summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2012-11-09Merge tag 'fixes-for-linus' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux Pull virtio and module fixes from Rusty Russell: "YA module signing build tweak, and two cc'd to stable." * tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux: virtio: Don't access index after unregister. modules: don't break modules_install on external modules with no key. module: fix out-by-one error in kallsyms
2012-11-09Merge tag 'for-linus-v3.7-rc5' of git://oss.sgi.com/xfs/xfsLinus Torvalds
Pull xfs bugfixes from Ben Myers: - fix for large transactions spanning multiple iclog buffers - zero the allocation_args structure on the stack before using it to determine whether to use a worker for allocation - move allocation stack switch to xfs_bmapi_allocate in order to prevent deadlock on AGF buffers - growfs no longer reads in garbage for new secondary superblocks - silence a build warning - ensure that invalid buffers never get written to disk while on free list - don't vmap inode cluster buffers during free - fix buffer shutdown reference count mismatch - fix reading of wrapped log data * tag 'for-linus-v3.7-rc5' of git://oss.sgi.com/xfs/xfs: xfs: fix reading of wrapped log data xfs: fix buffer shudown reference count mismatch xfs: don't vmap inode cluster buffers during free xfs: invalidate allocbt blocks moved to the free list xfs: silence uninitialised f.file warning. xfs: growfs: don't read garbage for new secondary superblocks xfs: move allocation stack switch up to xfs_bmapi_allocate xfs: introduce XFS_BMAPI_STACK_SWITCH xfs: zero allocation_args on the kernel stack xfs: only update the last_sync_lsn when a transaction completes
2012-11-09h8300: add missing L1_CACHE_SHIFTFengguang Wu
Fix the build error lib/atomic64.c: In function 'lock_addr': lib/atomic64.c:40:11: error: 'L1_CACHE_SHIFT' undeclared (first use in this function) lib/atomic64.c:40:11: note: each undeclared identifier is reported only once for each function it appears in Signed-off-by: Fengguang Wu <fengguang.wu@intel.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-11-09mm: bugfix: set current->reclaim_state to NULL while returning from kswapd()Takamori Yamaguchi
In kswapd(), set current->reclaim_state to NULL before returning, as current->reclaim_state holds reference to variable on kswapd()'s stack. In rare cases, while returning from kswapd() during memory offlining, __free_slab() and freepages() can access the dangling pointer of current->reclaim_state. Signed-off-by: Takamori Yamaguchi <takamori.yamaguchi@jp.sony.com> Signed-off-by: Aaditya Kumar <aaditya.kumar@ap.sony.com> Acked-by: David Rientjes <rientjes@google.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-11-09fanotify: fix missing breakEric Paris
Anders Blomdell noted in 2010 that Fanotify lost events and provided a test case. Eric Paris confirmed it was a bug and posted a fix to the list https://groups.google.com/forum/?fromgroups=#!topic/linux.kernel/RrJfTfyW2BE but never applied it. Repeated attempts over time to actually get him to apply it have never had a reply from anyone who has raised it So apply it anyway Signed-off-by: Alan Cox <alan@linux.intel.com> Reported-by: Anders Blomdell <anders.blomdell@control.lth.se> Cc: Eric Paris <eparis@redhat.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-11-09revert "epoll: support for disabling items, and a self-test app"Andrew Morton
Revert commit 03a7beb55b9f ("epoll: support for disabling items, and a self-test app") pending resolution of the issues identified by Michael Kerrisk, copied below. We'll revisit this for 3.8. : I've taken a look at this patch as it currently stands in 3.7-rc1, and : done a bit of testing. (By the way, the test program : tools/testing/selftests/epoll/test_epoll.c does not compile...) : : There are one or two places where the behavior seems a little strange, : so I have a question or two at the end of this mail. But other than : that, I want to check my understanding so that the interface can be : correctly documented. : : Just to go though my understanding, the problem is the following : scenario in a multithreaded application: : : 1. Multiple threads are performing epoll_wait() operations, : and maintaining a user-space cache that contains information : corresponding to each file descriptor being monitored by : epoll_wait(). : : 2. At some point, a thread wants to delete (EPOLL_CTL_DEL) : a file descriptor from the epoll interest list, and : delete the corresponding record from the user-space cache. : : 3. The problem with (2) is that some other thread may have : previously done an epoll_wait() that retrieved information : about the fd in question, and may be in the middle of using : information in the cache that relates to that fd. Thus, : there is a potential race. : : 4. The race can't solved purely in user space, because doing : so would require applying a mutex across the epoll_wait() : call, which would of course blow thread concurrency. : : Right? : : Your solution is the EPOLL_CTL_DISABLE operation. I want to : confirm my understanding about how to use this flag, since : the description that has accompanied the patches so far : has been a bit sparse : : 0. In the scenario you're concerned about, deleting a file : descriptor means (safely) doing the following: : (a) Deleting the file descriptor from the epoll interest list : using EPOLL_CTL_DEL : (b) Deleting the corresponding record in the user-space cache : : 1. It's only meaningful to use this EPOLL_CTL_DISABLE in : conjunction with EPOLLONESHOT. : : 2. Using EPOLL_CTL_DISABLE without using EPOLLONESHOT in : conjunction is a logical error. : : 3. The correct way to code multithreaded applications using : EPOLL_CTL_DISABLE and EPOLLONESHOT is as follows: : : a. All EPOLL_CTL_ADD and EPOLL_CTL_MOD operations should : should EPOLLONESHOT. : : b. When a thread wants to delete a file descriptor, it : should do the following: : : [1] Call epoll_ctl(EPOLL_CTL_DISABLE) : [2] If the return status from epoll_ctl(EPOLL_CTL_DISABLE) : was zero, then the file descriptor can be safely : deleted by the thread that made this call. : [3] If the epoll_ctl(EPOLL_CTL_DISABLE) fails with EBUSY, : then the descriptor is in use. In this case, the calling : thread should set a flag in the user-space cache to : indicate that the thread that is using the descriptor : should perform the deletion operation. : : Is all of the above correct? : : The implementation depends on checking on whether : (events & ~EP_PRIVATE_BITS) == 0 : This replies on the fact that EPOLL_CTL_AD and EPOLL_CTL_MOD always : set EPOLLHUP and EPOLLERR in the 'events' mask, and EPOLLONESHOT : causes those flags (as well as all others in ~EP_PRIVATE_BITS) to be : cleared. : : A corollary to the previous paragraph is that using EPOLL_CTL_DISABLE : is only useful in conjunction with EPOLLONESHOT. However, as things : stand, one can use EPOLL_CTL_DISABLE on a file descriptor that does : not have EPOLLONESHOT set in 'events' This results in the following : (slightly surprising) behavior: : : (a) The first call to epoll_ctl(EPOLL_CTL_DISABLE) returns 0 : (the indicator that the file descriptor can be safely deleted). : (b) The next call to epoll_ctl(EPOLL_CTL_DISABLE) fails with EBUSY. : : This doesn't seem particularly useful, and in fact is probably an : indication that the user made a logic error: they should only be using : epoll_ctl(EPOLL_CTL_DISABLE) on a file descriptor for which : EPOLLONESHOT was set in 'events'. If that is correct, then would it : not make sense to return an error to user space for this case? Cc: Michael Kerrisk <mtk.manpages@gmail.com> Cc: "Paton J. Lewis" <palewis@adobe.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-11-09checkpatch: improve network block comment style checkingJoe Perches
Some comment styles in net and drivers/net are flagged inappropriately. Avoid proclaiming inline comments like: int a = b; /* some comment */ and block comments like: /********************* * some comment ********************/ are defective. Tested with $ cat drivers/net/t.c /* foo */ /* * foo */ /* foo */ /* foo * bar */ /**************************** * some long block comment ***************************/ struct foo { int bar; /* another test */ }; $ Signed-off-by: Joe Perches <joe@perches.com> Reported-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: David Miller <davem@davemloft.net> Cc: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-11-09Merge branch 'drm-nouveau-fixes' of ↵Dave Airlie
git://anongit.freedesktop.org/git/nouveau/linux-2.6 into drm-fixes just some misc regression fixes and typo fixes. * 'drm-nouveau-fixes' of git://anongit.freedesktop.org/git/nouveau/linux-2.6: drm/nouveau: fix acpi edid retrieval drm/nvc0/disp: fix regression in vblank semaphore release drm/nv40/mpeg: fix context handling drm/nv40/graph: fix typo in type names drm/nv41/vm: fix typo in type name
2012-11-09virtio: Don't access index after unregister.Cornelia Huck
Virtio wants to release used indices after the corresponding virtio device has been unregistered. However, virtio does not hold an extra reference, giving up its last reference with device_unregister(), making accessing dev->index afterwards invalid. I actually saw problems when testing my (not-yet-merged) virtio-ccw code: - device_add virtio-net,id=xxx -> creates device virtio<n> with n>0 - device_del xxx -> deletes virtio<n>, but calls ida_simple_remove with an index of 0 - device_add virtio-net,id=xxx -> tries to add virtio0, which is still in use... So let's save the index we want to release before calling device_unregister(). Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com> Acked-by: Sjur Brændeland <sjur.brandeland@stericsson.com> Cc: stable@kernel.org Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2012-11-09drm/nouveau: fix acpi edid retrievalMaarten Lankhorst
Commit c0077061e7ea accidentally inverted the logic for nouveau_acpi_edid, causing it to only show a connector as connected when the edid could not be retrieved with acpi. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-11-09drm/nvc0/disp: fix regression in vblank semaphore releaseKelly Doran
Signed-off-by: Kelly Doran <kel.p.doran@gmail.com> Reviewed-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-11-09drm/nv40/mpeg: fix context handlingMarcin Slusarz
It slipped in thanks to typeless API. Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-11-09drm/nv40/graph: fix typo in type namesMarcin Slusarz
nv04_graph_priv / nv04_graph_chan are not defined in this context... Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-11-09drm/nv41/vm: fix typo in type nameMarcin Slusarz
It's a miracle it compiles at all - nv04_vm_priv does not exist anywhere in the tree. Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-11-09Merge branch 'drm-fixes-3.7' of git://people.freedesktop.org/~agd5f/linux ↵Dave Airlie
into drm-fixes Just some minor fixes for VM reg check and a regression fix for dce3 plls * 'drm-fixes-3.7' of git://people.freedesktop.org/~agd5f/linux: drm/radeon/si: add some missing regs to the VM reg checker drm/radeon/cayman: add some missing regs to the VM reg checker drm/radeon/dce3: switch back to old pll allocation order for discrete
2012-11-08ARM: OMAP2+: PRCM: remove obsolete prcm.[ch]Paul Walmsley
arch/arm/mach-omap2/prcm.c and arch/arm/plat-omap/include/plat/prcm.h are now completely unused and can be removed. Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com>
2012-11-08ARM: OMAP2+: hwmod: call to _omap4_disable_module() should use the ↵Paul Walmsley
SoC-specific call The hwmod code unconditionally calls _omap4_disable_module() on all SoCs when a module doesn't enable correctly. This "worked" due to the weak function omap4_cminst_wait_module_idle() in arch/arm/mach-omap2/prcm.c, which was a no-op. But now those weak functions are going away - they should not be used. So this patch will now call the SoC-specific disable_module code, assuming it exists. Needs to be done before the weak function is removed, otherwise AM33xx will crash early in boot. Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com>
2012-11-08ARM: OMAP2+: PRCM: consolidate PRCM-related timeout macrosPaul Walmsley
Consolidate all of the copies of MAX_MODULE_HARDRESET_WAIT and MAX_MODULE_SOFTRESET_WAIT into one place, arch/arm/mach-omap2/prm.h. Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com>
2012-11-08ARM: OMAP2+: PRCM: split and relocate the PRM/CM globals setupPaul Walmsley
Split omap2_set_globals_prcm() into PRM, CM, and PRCM_MPU variants, since these are all separate IP blocks. This should make it easier to move the PRM, CM, PRCM_MPU code into drivers/ in future patchsets. At this point arch/arm/plat-omap/include/plat/prcm.h is empty; a subsequent patch will remove it, and remove the #include from all the files that #include it. Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com>
2012-11-08ARM: OMAP2+: PRCM: remove omap2_cm_wait_idlest()Paul Walmsley
Now that all users of mach-omap2/omap2_cm_wait_idlest() have been removed, delete the function and its supporting macros and prototypes. Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com>
2012-11-08ARM: OMAP2+: CM/clock: convert _omap2_module_wait_ready() to use ↵Paul Walmsley
SoC-independent CM functions Convert the OMAP clock code's _omap2_module_wait_ready() to use SoC-independent CM functions that are provided by the CM code, rather than using a deprecated function from mach-omap2/prcm.c. This facilitates the future conversion of the CM code to a driver, and also removes a mach-omap2/prcm.c user. mach-omap2/prcm.c will be removed by a subsequent patch. Some modules have IDLEST registers that aren't in the CM module, such as the AM3517 IDLEST bits. So we also need a fallback function for these non-CM odd cases. Create a temporary one in mach-omap2/clock.c, intended to exist until the SCM drivers are ready. Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com>
2012-11-08ARM: OMAP2xxx: APLL/CM: convert to use omap2_cm_wait_module_ready()Paul Walmsley
Convert the OMAP2xxx APLL code to use omap2_cm_wait_module_ready(), and move the low-level CM register manipulation functions to mach-omap2/cm2xxx.c. The objectives here are to remove the dependency on the deprecated omap2_cm_wait_idlest() function in mach-omap2/prcm.c, so that code can be removed later; and move low-level register accesses to the CM IP block to the CM code, which will soon be moved into drivers/. Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com>
2012-11-08ARM: OMAP2+: board files: use SoC-specific system restart functionsPaul Walmsley
Modify the board files to use the SoC-specific system restart functions. At this point it's possible to remove omap_prcm_restart() from mach-omap2/prcm.c. While removing the prototypes for the now-unused restart functions, clean up a few more obsolete prototypes in mach-omap2/clock.h. Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com>
2012-11-08ARM: OMAP2+: PRCM: create SoC-specific chip restart functionsPaul Walmsley
Split omap_prcm_restart() from mach-omap2/prcm.c into SoC-specific variants. These functions need to be able to save the reboot reason into the scratchpad RAM. This implies a dependency on both the PRM and SCM IP blocks, so they've been moved into their own file. This will eventually call functions in the PRM and SCM drivers, once those are created. Vaibhav Hiremath <hvaibhav@ti.com> identified an unused prototype in the first version of this patch - now removed. Tony Lindgren <tony@atomide.com> noted a compile problem with some RMK Kconfigs; resolved in this patch. Signed-off-by: Paul Walmsley <paul@pwsan.com> Cc: Vaibhav Hiremath <hvaibhav@ti.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com> Cc: Tony Lindgren <tony@atomide.com>
2012-11-08ARM: OMAP2xxx: clock: move virt_prcm_set code into clkt2xxx_virt_prcm_set.cPaul Walmsley
Collect all of the virt_prcm_set-specific clocktype code into mach-omap2/clkt2xxx_virt_prcm_set.c. Remove its dependency on the 'sclk' and 'vclk' global variables. Those variables will be removed by subsequent patches. This is part of the process of cleaning up the OMAP2xxx clock code and preparing for the removal of the omap_prcm_restart() function. Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com>
2012-11-08ARM: OMAP2xxx: clock: remove global 'dclk' variablePaul Walmsley
Remove the global 'dclk' variable, instead replacing it with a variable local to the dpllcore clock type C file. This removes some of the special-case code surrounding the OMAP2xxx clock init. This patch is a prerequisite for the removal of the omap_prcm_restart() code from arch/arm/mach-omap2/prcm.c. It also cleans up some special-case OMAP2xxx clock code in the process. Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com>
2012-11-08ARM: OMAP2/3: PRM: add SoC reset functions (using the CORE DPLL method)Paul Walmsley
Add SoC reset functions into the PRM code. These functions are based on code from mach-omap2/prcm.c. They reset the SoC using the CORE DPLL reset method (as opposed to one of the other two or three chip reset methods). Adding them here will facilitate their removal from arch/arm/mach-omap2/prcm.c. (prcm.c is deprecated.) Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com>
2012-11-08ARM: OMAP2+: common: remove mach-omap2/common.c globals and map_common_io codePaul Walmsley
Get rid of the mach-omap2/common.c globals by moving the global initialization for IP block addresses that must occur early into mach-omap2/io.c. In the process, remove the *_map_common_io*() and SoC-specific *set_globals* functions. Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com>
2012-11-08ARM: OMAP2+: PRCM: remove omap_prcm_get_reset_sources()Paul Walmsley
omap_prcm_get_reset_sources() is now unused; so, remove it. Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Vaibhav Hiremath <hvaibhav@ti.com>
2012-11-08watchdog: OMAP: use standard GETBOOTSTATUS interface; use platform_data fn ptrPaul Walmsley
Previously the OMAP watchdog driver used a non-standard way to report the chip reset source via the GETBOOTSTATUS ioctl. This patch converts the driver to use the standard WDIOF_* flags for this purpose. This patch may break existing userspace code that uses the existing non-standard data format returned by the OMAP watchdog driver's GETBOOTSTATUS ioctl. To fetch detailed reset source information, userspace code will need to retrieve it directly from the CGRM or PRM drivers when those are completed. Previously, to fetch the reset source, the driver either read a register outside the watchdog IP block (OMAP1), or called a function exported directly from arch/arm/mach-omap2. Both approaches are broken. This patch also converts the driver to use a platform_data function pointer. This approach is temporary, and is due to the lack of drivers for the OMAP16xx+ Clock Generation and Reset Management IP block and the OMAP2+ Power and Reset Management IP block. Once drivers are available for those IP blocks, the watchdog driver can be converted to call exported functions from those drivers directly. At that point, the platform_data function pointer can be removed. In the short term, this patch is needed to allow the PRM code to be removed from arch/arm/mach-omap2 (it is being moved to a driver). This version integrates a fix from Jon Hunter <jon-hunter@ti.com> that avoids a NULL pointer dereference in a DT-only boot, and integrates a patch commit message fix from Felipe Balbi <balbi@ti.com>. Signed-off-by: Paul Walmsley <paul@pwsan.com> Cc: Wim Van Sebroeck <wim@iguana.be> Acked-by: Wim Van Sebroeck <wim@iguana.be> [paul@pwsan.com: integrated pdata fix from Jon Hunter] Cc: Jon Hunter <jon-hunter@ti.com> [paul@pwsan.com: integrated changelog fix from Felipe Balbi] Cc: Felipe Balbi <balbi@ti.com>
2012-11-08xfs: fix reading of wrapped log dataDave Chinner
Commit 4439647 ("xfs: reset buffer pointers before freeing them") in 3.0-rc1 introduced a regression when recovering log buffers that wrapped around the end of log. The second part of the log buffer at the start of the physical log was being read into the header buffer rather than the data buffer, and hence recovery was seeing garbage in the data buffer when it got to the region of the log buffer that was incorrectly read. Cc: <stable@vger.kernel.org> # 3.0.x, 3.2.x, 3.4.x 3.6.x Reported-by: Torsten Kaiser <just.for.lkml@googlemail.com> Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Mark Tinguely <tinguely@sgi.com> Signed-off-by: Ben Myers <bpm@sgi.com>
2012-11-08xfs: fix buffer shudown reference count mismatchDave Chinner
When we shut down the filesystem, we have to unpin and free all the buffers currently active in the CIL. To do this we unpin and remove them in one operation as a result of a failed iclogbuf write. For buffers, we do this removal via a simultated IO completion of after marking the buffer stale. At the time we do this, we have two references to the buffer - the active LRU reference and the buf log item. The LRU reference is removed by marking the buffer stale, and the active CIL reference is by the xfs_buf_iodone() callback that is run by xfs_buf_do_callbacks() during ioend processing (via the bp->b_iodone callback). However, ioend processing requires one more reference - that of the IO that it is completing. We don't have this reference, so we free the buffer prematurely and use it after it is freed. For buffers marked with XBF_ASYNC, this leads to assert failures in xfs_buf_rele() on debug kernels because the b_hold count is zero. Fix this by making sure we take the necessary IO reference before starting IO completion processing on the stale buffer, and set the XBF_ASYNC flag to ensure that IO completion processing removes all the active references from the buffer to ensure it is fully torn down. Cc: <stable@vger.kernel.org> Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Mark Tinguely <tinguely@sgi.com> Signed-off-by: Ben Myers <bpm@sgi.com>
2012-11-08xfs: don't vmap inode cluster buffers during freeDave Chinner
Inode buffers do not need to be mapped as inodes are read or written directly from/to the pages underlying the buffer. This fixes a regression introduced by commit 611c994 ("xfs: make XBF_MAPPED the default behaviour"). Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Mark Tinguely <tinguely@sgi.com> Signed-off-by: Ben Myers <bpm@sgi.com>
2012-11-08xfs: invalidate allocbt blocks moved to the free listDave Chinner
When we free a block from the alloc btree tree, we move it to the freelist held in the AGFL and mark it busy in the busy extent tree. This typically happens when we merge btree blocks. Once the transaction is committed and checkpointed, the block can remain on the free list for an indefinite amount of time. Now, this isn't the end of the world at this point - if the free list is shortened, the buffer is invalidated in the transaction that moves it back to free space. If the buffer is allocated as metadata from the free list, then all the modifications getted logged, and we have no issues, either. And if it gets allocated as userdata direct from the freelist, it gets invalidated and so will never get written. However, during the time it sits on the free list, pressure on the log can cause the AIL to be pushed and the buffer that covers the block gets pushed for write. IOWs, we end up writing a freed metadata block to disk. Again, this isn't the end of the world because we know from the above we are only writing to free space. The problem, however, is for validation callbacks. If the block was on old btree root block, then the level of the block is going to be higher than the current tree root, and so will fail validation. There may be other inconsistencies in the block as well, and currently we don't care because the block is in free space. Shutting down the filesystem because a freed block doesn't pass write validation, OTOH, is rather unfriendly. So, make sure we always invalidate buffers as they move from the free space trees to the free list so that we guarantee they never get written to disk while on the free list. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Phil White <pwhite@sgi.com> Reviewed-by: Mark Tinguely <tinguely@sgi.com> Signed-off-by: Ben Myers <bpm@sgi.com>
2012-11-08xfs: silence uninitialised f.file warning.Dave Chinner
Uninitialised variable build warning introduced by 2903ff0 ("switch simple cases of fget_light to fdget"), gcc is not smart enough to work out that the variable is not used uninitialised, and the commit removed the initialisation at declaration that the old variable had. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Mark Tinguely <tinguely@sgi.com> Signed-off-by: Ben Myers <bpm@sgi.com>
2012-11-08xfs: growfs: don't read garbage for new secondary superblocksDave Chinner
When updating new secondary superblocks in a growfs operation, the superblock buffer is read from the newly grown region of the underlying device. This is not guaranteed to be zero, so violates the underlying assumption that the unused parts of superblocks are zero filled. Get a new buffer for these secondary superblocks to ensure that the unused regions are zero filled correctly. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com> Signed-off-by: Ben Myers <bpm@sgi.com>
2012-11-08xfs: move allocation stack switch up to xfs_bmapi_allocateDave Chinner
Switching stacks are xfs_alloc_vextent can cause deadlocks when we run out of worker threads on the allocation workqueue. This can occur because xfs_bmap_btalloc can make multiple calls to xfs_alloc_vextent() and even if xfs_alloc_vextent() fails it can return with the AGF locked in the current allocation transaction. If we then need to make another allocation, and all the allocation worker contexts are exhausted because the are blocked waiting for the AGF lock, holder of the AGF cannot get it's xfs-alloc_vextent work completed to release the AGF. Hence allocation effectively deadlocks. To avoid this, move the stack switch one layer up to xfs_bmapi_allocate() so that all of the allocation attempts in a single switched stack transaction occur in a single worker context. This avoids the problem of an allocation being blocked waiting for a worker thread whilst holding the AGF. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Mark Tinguely <tinguely@sgi.com> Signed-off-by: Ben Myers <bpm@sgi.com>
2012-11-08xfs: introduce XFS_BMAPI_STACK_SWITCHDave Chinner
Certain allocation paths through xfs_bmapi_write() are in situations where we have limited stack available. These are almost always in the buffered IO writeback path when convertion delayed allocation extents to real extents. The current stack switch occurs for userdata allocations, which means we also do stack switches for preallocation, direct IO and unwritten extent conversion, even those these call chains have never been implicated in a stack overrun. Hence, let's target just the single stack overun offended for stack switches. To do that, introduce a XFS_BMAPI_STACK_SWITCH flag that the caller can pass xfs_bmapi_write() to indicate it should switch stacks if it needs to do allocation. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Mark Tinguely <tinguely@sgi.com> Signed-off-by: Ben Myers <bpm@sgi.com>
2012-11-08xfs: zero allocation_args on the kernel stackMark Tinguely
Zero the kernel stack space that makes up the xfs_alloc_arg structures. Signed-off-by: Mark Tinguely <tinguely@sgi.com> Reviewed-by: Ben Myers <bpm@sgi.com> Signed-off-by: Ben Myers <bpm@sgi.com>
2012-11-08xfs: only update the last_sync_lsn when a transaction completesDave Chinner
The log write code stamps each iclog with the current tail LSN in the iclog header so that recovery knows where to find the tail of thelog once it has found the head. Normally this is taken from the first item on the AIL - the log item that corresponds to the oldest active item in the log. The problem is that when the AIL is empty, the tail lsn is dervied from the the l_last_sync_lsn, which is the LSN of the last iclog to be written to the log. In most cases this doesn't happen, because the AIL is rarely empty on an active filesystem. However, when it does, it opens up an interesting case when the transaction being committed to the iclog spans multiple iclogs. That is, the first iclog is stamped with the l_last_sync_lsn, and IO is issued. Then the next iclog is setup, the changes copied into the iclog (takes some time), and then the l_last_sync_lsn is stamped into the header and IO is issued. This is still the same transaction, so the tail lsn of both iclogs must be the same for log recovery to find the entire transaction to be able to replay it. The problem arises in that the iclog buffer IO completion updates the l_last_sync_lsn with it's own LSN. Therefore, If the first iclog completes it's IO before the second iclog is filled and has the tail lsn stamped in it, it will stamp the LSN of the first iclog into it's tail lsn field. If the system fails at this point, log recovery will not see a complete transaction, so the transaction will no be replayed. The fix is simple - the l_last_sync_lsn is updated when a iclog buffer IO completes, and this is incorrect. The l_last_sync_lsn shoul dbe updated when a transaction is completed by a iclog buffer IO. That is, only iclog buffers that have transaction commit callbacks attached to them should update the l_last_sync_lsn. This means that the last_sync_lsn will only move forward when a commit record it written, not in the middle of a large transaction that is rolling through multiple iclog buffers. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Mark Tinguely <tinguely@sgi.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Ben Myers <bpm@sgi.com>
2012-11-08arm64: mm: fix booting on systems with no memory below 4GBWill Deacon
Booting on a system with all of its memory above the 4GB boundary breaks for two reasons: (1) We still try to create a non-empty DMA32 zone (2) no-bootmem limits allocations to 0xffffffff This patch fixes these issues for ARM64. Tested-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2012-11-08arm64: smp: add missing completion for secondary bootWill Deacon
Commit 149c24151e85 ("ARM: SMP: use a timing out completion for cpu hotplug") modified arm's CPU up path to use completions. It seems that we only got half of this patch for arm64, so add the missing call to complete. Reported-by: Jon Brawn <jon.brawn@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2012-11-08arm64: compat: select CONFIG_ARCH_WANT_COMPAT_IPC_PARSE_VERSIONWill Deacon
Commit c1d7e01d7877 ("ipc: use Kconfig options for __ARCH_WANT_[COMPAT_]IPC_PARSE_VERSION") replaced the __ARCH_WANT_COMPAT_IPC_PARSE_VERSION token with a corresponding Kconfig option instead. This patch updates arm64 to use the latter, rather than #define an unused token. Signed-off-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2012-11-08arm64: elf: fix core dumping definitions for GP and FP registersWill Deacon
struct user_fp does not exist for arm64, so use struct user_fpsimd_state instead for the ELF core dumping definitions. Furthermore, since we use regset-based core dumping, we do not need definitions for dump_task_regs and dump_fpu. Signed-off-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2012-11-08arm64: perf: use architected event for CPU cycle counterWill Deacon
We currently use a fake event encoding (0xFF) to indicate CPU cycles so that we don't waste an event counter and can target the hardware cycle counter instead. The problem with this approach is that the event space defined by the architecture permits an implementation to allocate 0xFF for some other event. This patch uses the architected cycle counter encoding (0x11) so that we avoid potentially clashing with event encodings on future CPU implementations. Signed-off-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2012-11-08drm/radeon/si: add some missing regs to the VM reg checkerAlex Deucher
This register is needed for streamout to work properly. Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
2012-11-08drm/radeon/cayman: add some missing regs to the VM reg checkerAlex Deucher
These regs were being wronly rejected leading to rendering issues. fixes: https://bugs.freedesktop.org/show_bug.cgi?id=56876 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
2012-11-08ALSA: Fix card refcount unbalanceTakashi Iwai
There are uncovered cases whether the card refcount introduced by the commit a0830dbd isn't properly increased or decreased: - OSS PCM and mixer success paths - When lookup function gets NULL This patch fixes these places. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=50251 Cc: <stable@vger.kernel.org> Signed-off-by: Takashi Iwai <tiwai@suse.de>
2012-11-08ALSA: hda - Add new codec ALC668 and ALC900 (default name ALC1150)Kailang Yang
Signed-off-by: Kailang Yang <kailang@realtek.com> Cc: <stable@vger.kernel.org> Signed-off-by: Takashi Iwai <tiwai@suse.de>
2012-11-08ALSA: hda - Improve HP depop when system enter to S3Kailang Yang
alc269_toggle_power_output() was only use in ALC269VB. I rename it to alc269vb_toggle_power_output(). Signed-off-by: Kailang Yang <kailang@realtek.com> Cc: <stable@vger.kernel.org> Signed-off-by: Takashi Iwai <tiwai@suse.de>