summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* m68k: Return -ENODEV if no device is foundGeert Uytterhoeven2008-07-214-5/+5
| | | | | | | | According to the tests in do_initcalls(), the proper error code in case no device is found is -ENODEV, not -ENXIO or -EIO. Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* ariadne: use netstats in net_device structurePaulius Zaleckas2008-07-211-25/+22
| | | | | | | | | Use net_device_stats from net_device structure instead of local. Signed-off-by: Paulius Zaleckas <paulius.zaleckas@teltonika.lt> Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Acked-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* m68k: remove CVS keywordsAdrian Bunk2008-07-2116-19/+10
| | | | | | | | | This patch removes CVS keywords that weren't updated for a long time from comments. Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Merge branch 'configfs-fixup-ptr-error' of ↵Linus Torvalds2008-07-218-109/+73
|\ | | | | | | | | | | | | | | git://oss.oracle.com/git/jlbec/linux-2.6 * 'configfs-fixup-ptr-error' of git://oss.oracle.com/git/jlbec/linux-2.6: configfs: Allow ->make_item() and ->make_group() to return detailed errors. Revert "configfs: Allow ->make_item() and ->make_group() to return detailed errors."
| * configfs: Allow ->make_item() and ->make_group() to return detailed errors.Joel Becker2008-07-187-37/+31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The configfs operations ->make_item() and ->make_group() currently return a new item/group. A return of NULL signifies an error. Because of this, -ENOMEM is the only return code bubbled up the stack. Multiple folks have requested the ability to return specific error codes when these operations fail. This patch adds that ability by changing the ->make_item/group() ops to return ERR_PTR() values. These errors are bubbled up appropriately. NULL returns are changed to -ENOMEM for compatibility. Also updated are the in-kernel users of configfs. This is a rework of reverted commit 11c3b79218390a139f2d474ee1e983a672d5839a. Signed-off-by: Joel Becker <joel.becker@oracle.com>
| * Revert "configfs: Allow ->make_item() and ->make_group() to return detailed ↵Joel Becker2008-07-178-94/+64
| | | | | | | | | | | | | | | | | | errors." This reverts commit 11c3b79218390a139f2d474ee1e983a672d5839a. The code will move to PTR_ERR(). Signed-off-by: Joel Becker <joel.becker@oracle.com>
* | APCI: revert another duplicated patchThomas Gleixner2008-07-211-3/+0
| | | | | | | | | | | | | | | | | | | | | | commit d1857056904d5f313f11184fcfa624652ff9620a ("ACPI: don't walk tables if ACPI was disabled") is another superfluous duplicate commit caused by git -> quilt -> git conversion. Revert it. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | APCI: revert duplicated patchThomas Gleixner2008-07-212-6/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | commit 816c2eda3ce8fa7eb62f22e01e2ec7a3f7d677c0 ("dock: bay: Don't call acpi_walk_namespace() when ACPI is disabled.") was merged between 2.6.26-rc8 and -rc9) Due to rebasing the ACPI tree via quilt the same patch got applied again via commit cc7e51666d82aedfd6b9a033ca1a10d71c21f1ca ("dock: bay: Don't call acpi_walk_namespace() when ACPI is disabled.") Revert it, as it is obviously bogus. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | Fix strip driver back up for ldisc/tty changesAlan Cox2008-07-211-1/+1
| | | | | | | | | | Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | tty: add more tty_port fieldsAlan Cox2008-07-2118-198/+169
| | | | | | | | | | | | | | Move more bits into the tty_port structure Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | termios: Termios defines for other platformsAlan Cox2008-07-213-1/+12
| | | | | | | | | | | | | | Fix up the termios of the people who have not yet got with the program Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | cyclades: use tty_portAlan Cox2008-07-212-164/+158
| | | | | | | | | | | | | | Switch cyclades to use the new tty_port structure Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | tty: Clean up tiocmsetAlan Cox2008-07-211-26/+22
| | | | | | | | | | | | | | Reverse the order of one test and it gets much more readable Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | synclink: use tty_portAlan Cox2008-07-213-310/+289
| | | | | | | | | | | | | | Switch the synclink ports to use the new tty_port structure Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | stallion: use tty_portAlan Cox2008-07-212-85/+81
| | | | | | | | | | | | | | Switch the stallion driver to use the tty_port structure Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | atmel_serial: Fix tty_port breakageHaavard Skinnemoen2008-07-211-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The tty pointer has been moved into a tty_port field, so we need to use ->info->port.tty instead of just ->info->tty. Fixes these build errors: David Brownell <david-b@pacbell.net> wrote: > drivers/serial/atmel_serial.c: In function 'atmel_rx_from_ring': > drivers/serial/atmel_serial.c:665: error: 'struct uart_info' has no member named 'tty' > drivers/serial/atmel_serial.c: In function 'atmel_rx_from_dma': > drivers/serial/atmel_serial.c:672: error: 'struct uart_info' has no member named 'tty' > drivers/serial/atmel_serial.c: In function 'atmel_startup': > drivers/serial/atmel_serial.c:797: error: 'struct uart_info' has no member named 'tty' > make[2]: *** [drivers/serial/atmel_serial.o] Error 1 Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com> Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | rocket: use tty_portAlan Cox2008-07-212-64/+61
| | | | | | | | | | | | | | Switch the rocketport to use the new tty_port structure Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | mxser: use tty_portAlan Cox2008-07-211-131/+126
| | | | | | | | | | | | | | Switch mxser to use the new tty_port structure Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | moxa: use tty_portAlan Cox2008-07-211-45/+42
| | | | | | | | | | | | | | Switch MOXA to use the new tty_port structure Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | istallion: use tty_portAlan Cox2008-07-212-63/+59
| | | | | | | | | | | | | | Switch istallion to use the new tty_port structure Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | isicom: use tty_portAlan Cox2008-07-211-102/+79
| | | | | | | | | | | | | | Switch isicom to use a tty_port structure for some fields Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | riscom8: remove bogus checksAlan Cox2008-07-211-24/+8
| | | | | | | | | | | | | | | | | | Chris Malley posted a patch removing a NULL check in the riscom8 driver. Further analysis shows that even more of the tests are irrelevant so we can delete lots of stuff Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | riscom8: use tty_portAlan Cox2008-07-212-85/+68
| | | | | | | | | | | | | | Switch riscom8 to use the new tty_port structure Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | gs: use tty_portAlan Cox2008-07-218-173/+166
| | | | | | | | | | | | | | | | Switch drivers using the old "generic serial" driver to use the tty_port structures Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | esp: use tty_portAlan Cox2008-07-212-143/+138
| | | | | | | | | | | | | | Switch esp to use the new tty_port structures Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | epca: use tty_portAlan Cox2008-07-212-59/+54
| | | | | | | | | | | | | | Switch the EPCA driver to include and begin using a tty_port structure Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | tty.h: clean upAlan Cox2008-07-211-80/+85
| | | | | | | | | | | | | | Coding style clean up and white space tidy Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | Fix compile errors in SGI console drivers (linux-next tree)Takashi Iwai2008-07-2133-147/+151
| | | | | | | | | | | | | | | | | | | | The below is the patch to replace blindly all possible places, including Jack's fixes. Signed-off-by: Takashi Iwai <tiwai@suse.de> (Reviewed and checked rather than blindly added) Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | - Fix compile errors in SGI console drivers (linux-next tree)Jack Steiner2008-07-212-8/+8
| | | | | | | | | | | | | | | | | | Fix compile errors in SGI console drivers caused by changes to the tty_port structures in the linux-next tree. Signed-off-by: Jack Steiner <steiner@sgi.com> Acked-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | ttydev: fix pamc_zilog for tty pointer moveStephen Rothwell2008-07-211-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Today's linux-next build (powerpc allyesconfig) failed like this: drivers/serial/pmac_zilog.c: In function 'pmz_receive_chars': drivers/serial/pmac_zilog.c:245: error: 'struct uart_info' has no member named 'tty' drivers/serial/pmac_zilog.c:250: error: 'struct uart_info' has no member named 'tty' I applied the patch below (which builds but may, or may not, be correct). -- Cheers, Stephen Rothwell sfr@canb.auug.org.au http://www.canb.auug.org.au/~sfr/ Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | ttydev: fix pamc_zilog for tty pointer moveStephen Rothwell2008-07-214-5/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Today's linux-next build (sparc64 defconfig) failed like this: drivers/serial/sunhv.c: In function `receive_chars': drivers/serial/sunhv.c:188: error: structure has no member named `tty' drivers/serial/sunsu.c: In function `receive_chars': drivers/serial/sunsu.c:314: error: structure has no member named `tty' drivers/serial/sunsab.c: In function `receive_chars': drivers/serial/sunsab.c:121: error: structure has no member named `tty' I applied the following patch (which, again, may not be correct). Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | MN10300: Fix MN10300's serial port driver to get at its tty_structDavid Howells2008-07-211-1/+1
| | | | | | | | | | | | | | | | | | Fix MN10300's serial port driver to get at its tty_struct as this moved from struct uart_info into struct tty_port in patch: Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | serial: use tty_portAlan Cox2008-07-215-58/+60
| | | | | | | | | | | | | | | | | | Switch the serial_core based drivers to use the new tty_port structure. We can't quite use all of it yet because of the dynamically allocated extras in the serial_core layer. Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | tty: Introduce a tty_port common structureAlan Cox2008-07-212-1/+63
| | | | | | | | | | | | | | | | | | | | | | Every tty driver has its own concept of a port structure and because they all differ we cannot extract commonality. Begin fixing this by creating a structure drivers can elect to use so that over time we can push fields into this and create commonality and then introduce common methods. Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | drivers/serial/: remove CVS keywordsAdrian Bunk2008-07-2115-40/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | This patch removes CVS keywords that weren't updated for a long time in comments, printk's and MODULE_DESCRIPTION's (no printk's or MODULE_DESCRIPTION's are completely removed). While doing this I also found and fixed a missing \n in a printk in m32r_sio.c Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | drivers/char/rio/: remove VCS tagsAdrian Bunk2008-07-2126-121/+0
| | | | | | | | | | | | | | | | | | This patch removes ancient VCS tags (either protected by #ifdef SCCS_LABELS or commented out). Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | tty: Ldisc revampAlan Cox2008-07-2132-246/+286
| | | | | | | | | | | | | | | | | | | | | | | | Move the line disciplines towards a conventional ->ops arrangement. For the moment the actual 'tty_ldisc' struct in the tty is kept as part of the tty struct but this can then be changed if it turns out that when it all settles down we want to refcount ldiscs separately to the tty. Pull the ldisc code out of /proc and put it with our ldisc code. Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | tty: isicom, enable/disable pci deviceJiri Slaby2008-07-211-0/+8
| | | | | | | | | | | | | | | | | | | | Don't forget to enable and disable PCI devices. The device might be unusable without that. Signed-off-by: Jiri Slaby <jirislaby@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | Subject: [PATCH 2/2] atmel_serial: Implement flush_buffer() hookHaavard Skinnemoen2008-07-211-0/+15
| | | | | | | | | | | | | | | | | | Avoid dumping garbage to the serial port when the tty is flushed. This tends to happen when rebooting from a serial console. Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com> Acked-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | Subject: [PATCH 1/2] serial: Add flush_buffer() operation to uart_opsHaavard Skinnemoen2008-07-213-0/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Serial drivers using DMA (like the atmel_serial driver) tend to get very confused when the xmit buffer is flushed and nobody told them. They also tend to spew a lot of garbage since the DMA engine keeps running after the buffer is flushed and possibly refilled with unrelated data. This patch adds a new flush_buffer operation to the uart_ops struct, along with a call to it from uart_flush_buffer() right after the xmit buffer has been cleared. The driver can implement this in order to syncronize its internal DMA state with the xmit buffer when the buffer is flushed. Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com> Acked-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | simserial: Fix up for ldisc changesAlan Cox2008-07-211-43/+3
|/ | | | | | | | Noted by Tony Luck although I've done the patches differently and also removed some other bogus oddments. Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Merge branch 'upstream-linus' of ↵Linus Torvalds2008-07-1722-127/+383
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/mfasheh/ocfs2 * 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mfasheh/ocfs2: [PATCH] ocfs2: fix oops in mmap_truncate testing configfs: call drop_link() to cleanup after create_link() failure configfs: Allow ->make_item() and ->make_group() to return detailed errors. configfs: Fix failing mkdir() making racing rmdir() fail configfs: Fix deadlock with racing rmdir() and rename() configfs: Make configfs_new_dirent() return error code instead of NULL configfs: Protect configfs_dirent s_links list mutations configfs: Introduce configfs_dirent_lock ocfs2: Don't snprintf() without a format. ocfs2: Fix CONFIG_OCFS2_DEBUG_FS #ifdefs ocfs2/net: Silence build warnings on sparc64 ocfs2: Handle error during journal load ocfs2: Silence an error message in ocfs2_file_aio_read() ocfs2: use simple_read_from_buffer() ocfs2: fix printk format warnings with OCFS2_FS_STATS=n [PATCH 2/2] ocfs2: Instrument fs cluster locks [PATCH 1/2] ocfs2: Add CONFIG_OCFS2_FS_STATS config option
| * [PATCH] ocfs2: fix oops in mmap_truncate testingColy Li2008-07-171-3/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch fixes a mmap_truncate bug which was found by ocfs2 test suite. In an ocfs2 cluster more than 1 node, run program mmap_truncate, which races mmap writes and truncates from multiple processes. While the test is running, a stat from another node forces writeout, causing an oops in ocfs2_get_block() because it sees a buffer to write which isn't allocated. This patch fixed the bug by clear dirty and uptodate bits in buffer, leave the buffer unmapped and return. Fix is suggested by Mark Fasheh, and I code up the patch. Signed-off-by: Coly Li <coyli@suse.de> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
| * configfs: call drop_link() to cleanup after create_link() failureLouis Rilling2008-07-141-1/+5
| | | | | | | | | | | | | | | | | | | | When allow_link() succeeds but create_link() fails, the subsystem is not informed of the failure. This patch fixes this by calling drop_link() on create_link() failures. Signed-off-by: Louis Rilling <Louis.Rilling@kerlabs.com> Signed-off-by: Joel Becker <joel.becker@oracle.com>
| * configfs: Allow ->make_item() and ->make_group() to return detailed errors.Joel Becker2008-07-148-64/+94
| | | | | | | | | | | | | | | | | | | | | | | | | | | | The configfs operations ->make_item() and ->make_group() currently return a new item/group. A return of NULL signifies an error. Because of this, -ENOMEM is the only return code bubbled up the stack. Multiple folks have requested the ability to return specific error codes when these operations fail. This patch adds that ability by changing the ->make_item/group() ops to return an int. Also updated are the in-kernel users of configfs. Signed-off-by: Joel Becker <joel.becker@oracle.com>
| * configfs: Fix failing mkdir() making racing rmdir() failLouis Rilling2008-07-142-10/+44
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When fixing the rename() vs rmdir() deadlock, we stopped locking default groups' inodes in configfs_detach_prep(), letting racing mkdir() in default groups proceed concurrently. This enables races like below happen, which leads to a failing mkdir() making rmdir() fail, despite the group to remove having no user-created directory under it in the end. process A: process B: /* PWD=A/B */ mkdir("C") make_item("C") attach_group("C") rmdir("A") detach_prep("A") detach_prep("B") error because of "C" return -ENOTEMPTY attach_group("C/D") error (eg -ENOMEM) return -ENOMEM This patch prevents such scenarii by making rmdir() wait as long as detach_prep() fails because a racing mkdir() is in the middle of attach_group(). To achieve this, mkdir() sets a flag CONFIGFS_USET_IN_MKDIR in parent's configfs_dirent before calling attach_group(), and clears the flag once attach_group() is done. detach_prep() fails with -EAGAIN whenever the flag is hit and returns the guilty inode's mutex so that rmdir() can wait on it. Signed-off-by: Louis Rilling <Louis.Rilling@kerlabs.com> Signed-off-by: Joel Becker <joel.becker@oracle.com>
| * configfs: Fix deadlock with racing rmdir() and rename()Louis Rilling2008-07-141-20/+25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch fixes the deadlock between racing sys_rename() and configfs_rmdir(). The idea is to avoid locking i_mutexes of default groups in configfs_detach_prep(), and rely instead on the new configfs_dirent_lock to protect against configfs_dirent's linkage mutations. To ensure that an mkdir() racing with rmdir() will not create new items in a to-be-removed default group, we make configfs_new_dirent() check for the CONFIGFS_USET_DROPPING flag right before linking the new dirent, and return error if the flag is set. This makes racing mkdir()/symlink()/dir_open() fail in places where errors could already happen, resp. in (attach_item()|attach_group())/create_link()/new_dirent(). configfs_depend() remains safe since it locks all the path from configfs root, and is thus mutually exclusive with rmdir(). An advantage of this is that now detach_groups() unconditionnaly takes the default groups i_mutex, which makes it more consistent with populate_groups(). Signed-off-by: Louis Rilling <Louis.Rilling@kerlabs.com> Signed-off-by: Joel Becker <joel.becker@oracle.com>
| * configfs: Make configfs_new_dirent() return error code instead of NULLLouis Rilling2008-07-141-4/+5
| | | | | | | | | | | | | | | | This patch makes configfs_new_dirent return negative error code instead of NULL, which will be useful in the next patch to differentiate ENOMEM from ENOENT. Signed-off-by: Louis Rilling <Louis.Rilling@kerlabs.com> Signed-off-by: Joel Becker <joel.becker@oracle.com>
| * configfs: Protect configfs_dirent s_links list mutationsLouis Rilling2008-07-142-4/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Symlinks to a config_item are listed under its configfs_dirent s_links, but the list mutations are not protected by any common lock. This patch uses the configfs_dirent_lock spinlock to add the necessary protection. Note: we should also protect the list_empty() test in configfs_detach_prep() but 1/ the lock should not be released immediately because nothing would prevent the list from being filled after a successful list_empty() test, making the problem tricky, 2/ this will be solved by the rmdir() vs rename() deadlock bugfix. Signed-off-by: Louis Rilling <Louis.Rilling@kerlabs.com> Signed-off-by: Joel Becker <joel.becker@oracle.com>
| * configfs: Introduce configfs_dirent_lockLouis Rilling2008-07-144-1/+34
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch introduces configfs_dirent_lock spinlock to protect configfs_dirent traversals against linkage mutations (add/del/move). This will allow configfs_detach_prep() to avoid locking i_mutexes. Locking rules for configfs_dirent linkage mutations are the same plus the requirement of taking configfs_dirent_lock. For configfs_dirent walking, one can either take appropriate i_mutex as before, or take configfs_dirent_lock. The spinlock could actually be a mutex, but the critical sections are either O(1) or should not be too long (default groups walking in last patch). ChangeLog: - Clarify the comment on configfs_dirent_lock usage - Move sd->s_element init before linking the new dirent - In lseek(), do not release configfs_dirent_lock before the dirent is relinked. Signed-off-by: Louis Rilling <Louis.Rilling@kerlabs.com> Signed-off-by: Joel Becker <joel.becker@oracle.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-25 13:31:06 +0100