.fe_len; + ac->ac_pa->pa_len -= ac->ac_b_ex.fe_len; + spin_unlock(&ac->ac_pa->pa_lock); + } + ext4_mb_put_pa(ac, ac->ac_sb, ac->ac_pa); + } + if (ac->ac_bitmap_page) + page_cache_release(ac->ac_bitmap_page); + if (ac->ac_buddy_page) + page_cache_release(ac->ac_buddy_page); + if (ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC) + mutex_unlock(&ac->ac_lg->lg_mutex); + ext4_mb_collect_stats(ac); + return 0; +} + +static int ext4_mb_discard_preallocations(struct super_block *sb, int needed) +{ + ext4_group_t i; + int ret; + int freed = 0; + + for (i = 0; i < EXT4_SB(sb)->s_groups_count && needed > 0; i++) { + ret = ext4_mb_discard_group_preallocations(sb, i, needed); + freed += ret; + needed -= ret; + } + + return freed; +} + +/* + * Main entry point into mballoc to allocate blocks + * it tries to use preallocation first, then falls back + * to usual allocation + */ +ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle, + struct ext4_allocation_request *ar, int *errp) +{ + struct ext4_allocation_context ac; + struct ext4_sb_info *sbi; + struct super_block *sb; + ext4_fsblk_t block = 0; + int freed; + int inquota; + + sb = ar->inode->i_sb; + sbi = EXT4_SB(sb); + + if (!test_opt(sb, MBALLOC)) { + block = ext4_new_blocks_old(handle, ar->inode, ar->goal, + &(ar->len), errp); + return block; + } + + while (ar->len && DQUOT_ALLOC_BLOCK(ar->inode, ar->len)) { + ar->flags |= EXT4_MB_HINT_NOPREALLOC; + ar->len--; + } + if (ar->len == 0) { + *errp = -EDQUOT; + return 0; + } + inquota = ar->len; + + ext4_mb_poll_new_transaction(sb, handle); + + *errp = ext4_mb_initialize_context(&ac, ar); + if (*errp) { + ar->len = 0; + goto out; + } + + ac.ac_op = EXT4_MB_HISTORY_PREALLOC; + if (!ext4_mb_use_preallocated(&ac)) { + + ac.ac_op = EXT4_MB_HISTORY_ALLOC; + ext4_mb_normalize_request(&ac, ar); + +repeat: + /* allocate space in core */ + ext4_mb_regular_allocator(&ac); + + /* as we've just preallocated more space than + * user requested orinally, we store allocated + * space in a special descriptor */ + if (ac.ac_status == AC_STATUS_FOUND && + ac.ac_o_ex.fe_len < ac.ac_b_ex.fe_len) + ext4_mb_new_preallocation(&ac); + } + + if (likely(ac.ac_status == AC_STATUS_FOUND)) { + ext4_mb_mark_diskspace_used(&ac, handle); + *errp = 0; + block = ext4_grp_offs_to_block(sb, &ac.ac_b_ex); + ar->len = ac.ac_b_ex.fe_len; + } else { + freed = ext4_mb_discard_preallocations(sb, ac.ac_o_ex.fe_len); + if (freed) + goto repeat; + *errp = -ENOSPC; + ac.ac_b_ex.fe_len = 0; + ar->len = 0; + ext4_mb_show_ac(&ac); + } + + ext4_mb_release_context(&ac); + +out: + if (ar->len < inquota) + DQUOT_FREE_BLOCK(ar->inode, inquota - ar->len); + + return block; +} +static void ext4_mb_poll_new_transaction(struct super_block *sb, + handle_t *handle) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + + if (sbi->s_last_transaction == handle->h_transaction->t_tid) + return; + + /* new transaction! time to close last one and free blocks for + * committed transaction. we know that only transaction can be + * active, so previos transaction can be being logged and we + * know that transaction before previous is known to be already + * logged. this means that now we may free blocks freed in all + * transactions before previous one. hope I'm clear enough ... */ + + spin_lock(&sbi->s_md_lock); + if (sbi->s_last_transaction != handle->h_transaction->t_tid) { + mb_debug("new transaction %lu, old %lu\n", + (unsigned long) handle->h_transaction->t_tid, + (unsigned long) sbi->s_last_transaction); + list_splice_init(&sbi->s_closed_transaction, + &sbi->s_committed_transaction); + list_splice_init(&sbi->s_active_transaction, + &sbi->s_closed_transaction); + sbi->s_last_transaction = handle->h_transaction->t_tid; + } + spin_unlock(&sbi->s_md_lock); + + ext4_mb_free_committed_blocks(sb); +} + +static int ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b, + ext4_group_t group, ext4_grpblk_t block, int count) +{ + struct ext4_group_info *db = e4b->bd_info; + struct super_block *sb = e4b->bd_sb; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_free_metadata *md; + int i; + + BUG_ON(e4b->bd_bitmap_page == NULL); + BUG_ON(e4b->bd_buddy_page == NULL); + + ext4_lock_group(sb, group); + for (i = 0; i < count; i++) { + md = db->bb_md_cur; + if (md && db->bb_tid != handle->h_transaction->t_tid) { + db->bb_md_cur = NULL; + md = NULL; + } + + if (md == NULL) { + ext4_unlock_group(sb, group); + md = kmalloc(sizeof(*md), GFP_NOFS); + if (md == NULL) + return -ENOMEM; + md->num = 0; + md->group = group; + + ext4_lock_group(sb, group); + if (db->bb_md_cur == NULL) { + spin_lock(&sbi->s_md_lock); + list_add(&md->list, &sbi->s_active_transaction); + spin_unlock(&sbi->s_md_lock); + /* protect buddy cache from being freed, + * otherwise we'll refresh it from + * on-disk bitmap and lose not-yet-available + * blocks */ + page_cache_get(e4b->bd_buddy_page); + page_cache_get(e4b->bd_bitmap_page); + db->bb_md_cur = md; + db->bb_tid = handle->h_transaction->t_tid; + mb_debug("new md 0x%p for group %lu\n", + md, md->group); + } else { + kfree(md); + md = db->bb_md_cur; + } + } + + BUG_ON(md->num >= EXT4_BB_MAX_BLOCKS); + md->blocks[md->num] = block + i; + md->num++; + if (md->num == EXT4_BB_MAX_BLOCKS) { + /* no more space, put full container on a sb's list */ + db->bb_md_cur = NULL; + } + } + ext4_unlock_group(sb, group); + return 0; +} + +/* + * Main entry point into mballoc to free blocks + */ +void ext4_mb_free_blocks(handle_t *handle, struct inode *inode, + unsigned long block, unsigned long count, + int metadata, unsigned long *freed) +{ + struct buffer_head *bitmap_bh = 0; + struct super_block *sb = inode->i_sb; + struct ext4_allocation_context ac; + struct ext4_group_desc *gdp; + struct ext4_super_block *es; + unsigned long overflow; + ext4_grpblk_t bit; + struct buffer_head *gd_bh; + ext4_group_t block_group; + struct ext4_sb_info *sbi; + struct ext4_buddy e4b; + int err = 0; + int ret; + + *freed = 0; + + ext4_mb_poll_new_transaction(sb, handle); + + sbi = EXT4_SB(sb); + es = EXT4_SB(sb)->s_es; + if (block < le32_to_cpu(es->s_first_data_block) || + block + count < block || + block + count > ext4_blocks_count(es)) { + ext4_error(sb, __FUNCTION__, + "Freeing blocks not in datazone - " + "block = %lu, count = %lu", block, count); + goto error_return; + } + + ext4_debug("freeing block %lu\n", block); + + ac.ac_op = EXT4_MB_HISTORY_FREE; + ac.ac_inode = inode; + ac.ac_sb = sb; + +do_more: + overflow = 0; + ext4_get_group_no_and_offset(sb, block, &block_group, &bit); + + /* + * Check to see if we are freeing blocks across a group + * boundary. + */ + if (bit + count > EXT4_BLOCKS_PER_GROUP(sb)) { + overflow = bit + count - EXT4_BLOCKS_PER_GROUP(sb); + count -= overflow; + } + bitmap_bh = read_block_bitmap(sb, block_group); + if (!bitmap_bh) + goto error_return; + gdp = ext4_get_group_desc(sb, block_group, &gd_bh); + if (!gdp) + goto error_return; + + if (in_range(ext4_block_bitmap(sb, gdp), block, count) || + in_range(ext4_inode_bitmap(sb, gdp), block, count) || + in_range(block, ext4_inode_table(sb, gdp), + EXT4_SB(sb)->s_itb_per_group) || + in_range(block + count - 1, ext4_inode_table(sb, gdp), + EXT4_SB(sb)->s_itb_per_group)) { + + ext4_error(sb, __FUNCTION__, + "Freeing blocks in system zone - " + "Block = %lu, count = %lu", block, count); + } + + BUFFER_TRACE(bitmap_bh, "getting write access"); + err = ext4_journal_get_write_access(handle, bitmap_bh); + if (err) + goto error_return; + + /* + * We are about to modify some metadata. Call the journal APIs + * to unshare ->b_data if a currently-committing transaction is + * using it + */ + BUFFER_TRACE(gd_bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, gd_bh); + if (err) + goto error_return; + + err = ext4_mb_load_buddy(sb, block_group, &e4b); + if (err) + goto error_return; + +#ifdef AGGRESSIVE_CHECK + { + int i; + for (i = 0; i < count; i++) + BUG_ON(!mb_test_bit(bit + i, bitmap_bh->b_data)); + } +#endif + mb_clear_bits(sb_bgl_lock(sbi, block_group), bitmap_bh->b_data, + bit, count); + + /* We dirtied the bitmap block */ + BUFFER_TRACE(bitmap_bh, "dirtied bitmap block"); + err = ext4_journal_dirty_metadata(handle, bitmap_bh); + + ac.ac_b_ex.fe_group = block_group; + ac.ac_b_ex.fe_start = bit; + ac.ac_b_ex.fe_len = count; + ext4_mb_store_history(&ac); + + if (metadata) { + /* blocks being freed are metadata. these blocks shouldn't + * be used until this transaction is committed */ + ext4_mb_free_metadata(handle, &e4b, block_group, bit, count); + } else { + ext4_lock_group(sb, block_group); + err = mb_free_blocks(inode, &e4b, bit, count); + ext4_mb_return_to_preallocation(inode, &e4b, block, count); + ext4_unlock_group(sb, block_group); + BUG_ON(err != 0); + } + + spin_lock(sb_bgl_lock(sbi, block_group)); + gdp->bg_free_blocks_count = + cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) + count); + gdp->bg_checksum = ext4_group_desc_csum(sbi, block_group, gdp); + spin_unlock(sb_bgl_lock(sbi, block_group)); + percpu_counter_add(&sbi->s_freeblocks_counter, count); + + ext4_mb_release_desc(&e4b); + + *freed += count; + + /* And the group descriptor block */ + BUFFER_TRACE(gd_bh, "dirtied group descriptor block"); + ret = ext4_journal_dirty_metadata(handle, gd_bh); + if (!err) + err = ret; + + if (overflow && !err) { + block += count; + count = overflow; + put_bh(bitmap_bh); + goto do_more; + } + sb->s_dirt = 1; +error_return: + brelse(bitmap_bh); + ext4_std_error(sb, err); + return; +} diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c index ec7cb567a7da..3ebc2332f52e 100644 --- a/fs/ext4/migrate.c +++ b/fs/ext4/migrate.c @@ -236,10 +236,10 @@ static int free_dind_blocks(handle_t *handle, for (i = 0; i < max_entries; i++) { if (tmp_idata[i]) ext4_free_blocks(handle, inode, - le32_to_cpu(tmp_idata[i]), 1); + le32_to_cpu(tmp_idata[i]), 1, 1); } put_bh(bh); - ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1); + ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1, 1); return 0; } @@ -267,7 +267,7 @@ static int free_tind_blocks(handle_t *handle, } } put_bh(bh); - ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1); + ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1, 1); return 0; } @@ -278,7 +278,7 @@ static int free_ind_block(handle_t *handle, struct inode *inode) if (ei->i_data[EXT4_IND_BLOCK]) ext4_free_blocks(handle, inode, - le32_to_cpu(ei->i_data[EXT4_IND_BLOCK]), 1); + le32_to_cpu(ei->i_data[EXT4_IND_BLOCK]), 1, 1); if (ei->i_data[EXT4_DIND_BLOCK]) { retval = free_dind_blocks(handle, inode, @@ -365,7 +365,7 @@ static int free_ext_idx(handle_t *handle, struct inode *inode, } } put_bh(bh); - ext4_free_blocks(handle, inode, block, 1); + ext4_free_blocks(handle, inode, block, 1, 1); return retval; } diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 64fc7f111734..3a51ffc47790 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -503,6 +503,7 @@ static void ext4_put_super (struct super_block * sb) struct ext4_super_block *es = sbi->s_es; int i; + ext4_mb_release(sb); ext4_ext_release(sb); ext4_xattr_put_super(sb); jbd2_journal_destroy(sbi->s_journal); @@ -569,6 +570,8 @@ static struct inode *ext4_alloc_inode(struct super_block *sb) ei->i_block_alloc_info = NULL; ei->vfs_inode.i_version = 1; memset(&ei->i_cached_extent, 0, sizeof(struct ext4_ext_cache)); + INIT_LIST_HEAD(&ei->i_prealloc_list); + spin_lock_init(&ei->i_prealloc_lock); return &ei->vfs_inode; } @@ -881,6 +884,7 @@ enum { Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota, Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota, Opt_grpquota, Opt_extents, Opt_noextents, Opt_i_version, + Opt_mballoc, Opt_nomballoc, Opt_stripe, }; static match_table_t tokens = { @@ -935,6 +939,9 @@ static match_table_t tokens = { {Opt_extents, "extents"}, {Opt_noextents, "noextents"}, {Opt_i_version, "i_version"}, + {Opt_mballoc, "mballoc"}, + {Opt_nomballoc, "nomballoc"}, + {Opt_stripe, "stripe=%u"}, {Opt_err, NULL}, {Opt_resize, "resize"}, }; @@ -1284,6 +1291,19 @@ clear_qf_name: set_opt(sbi->s_mount_opt, I_VERSION); sb->s_flags |= MS_I_VERSION; break; + case Opt_mballoc: + set_opt(sbi->s_mount_opt, MBALLOC); + break; + case Opt_nomballoc: + clear_opt(sbi->s_mount_opt, MBALLOC); + break; + case Opt_stripe: + if (match_int(&args[0], &option)) + return 0; + if (option < 0) + return 0; + sbi->s_stripe = option; + break; default: printk (KERN_ERR "EXT4-fs: Unrecognized mount option \"%s\" " @@ -1742,6 +1762,34 @@ static ext4_fsblk_t descriptor_loc(struct super_block *sb, return (has_super + ext4_group_first_block_no(sb, bg)); } +/** + * ext4_get_stripe_size: Get the stripe size. + * @sbi: In memory super block info + * + * If we have specified it via mount option, then + * use the mount option value. If the value specified at mount time is + * greater than the blocks per group use the super block value. + * If the super block value is greater than blocks per group return 0. + * Allocator needs it be less than blocks per group. + * + */ +static unsigned long ext4_get_stripe_size(struct ext4_sb_info *sbi) +{ + unsigned long stride = le16_to_cpu(sbi->s_es->s_raid_stride); + unsigned long stripe_width = + le32_to_cpu(sbi->s_es->s_raid_stripe_width); + + if (sbi->s_stripe && sbi->s_stripe <= sbi->s_blocks_per_group) + return sbi->s_stripe; + + if (stripe_width <= sbi->s_blocks_per_group) + return stripe_width; + + if (stride <= sbi->s_blocks_per_group) + return stride; + + return 0; +} static int ext4_fill_super (struct super_block *sb, void *data, int silent) __releases(kernel_sem) @@ -2091,6 +2139,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) sbi->s_rsv_window_head.rsv_goal_size = 0; ext4_rsv_window_add(sb, &sbi->s_rsv_window_head); + sbi->s_stripe = ext4_get_stripe_size(sbi); + /* * set up enough so that it can read an inode */ @@ -2250,6 +2300,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) "writeback"); ext4_ext_init(sb); + ext4_mb_init(sb, needs_recovery); lock_kernel(); return 0; @@ -3232,9 +3283,15 @@ static struct file_system_type ext4dev_fs_type = { static int __init init_ext4_fs(void) { - int err = init_ext4_xattr(); + int err; + + err = init_ext4_mballoc(); if (err) return err; + + err = init_ext4_xattr(); + if (err) + goto out2; err = init_inodecache(); if (err) goto out1; @@ -3246,6 +3303,8 @@ out: destroy_inodecache(); out1: exit_ext4_xattr(); +out2: + exit_ext4_mballoc(); return err; } @@ -3254,6 +3313,7 @@ static void __exit exit_ext4_fs(void) unregister_filesystem(&ext4dev_fs_type); destroy_inodecache(); exit_ext4_xattr(); + exit_ext4_mballoc(); } MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others"); diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 86387302c2a9..d7962139c010 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -480,7 +480,7 @@ ext4_xattr_release_block(handle_t *handle, struct inode *inode, ea_bdebug(bh, "refcount now=0; freeing"); if (ce) mb_cache_entry_free(ce); - ext4_free_blocks(handle, inode, bh->b_blocknr, 1); + ext4_free_blocks(handle, inode, bh->b_blocknr, 1, 1); get_bh(bh); ext4_forget(handle, 1, inode, bh, bh->b_blocknr); } else { @@ -821,7 +821,7 @@ inserted: new_bh = sb_getblk(sb, block); if (!new_bh) { getblk_failed: - ext4_free_blocks(handle, inode, block, 1); + ext4_free_blocks(handle, inode, block, 1, 1); error = -EIO; goto cleanup; } diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h index d0b7ca99b91f..1852313fc7c7 100644 --- a/include/linux/ext4_fs.h +++ b/include/linux/ext4_fs.h @@ -20,6 +20,8 @@ #include #include +#include + /* * The second extended filesystem constants/structures */ @@ -51,6 +53,50 @@ #define ext4_debug(f, a...) do {} while (0) #endif +#define EXT4_MULTIBLOCK_ALLOCATOR 1 + +/* prefer goal again. length */ +#define EXT4_MB_HINT_MERGE 1 +/* blocks already reserved */ +#define EXT4_MB_HINT_RESERVED 2 +/* metadata is being allocated */ +#define EXT4_MB_HINT_METADATA 4 +/* first blocks in the file */ +#define EXT4_MB_HINT_FIRST 8 +/* search for the best chunk */ +#define EXT4_MB_HINT_BEST 16 +/* data is being allocated */ +#define EXT4_MB_HINT_DATA 32 +/* don't preallocate (for tails) */ +#define EXT4_MB_HINT_NOPREALLOC 64 +/* allocate for locality group */ +#define EXT4_MB_HINT_GROUP_ALLOC 128 +/* allocate goal blocks or none */ +#define EXT4_MB_HINT_GOAL_ONLY 256 +/* goal is meaningful */ +#define EXT4_MB_HINT_TRY_GOAL 512 + +struct ext4_allocation_request { + /* target inode for block we're allocating */ + struct inode *inode; + /* logical block in target inode */ + ext4_lblk_t logical; + /* phys. target (a hint) */ + ext4_fsblk_t goal; + /* the closest logical allocated block to the left */ + ext4_lblk_t lleft; + /* phys. block for ^^^ */ + ext4_fsblk_t pleft; + /* the closest logical allocated block to the right */ + ext4_lblk_t lright; + /* phys. block for ^^^ */ + ext4_fsblk_t pright; + /* how many blocks we want to allocate */ + unsigned long len; + /* flags. see above EXT4_MB_HINT_* */ + unsigned long flags; +}; + /* * Special inodes numbers */ @@ -474,6 +520,7 @@ do { \ #define EXT4_MOUNT_JOURNAL_CHECKSUM 0x800000 /* Journal checksums */ #define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT 0x1000000 /* Journal Async Commit */ #define EXT4_MOUNT_I_VERSION 0x2000000 /* i_version support */ +#define EXT4_MOUNT_MBALLOC 0x4000000 /* Buddy allocation support */ /* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */ #ifndef _LINUX_EXT2_FS_H #define clear_opt(o, opt) o &= ~EXT4_MOUNT_##opt @@ -912,7 +959,7 @@ extern ext4_fsblk_t ext4_new_blocks (handle_t *handle, struct inode *inode, extern ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode, ext4_fsblk_t goal, unsigned long *count, int *errp); extern void ext4_free_blocks (handle_t *handle, struct inode *inode, - ext4_fsblk_t block, unsigned long count); + ext4_fsblk_t block, unsigned long count, int metadata); extern void ext4_free_blocks_sb (handle_t *handle, struct super_block *sb, ext4_fsblk_t block, unsigned long count, unsigned long *pdquot_freed_blocks); @@ -950,6 +997,20 @@ extern unsigned long ext4_count_dirs (struct super_block *); extern void ext4_check_inodes_bitmap (struct super_block *); extern unsigned long ext4_count_free (struct buffer_head *, unsigned); +/* mballoc.c */ +extern long ext4_mb_stats; +extern long ext4_mb_max_to_scan; +extern int ext4_mb_init(struct super_block *, int); +extern int ext4_mb_release(struct super_block *); +extern ext4_fsblk_t ext4_mb_new_blocks(handle_t *, + struct ext4_allocation_request *, int *); +extern int ext4_mb_reserve_blocks(struct super_block *, int); +extern void ext4_mb_discard_inode_preallocations(struct inode *); +extern int __init init_ext4_mballoc(void); +extern void exit_ext4_mballoc(void); +extern void ext4_mb_free_blocks(handle_t *, struct inode *, + unsigned long, unsigned long, int, unsigned long *); + /* inode.c */ int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode, @@ -1080,6 +1141,19 @@ static inline void ext4_isize_set(struct ext4_inode *raw_inode, loff_t i_size) raw_inode->i_size_high = cpu_to_le32(i_size >> 32); } +static inline +struct ext4_group_info *ext4_get_group_info(struct super_block *sb, + ext4_group_t group) +{ + struct ext4_group_info ***grp_info; + long indexv, indexh; + grp_info = EXT4_SB(sb)->s_group_info; + indexv = group >> (EXT4_DESC_PER_BLOCK_BITS(sb)); + indexh = group & ((EXT4_DESC_PER_BLOCK(sb)) - 1); + return grp_info[indexv][indexh]; +} + + #define ext4_std_error(sb, errno) \ do { \ if ((errno)) \ diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h index 4377d249d378..d5508d3cf290 100644 --- a/include/linux/ext4_fs_i.h +++ b/include/linux/ext4_fs_i.h @@ -158,6 +158,10 @@ struct ext4_inode_info { * struct timespec i_{a,c,m}time in the generic inode. */ struct timespec i_crtime; + + /* mballoc */ + struct list_head i_prealloc_list; + spinlock_t i_prealloc_lock; }; #endif /* _LINUX_EXT4_FS_I */ diff --git a/include/linux/ext4_fs_sb.h b/include/linux/ext4_fs_sb.h index 38a47ec06df9..abaae2c8cccf 100644 --- a/include/linux/ext4_fs_sb.h +++ b/include/linux/ext4_fs_sb.h @@ -91,6 +91,58 @@ struct ext4_sb_info { unsigned long s_ext_blocks; unsigned long s_ext_extents; #endif + + /* for buddy allocator */ + struct ext4_group_info ***s_group_info; + struct inode *s_buddy_cache; + long s_blocks_reserved; + spinlock_t s_reserve_lock; + struct list_head s_active_transaction; + struct list_head s_closed_transaction; + struct list_head s_committed_transaction; + spinlock_t s_md_lock; + tid_t s_last_transaction; + unsigned short *s_mb_offsets, *s_mb_maxs; + + /* tunables */ + unsigned long s_stripe; + unsigned long s_mb_stream_request; + unsigned long s_mb_max_to_scan; + unsigned long s_mb_min_to_scan; + unsigned long s_mb_stats; + unsigned long s_mb_order2_reqs; + unsigned long s_mb_group_prealloc; + /* where last allocation was done - for stream allocation */ + unsigned long s_mb_last_group; + unsigned long s_mb_last_start; + + /* history to debug policy */ + struct ext4_mb_history *s_mb_history; + int s_mb_history_cur; + int s_mb_history_max; + int s_mb_history_num; + struct proc_dir_entry *s_mb_proc; + spinlock_t s_mb_history_lock; + int s_mb_history_filter; + + /* stats for buddy allocator */ + spinlock_t s_mb_pa_lock; + atomic_t s_bal_reqs; /* number of reqs with len > 1 */ + atomic_t s_bal_success; /* we found long enough chunks */ + atomic_t s_bal_allocated; /* in blocks */ + atomic_t s_bal_ex_scanned; /* total extents scanned */ + atomic_t s_bal_goals; /* goal hits */ + atomic_t s_bal_breaks; /* too long searches */ + atomic_t s_bal_2orders; /* 2^order hits */ + spinlock_t s_bal_lock; + unsigned long s_mb_buddies_generated; + unsigned long long s_mb_generation_time; + atomic_t s_mb_lost_chunks; + atomic_t s_mb_preallocated; + atomic_t s_mb_discarded; + + /* locality groups */ + struct ext4_locality_group *s_locality_groups; }; #endif /* _LINUX_EXT4_FS_SB */ -- cgit v1.2.3 From a9d2517c7a5c8028fbc5296d3af3c75597d3d180 Mon Sep 17 00:00:00 2001 From: Dmitri Vorobiev Date: Thu, 24 Jan 2008 19:52:57 +0300 Subject: [MIPS] remove Documentation/mips/GT64120.README Based upon the 2.4 kernel, the information presented in the Documentation/mips/GT64120.README file is outdated. Worse, the document contents are plain misleading nowadays because the text mentions files and directories, which have been deleted, moved or restructured for 2.6. This patch removes the documentation, which is no more valid. Signed-off-by: Dmitri Vorobiev Signed-off-by: Ralf Baechle --- Documentation/mips/00-INDEX | 2 -- Documentation/mips/GT64120.README | 65 --------------------------------------- 2 files changed, 67 deletions(-) delete mode 100644 Documentation/mips/GT64120.README (limited to 'Documentation') diff --git a/Documentation/mips/00-INDEX b/Documentation/mips/00-INDEX index 3f13bf8043d2..8ae9cffc2262 100644 --- a/Documentation/mips/00-INDEX +++ b/Documentation/mips/00-INDEX @@ -2,5 +2,3 @@ - this file. AU1xxx_IDE.README - README for MIPS AU1XXX IDE driver. -GT64120.README - - README for dir with info on MIPS boards using GT-64120 or GT-64120A. diff --git a/Documentation/mips/GT64120.README b/Documentation/mips/GT64120.README deleted file mode 100644 index 2d0eec91dc59..000000000000 --- a/Documentation/mips/GT64120.README +++ /dev/null @@ -1,65 +0,0 @@ -README for arch/mips/gt64120 directory and subdirectories - -Jun Sun, jsun@mvista.com or jsun@junsun.net -01/27, 2001 - -MOTIVATION ----------- - -Many MIPS boards share the same system controller (or CPU companian chip), -such as GT-64120. It is highly desirable to let these boards share -the same controller code instead of duplicating them. - -This directory is meant to hold all MIPS boards that use GT-64120 or GT-64120A. - - -HOW TO ADD A BOARD ------------------- - -. Create a subdirectory include/asm/gt64120/. - -. Create a file called gt64120_dep.h under that directory. - -. Modify include/asm/gt64120/gt64120.h file to include the new gt64120_dep.h - based on config options. The board-dep section is at the end of - include/asm/gt64120/gt64120.h file. There you can find all required - definitions include/asm/gt64120//gt64120_dep.h file must supply. - -. Create a subdirectory arch/mips/gt64120/ directory to hold - board specific routines. - -. The GT-64120 common code is supplied under arch/mips/gt64120/common directory. - It includes: - 1) arch/mips/gt64120/pci.c - - common PCI routine, include the top-level pcibios_init() - 2) arch/mips/gt64120/irq.c - - common IRQ routine, include the top-level do_IRQ() - [This part really belongs to arch/mips/kernel. jsun] - 3) arch/mips/gt64120/gt_irq.c - - common IRQ routines for GT-64120 chip. Currently it only handles - the timer interrupt. - -. Board-specific routines are supplied under arch/mips/gt64120/ dir. - 1) arch/mips/gt64120//pci.c - it provides bus fixup routine - 2) arch/mips/gt64120//irq.c - it provides enable/disable irqs - and board irq setup routine (irq_setup) - 3) arch/mips/gt64120//int-handler.S - - The first-level interrupt dispatching routine. - 4) a bunch of other "normal" stuff (setup, prom, dbg_io, reset, etc) - -. Follow other "normal" procedure to modify configuration files, etc. - - -TO-DO LIST ----------- - -. Expand arch/mips/gt64120/gt_irq.c to handle all GT-64120 interrupts. - We probably need to introduce GT_IRQ_BASE in board-dep header file, - which is used the starting irq_nr for all GT irqs. - - A function, gt64120_handle_irq(), will be added so that the first-level - irq dispatcher will call this function if it detects an interrupt - from GT-64120. - -. More support for GT-64120 PCI features (2nd PCI bus, perhaps) - -- cgit v1.2.3 From ec04b13f67be3c90b38c625f4b8bdfea54c1ff60 Mon Sep 17 00:00:00 2001 From: Balaji Rao Date: Fri, 28 Dec 2007 14:26:24 +0530 Subject: lguest: Reboot support Reboot Implemented (Prevent fd leak, fix style and fix documentation --RR) Signed-off-by: Balaji Rao Signed-off-by: Rusty Russell --- Documentation/lguest/lguest.c | 32 ++++++++++++++++++++++++++++++-- arch/x86/lguest/boot.c | 11 +++++++++-- drivers/lguest/core.c | 2 ++ drivers/lguest/hypercalls.c | 6 ++++-- include/asm-x86/lguest_hcall.h | 6 +++++- 5 files changed, 50 insertions(+), 7 deletions(-) (limited to 'Documentation') diff --git a/Documentation/lguest/lguest.c b/Documentation/lguest/lguest.c index 9b0e322118b5..86cac3e622ab 100644 --- a/Documentation/lguest/lguest.c +++ b/Documentation/lguest/lguest.c @@ -153,6 +153,9 @@ struct virtqueue void (*handle_output)(int fd, struct virtqueue *me); }; +/* Remember the arguments to the program so we can "reboot" */ +static char **main_args; + /* Since guest is UP and we don't run at the same time, we don't need barriers. * But I include them in the code in case others copy it. */ #define wmb() @@ -1489,7 +1492,9 @@ static void setup_block_file(const char *filename) /* Create stack for thread and run it */ stack = malloc(32768); - if (clone(io_thread, stack + 32768, CLONE_VM, dev) == -1) + /* SIGCHLD - We dont "wait" for our cloned thread, so prevent it from + * becoming a zombie. */ + if (clone(io_thread, stack + 32768, CLONE_VM | SIGCHLD, dev) == -1) err(1, "Creating clone"); /* We don't need to keep the I/O thread's end of the pipes open. */ @@ -1499,7 +1504,21 @@ static void setup_block_file(const char *filename) verbose("device %u: virtblock %llu sectors\n", devices.device_num, cap); } -/* That's the end of device setup. */ +/* That's the end of device setup. :*/ + +/* Reboot */ +static void __attribute__((noreturn)) restart_guest(void) +{ + unsigned int i; + + /* Closing pipes causes the waker thread and io_threads to die, and + * closing /dev/lguest cleans up the Guest. Since we don't track all + * open fds, we simply close everything beyond stderr. */ + for (i = 3; i < FD_SETSIZE; i++) + close(i); + execv(main_args[0], main_args); + err(1, "Could not exec %s", main_args[0]); +} /*L:220 Finally we reach the core of the Launcher, which runs the Guest, serves * its input and output, and finally, lays it to rest. */ @@ -1523,6 +1542,9 @@ static void __attribute__((noreturn)) run_guest(int lguest_fd) char reason[1024] = { 0 }; read(lguest_fd, reason, sizeof(reason)-1); errx(1, "%s", reason); + /* ERESTART means that we need to reboot the guest */ + } else if (errno == ERESTART) { + restart_guest(); /* EAGAIN means the Waker wanted us to look at some input. * Anything else means a bug or incompatible change. */ } else if (errno != EAGAIN) @@ -1571,6 +1593,12 @@ int main(int argc, char *argv[]) /* If they specify an initrd file to load. */ const char *initrd_name = NULL; + /* Save the args: we "reboot" by execing ourselves again. */ + main_args = argv; + /* We don't "wait" for the children, so prevent them from becoming + * zombies. */ + signal(SIGCHLD, SIG_IGN); + /* First we initialize the device list. Since console and network * device receive input from a file descriptor, we keep an fdset * (infds) and the maximum fd number (max_infd) with the head of the diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index 92c56117eae5..d6b18e2e5431 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -67,6 +67,7 @@ #include #include #include +#include /* for struct machine_ops */ /*G:010 Welcome to the Guest! * @@ -812,7 +813,7 @@ static void lguest_safe_halt(void) * rather than virtual addresses, so we use __pa() here. */ static void lguest_power_off(void) { - hcall(LHCALL_CRASH, __pa("Power down"), 0, 0); + hcall(LHCALL_SHUTDOWN, __pa("Power down"), LGUEST_SHUTDOWN_POWEROFF, 0); } /* @@ -822,7 +823,7 @@ static void lguest_power_off(void) */ static int lguest_panic(struct notifier_block *nb, unsigned long l, void *p) { - hcall(LHCALL_CRASH, __pa(p), 0, 0); + hcall(LHCALL_SHUTDOWN, __pa(p), LGUEST_SHUTDOWN_POWEROFF, 0); /* The hcall won't return, but to keep gcc happy, we're "done". */ return NOTIFY_DONE; } @@ -926,6 +927,11 @@ static unsigned lguest_patch(u8 type, u16 clobber, void *ibuf, return insn_len; } +static void lguest_restart(char *reason) +{ + hcall(LHCALL_SHUTDOWN, __pa(reason), LGUEST_SHUTDOWN_RESTART, 0); +} + /*G:030 Once we get to lguest_init(), we know we're a Guest. The pv_ops * structures in the kernel provide points for (almost) every routine we have * to override to avoid privileged instructions. */ @@ -1059,6 +1065,7 @@ __init void lguest_init(void) * the Guest routine to power off. */ pm_power_off = lguest_power_off; + machine_ops.restart = lguest_restart; /* Now we're set up, call start_kernel() in init/main.c and we proceed * to boot as normal. It never returns. */ start_kernel(); diff --git a/drivers/lguest/core.c b/drivers/lguest/core.c index f10abc8d7635..c1069bceba11 100644 --- a/drivers/lguest/core.c +++ b/drivers/lguest/core.c @@ -235,6 +235,8 @@ int run_guest(struct lguest *lg, unsigned long __user *user) lguest_arch_handle_trap(lg); } + if (lg->dead == ERR_PTR(-ERESTART)) + return -ERESTART; /* The Guest is dead => "No such file or directory" */ return -ENOENT; } diff --git a/drivers/lguest/hypercalls.c b/drivers/lguest/hypercalls.c index b478affe8f91..05fad6fa8049 100644 --- a/drivers/lguest/hypercalls.c +++ b/drivers/lguest/hypercalls.c @@ -41,8 +41,8 @@ static void do_hcall(struct lguest *lg, struct hcall_args *args) * do that. */ kill_guest(lg, "already have lguest_data"); break; - case LHCALL_CRASH: { - /* Crash is such a trivial hypercall that we do it in four + case LHCALL_SHUTDOWN: { + /* Shutdown is such a trivial hypercall that we do it in four * lines right here. */ char msg[128]; /* If the lgread fails, it will call kill_guest() itself; the @@ -50,6 +50,8 @@ static void do_hcall(struct lguest *lg, struct hcall_args *args) __lgread(lg, msg, args->arg1, sizeof(msg)); msg[sizeof(msg)-1] = '\0'; kill_guest(lg, "CRASH: %s", msg); + if (args->arg2 == LGUEST_SHUTDOWN_RESTART) + lg->dead = ERR_PTR(-ERESTART); break; } case LHCALL_FLUSH_TLB: diff --git a/include/asm-x86/lguest_hcall.h b/include/asm-x86/lguest_hcall.h index 2091779e91fb..758b9a5d4539 100644 --- a/include/asm-x86/lguest_hcall.h +++ b/include/asm-x86/lguest_hcall.h @@ -4,7 +4,7 @@ #define LHCALL_FLUSH_ASYNC 0 #define LHCALL_LGUEST_INIT 1 -#define LHCALL_CRASH 2 +#define LHCALL_SHUTDOWN 2 #define LHCALL_LOAD_GDT 3 #define LHCALL_NEW_PGTABLE 4 #define LHCALL_FLUSH_TLB 5 @@ -20,6 +20,10 @@ #define LGUEST_TRAP_ENTRY 0x1F +/* Argument number 3 to LHCALL_LGUEST_SHUTDOWN */ +#define LGUEST_SHUTDOWN_POWEROFF 1 +#define LGUEST_SHUTDOWN_RESTART 2 + #ifndef __ASSEMBLY__ #include -- cgit v1.2.3 From e3283fa0cc5c4f9bde52339a40da89297e51b481 Mon Sep 17 00:00:00 2001 From: Glauber de Oliveira Costa Date: Mon, 7 Jan 2008 11:05:23 -0200 Subject: lguest: adapt launcher to per-cpuness This patch makes uses of pread() and pwrite() in lguest launcher to communicate the vcpu id to the lguest driver. The id is kept in a thread variable, which means we'll span in the future, vcpus as threads. But right now, only the infrastructure is out there. Signed-off-by: Glauber de Oliveira Costa Signed-off-by: Rusty Russell --- Documentation/lguest/lguest.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) (limited to 'Documentation') diff --git a/Documentation/lguest/lguest.c b/Documentation/lguest/lguest.c index 86cac3e622ab..6c8a2386cd50 100644 --- a/Documentation/lguest/lguest.c +++ b/Documentation/lguest/lguest.c @@ -79,6 +79,9 @@ static void *guest_base; /* The maximum guest physical address allowed, and maximum possible. */ static unsigned long guest_limit, guest_max; +/* a per-cpu variable indicating whose vcpu is currently running */ +static unsigned int __thread cpu_id; + /* This is our list of devices. */ struct device_list { @@ -557,7 +560,7 @@ static void wake_parent(int pipefd, int lguest_fd) else FD_CLR(-fd - 1, &devices.infds); } else /* Send LHREQ_BREAK command. */ - write(lguest_fd, args, sizeof(args)); + pwrite(lguest_fd, args, sizeof(args), cpu_id); } } @@ -1530,7 +1533,8 @@ static void __attribute__((noreturn)) run_guest(int lguest_fd) int readval; /* We read from the /dev/lguest device to run the Guest. */ - readval = read(lguest_fd, ¬ify_addr, sizeof(notify_addr)); + readval = pread(lguest_fd, ¬ify_addr, + sizeof(notify_addr), cpu_id); /* One unsigned long means the Guest did HCALL_NOTIFY */ if (readval == sizeof(notify_addr)) { @@ -1540,7 +1544,7 @@ static void __attribute__((noreturn)) run_guest(int lguest_fd) /* ENOENT means the Guest died. Reading tells us why. */ } else if (errno == ENOENT) { char reason[1024] = { 0 }; - read(lguest_fd, reason, sizeof(reason)-1); + pread(lguest_fd, reason, sizeof(reason)-1, cpu_id); errx(1, "%s", reason); /* ERESTART means that we need to reboot the guest */ } else if (errno == ERESTART) { @@ -1550,9 +1554,13 @@ static void __attribute__((noreturn)) run_guest(int lguest_fd) } else if (errno != EAGAIN) err(1, "Running guest failed"); + /* Only service input on thread for CPU 0. */ + if (cpu_id != 0) + continue; + /* Service input, then unset the BREAK to release the Waker. */ handle_input(lguest_fd); - if (write(lguest_fd, args, sizeof(args)) < 0) + if (pwrite(lguest_fd, args, sizeof(args), cpu_id) < 0) err(1, "Resetting break"); } } @@ -1610,6 +1618,7 @@ int main(int argc, char *argv[]) devices.lastdev = &devices.dev; devices.next_irq = 1; + cpu_id = 0; /* We need to know how much memory so we can set up the device * descriptor and memory pages for the devices as we parse the command * line. So we quickly look through the arguments to find the amount -- cgit v1.2.3 From b02aae9cf52956dfe1bec73f77f81a3d05d3902b Mon Sep 17 00:00:00 2001 From: Rene Herman Date: Wed, 30 Jan 2008 13:30:05 +0100 Subject: x86: provide a DMI based port 0x80 I/O delay override. x86: provide a DMI based port 0x80 I/O delay override. Certain (HP) laptops experience trouble from our port 0x80 I/O delay writes. This patch provides for a DMI based switch to the "alternate diagnostic port" 0xed (as used by some BIOSes as well) for these. David P. Reed confirmed that port 0xed works for him and provides a proper delay. The symptoms of _not_ working are a hanging machine, with "hwclock" use being a direct trigger. Earlier versions of this attempted to simply use udelay(2), with the 2 being a value tested to be a nicely conservative upper-bound with help from many on the linux-kernel mailinglist but that approach has two problems. First, pre-loops_per_jiffy calibration (which is post PIT init while some implementations of the PIT are actually one of the historically problematic devices that need the delay) udelay() isn't particularly well-defined. We could initialise loops_per_jiffy conservatively (and based on CPU family so as to not unduly delay old machines) which would sort of work, but... Second, delaying isn't the only effect that a write to port 0x80 has. It's also a PCI posting barrier which some devices may be explicitly or implicitly relying on. Alan Cox did a survey and found evidence that additionally some drivers may be racy on SMP without the bus locking outb. Switching to an inb() makes the timing too unpredictable and as such, this DMI based switch should be the safest approach for now. Any more invasive changes should get more rigid testing first. It's moreover only very few machines with the problem and a DMI based hack seems to fit that situation. This also introduces a command-line parameter "io_delay" to override the DMI based choice again: io_delay= where "standard" means using the standard port 0x80 and "alternate" port 0xed. This retains the udelay method as a config (CONFIG_UDELAY_IO_DELAY) and command-line ("io_delay=udelay") choice for testing purposes as well. This does not change the io_delay() in the boot code which is using the same port 0x80 I/O delay but those do not appear to be a problem as David P. Reed reported the problem was already gone after using the udelay version. He moreover reported that booting with "acpi=off" also fixed things and seeing as how ACPI isn't touched until after this DMI based I/O port switch I believe it's safe to leave the ones in the boot code be. The DMI strings from David's HP Pavilion dv9000z are in there already and we need to get/verify the DMI info from other machines with the problem, notably the HP Pavilion dv6000z. This patch is partly based on earlier patches from Pavel Machek and David P. Reed. Signed-off-by: Rene Herman Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner --- Documentation/kernel-parameters.txt | 8 +++ arch/x86/Kconfig.debug | 9 +++ arch/x86/boot/compressed/misc_32.c | 8 +-- arch/x86/boot/compressed/misc_64.c | 8 +-- arch/x86/kernel/Makefile_32 | 2 +- arch/x86/kernel/Makefile_64 | 2 +- arch/x86/kernel/io_delay.c | 106 ++++++++++++++++++++++++++++++++++++ arch/x86/kernel/setup_32.c | 2 + arch/x86/kernel/setup_64.c | 2 + include/asm-x86/io_32.h | 8 ++- include/asm-x86/io_64.h | 33 +++++++---- 11 files changed, 165 insertions(+), 23 deletions(-) create mode 100644 arch/x86/kernel/io_delay.c (limited to 'Documentation') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 880f882160e2..9e6056058425 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -794,6 +794,14 @@ and is between 256 and 4096 characters. It is defined in the file for translation below 32 bit and if not available then look in the higher range. + io_delay= [X86-32,X86-64] I/O delay method + standard + Standard port 0x80 delay + alternate + Alternate port 0xed delay + udelay + Simple two microsecond delay + io7= [HW] IO7 for Marvel based alpha systems See comment before marvel_specify_io7 in arch/alpha/kernel/core_marvel.c. diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index 761ca7b5f120..40aba670fb37 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -112,4 +112,13 @@ config IOMMU_LEAK Add a simple leak tracer to the IOMMU code. This is useful when you are debugging a buggy device driver that leaks IOMMU mappings. +config UDELAY_IO_DELAY + bool "Delay I/O through udelay instead of outb" + depends on DEBUG_KERNEL + help + Make inb_p/outb_p use udelay() based delays by default. Please note + that udelay() does not have the same bus-level side-effects that + the normal outb based delay does meaning this could cause drivers + to change behaviour and/or bugs to surface. + endmenu diff --git a/arch/x86/boot/compressed/misc_32.c b/arch/x86/boot/compressed/misc_32.c index b74d60d1b2fa..288e16283ef9 100644 --- a/arch/x86/boot/compressed/misc_32.c +++ b/arch/x86/boot/compressed/misc_32.c @@ -276,10 +276,10 @@ static void putstr(const char *s) RM_SCREEN_INFO.orig_y = y; pos = (x + cols * y) * 2; /* Update cursor position */ - outb_p(14, vidport); - outb_p(0xff & (pos >> 9), vidport+1); - outb_p(15, vidport); - outb_p(0xff & (pos >> 1), vidport+1); + outb(14, vidport); + outb(0xff & (pos >> 9), vidport+1); + outb(15, vidport); + outb(0xff & (pos >> 1), vidport+1); } static void* memset(void* s, int c, unsigned n) diff --git a/arch/x86/boot/compressed/misc_64.c b/arch/x86/boot/compressed/misc_64.c index 6ea015aa65e4..43e5fcc37be9 100644 --- a/arch/x86/boot/compressed/misc_64.c +++ b/arch/x86/boot/compressed/misc_64.c @@ -269,10 +269,10 @@ static void putstr(const char *s) RM_SCREEN_INFO.orig_y = y; pos = (x + cols * y) * 2; /* Update cursor position */ - outb_p(14, vidport); - outb_p(0xff & (pos >> 9), vidport+1); - outb_p(15, vidport); - outb_p(0xff & (pos >> 1), vidport+1); + outb(14, vidport); + outb(0xff & (pos >> 9), vidport+1); + outb(15, vidport); + outb(0xff & (pos >> 1), vidport+1); } static void* memset(void* s, int c, unsigned n) diff --git a/arch/x86/kernel/Makefile_32 b/arch/x86/kernel/Makefile_32 index a7bc93c27662..0cc1981d1e38 100644 --- a/arch/x86/kernel/Makefile_32 +++ b/arch/x86/kernel/Makefile_32 @@ -8,7 +8,7 @@ CPPFLAGS_vmlinux.lds += -Ui386 obj-y := process_32.o signal_32.o entry_32.o traps_32.o irq_32.o \ ptrace_32.o time_32.o ioport_32.o ldt_32.o setup_32.o i8259_32.o sys_i386_32.o \ pci-dma_32.o i386_ksyms_32.o i387_32.o bootflag.o e820_32.o\ - quirks.o i8237.o topology.o alternative.o i8253.o tsc_32.o + quirks.o i8237.o topology.o alternative.o i8253.o tsc_32.o io_delay.o obj-$(CONFIG_STACKTRACE) += stacktrace.o obj-y += cpu/ diff --git a/arch/x86/kernel/Makefile_64 b/arch/x86/kernel/Makefile_64 index 5a88890d8ee9..08a68f0d8fda 100644 --- a/arch/x86/kernel/Makefile_64 +++ b/arch/x86/kernel/Makefile_64 @@ -11,7 +11,7 @@ obj-y := process_64.o signal_64.o entry_64.o traps_64.o irq_64.o \ x8664_ksyms_64.o i387_64.o syscall_64.o vsyscall_64.o \ setup64.o bootflag.o e820_64.o reboot_64.o quirks.o i8237.o \ pci-dma_64.o pci-nommu_64.o alternative.o hpet.o tsc_64.o bugs_64.o \ - i8253.o + i8253.o io_delay.o obj-$(CONFIG_STACKTRACE) += stacktrace.o obj-y += cpu/ diff --git a/arch/x86/kernel/io_delay.c b/arch/x86/kernel/io_delay.c new file mode 100644 index 000000000000..4d955e74b974 --- /dev/null +++ b/arch/x86/kernel/io_delay.c @@ -0,0 +1,106 @@ +/* + * I/O delay strategies for inb_p/outb_p + */ +#include +#include +#include +#include +#include +#include + +/* + * Allow for a DMI based override of port 0x80 needed for certain HP laptops + */ +#define IO_DELAY_PORT_STD 0x80 +#define IO_DELAY_PORT_ALT 0xed + +static void standard_io_delay(void) +{ + asm volatile ("outb %%al, %0" : : "N" (IO_DELAY_PORT_STD)); +} + +static void alternate_io_delay(void) +{ + asm volatile ("outb %%al, %0" : : "N" (IO_DELAY_PORT_ALT)); +} + +/* + * 2 usecs is an upper-bound for the outb delay but note that udelay doesn't + * have the bus-level side-effects that outb does + */ +#define IO_DELAY_USECS 2 + +/* + * High on a hill was a lonely goatherd + */ +static void udelay_io_delay(void) +{ + udelay(IO_DELAY_USECS); +} + +#ifndef CONFIG_UDELAY_IO_DELAY +static void (*io_delay)(void) = standard_io_delay; +#else +static void (*io_delay)(void) = udelay_io_delay; +#endif + +/* + * Paravirt wants native_io_delay to be a constant. + */ +void native_io_delay(void) +{ + io_delay(); +} +EXPORT_SYMBOL(native_io_delay); + +#ifndef CONFIG_UDELAY_IO_DELAY +static int __init dmi_alternate_io_delay_port(const struct dmi_system_id *id) +{ + printk(KERN_NOTICE "%s: using alternate I/O delay port\n", id->ident); + io_delay = alternate_io_delay; + return 0; +} + +static struct dmi_system_id __initdata alternate_io_delay_port_dmi_table[] = { + { + .callback = dmi_alternate_io_delay_port, + .ident = "HP Pavilion dv9000z", + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "Quanta"), + DMI_MATCH(DMI_BOARD_NAME, "30B9") + } + }, + { + } +}; + +static int __initdata io_delay_override; + +void __init io_delay_init(void) +{ + if (!io_delay_override) + dmi_check_system(alternate_io_delay_port_dmi_table); +} +#endif + +static int __init io_delay_param(char *s) +{ + if (!s) + return -EINVAL; + + if (!strcmp(s, "standard")) + io_delay = standard_io_delay; + else if (!strcmp(s, "alternate")) + io_delay = alternate_io_delay; + else if (!strcmp(s, "udelay")) + io_delay = udelay_io_delay; + else + return -EINVAL; + +#ifndef CONFIG_UDELAY_IO_DELAY + io_delay_override = 1; +#endif + return 0; +} + +early_param("io_delay", io_delay_param); diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c index 9c24b45b513c..51bdc0b1b72e 100644 --- a/arch/x86/kernel/setup_32.c +++ b/arch/x86/kernel/setup_32.c @@ -648,6 +648,8 @@ void __init setup_arch(char **cmdline_p) dmi_scan_machine(); + io_delay_init();; + #ifdef CONFIG_X86_GENERICARCH generic_apic_probe(); #endif diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c index 30d94d1d5f5f..ec976edf0399 100644 --- a/arch/x86/kernel/setup_64.c +++ b/arch/x86/kernel/setup_64.c @@ -311,6 +311,8 @@ void __init setup_arch(char **cmdline_p) dmi_scan_machine(); + io_delay_init(); + #ifdef CONFIG_SMP /* setup to use the static apicid table during kernel startup */ x86_cpu_to_apicid_ptr = (void *)&x86_cpu_to_apicid_init; diff --git a/include/asm-x86/io_32.h b/include/asm-x86/io_32.h index fe881cd1e6f4..a8d25c38b91c 100644 --- a/include/asm-x86/io_32.h +++ b/include/asm-x86/io_32.h @@ -250,10 +250,14 @@ static inline void flush_write_buffers(void) #endif /* __KERNEL__ */ -static inline void native_io_delay(void) +#ifndef CONFIG_UDELAY_IO_DELAY +extern void io_delay_init(void); +#else +static inline void io_delay_init(void) { - asm volatile("outb %%al,$0x80" : : : "memory"); } +#endif +extern void native_io_delay(void); #if defined(CONFIG_PARAVIRT) #include diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h index a037b0794332..5bebaf961692 100644 --- a/include/asm-x86/io_64.h +++ b/include/asm-x86/io_64.h @@ -35,13 +35,24 @@ * - Arnaldo Carvalho de Melo */ -#define __SLOW_DOWN_IO "\noutb %%al,$0x80" +#ifndef CONFIG_UDELAY_IO_DELAY +extern void io_delay_init(void); +#else +static inline void io_delay_init(void) +{ +} +#endif +extern void native_io_delay(void); +static inline void slow_down_io(void) +{ + native_io_delay(); #ifdef REALLY_SLOW_IO -#define __FULL_SLOW_DOWN_IO __SLOW_DOWN_IO __SLOW_DOWN_IO __SLOW_DOWN_IO __SLOW_DOWN_IO -#else -#define __FULL_SLOW_DOWN_IO __SLOW_DOWN_IO + native_io_delay(); + native_io_delay(); + native_io_delay(); #endif +} /* * Talk about misusing macros.. @@ -50,21 +61,21 @@ static inline void out##s(unsigned x value, unsigned short port) { #define __OUT2(s,s1,s2) \ -__asm__ __volatile__ ("out" #s " %" s1 "0,%" s2 "1" +__asm__ __volatile__ ("out" #s " %" s1 "0,%" s2 "1" : : "a" (value), "Nd" (port)) #define __OUT(s,s1,x) \ -__OUT1(s,x) __OUT2(s,s1,"w") : : "a" (value), "Nd" (port)); } \ -__OUT1(s##_p,x) __OUT2(s,s1,"w") __FULL_SLOW_DOWN_IO : : "a" (value), "Nd" (port));} \ +__OUT1(s,x) __OUT2(s,s1,"w"); } \ +__OUT1(s##_p,x) __OUT2(s,s1,"w"); slow_down_io(); } #define __IN1(s) \ static inline RETURN_TYPE in##s(unsigned short port) { RETURN_TYPE _v; #define __IN2(s,s1,s2) \ -__asm__ __volatile__ ("in" #s " %" s2 "1,%" s1 "0" +__asm__ __volatile__ ("in" #s " %" s2 "1,%" s1 "0" : "=a" (_v) : "Nd" (port)) -#define __IN(s,s1,i...) \ -__IN1(s) __IN2(s,s1,"w") : "=a" (_v) : "Nd" (port) ,##i ); return _v; } \ -__IN1(s##_p) __IN2(s,s1,"w") __FULL_SLOW_DOWN_IO : "=a" (_v) : "Nd" (port) ,##i ); return _v; } \ +#define __IN(s,s1) \ +__IN1(s) __IN2(s,s1,"w"); return _v; } \ +__IN1(s##_p) __IN2(s,s1,"w"); slow_down_io(); return _v; } #define __INS(s) \ static inline void ins##s(unsigned short port, void * addr, unsigned long count) \ -- cgit v1.2.3 From 6e7c402590b75b6b45138792445ee0f0315a8473 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 30 Jan 2008 13:30:05 +0100 Subject: x86: various changes and cleanups to in_p/out_p delay details various changes to the in_p/out_p delay details: - add the io_delay=none method - make each method selectable from the kernel config - simplify the delay code a bit by getting rid of an indirect function call - add the /proc/sys/kernel/io_delay_type sysctl - change 'io_delay=standard|alternate' to io_delay=0x80 and io_delay=0xed - make the io delay config not depend on CONFIG_DEBUG_KERNEL Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner Tested-by: "David P. Reed" --- Documentation/kernel-parameters.txt | 12 ++-- arch/x86/Kconfig.debug | 79 ++++++++++++++++++++++++--- arch/x86/kernel/io_delay.c | 106 +++++++++++++++++------------------- include/asm-x86/io_32.h | 10 +--- include/asm-x86/io_64.h | 10 +--- kernel/sysctl.c | 9 +++ 6 files changed, 143 insertions(+), 83 deletions(-) (limited to 'Documentation') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 9e6056058425..b427b7c0e5d0 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -795,12 +795,14 @@ and is between 256 and 4096 characters. It is defined in the file then look in the higher range. io_delay= [X86-32,X86-64] I/O delay method - standard - Standard port 0x80 delay - alternate - Alternate port 0xed delay + 0x80 + Standard port 0x80 based delay + 0xed + Alternate port 0xed based delay (needed on some systems) udelay - Simple two microsecond delay + Simple two microseconds delay + none + No delay io7= [HW] IO7 for Marvel based alpha systems See comment before marvel_specify_io7 in diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index 40aba670fb37..77eda46f97b8 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -112,13 +112,78 @@ config IOMMU_LEAK Add a simple leak tracer to the IOMMU code. This is useful when you are debugging a buggy device driver that leaks IOMMU mappings. -config UDELAY_IO_DELAY - bool "Delay I/O through udelay instead of outb" - depends on DEBUG_KERNEL +# +# IO delay types: +# + +config IO_DELAY_TYPE_0X80 + int + default "0" + +config IO_DELAY_TYPE_0XED + int + default "1" + +config IO_DELAY_TYPE_UDELAY + int + default "2" + +config IO_DELAY_TYPE_NONE + int + default "3" + +choice + prompt "IO delay type" + default IO_DELAY_0X80 + +config IO_DELAY_0X80 + bool "port 0x80 based port-IO delay [recommended]" + help + This is the traditional Linux IO delay used for in/out_p. + It is the most tested hence safest selection here. + +config IO_DELAY_0XED + bool "port 0xed based port-IO delay" + help + Use port 0xed as the IO delay. This frees up port 0x80 which is + often used as a hardware-debug port. + +config IO_DELAY_UDELAY + bool "udelay based port-IO delay" + help + Use udelay(2) as the IO delay method. This provides the delay + while not having any side-effect on the IO port space. + +config IO_DELAY_NONE + bool "no port-IO delay" help - Make inb_p/outb_p use udelay() based delays by default. Please note - that udelay() does not have the same bus-level side-effects that - the normal outb based delay does meaning this could cause drivers - to change behaviour and/or bugs to surface. + No port-IO delay. Will break on old boxes that require port-IO + delay for certain operations. Should work on most new machines. + +endchoice + +if IO_DELAY_0X80 +config DEFAULT_IO_DELAY_TYPE + int + default IO_DELAY_TYPE_0X80 +endif + +if IO_DELAY_0XED +config DEFAULT_IO_DELAY_TYPE + int + default IO_DELAY_TYPE_0XED +endif + +if IO_DELAY_UDELAY +config DEFAULT_IO_DELAY_TYPE + int + default IO_DELAY_TYPE_UDELAY +endif + +if IO_DELAY_NONE +config DEFAULT_IO_DELAY_TYPE + int + default IO_DELAY_TYPE_NONE +endif endmenu diff --git a/arch/x86/kernel/io_delay.c b/arch/x86/kernel/io_delay.c index 4d955e74b974..f052e34dc94c 100644 --- a/arch/x86/kernel/io_delay.c +++ b/arch/x86/kernel/io_delay.c @@ -1,5 +1,9 @@ /* * I/O delay strategies for inb_p/outb_p + * + * Allow for a DMI based override of port 0x80, needed for certain HP laptops + * and possibly other systems. Also allow for the gradual elimination of + * outb_p/inb_p API uses. */ #include #include @@ -8,98 +12,86 @@ #include #include -/* - * Allow for a DMI based override of port 0x80 needed for certain HP laptops - */ -#define IO_DELAY_PORT_STD 0x80 -#define IO_DELAY_PORT_ALT 0xed - -static void standard_io_delay(void) -{ - asm volatile ("outb %%al, %0" : : "N" (IO_DELAY_PORT_STD)); -} - -static void alternate_io_delay(void) -{ - asm volatile ("outb %%al, %0" : : "N" (IO_DELAY_PORT_ALT)); -} - -/* - * 2 usecs is an upper-bound for the outb delay but note that udelay doesn't - * have the bus-level side-effects that outb does - */ -#define IO_DELAY_USECS 2 - -/* - * High on a hill was a lonely goatherd - */ -static void udelay_io_delay(void) -{ - udelay(IO_DELAY_USECS); -} +int io_delay_type __read_mostly = CONFIG_DEFAULT_IO_DELAY_TYPE; +EXPORT_SYMBOL_GPL(io_delay_type); -#ifndef CONFIG_UDELAY_IO_DELAY -static void (*io_delay)(void) = standard_io_delay; -#else -static void (*io_delay)(void) = udelay_io_delay; -#endif +static int __initdata io_delay_override; /* * Paravirt wants native_io_delay to be a constant. */ void native_io_delay(void) { - io_delay(); + switch (io_delay_type) { + default: + case CONFIG_IO_DELAY_TYPE_0X80: + asm volatile ("outb %al, $0x80"); + break; + case CONFIG_IO_DELAY_TYPE_0XED: + asm volatile ("outb %al, $0xed"); + break; + case CONFIG_IO_DELAY_TYPE_UDELAY: + /* + * 2 usecs is an upper-bound for the outb delay but + * note that udelay doesn't have the bus-level + * side-effects that outb does, nor does udelay() have + * precise timings during very early bootup (the delays + * are shorter until calibrated): + */ + udelay(2); + case CONFIG_IO_DELAY_TYPE_NONE: + break; + } } EXPORT_SYMBOL(native_io_delay); -#ifndef CONFIG_UDELAY_IO_DELAY -static int __init dmi_alternate_io_delay_port(const struct dmi_system_id *id) +static int __init dmi_io_delay_0xed_port(const struct dmi_system_id *id) { - printk(KERN_NOTICE "%s: using alternate I/O delay port\n", id->ident); - io_delay = alternate_io_delay; + if (io_delay_type == CONFIG_IO_DELAY_TYPE_0X80) { + printk(KERN_NOTICE "%s: using 0xed I/O delay port\n", + id->ident); + io_delay_type = CONFIG_IO_DELAY_TYPE_0XED; + } + return 0; } -static struct dmi_system_id __initdata alternate_io_delay_port_dmi_table[] = { +/* + * Quirk table for systems that misbehave (lock up, etc.) if port + * 0x80 is used: + */ +static struct dmi_system_id __initdata io_delay_0xed_port_dmi_table[] = { { - .callback = dmi_alternate_io_delay_port, + .callback = dmi_io_delay_0xed_port, .ident = "HP Pavilion dv9000z", .matches = { DMI_MATCH(DMI_BOARD_VENDOR, "Quanta"), DMI_MATCH(DMI_BOARD_NAME, "30B9") } }, - { - } + { } }; -static int __initdata io_delay_override; - void __init io_delay_init(void) { if (!io_delay_override) - dmi_check_system(alternate_io_delay_port_dmi_table); + dmi_check_system(io_delay_0xed_port_dmi_table); } -#endif static int __init io_delay_param(char *s) { - if (!s) - return -EINVAL; - - if (!strcmp(s, "standard")) - io_delay = standard_io_delay; - else if (!strcmp(s, "alternate")) - io_delay = alternate_io_delay; + if (!strcmp(s, "0x80")) + io_delay_type = CONFIG_IO_DELAY_TYPE_0X80; + else if (!strcmp(s, "0xed")) + io_delay_type = CONFIG_IO_DELAY_TYPE_0XED; else if (!strcmp(s, "udelay")) - io_delay = udelay_io_delay; + io_delay_type = CONFIG_IO_DELAY_TYPE_UDELAY; + else if (!strcmp(s, "none")) + io_delay_type = CONFIG_IO_DELAY_TYPE_NONE; else return -EINVAL; -#ifndef CONFIG_UDELAY_IO_DELAY io_delay_override = 1; -#endif return 0; } diff --git a/include/asm-x86/io_32.h b/include/asm-x86/io_32.h index a8d25c38b91c..2a04bd17eac5 100644 --- a/include/asm-x86/io_32.h +++ b/include/asm-x86/io_32.h @@ -250,15 +250,11 @@ static inline void flush_write_buffers(void) #endif /* __KERNEL__ */ -#ifndef CONFIG_UDELAY_IO_DELAY -extern void io_delay_init(void); -#else -static inline void io_delay_init(void) -{ -} -#endif extern void native_io_delay(void); +extern int io_delay_type; +extern void io_delay_init(void); + #if defined(CONFIG_PARAVIRT) #include #else diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h index 5bebaf961692..dbcc03aa1c6a 100644 --- a/include/asm-x86/io_64.h +++ b/include/asm-x86/io_64.h @@ -35,15 +35,11 @@ * - Arnaldo Carvalho de Melo */ -#ifndef CONFIG_UDELAY_IO_DELAY -extern void io_delay_init(void); -#else -static inline void io_delay_init(void) -{ -} -#endif extern void native_io_delay(void); +extern int io_delay_type; +extern void io_delay_init(void); + static inline void slow_down_io(void) { native_io_delay(); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 4bc8e48434a7..357b68ba23ec 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -53,6 +53,7 @@ #ifdef CONFIG_X86 #include #include +#include #endif static int deprecated_sysctl_warning(struct __sysctl_args *args); @@ -727,6 +728,14 @@ static struct ctl_table kern_table[] = { .mode = 0644, .proc_handler = &proc_dointvec, }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "io_delay_type", + .data = &io_delay_type, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, #endif #if defined(CONFIG_MMU) { -- cgit v1.2.3 From af65d64845a90c8f2fc90b97e2148ff74672e979 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Wed, 30 Jan 2008 13:30:43 +0100 Subject: x86 vDSO: consolidate vdso32 This makes x86_64's ia32 emulation support share the sources used in the 32-bit kernel for the 32-bit vDSO and much of its setup code. The 32-bit vDSO mapping now behaves the same on x86_64 as on native 32-bit. The abi.syscall32 sysctl on x86_64 now takes the same values that vm.vdso_enabled takes on the 32-bit kernel. That is, 1 means a randomized vDSO location, 2 means the fixed old address. The CONFIG_COMPAT_VDSO option is now available to make this the default setting, the same meaning it has for the 32-bit kernel. (This does not affect the 64-bit vDSO.) The argument vdso32=[012] can be used on both 32-bit and 64-bit kernels to set this paramter at boot time. The vdso=[012] argument still does this same thing on the 32-bit kernel. Signed-off-by: Roland McGrath Cc: Andrew Morton Cc: Linus Torvalds Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner --- Documentation/kernel-parameters.txt | 5 ++ arch/x86/Kconfig | 4 +- arch/x86/ia32/Makefile | 2 +- arch/x86/ia32/ia32_binfmt.c | 9 ++- arch/x86/ia32/ia32_signal.c | 22 ++++--- arch/x86/vdso/Makefile | 2 +- arch/x86/vdso/vdso32-setup.c | 119 ++++++++++++++++++++++++++++-------- arch/x86/vdso/vdso32.S | 16 +++-- arch/x86/xen/setup.c | 4 +- include/asm-x86/elf.h | 23 ++++--- 10 files changed, 146 insertions(+), 60 deletions(-) (limited to 'Documentation') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index b427b7c0e5d0..b3f20beea411 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1988,6 +1988,11 @@ and is between 256 and 4096 characters. It is defined in the file vdso=1: enable VDSO (default) vdso=0: disable VDSO mapping + vdso32= [X86-32,X86-64] + vdso32=2: enable compat VDSO (default with COMPAT_VDSO) + vdso32=1: enable 32-bit VDSO (default) + vdso32=0: disable 32-bit VDSO mapping + vector= [IA-64,SMP] vector=percpu: enable percpu vector domain diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 80b7ba4056db..2f4d88babd36 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1191,9 +1191,9 @@ config HOTPLUG_CPU config COMPAT_VDSO bool "Compat VDSO support" default y - depends on X86_32 + depends on X86_32 || IA32_EMULATION help - Map the VDSO to the predictable old-style address too. + Map the 32-bit VDSO to the predictable old-style address too. ---help--- Say N here if you are running a sufficiently recent glibc version (2.3.3 or later), to remove the high-mapped diff --git a/arch/x86/ia32/Makefile b/arch/x86/ia32/Makefile index a3c997e9f39a..1f58a21a41dc 100644 --- a/arch/x86/ia32/Makefile +++ b/arch/x86/ia32/Makefile @@ -3,7 +3,7 @@ # obj-$(CONFIG_IA32_EMULATION) := ia32entry.o sys_ia32.o ia32_signal.o tls32.o \ - ia32_binfmt.o fpu32.o ptrace32.o syscall32.o syscall32_syscall.o + ia32_binfmt.o fpu32.o ptrace32.o sysv-$(CONFIG_SYSVIPC) := ipc32.o obj-$(CONFIG_IA32_EMULATION) += $(sysv-y) diff --git a/arch/x86/ia32/ia32_binfmt.c b/arch/x86/ia32/ia32_binfmt.c index 55822d2cf053..e32974c3dd3b 100644 --- a/arch/x86/ia32/ia32_binfmt.c +++ b/arch/x86/ia32/ia32_binfmt.c @@ -26,7 +26,7 @@ #include #include #include -#include +#include #undef ELF_ARCH #undef ELF_CLASS @@ -47,14 +47,13 @@ #define AT_SYSINFO 32 #define AT_SYSINFO_EHDR 33 -int sysctl_vsyscall32 = 1; +extern int sysctl_vsyscall32; #undef ARCH_DLINFO #define ARCH_DLINFO do { \ if (sysctl_vsyscall32) { \ - current->mm->context.vdso = (void *)VSYSCALL32_BASE; \ - NEW_AUX_ENT(AT_SYSINFO, (u32)(u64)VSYSCALL32_VSYSCALL); \ - NEW_AUX_ENT(AT_SYSINFO_EHDR, VSYSCALL32_BASE); \ + NEW_AUX_ENT(AT_SYSINFO, (u32)VDSO_ENTRY); \ + NEW_AUX_ENT(AT_SYSINFO_EHDR, (u32)VDSO_CURRENT_BASE); \ } \ } while(0) diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c index 0fc5d8563e19..39356a756b28 100644 --- a/arch/x86/ia32/ia32_signal.c +++ b/arch/x86/ia32/ia32_signal.c @@ -31,7 +31,7 @@ #include #include #include -#include +#include #define DEBUG_SIG 0 @@ -465,13 +465,16 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka, goto give_sigsegv; } - /* Return stub is in 32bit vsyscall page */ - if (current->binfmt->hasvdso) - restorer = VSYSCALL32_SIGRETURN; - else - restorer = (void *)&frame->retcode; - if (ka->sa.sa_flags & SA_RESTORER) + if (ka->sa.sa_flags & SA_RESTORER) { restorer = ka->sa.sa_restorer; + } else { + /* Return stub is in 32bit vsyscall page */ + if (current->binfmt->hasvdso) + restorer = VDSO32_SYMBOL(current->mm->context.vdso, + sigreturn); + else + restorer = (void *)&frame->retcode; + } err |= __put_user(ptr_to_compat(restorer), &frame->pretcode); /* @@ -519,7 +522,7 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, { struct rt_sigframe __user *frame; struct exec_domain *ed = current_thread_info()->exec_domain; - void __user *restorer = VSYSCALL32_RTSIGRETURN; + void __user *restorer; int err = 0; /* __copy_to_user optimizes that into a single 8 byte store */ @@ -564,6 +567,9 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, if (ka->sa.sa_flags & SA_RESTORER) restorer = ka->sa.sa_restorer; + else + restorer = VDSO32_SYMBOL(current->mm->context.vdso, + rt_sigreturn); err |= __put_user(ptr_to_compat(restorer), &frame->pretcode); /* diff --git a/arch/x86/vdso/Makefile b/arch/x86/vdso/Makefile index 1127c716df02..47bc2760e6a8 100644 --- a/arch/x86/vdso/Makefile +++ b/arch/x86/vdso/Makefile @@ -15,7 +15,7 @@ vobjs-y := vdso-note.o vclock_gettime.o vgetcpu.o vvar.o # files to link into kernel obj-$(VDSO64-y) += vma.o vdso.o -obj-$(CONFIG_X86_32) += vdso32.o vdso32-setup.o +obj-$(VDSO32-y) += vdso32.o vdso32-setup.o vobjs := $(foreach F,$(vobjs-y),$(obj)/$F) diff --git a/arch/x86/vdso/vdso32-setup.c b/arch/x86/vdso/vdso32-setup.c index fb71a93c5dce..d97a6d7d062b 100644 --- a/arch/x86/vdso/vdso32-setup.c +++ b/arch/x86/vdso/vdso32-setup.c @@ -24,6 +24,7 @@ #include #include #include +#include enum { VDSO_DISABLED = 0, @@ -37,14 +38,24 @@ enum { #define VDSO_DEFAULT VDSO_ENABLED #endif +#ifdef CONFIG_X86_64 +#define vdso_enabled sysctl_vsyscall32 +#define arch_setup_additional_pages syscall32_setup_pages +#endif + +/* + * This is the difference between the prelinked addresses in the vDSO images + * and the VDSO_HIGH_BASE address where CONFIG_COMPAT_VDSO places the vDSO + * in the user address space. + */ +#define VDSO_ADDR_ADJUST (VDSO_HIGH_BASE - (unsigned long)VDSO32_PRELINK) + /* * Should the kernel map a VDSO page into processes and pass its * address down to glibc upon exec()? */ unsigned int __read_mostly vdso_enabled = VDSO_DEFAULT; -EXPORT_SYMBOL_GPL(vdso_enabled); - static int __init vdso_setup(char *s) { vdso_enabled = simple_strtoul(s, NULL, 0); @@ -52,9 +63,18 @@ static int __init vdso_setup(char *s) return 1; } -__setup("vdso=", vdso_setup); +/* + * For consistency, the argument vdso32=[012] affects the 32-bit vDSO + * behavior on both 64-bit and 32-bit kernels. + * On 32-bit kernels, vdso=[012] means the same thing. + */ +__setup("vdso32=", vdso_setup); -extern asmlinkage void ia32_sysenter_target(void); +#ifdef CONFIG_X86_32 +__setup_param("vdso=", vdso32_setup, vdso_setup, 0); + +EXPORT_SYMBOL_GPL(vdso_enabled); +#endif static __init void reloc_symtab(Elf32_Ehdr *ehdr, unsigned offset, unsigned size) @@ -79,7 +99,7 @@ static __init void reloc_symtab(Elf32_Ehdr *ehdr, case STT_FUNC: case STT_SECTION: case STT_FILE: - sym->st_value += VDSO_HIGH_BASE; + sym->st_value += VDSO_ADDR_ADJUST; } } } @@ -105,7 +125,7 @@ static __init void reloc_dyn(Elf32_Ehdr *ehdr, unsigned offset) case DT_VERNEED: case DT_ADDRRNGLO ... DT_ADDRRNGHI: /* definitely pointers needing relocation */ - dyn->d_un.d_ptr += VDSO_HIGH_BASE; + dyn->d_un.d_ptr += VDSO_ADDR_ADJUST; break; case DT_ENCODING ... OLD_DT_LOOS-1: @@ -114,7 +134,7 @@ static __init void reloc_dyn(Elf32_Ehdr *ehdr, unsigned offset) they're even */ if (dyn->d_tag >= DT_ENCODING && (dyn->d_tag & 1) == 0) - dyn->d_un.d_ptr += VDSO_HIGH_BASE; + dyn->d_un.d_ptr += VDSO_ADDR_ADJUST; break; case DT_VERDEFNUM: @@ -143,15 +163,15 @@ static __init void relocate_vdso(Elf32_Ehdr *ehdr) int i; BUG_ON(memcmp(ehdr->e_ident, ELFMAG, 4) != 0 || - !elf_check_arch(ehdr) || + !elf_check_arch_ia32(ehdr) || ehdr->e_type != ET_DYN); - ehdr->e_entry += VDSO_HIGH_BASE; + ehdr->e_entry += VDSO_ADDR_ADJUST; /* rebase phdrs */ phdr = (void *)ehdr + ehdr->e_phoff; for (i = 0; i < ehdr->e_phnum; i++) { - phdr[i].p_vaddr += VDSO_HIGH_BASE; + phdr[i].p_vaddr += VDSO_ADDR_ADJUST; /* relocate dynamic stuff */ if (phdr[i].p_type == PT_DYNAMIC) @@ -164,7 +184,7 @@ static __init void relocate_vdso(Elf32_Ehdr *ehdr) if (!(shdr[i].sh_flags & SHF_ALLOC)) continue; - shdr[i].sh_addr += VDSO_HIGH_BASE; + shdr[i].sh_addr += VDSO_ADDR_ADJUST; if (shdr[i].sh_type == SHT_SYMTAB || shdr[i].sh_type == SHT_DYNSYM) @@ -173,6 +193,45 @@ static __init void relocate_vdso(Elf32_Ehdr *ehdr) } } +/* + * These symbols are defined by vdso32.S to mark the bounds + * of the ELF DSO images included therein. + */ +extern const char vdso32_default_start, vdso32_default_end; +extern const char vdso32_sysenter_start, vdso32_sysenter_end; +static struct page *vdso32_pages[1]; + +#ifdef CONFIG_X86_64 + +static int use_sysenter __read_mostly = -1; + +#define vdso32_sysenter() (use_sysenter > 0) + +/* May not be __init: called during resume */ +void syscall32_cpu_init(void) +{ + if (use_sysenter < 0) + use_sysenter = (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL); + + /* Load these always in case some future AMD CPU supports + SYSENTER from compat mode too. */ + checking_wrmsrl(MSR_IA32_SYSENTER_CS, (u64)__KERNEL_CS); + checking_wrmsrl(MSR_IA32_SYSENTER_ESP, 0ULL); + checking_wrmsrl(MSR_IA32_SYSENTER_EIP, (u64)ia32_sysenter_target); + + wrmsrl(MSR_CSTAR, ia32_cstar_target); +} + +#define compat_uses_vma 1 + +static inline void map_compat_vdso(int map) +{ +} + +#else /* CONFIG_X86_32 */ + +#define vdso32_sysenter() (boot_cpu_has(X86_FEATURE_SEP)) + void enable_sep_cpu(void) { int cpu = get_cpu(); @@ -210,13 +269,7 @@ static int __init gate_vma_init(void) return 0; } -/* - * These symbols are defined by vsyscall.o to mark the bounds - * of the ELF DSO images included therein. - */ -extern const char vsyscall_int80_start, vsyscall_int80_end; -extern const char vsyscall_sysenter_start, vsyscall_sysenter_end; -static struct page *syscall_pages[1]; +#define compat_uses_vma 0 static void map_compat_vdso(int map) { @@ -227,31 +280,35 @@ static void map_compat_vdso(int map) vdso_mapped = map; - __set_fixmap(FIX_VDSO, page_to_pfn(syscall_pages[0]) << PAGE_SHIFT, + __set_fixmap(FIX_VDSO, page_to_pfn(vdso32_pages[0]) << PAGE_SHIFT, map ? PAGE_READONLY_EXEC : PAGE_NONE); /* flush stray tlbs */ flush_tlb_all(); } +#endif /* CONFIG_X86_64 */ + int __init sysenter_setup(void) { void *syscall_page = (void *)get_zeroed_page(GFP_ATOMIC); const void *vsyscall; size_t vsyscall_len; - syscall_pages[0] = virt_to_page(syscall_page); + vdso32_pages[0] = virt_to_page(syscall_page); +#ifdef CONFIG_X86_32 gate_vma_init(); printk("Compat vDSO mapped to %08lx.\n", __fix_to_virt(FIX_VDSO)); +#endif - if (!boot_cpu_has(X86_FEATURE_SEP)) { - vsyscall = &vsyscall_int80_start; - vsyscall_len = &vsyscall_int80_end - &vsyscall_int80_start; + if (!vdso32_sysenter()) { + vsyscall = &vdso32_default_start; + vsyscall_len = &vdso32_default_end - &vdso32_default_start; } else { - vsyscall = &vsyscall_sysenter_start; - vsyscall_len = &vsyscall_sysenter_end - &vsyscall_sysenter_start; + vsyscall = &vdso32_sysenter_start; + vsyscall_len = &vdso32_sysenter_end - &vdso32_sysenter_start; } memcpy(syscall_page, vsyscall, vsyscall_len); @@ -284,7 +341,9 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack) ret = addr; goto up_fail; } + } + if (compat_uses_vma || !compat) { /* * MAYWRITE to allow gdb to COW and set breakpoints * @@ -298,7 +357,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack) VM_READ|VM_EXEC| VM_MAYREAD|VM_MAYWRITE|VM_MAYEXEC| VM_ALWAYSDUMP, - syscall_pages); + vdso32_pages); if (ret) goto up_fail; @@ -314,6 +373,12 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack) return ret; } +#ifdef CONFIG_X86_64 + +__initcall(sysenter_setup); + +#else /* CONFIG_X86_32 */ + const char *arch_vma_name(struct vm_area_struct *vma) { if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso) @@ -342,3 +407,5 @@ int in_gate_area_no_task(unsigned long addr) { return 0; } + +#endif /* CONFIG_X86_64 */ diff --git a/arch/x86/vdso/vdso32.S b/arch/x86/vdso/vdso32.S index cab020c99c3d..1e36f72cab86 100644 --- a/arch/x86/vdso/vdso32.S +++ b/arch/x86/vdso/vdso32.S @@ -2,14 +2,18 @@ __INITDATA - .globl vsyscall_int80_start, vsyscall_int80_end -vsyscall_int80_start: + .globl vdso32_default_start, vdso32_default_end +vdso32_default_start: +#ifdef CONFIG_X86_32 .incbin "arch/x86/vdso/vdso32-int80.so" -vsyscall_int80_end: +#else + .incbin "arch/x86/vdso/vdso32-syscall.so" +#endif +vdso32_default_end: - .globl vsyscall_sysenter_start, vsyscall_sysenter_end -vsyscall_sysenter_start: + .globl vdso32_sysenter_start, vdso32_sysenter_end +vdso32_sysenter_start: .incbin "arch/x86/vdso/vdso32-sysenter.so" -vsyscall_sysenter_end: +vdso32_sysenter_end: __FINIT diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index fd91568090f4..7d6d0ef55890 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -62,8 +62,8 @@ static void xen_idle(void) */ static void fiddle_vdso(void) { - extern char vsyscall_int80_start; - u32 *mask = VDSO32_SYMBOL(&vsyscall_int80_start, NOTE_MASK); + extern const char vdso32_default_start; + u32 *mask = VDSO32_SYMBOL(&vdso32_default_start, NOTE_MASK); *mask |= 1 << VDSO_NOTE_NONEGSEG_BIT; } diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h index 70edff2d5671..60f5101d9483 100644 --- a/include/asm-x86/elf.h +++ b/include/asm-x86/elf.h @@ -74,17 +74,19 @@ typedef struct user_fxsr_struct elf_fpxregset_t; #ifdef __KERNEL__ +/* + * This is used to ensure we don't load something for the wrong architecture. + */ +#define elf_check_arch_ia32(x) \ + (((x)->e_machine == EM_386) || ((x)->e_machine == EM_486)) + #ifdef CONFIG_X86_32 #include #include /* for savesegment */ #include #include -/* - * This is used to ensure we don't load something for the wrong architecture. - */ -#define elf_check_arch(x) \ - (((x)->e_machine == EM_386) || ((x)->e_machine == EM_486)) +#define elf_check_arch(x) elf_check_arch_ia32(x) /* SVR4/i386 ABI (pages 3-31, 3-32) says that when the program starts %edx contains a pointer to a function which might be registered using `atexit'. @@ -247,10 +249,6 @@ extern int dump_task_extended_fpu (struct task_struct *, #define ELF_CORE_XFPREG_TYPE NT_PRXFPREG #define VDSO_HIGH_BASE (__fix_to_virt(FIX_VDSO)) -#define VDSO_CURRENT_BASE ((unsigned long)current->mm->context.vdso) - -#define VDSO_ENTRY \ - ((unsigned long) VDSO32_SYMBOL(VDSO_CURRENT_BASE, vsyscall)) /* update AT_VECTOR_SIZE_ARCH if the number of NEW_AUX_ENT entries changes */ @@ -262,6 +260,8 @@ do if (vdso_enabled) { \ #else /* CONFIG_X86_32 */ +#define VDSO_HIGH_BASE 0xffffe000U /* CONFIG_COMPAT_VDSO address */ + /* 1GB for 64bit, 8MB for 32bit */ #define STACK_RND_MASK (test_thread_flag(TIF_IA32) ? 0x7ff : 0x3fffff) @@ -272,6 +272,11 @@ do if (vdso_enabled) { \ #endif /* !CONFIG_X86_32 */ +#define VDSO_CURRENT_BASE ((unsigned long)current->mm->context.vdso) + +#define VDSO_ENTRY \ + ((unsigned long) VDSO32_SYMBOL(VDSO_CURRENT_BASE, vsyscall)) + struct linux_binprm; #define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1 -- cgit v1.2.3 From fa20efd2fcd9349770113c6f72fc76ce437b62f5 Mon Sep 17 00:00:00 2001 From: Aaron Durbin Date: Wed, 30 Jan 2008 13:31:17 +0100 Subject: x86: add ACPI reboot option Add the ability to reboot an x86_64 based machine using the RESET_REG in the FADT ACPI table. Signed-off-by: Aaron Durbin Cc: Len Brown Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner --- Documentation/x86_64/boot-options.txt | 5 ++++- arch/x86/kernel/reboot_64.c | 11 ++++++++++- 2 files changed, 14 insertions(+), 2 deletions(-) (limited to 'Documentation') diff --git a/Documentation/x86_64/boot-options.txt b/Documentation/x86_64/boot-options.txt index 945311840a10..d7a333258682 100644 --- a/Documentation/x86_64/boot-options.txt +++ b/Documentation/x86_64/boot-options.txt @@ -110,12 +110,15 @@ Idle loop Rebooting - reboot=b[ios] | t[riple] | k[bd] [, [w]arm | [c]old] + reboot=b[ios] | t[riple] | k[bd] | a[cpi] [, [w]arm | [c]old] bios Use the CPU reboot vector for warm reset warm Don't set the cold reboot flag cold Set the cold reboot flag triple Force a triple fault (init) kbd Use the keyboard controller. cold reset (default) + acpi Use the ACPI RESET_REG in the FADT. If ACPI is not configured or the + ACPI reset does not work, the reboot path attempts the reset using + the keyboard controller. Using warm reset will be much faster especially on big memory systems because the BIOS will not go through the memory check. diff --git a/arch/x86/kernel/reboot_64.c b/arch/x86/kernel/reboot_64.c index 53620a92a8fd..307f996a3933 100644 --- a/arch/x86/kernel/reboot_64.c +++ b/arch/x86/kernel/reboot_64.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include @@ -29,7 +30,8 @@ EXPORT_SYMBOL(pm_power_off); static long no_idt[3]; static enum { BOOT_TRIPLE = 't', - BOOT_KBD = 'k' + BOOT_KBD = 'k', + BOOT_ACPI = 'a' } reboot_type = BOOT_KBD; static int reboot_mode = 0; int reboot_force; @@ -39,6 +41,7 @@ int reboot_force; cold Set the cold reboot flag triple Force a triple fault (init) kbd Use the keyboard controller. cold reset (default) + acpi Use the RESET_REG in the FADT force Avoid anything that could hang. */ static int __init reboot_setup(char *str) @@ -54,6 +57,7 @@ static int __init reboot_setup(char *str) break; case 't': + case 'a': case 'b': case 'k': reboot_type = *str; @@ -146,6 +150,11 @@ void machine_emergency_restart(void) reboot_type = BOOT_KBD; break; + + case BOOT_ACPI: + acpi_reboot(); + reboot_type = BOOT_KBD; + break; } } } -- cgit v1.2.3 From 9ad65e4748f55e3159283d7fa9d54fb30c086113 Mon Sep 17 00:00:00 2001 From: "Huang, Ying" Date: Wed, 30 Jan 2008 13:31:19 +0100 Subject: x86: EFI runtime service support: document for EFI runtime services This patch adds document for EFI x86_64 runtime services support. Signed-off-by: Chandramouli Narayanan Signed-off-by: Huang Ying Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner --- Documentation/x86_64/boot-options.txt | 9 ++++++++- Documentation/x86_64/uefi.txt | 9 +++++++++ 2 files changed, 17 insertions(+), 1 deletion(-) (limited to 'Documentation') diff --git a/Documentation/x86_64/boot-options.txt b/Documentation/x86_64/boot-options.txt index d7a333258682..638bf46ca059 100644 --- a/Documentation/x86_64/boot-options.txt +++ b/Documentation/x86_64/boot-options.txt @@ -110,7 +110,7 @@ Idle loop Rebooting - reboot=b[ios] | t[riple] | k[bd] | a[cpi] [, [w]arm | [c]old] + reboot=b[ios] | t[riple] | k[bd] | a[cpi] | e[fi] [, [w]arm | [c]old] bios Use the CPU reboot vector for warm reset warm Don't set the cold reboot flag cold Set the cold reboot flag @@ -119,6 +119,9 @@ Rebooting acpi Use the ACPI RESET_REG in the FADT. If ACPI is not configured or the ACPI reset does not work, the reboot path attempts the reset using the keyboard controller. + efi Use efi reset_system runtime service. If EFI is not configured or the + EFI reset does not work, the reboot path attempts the reset using + the keyboard controller. Using warm reset will be much faster especially on big memory systems because the BIOS will not go through the memory check. @@ -303,4 +306,8 @@ Debugging newfallback: use new unwinder but fall back to old if it gets stuck (default) +EFI + + noefi Disable EFI support + Miscellaneous diff --git a/Documentation/x86_64/uefi.txt b/Documentation/x86_64/uefi.txt index 91a98edfb588..7d77120a5184 100644 --- a/Documentation/x86_64/uefi.txt +++ b/Documentation/x86_64/uefi.txt @@ -19,6 +19,10 @@ Mechanics: - Build the kernel with the following configuration. CONFIG_FB_EFI=y CONFIG_FRAMEBUFFER_CONSOLE=y + If EFI runtime services are expected, the following configuration should + be selected. + CONFIG_EFI=y + CONFIG_EFI_VARS=y or m # optional - Create a VFAT partition on the disk - Copy the following to the VFAT partition: elilo bootloader with x86_64 support, elilo configuration file, @@ -27,3 +31,8 @@ Mechanics: can be found in the elilo sourceforge project. - Boot to EFI shell and invoke elilo choosing the kernel image built in first step. +- If some or all EFI runtime services don't work, you can try following + kernel command line parameters to turn off some or all EFI runtime + services. + noefi turn off all EFI runtime services + reboot_type=k turn off EFI reboot runtime service -- cgit v1.2.3 From 8b2cb7a8f531d6ca72a8aff873b9bb1c6b3122ba Mon Sep 17 00:00:00 2001 From: "Huang, Ying" Date: Wed, 30 Jan 2008 13:32:11 +0100 Subject: x86: 32-bit EFI runtime service support: fixes in sync with 64-bit support support according to fixes of x86_64 support. - Delete efi_rt_lock because it is used during system early boot, before SMP is initialized. - Change local_flush_tlb() to __flush_tlb_all() to flush global page mapping. - Clean up includes. - Revise Kconfig description. - Enable noefi kernel parameter on i386. Signed-off-by: Huang Ying Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner --- Documentation/kernel-parameters.txt | 2 ++ Documentation/x86_64/boot-options.txt | 4 ---- arch/x86/Kconfig | 19 ++++++++----------- arch/x86/kernel/efi.c | 7 +++++++ arch/x86/kernel/efi_32.c | 25 +++++-------------------- arch/x86/kernel/efi_64.c | 7 ------- arch/x86/kernel/setup_32.c | 6 +++--- 7 files changed, 25 insertions(+), 45 deletions(-) (limited to 'Documentation') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index b3f20beea411..e7910f8b4fcc 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1169,6 +1169,8 @@ and is between 256 and 4096 characters. It is defined in the file nodisconnect [HW,SCSI,M68K] Disables SCSI disconnects. + noefi [X86-32,X86-64] Disable EFI runtime services support. + noexec [IA-64] noexec [X86-32,X86-64] diff --git a/Documentation/x86_64/boot-options.txt b/Documentation/x86_64/boot-options.txt index 638bf46ca059..34abae4e9442 100644 --- a/Documentation/x86_64/boot-options.txt +++ b/Documentation/x86_64/boot-options.txt @@ -306,8 +306,4 @@ Debugging newfallback: use new unwinder but fall back to old if it gets stuck (default) -EFI - - noefi Disable EFI support - Miscellaneous diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index d89b94528153..fa6fa52248d9 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -961,21 +961,18 @@ config MTRR config EFI def_bool n - prompt "Boot from EFI support" + prompt "EFI runtime service support" depends on ACPI ---help--- - This enables the kernel to boot on EFI platforms using - system configuration information passed to it from the firmware. - This also enables the kernel to use any EFI runtime services that are + This enables the kernel to use EFI runtime services that are available (such as the EFI variable services). - This option is only useful on systems that have EFI firmware - and will result in a kernel image that is ~8k larger. In addition, - you must use the latest ELILO loader available at - in order to take advantage of - kernel initialization using EFI information (neither GRUB nor LILO know - anything about EFI). However, even with this option, the resultant - kernel should continue to boot on existing non-EFI platforms. + This option is only useful on systems that have EFI firmware. + In addition, you should use the latest ELILO loader available + at in order to take advantage + of EFI runtime services. However, even with this option, the + resultant kernel should continue to boot on existing non-EFI + platforms. config IRQBALANCE def_bool y diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c index 0a61522e85c7..2939b015c2ed 100644 --- a/arch/x86/kernel/efi.c +++ b/arch/x86/kernel/efi.c @@ -55,6 +55,13 @@ struct efi_memory_map memmap; struct efi efi_phys __initdata; static efi_system_table_t efi_systab __initdata; +static int __init setup_noefi(char *arg) +{ + efi_enabled = 0; + return 0; +} +early_param("noefi", setup_noefi); + static efi_status_t virt_efi_get_time(efi_time_t *tm, efi_time_cap_t *tc) { return efi_call_virt2(get_time, tm, tc); diff --git a/arch/x86/kernel/efi_32.c b/arch/x86/kernel/efi_32.c index 30f937116288..afd2c3b039d6 100644 --- a/arch/x86/kernel/efi_32.c +++ b/arch/x86/kernel/efi_32.c @@ -20,27 +20,15 @@ */ #include -#include -#include #include -#include -#include -#include #include -#include #include -#include -#include #include #include #include -#include -#include #include -#define PFX "EFI: " - /* * To make EFI call EFI runtime service in physical addressing mode we need * prelog/epilog before/after the invocation to disable interrupt, to @@ -49,16 +37,14 @@ */ static unsigned long efi_rt_eflags; -static DEFINE_SPINLOCK(efi_rt_lock); static pgd_t efi_bak_pg_dir_pointer[2]; -void efi_call_phys_prelog(void) __acquires(efi_rt_lock) +void efi_call_phys_prelog(void) { unsigned long cr4; unsigned long temp; struct desc_ptr gdt_descr; - spin_lock(&efi_rt_lock); local_irq_save(efi_rt_eflags); /* @@ -88,14 +74,14 @@ void efi_call_phys_prelog(void) __acquires(efi_rt_lock) /* * After the lock is released, the original page table is restored. */ - local_flush_tlb(); + __flush_tlb_all(); gdt_descr.address = __pa(get_cpu_gdt_table(0)); gdt_descr.size = GDT_SIZE - 1; load_gdt(&gdt_descr); } -void efi_call_phys_epilog(void) __releases(efi_rt_lock) +void efi_call_phys_epilog(void) { unsigned long cr4; struct desc_ptr gdt_descr; @@ -119,10 +105,9 @@ void efi_call_phys_epilog(void) __releases(efi_rt_lock) /* * After the lock is released, the original page table is restored. */ - local_flush_tlb(); + __flush_tlb_all(); local_irq_restore(efi_rt_eflags); - spin_unlock(&efi_rt_lock); } /* @@ -135,7 +120,7 @@ void __init efi_map_memmap(void) memmap.map = bt_ioremap((unsigned long) memmap.phys_map, (memmap.nr_map * memmap.desc_size)); if (memmap.map == NULL) - printk(KERN_ERR PFX "Could not remap the EFI memmap!\n"); + printk(KERN_ERR "Could not remap the EFI memmap!\n"); memmap.map_end = memmap.map + (memmap.nr_map * memmap.desc_size); } diff --git a/arch/x86/kernel/efi_64.c b/arch/x86/kernel/efi_64.c index f2000dbc7195..269de2e049a3 100644 --- a/arch/x86/kernel/efi_64.c +++ b/arch/x86/kernel/efi_64.c @@ -40,13 +40,6 @@ static pgd_t save_pgd __initdata; static unsigned long efi_flags __initdata; -static int __init setup_noefi(char *arg) -{ - efi_enabled = 0; - return 0; -} -early_param("noefi", setup_noefi); - static void __init early_mapping_set_exec(unsigned long start, unsigned long end, int executable) diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c index 2e805da337a2..704550fdb84c 100644 --- a/arch/x86/kernel/setup_32.c +++ b/arch/x86/kernel/setup_32.c @@ -648,9 +648,6 @@ void __init setup_arch(char **cmdline_p) printk(KERN_INFO "BIOS-provided physical RAM map:\n"); print_memory_map(memory_setup()); - if (efi_enabled) - efi_init(); - copy_edd(); if (!boot_params.hdr.root_flags) @@ -677,6 +674,9 @@ void __init setup_arch(char **cmdline_p) strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE); *cmdline_p = command_line; + if (efi_enabled) + efi_init(); + max_low_pfn = setup_memory(); #ifdef CONFIG_VMI -- cgit v1.2.3 From a25bd94964e87b1b93903a822fba5025d995d4da Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Wed, 30 Jan 2008 13:33:08 +0100 Subject: x86: add the "print code before the trapping instruction" feature to 64 bit The 32 bit x86 tree has a very useful feature that prints the Code: line for the code even before the trapping instrution (and the start of the trapping instruction is then denoted with a <>). Unfortunately, the 64 bit x86 tree does not yet have this feature, making diagnosing backtraces harder than needed. This patch adds this feature in the same was as the 32 bit tree has (including the same kernel boot parameter), and including a bugfix to make the code use probe_kernel_address() rarther than a buggy (deadlocking) __get_user. Signed-off-by: Arjan van de Ven Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner --- Documentation/kernel-parameters.txt | 4 ++-- arch/x86/kernel/traps_64.c | 44 +++++++++++++++++++++++++++---------- 2 files changed, 35 insertions(+), 13 deletions(-) (limited to 'Documentation') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index e7910f8b4fcc..40db7dd1d92a 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -416,8 +416,8 @@ and is between 256 and 4096 characters. It is defined in the file [SPARC64] tick [X86-64] hpet,tsc - code_bytes [IA32] How many bytes of object code to print in an - oops report. + code_bytes [IA32/X86_64] How many bytes of object code to print + in an oops report. Range: 0 - 8192 Default: 64 diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c index 304ca6b4a1ca..0bba7924604e 100644 --- a/arch/x86/kernel/traps_64.c +++ b/arch/x86/kernel/traps_64.c @@ -74,6 +74,8 @@ asmlinkage void alignment_check(void); asmlinkage void machine_check(void); asmlinkage void spurious_interrupt_bug(void); +static unsigned int code_bytes = 64; + static inline void conditional_sti(struct pt_regs *regs) { if (regs->flags & X86_EFLAGS_IF) @@ -459,12 +461,15 @@ EXPORT_SYMBOL(dump_stack); void show_registers(struct pt_regs *regs) { int i; - int in_kernel = !user_mode(regs); unsigned long sp; const int cpu = smp_processor_id(); struct task_struct *cur = cpu_pda(cpu)->pcurrent; + u8 *ip; + unsigned int code_prologue = code_bytes * 43 / 64; + unsigned int code_len = code_bytes; sp = regs->sp; + ip = (u8 *) regs->ip - code_prologue; printk("CPU %d ", cpu); __show_regs(regs); printk("Process %s (pid: %d, threadinfo %p, task %p)\n", @@ -474,22 +479,28 @@ void show_registers(struct pt_regs *regs) * When in-kernel, we also print out the stack and code at the * time of the fault.. */ - if (in_kernel) { + if (!user_mode(regs)) { + unsigned char c; printk("Stack: "); _show_stack(NULL, regs, (unsigned long *)sp, regs->bp); + printk("\n"); - printk("\nCode: "); - if (regs->ip < PAGE_OFFSET) - goto bad; - - for (i=0; i<20; i++) { - unsigned char c; - if (__get_user(c, &((unsigned char*)regs->ip)[i])) { -bad: + printk(KERN_EMERG "Code: "); + if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) { + /* try starting at RIP */ + ip = (u8 *) regs->ip; + code_len = code_len - code_prologue + 1; + } + for (i = 0; i < code_len; i++, ip++) { + if (ip < (u8 *)PAGE_OFFSET || + probe_kernel_address(ip, c)) { printk(" Bad RIP value."); break; } - printk("%02x ", c); + if (ip == (u8 *)regs->ip) + printk("<%02x> ", c); + else + printk("%02x ", c); } } printk("\n"); @@ -1164,3 +1175,14 @@ static int __init kstack_setup(char *s) return 0; } early_param("kstack", kstack_setup); + + +static int __init code_bytes_setup(char *s) +{ + code_bytes = simple_strtoul(s, NULL, 0); + if (code_bytes > 8192) + code_bytes = 8192; + + return 1; +} +__setup("code_bytes=", code_bytes_setup); -- cgit v1.2.3 From aaf230424204864e2833dcc1da23e2cb0b9f39cd Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 30 Jan 2008 13:33:09 +0100 Subject: x86: disable the GART early, 64-bit For K8 system: 4G RAM with memory hole remapping enabled, or more than 4G RAM installed. when try to use kexec second kernel, and the first doesn't include gart_shutdown. the second kernel could have different aper position than the first kernel. and second kernel could use that hole as RAM that is still used by GART set by the first kernel. esp. when try to kexec 2.6.24 with sparse mem enable from previous kernel (from RHEL 5 or SLES 10). the new kernel will use aper by GART (set by first kernel) for vmemmap. and after new kernel setting one new GART. the position will be real RAM. the _mapcount set is lost. Bad page state in process 'swapper' page:ffffe2000e600020 flags:0x0000000000000000 mapping:0000000000000000 mapcount:1 count:0 Trying to fix it up, but a reboot is needed Backtrace: Pid: 0, comm: swapper Not tainted 2.6.24-rc7-smp-gcdf71a10-dirty #13 Call Trace: [] bad_page+0x63/0x8d [] __free_pages_ok+0x7c/0x2a5 [] free_all_bootmem_core+0xd0/0x198 [] numa_free_all_bootmem+0x3b/0x76 [] mem_init+0x3b/0x152 [] start_kernel+0x236/0x2c2 [] _sinittext+0x11a/0x121 and [ffffe2000e600000-ffffe2000e7fffff] PMD ->ffff81001c200000 on node 0 phys addr is : 0x1c200000 RHEL 5.1 kernel -53 said: PCI-DMA: aperture base @ 1c000000 size 65536 KB new kernel said: Mapping aperture over 65536 KB of RAM @ 3c000000 So could try to disable that GART if possible. According to Ingo > hm, i'm wondering, instead of modifying the GART, why dont we simply > _detect_ whatever GART settings we have inherited, and propagate that > into our e820 maps? I.e. if there's inconsistency, then punch that out > from the memory maps and just dont use that memory. > > that way it would not matter whether the GART settings came from a [old > or crashing] Linux kernel that has not called gart_iommu_shutdown(), or > whether it's a BIOS that has set up an aperture hole inconsistent with > the memory map it passed. (or the memory map we _think_ i tried to pass > us) > > it would also be more robust to only read and do a memory map quirk > based on that, than actively trying to change the GART so early in the > bootup. Later on we have to re-enable the GART _anyway_ and have to > punch a hole for it. > > and as a bonus, we would have shored up our defenses against crappy > BIOSes as well. add e820 modification for gart inconsistent setting. gart_fix_e820=off could be used to disable e820 fix. Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner --- Documentation/kernel-parameters.txt | 4 ++ arch/x86/kernel/aperture_64.c | 89 +++++++++++++++++++++++++++++++++++++ arch/x86/kernel/e820_64.c | 12 +++++ arch/x86/kernel/setup_64.c | 3 ++ include/asm-x86/e820_64.h | 1 + include/asm-x86/gart.h | 5 +++ 6 files changed, 114 insertions(+) (limited to 'Documentation') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 40db7dd1d92a..860a90875491 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -660,6 +660,10 @@ and is between 256 and 4096 characters. It is defined in the file gamma= [HW,DRM] + gart_fix_e820= [X86_64] disable the fix e820 for K8 GART + Format: off | on + default: on + gdth= [HW,SCSI] See header of drivers/scsi/gdth.c. diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index 52d2beac4556..bf1b469d5847 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -218,6 +218,95 @@ static __u32 __init search_agp_bridge(u32 *order, int *valid_agp) return 0; } +static int gart_fix_e820 __initdata = 1; + +static int __init parse_gart_mem(char *p) +{ + if (!p) + return -EINVAL; + + if (!strncmp(p, "off", 3)) + gart_fix_e820 = 0; + else if (!strncmp(p, "on", 2)) + gart_fix_e820 = 1; + + return 0; +} +early_param("gart_fix_e820", parse_gart_mem); + +void __init early_gart_iommu_check(void) +{ + /* + * in case it is enabled before, esp for kexec/kdump, + * previous kernel already enable that. memset called + * by allocate_aperture/__alloc_bootmem_nopanic cause restart. + * or second kernel have different position for GART hole. and new + * kernel could use hole as RAM that is still used by GART set by + * first kernel + * or BIOS forget to put that in reserved. + * try to update e820 to make that region as reserved. + */ + int fix, num; + u32 ctl; + u32 aper_size = 0, aper_order = 0, last_aper_order = 0; + u64 aper_base = 0, last_aper_base = 0; + int aper_enabled = 0, last_aper_enabled = 0; + + if (!early_pci_allowed()) + return; + + fix = 0; + for (num = 24; num < 32; num++) { + if (!early_is_k8_nb(read_pci_config(0, num, 3, 0x00))) + continue; + + ctl = read_pci_config(0, num, 3, 0x90); + aper_enabled = ctl & 1; + aper_order = (ctl >> 1) & 7; + aper_size = (32 * 1024 * 1024) << aper_order; + aper_base = read_pci_config(0, num, 3, 0x94) & 0x7fff; + aper_base <<= 25; + + if ((last_aper_order && aper_order != last_aper_order) || + (last_aper_base && aper_base != last_aper_base) || + (last_aper_enabled && aper_enabled != last_aper_enabled)) { + fix = 1; + break; + } + last_aper_order = aper_order; + last_aper_base = aper_base; + last_aper_enabled = aper_enabled; + } + + if (!fix && !aper_enabled) + return; + + if (!aper_base || !aper_size || aper_base + aper_size > 0x100000000UL) + fix = 1; + + if (gart_fix_e820 && !fix && aper_enabled) { + if (e820_any_mapped(aper_base, aper_base + aper_size, + E820_RAM)) { + /* reserved it, so we can resuse it in second kernel */ + printk(KERN_INFO "update e820 for GART\n"); + add_memory_region(aper_base, aper_size, E820_RESERVED); + update_e820(); + } + return; + } + + /* different nodes have different setting, disable them all at first*/ + for (num = 24; num < 32; num++) { + if (!early_is_k8_nb(read_pci_config(0, num, 3, 0x00))) + continue; + + ctl = read_pci_config(0, num, 3, 0x90); + ctl &= ~1; + write_pci_config(0, num, 3, 0x90, ctl); + } + +} + void __init gart_iommu_hole_init(void) { u32 aper_size, aper_alloc = 0, aper_order = 0, last_aper_order = 0; diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c index abc473bcabe8..07cfaae7ab07 100644 --- a/arch/x86/kernel/e820_64.c +++ b/arch/x86/kernel/e820_64.c @@ -728,6 +728,18 @@ void __init finish_e820_parsing(void) } } +void __init update_e820(void) +{ + u8 nr_map; + + nr_map = e820.nr_map; + if (sanitize_e820_map(e820.map, &nr_map)) + return; + e820.nr_map = nr_map; + printk(KERN_INFO "modified physical RAM map:\n"); + e820_print_map("modified"); +} + unsigned long pci_mem_start = 0xaeedbabe; EXPORT_SYMBOL(pci_mem_start); diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c index 07547febac7a..12bad27d66f8 100644 --- a/arch/x86/kernel/setup_64.c +++ b/arch/x86/kernel/setup_64.c @@ -53,6 +53,7 @@ #include