summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* drivers/net: replace BUG() with BUG_ON() if possibleAlexander Beregalov2009-04-1413-42/+22
| | | | | Signed-off-by: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* kaweth: Clean up codeLarry Finger2009-04-141-8/+25
| | | | | | | | | The driver kaweth yields a -EBUSY error when starting, and a -ETIME error when shutting down. These errors are avoided, and the RX status is further checked for other potential errors. Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: David S. Miller <davem@davemloft.net>
* at76c50x-usb: remove pointless conditional before kfree_skb()Wei Yongjun2009-04-141-2/+1
| | | | | | | Remove pointless conditional before kfree_skb(). Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* be2net: changes for checksum and segmentation offloadAjit Khaparde2009-04-141-6/+22
| | | | | | | | | - Enables Rx checksum feature by default. - Disables support for ipv6 tso. - Changes in Rx path to handle Rx completions with various checksum options. Signed-off-by: Ajit Khaparde <ajitk@serverengines.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ixgbe: Be explict with what we are !'ing in ixgbe_sfp_config_module_task()Tony Breeds2009-04-111-1/+1
| | | | | | | | | | | | | GCC warns: drivers/net/ixgbe/ixgbe_main.c: In function 'ixgbe_sfp_config_module_task': drivers/net/ixgbe/ixgbe_main.c:3920: warning: suggest parantheses around operand of '!' or change '&' to '&&' or '!' to '~' Which I think is right. Bracket to remove ambiguity. Signed-off-by: Tony Breeds <tony@bakeyournoodle.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ixgbe: Update get_physical_layer() calls, plus a version bumpPeter P Waskiewicz Jr2009-04-115-70/+169
| | | | | | | | | | | | | | Not all physical connection types are being correctly identified. This fixes that issue, and cleans up the logic to make it more maintainable. Also clean up the code for device capabilities from the EEPROM to support multiple SKUs of the same hardware. Bump the version to reflect all the updates since the 82599 merge. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ixgbe: Update the usage of orig_autoc to be more consistentPJ Waskiewicz2009-04-112-13/+39
| | | | | | | | | | The orig_autoc variable tracks the original setting of the autonegotiate state prior to trying a new speed. The usage is inconsistent and not very maintainable. This patch updates the usage to make it more consistent. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ixgbe: Remove unnecessary PHY reset, properly identify multispeed fiber modulesPJ Waskiewicz2009-04-114-8/+29
| | | | | | | | | | | This patch does two things: 1) On 82599, the PHY is emedded in the MAC. On 82598, the SFP+ NIC has an external PHY. The reset in the SFP+ setup patch for 82598 is unnecessary on 82599, and adds extra dead time to device initialization. This removes that PHY reset for 82599 only. 2) On 82599, the SFP+ modules are multispeed fiber modules (10G/1G). We need to make sure to identify them properly for the remaining init sections to properly set them up. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ixgbe: Move PHY ops initialization to centralize bus accessesPJ Waskiewicz2009-04-114-50/+107
| | | | | | | | | | | When PHY operations are determined, the PHY must be identified. This identification causes bus access, and should be contained within its own routines. This also helps the 82599 PHY init paths for both SFP+ and KX/KX4 devices to be easier to maintain. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ixgbe: Don't return error in flow control configuration if FC is offPJ Waskiewicz2009-04-112-6/+10
| | | | | | | | | When flow control is disabled, an invalid low/high watermark configuration should not matter. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ixgbe: add define to support 82599 64 IVAR registersDon Skidmore2009-04-111-0/+1
| | | | | | | | | | 82599 supports 64 IVAR registers this patch adds a define to allow us to access them. Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com> Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ixgbe: add support for addition VT_CTL definesDon Skidmore2009-04-111-0/+2
| | | | | | | | | | | These defines allow for definition of target pool for packets that pass L2 filtering but didn't pass any of the pool filters. They are needed to reset the default pool. Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com> Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ixgbe: fix IXGBE_MTQC valuesDon Skidmore2009-04-111-1/+2
| | | | | | | | | | IXGBE_MTQC_64VF was wrong and 32VF not defined at all. This patch corrects that. Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com> Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ixgbe: Enable another bit for flow control operationPJ Waskiewicz2009-04-112-0/+2
| | | | | | | | | The discard pause frames bit was not enabled, so flow control frames could be passed up to the host. Enabled DPF to keep the frames off the stack. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* MAINTAINERS: rds list is moderatedKyle McMartin2009-04-101-1/+1
| | | | | | Signed-off-by: Kyle McMartin <kyle@redhat.com> Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ERR_PTR() dereference in net/rds/ib.cDan Carpenter2009-04-101-2/+2
| | | | | | | | | | | | | rdma_create_id() doesn't return NULL, only ERR_PTR(). Found by smatch (http://repo.or.cz/w/smatch.git). Compile tested. regards, dan carpenter Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ERR_PTR() dereference in net/rds/iw.cDan Carpenter2009-04-101-2/+2
| | | | | | | | | | | | | rdma_create_id() returns ERR_PTR() not null. Found by smatch (http://repo.or.cz/w/smatch.git). Compile tested. regards, dan carpenter Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* rds: use kmem_cache_zalloc instead of kmem_cache_alloc/memsetWei Yongjun2009-04-101-3/+1
| | | | | | | | Use kmem_cache_zalloc instead of kmem_cache_alloc/memset. Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com> Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* RDS: remove unused #include <version.h>Huang Weiyi2009-04-101-1/+0
| | | | | | | | Remove unused #include <version.h> in net/rds/af_rds.c. Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com> Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* RDS: use get_user_pages_fast()Andy Grover2009-04-102-8/+2
| | | | | | | Use the new function that is simpler and faster. Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* RDS: Establish connection before parsing CMSGsAndy Grover2009-04-101-5/+5
| | | | | | | | | The first message to a remote node should prompt a new connection. Even an RDMA op via CMSG. Therefore move CMSG parsing to after connection establishment. Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* RDS: Fix ordering in a conditionalAndy Grover2009-04-101-1/+1
| | | | | | | | | | | Putting the constant first is a supposed "best practice" that actually makes the code harder to read. Thanks to Roland Dreier for finding a bug in this "simple, obviously correct" patch. Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* RDS/IW+IB: Allow max credit advertise window.Steve Wise2009-04-107-13/+13
| | | | | | | | Fix hack that restricts the credit advertisement to 127. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* RDS/IW+IB: Set the RDS_LL_SEND_FULL bit when we're throttled.Steve Wise2009-04-102-2/+2
| | | | | | | | | | The RDS_LL_SEND_FULL bit should be set when we stop transmitted due to flow control. Otherwise the send worker will keep trying as opposed to sleeping until we unthrottle. Saves CPU. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* RDS: Correct some iw references in rdma_transport.cAndy Grover2009-04-101-6/+6
| | | | | | | | Had some lingering instances of _iw_ variable names from when the listen code was centralized into rdma_transport.c Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* RDS/IW+IB: Set recv ring low water mark to 1/2 full.Steve Wise2009-04-102-2/+2
| | | | | | | | | | | | Currently the recv ring low water mark is 1/4 the depth. Performance measurements show that this limits iWARP throughput by flow controlling the rds-stress senders. Setting it to 1/2 seems to max the T3 performance. I tried even higher levels but that didn't help and it started to increase the rds thread cpu utilization. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ISDN: update Documentation/isdn/00-INDEXTilman Schmidt2009-04-091-19/+26
| | | | | | | | | After the merging of mISDN, state which files refer only to the old isdn4linux subsystem. Also add a few missing files and sort alphabetically. Signed-off-by: Tilman Schmidt <tilman@imap.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Don't specify unexistent IRQJean Delvare2009-04-092-5/+0
| | | | | | | | | Neither the lm90 driver nor the lm87 driver do support interrupts, so there is no point in specifying one when declaring the devices. Signed-off-by: Jean Delvare <khali@linux-fr.org> Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* netxen: cache align register map tableDhananjay Phadke2009-04-091-1/+2
| | | | | | | | Aligning register offset translation table imporves performance on rx side. Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* netxen: enable GRO supportAmit Kumar Salecha2009-04-092-3/+5
| | | | | | Signed-off-by: Amit Kumar Salecha <amit@netxen.com> Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* netxen: enable rss for NX2031Dhananjay Phadke2009-04-094-43/+137
| | | | | | | | | Enable multiple rx rings for older NX2031 chip, firmware 3.4.336 or newer supports this feature. Signed-off-by: Amit Kumar Salecha <amit@netxen.com> Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* netxen: sanitize function namesDhananjay Phadke2009-04-097-406/+233
| | | | | | | | | | | | | Replace superfluous wrapper functions with two macros: NXWR32 replaces netxen_nic_reg_write, netxen_nic_write_w0, netxen_nic_read_w1, netxen_crb_writelit_adapter. NXRD32 replaces netxen_nic_reg_read, netxen_nic_read_w0, netxen_nic_read_w1. Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* netxen: annotate register access functionsDhananjay Phadke2009-04-097-344/+239
| | | | | | | | | | o remove unnecessary length parameter since register access width is fixed 4 byte. o remove superfluous pci_read_normalize and pci_write_normalize functions. Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* netxen: allocate status rings dynamicallyDhananjay Phadke2009-04-092-6/+30
| | | | | | | This reduces netxen_adapter footprint when rss (msi-x) is disabled. Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* netxen: async link event handlingDhananjay Phadke2009-04-096-57/+263
| | | | | | | | | | | Add support for asynchronous events from firmware, received over one of the rx rings. Add support for event based phy interrupts, enhanced links status reporting from firmware. Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* netxen: defer firmware handshakeDhananjay Phadke2009-04-093-42/+22
| | | | | | | | Removed duplicate firmware handshake, defer it until first port (interface) is brought up. Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* netxen: refactor transmit codeDhananjay Phadke2009-04-095-120/+108
| | | | | | | | | o move tx stuff into nx_host_tx_ring structure, this will help managing multiple tx rings in future. o sanitize some variable names Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* netxen: refactor netxen_adapterDhananjay Phadke2009-04-093-59/+39
| | | | | | | Rearrange members to align them at right offset. Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* netxen: code cleanupDhananjay Phadke2009-04-098-256/+146
| | | | | | | | | | o remove unused structure defs. o remove unnecessary includes. o replace enums with specific #defines. o reduce footprint of stats structure. Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* Linux 2.6.30-rc1v2.6.30-rc1Linus Torvalds2009-04-071-2/+2
|
* Merge branch 'core/softlockup' of ↵Linus Torvalds2009-04-077-114/+283
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip * 'core/softlockup' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: softlockup: make DETECT_HUNG_TASK default depend on DETECT_SOFTLOCKUP softlockup: move 'one' to the softlockup section in sysctl.c softlockup: ensure the task has been switched out once softlockup: remove timestamp checking from hung_task softlockup: convert read_lock in hung_task to rcu_read_lock softlockup: check all tasks in hung_task softlockup: remove unused definition for spawn_softlockup_task softlockup: fix potential race in hung_task when resetting timeout softlockup: fix to allow compiling with !DETECT_HUNG_TASK softlockup: decouple hung tasks check from softlockup detection
| * Merge branch 'linus' into core/softlockupIngo Molnar2009-04-078557-359622/+934152
| |\ | | | | | | | | | | | | Conflicts: kernel/sysctl.c
| * | softlockup: make DETECT_HUNG_TASK default depend on DETECT_SOFTLOCKUPIngo Molnar2009-04-071-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Don't offer a default-y option when the user has turned off CONFIG_DETECT_SOFTLOCKUP already. Do offer it as 'y' only if DETECT_SOFTLOCKUP is on already. This makes it match previous behavior - where the hung-task check was embedded i CONFIG_DETECT_SOFTLOCKUP code. Cc: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * | softlockup: move 'one' to the softlockup section in sysctl.cIngo Molnar2009-02-121-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | CONFIG_SOFTLOCKUP=y || CONFIG_DETECT_HUNG_TASKS=y is now the only user of the 'one' constant in kernel/sysctl.c. Move it to the softlockup block of constants. This fixes a GCC warning. Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * | Merge branch 'linus' into core/softlockupIngo Molnar2009-02-122296-21370/+47219
| |\ \
| * | | softlockup: ensure the task has been switched out onceFrederic Weisbecker2009-02-111-1/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When we check if a task has been switched out since the last scan, we might have a race condition on the following scenario: - the task is freshly created and scheduled - it puts its state to TASK_UNINTERRUPTIBLE and is not yet switched out - check_hung_task() scans this task and will report a false positive because t->nvcsw + t->nivcsw == t->last_switch_count == 0 Add a check for such cases. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Acked-by: Mandeep Singh Baines <msb@google.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * | | softlockup: remove timestamp checking from hung_taskMandeep Singh Baines2009-02-093-45/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Impact: saves sizeof(long) bytes per task_struct By guaranteeing that sysctl_hung_task_timeout_secs have elapsed between tasklist scans we can avoid using timestamps. Signed-off-by: Mandeep Singh Baines <msb@google.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * | | softlockup: convert read_lock in hung_task to rcu_read_lockMandeep Singh Baines2009-02-051-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since the tasklist is protected by rcu list operations, it is safe to convert the read_lock()s to rcu_read_lock(). Suggested-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Mandeep Singh Baines <msb@google.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * | | softlockup: check all tasks in hung_taskMandeep Singh Baines2009-02-051-2/+37
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Impact: extend the scope of hung-task checks Changed the default value of hung_task_check_count to PID_MAX_LIMIT. hung_task_batch_count added to put an upper bound on the critical section. Every hung_task_batch_count checks, the rcu lock is never held for a too long time. Keeping the critical section small minimizes time preemption is disabled and keeps rcu grace periods small. To prevent following a stale pointer, get_task_struct is called on g and t. To verify that g and t have not been unhashed while outside the critical section, the task states are checked. The design was proposed by Frédéric Weisbecker. Signed-off-by: Mandeep Singh Baines <msb@google.com> Suggested-by: Frédéric Weisbecker <fweisbec@gmail.com> Acked-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * | | softlockup: remove unused definition for spawn_softlockup_taskMandeep Singh Baines2009-01-311-3/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The definition of spawn_softlockup_task in sched.h became unnecessary once it was converted to the early_initcall() interface. Signed-off-by: Mandeep Singh Baines <msb@google.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-28 05:04:51 +0100