- 26 Jan, 2017 34 commits
-
-
Ondrej Kozina authored
commit 265e9098 upstream. In crypt_set_key(), if a failure occurs while replacing the old key (e.g. tfm->setkey() fails) the key must not have DM_CRYPT_KEY_VALID flag set. Otherwise, the crypto layer would have an invalid key that still has DM_CRYPT_KEY_VALID flag set. Signed-off-by:
Ondrej Kozina <okozina@redhat.com> Reviewed-by:
Mikulas Patocka <mpatocka@redhat.com> Signed-off-by:
Mike Snitzer <snitzer@redhat.com> Signed-off-by:
Jiri Slaby <jslaby@suse.cz>
-
Aleksa Sarai authored
commit 613cc2b6 upstream. If you have a process that has set itself to be non-dumpable, and it then undergoes exec(2), any CLOEXEC file descriptors it has open are "exposed" during a race window between the dumpable flags of the process being reset for exec(2) and CLOEXEC being applied to the file descriptors. This can be exploited by a process by attempting to access /proc/<pid>/fd/... during this window, without requiring CAP_SYS_PTRACE. The race in question is after set_dumpable has been (for get_link, though the trace is basically the same for readlink): [vfs] -> proc_pid_link_inode_operations.get_link -> proc_pid_get_link -> proc_fd_access_allowed -> ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS); Which will return 0, during the race window and CLOEXEC file descriptors will still be open during this window because do_close_on_exec has not been called yet. As a result, the ordering of these calls should be reversed to avoid this race window. This is of particular concern to container runtimes, where joining a PID namespace with file descriptors referring to the host filesystem can result in security issues (since PRCTL_SET_DUMPABLE doesn't protect against access of CLOEXEC file descriptors -- file descriptors which may reference filesystem objects the container shouldn't have access to). Cc: dev@opencontainers.org Reported-by:
Michael Crosby <crosbymichael@gmail.com> Signed-off-by:
Aleksa Sarai <asarai@suse.de> Signed-off-by:
Al Viro <viro@zeniv.linux.org.uk> Signed-off-by:
-
Shaohua Li authored
commit 5f33a080 upstream. Our system uses significantly more slab memory with memcg enabled with the latest kernel. With 3.10 kernel, slab uses 2G memory, while with 4.6 kernel, 6G memory is used. The shrinker has problem. Let's see we have two memcg for one shrinker. In do_shrink_slab: 1. Check cg1. nr_deferred = 0, assume total_scan = 700. batch size is 1024, then no memory is freed. nr_deferred = 700 2. Check cg2. nr_deferred = 700. Assume freeable = 20, then total_scan = 10 or 40. Let's assume it's 10. No memory is freed. nr_deferred = 10. The deferred share of cg1 is lost in this case. kswapd will free no memory even run above steps again and again. The fix makes sure one memcg's deferred share isn't lost. Link: http://lkml.kernel.org/r/2414be961b5d25892060315fbb56bb19d81d0c07.1476227351.git.shli@fb.comSigned-off-by:
Shaohua Li <shli@fb.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
-
Nicolai Stange authored
commit 05e6ea26 upstream. The struct file_operations instance serving the f2fs/status debugfs file lacks an initialization of its ->owner. This means that although that file might have been opened, the f2fs module can still get removed. Any further operation on that opened file, releasing included, will cause accesses to unmapped memory. Indeed, Mike Marshall reported the following: BUG: unable to handle kernel paging request at ffffffffa0307430 IP: [<ffffffff8132a224>] full_proxy_release+0x24/0x90 <...> Call Trace: [] __fput+0xdf/0x1d0 [] ____fput+0xe/0x10 [] task_work_run+0x8e/0xc0 [] do_exit+0x2ae/0xae0 [] ? __audit_syscall_entry+0xae/0x100 [] ? syscall_trace_enter+0x1ca/0x310 [] do_group_exit+0x44/0xc0 [] SyS_exit_group+0x14/0x20 [] do_syscall_64+0x61/0x150 [] entry_SYSCALL64_slow_path+0x25/0x25 <...> ---[ end trace f22ae883fa3ea6b8 ]--- Fixing recursive fault but reboot is needed! Fix this by initializing the f2fs/status file_operations' ->owner with THIS_MODULE. This will allow debugfs to grab a reference to the f2fs module upon any open on that file, thus preventing it from getting removed. Fixes: 902829aa ("f2fs: move proc files to debugfs") Reported-by:
Mike Marshall <hubcap@omnibond.com> Reported-by:
Martin Brandenburg <martin@omnibond.com> Signed-off-by:
Nicolai Stange <nicstange@gmail.com> Signed-off-by:
Jaegeuk Kim <jaegeuk@kernel.org> Signed-off-by:
-
Dan Carpenter authored
commit 578620f4 upstream. We should set the error code if kzalloc() fails. Fixes: 67cf5b09 ("ext4: add the basic function for inline data support") Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
Theodore Ts'o <tytso@mit.edu> Signed-off-by:
-
Darrick J. Wong authored
commit 7e6e1ef4 upstream. Don't load an inode with a negative size; this causes integer overflow problems in the VFS. [ Added EXT4_ERROR_INODE() to mark file system as corrupted. -TYT] js: use EIO for 3.12 instead of EFSCORRUPTED. Fixes: a48380f7 (ext4: rename i_dir_acl to i_size_high) Signed-off-by:
Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by:
-
Theodore Ts'o authored
commit c48ae41b upstream. The commit "ext4: sanity check the block and cluster size at mount time" should prevent any problems, but in case the superblock is modified while the file system is mounted, add an extra safety check to make sure we won't overrun the allocated buffer. Signed-off-by:
-
Theodore Ts'o authored
commit 5aee0f8a upstream. Fix a large number of problems with how we handle mount options in the superblock. For one, if the string in the superblock is long enough that it is not null terminated, we could run off the end of the string and try to interpret superblocks fields as characters. It's unlikely this will cause a security problem, but it could result in an invalid parse. Also, parse_options is destructive to the string, so in some cases if there is a comma-separated string, it would be modified in the superblock. (Fortunately it only happens on file systems with a 1k block size.) Signed-off-by:
-
Theodore Ts'o authored
commit cd6bb35b upstream. Centralize the checks for inodes_per_block and be more strict to make sure the inodes_per_block_group can't end up being zero. Signed-off-by:
Andreas Dilger <adilger@dilger.ca> Signed-off-by:
-
Chandan Rajendra authored
commit 30a9d7af upstream. The number of 'counters' elements needed in 'struct sg' is super_block->s_blocksize_bits + 2. Presently we have 16 'counters' elements in the array. This is insufficient for block sizes >= 32k. In such cases the memcpy operation performed in ext4_mb_seq_groups_show() would cause stack memory corruption. Fixes: c9de560dSigned-off-by:
Chandan Rajendra <chandan@linux.vnet.ibm.com> Signed-off-by:
Jan Kara <jack@suse.cz> Signed-off-by:
-
Chandan Rajendra authored
commit 69e43e8c upstream. 'border' variable is set to a value of 2 times the block size of the underlying filesystem. With 64k block size, the resulting value won't fit into a 16-bit variable. Hence this commit changes the data type of 'border' to 'unsigned int'. Fixes: c9de560dSigned-off-by:
-
Alex Porosanu authored
commit d128af17 upstream. The AEAD givenc descriptor relies on moving the IV through the output FIFO and then back to the CTX2 for authentication. The SEQ FIFO STORE could be scheduled before the data can be read from OFIFO, especially since the SEQ FIFO LOAD needs to wait for the SEQ FIFO LOAD SKIP to finish first. The SKIP takes more time when the input is SG than when it's a contiguous buffer. If the SEQ FIFO LOAD is not scheduled before the STORE, the DECO will hang waiting for data to be available in the OFIFO so it can be transferred to C2. In order to overcome this, first force transfer of IV to C2 by starting the "cryptlen" transfer first and then starting to store data from OFIFO to the output buffer. Fixes: 1acebad3 ("crypto: caam - faster aead implementation") Signed-off-by:
Alex Porosanu <alexandru.porosanu@nxp.com> Signed-off-by:
Horia Geantă <horia.geanta@nxp.com> Signed-off-by:
Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by:
-
NeilBrown authored
commit bcc7f5b4 upstream. bdev->bd_contains is not stable before calling __blkdev_get(). When __blkdev_get() is called on a parition with ->bd_openers == 0 it sets bdev->bd_contains = bdev; which is not correct for a partition. After a call to __blkdev_get() succeeds, ->bd_openers will be > 0 and then ->bd_contains is stable. When FMODE_EXCL is used, blkdev_get() calls bd_start_claiming() -> bd_prepare_to_claim() -> bd_may_claim() This call happens before __blkdev_get() is called, so ->bd_contains is not stable. So bd_may_claim() cannot safely use ->bd_contains. It currently tries to use it, and this can lead to a BUG_ON(). This happens when a whole device is already open with a bd_holder (in use by dm in my particular example) and two threads race to open a partition of that device for the first time, one opening with O_EXCL and one without. The thread that doesn't use O_EXCL gets through blkdev_get() to __blkdev_get(), gains the ->bd_mutex, and sets bdev->bd_contains = bdev; Immediately thereafter the other thread, using FMODE_EXCL, calls bd_start_claiming() from blkdev_get(). This should fail because the whole device has a holder, but because bdev->bd_contains == bdev bd_may_claim() incorrectly reports success. This thread continues and blocks on bd_mutex. The first thread then sets bdev->bd_contains correctly and drops the mutex. The thread using FMODE_EXCL then continues and when it calls bd_may_claim() again in: BUG_ON(!bd_may_claim(bdev, whole, holder)); The BUG_ON fires. Fix this by removing the dependency on ->bd_contains in bd_may_claim(). As bd_may_claim() has direct access to the whole device, it can simply test if the target bdev is the whole device. Fixes: 6b4517a7 ("block: implement bd_claiming and claiming block") Signed-off-by:
NeilBrown <neilb@suse.com> Signed-off-by:
Jens Axboe <axboe@fb.com> Signed-off-by:
-
Liu Bo authored
commit 2571e739 upstream. So we can read a btree block via readahead or intentional read, and we can end up with a memory leak when something happens as follows, 1) readahead starts to read block A but does not wait for read completion, 2) btree_readpage_end_io_hook finds that block A is corrupted, and it needs to clear all block A's pages' uptodate bit. 3) meanwhile an intentional read kicks in and checks block A's pages' uptodate to decide which page needs to be read. 4) when some pages have the uptodate bit during 3)'s check so 3) doesn't count them for eb->io_pages, but they are later cleared by 2) so we has to readpage on the page, we get the wrong eb->io_pages which results in a memory leak of this block. This fixes the problem by firstly getting all pages's locking and then checking pages' uptodate bit. t1(readahead) t2(readahead endio) t3(the following read) read_extent_buffer_pages end_bio_extent_readpage for pg in eb: for page 0,1,2 in eb: if pg is uptodate: btree_readpage_end_io_hook(pg) num_reads++ if uptodate: eb->io_pages = num_reads SetPageUptodate(pg) _______________ for pg in eb: for page 3 in eb: read_extent_buffer_pages if pg is NOT uptodate: btree_readpage_end_io_hook(pg) for pg in eb: __extent_read_full_page(pg) sanity check reports something wrong if pg is uptodate: clear_extent_buffer_uptodate(eb) num_reads++ for pg in eb: eb->io_pages = num_reads ClearPageUptodate(page) _______________ for pg in eb: if pg is NOT uptodate: __extent_read_full_page(pg) So t3's eb->io_pages is not consistent with the number of pages it's reading, and during endio(), atomic_dec_and_test(&eb->io_pages) will get a negative number so that we're not able to free the eb. Signed-off-by:
Liu Bo <bo.li.liu@oracle.com> Reviewed-by:
David Sterba <dsterba@suse.com> Signed-off-by:
-
Takashi Iwai authored
commit f73cd43a upstream. HP Z1 Gen3 AiO with Conexant codec doesn't give an unsolicited event to the headset mic pin upon the jack plugging, it reports only to the headphone pin. It results in the missing mic switching. Let's fix up by simply gating the jack event. Signed-off-by:
Takashi Iwai <tiwai@suse.de> Signed-off-by:
-
Jussi Laako authored
commit 995c6a7f upstream. Sampling rate changes after first set one are not reflected to the hardware, while driver and ALSA think the rate has been changed. Fix the problem by properly stopping the interface at the beginning of prepare call, allowing new rate to be set to the hardware. This keeps the hardware in sync with the driver. Signed-off-by:
Jussi Laako <jussi@sonarnerd.net> Signed-off-by:
-
Con Kolivas authored
commit 82ffb6fc upstream. The Logitech QuickCam Communicate Deluxe/S7500 microphone fails with the following warning. [ 6.778995] usb 2-1.2.2.2: Warning! Unlikely big volume range (=3072), cval->res is probably wrong. [ 6.778996] usb 2-1.2.2.2: [5] FU [Mic Capture Volume] ch = 1, val = 4608/7680/1 Adding it to the list of devices in volume_control_quirks makes it work properly, fixing related typo. Signed-off-by:
Con Kolivas <kernel@kolivas.org> Signed-off-by:
-
Alan Stern authored
commit ccdb6be9 upstream. The UHCI controllers in Intel chipsets rely on a platform-specific non-PME mechanism for wakeup signalling. They can generate wakeup signals even though they don't support PME. We need to let the USB core know this so that it will enable runtime suspend for UHCI controllers. Signed-off-by:
Alan Stern <stern@rowland.harvard.edu> Signed-off-by:
Bjorn Helgaas <bhelgaas@google.com> Acked-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by:
-
Felipe Balbi authored
commit e8f29bb7 upstream. usb_endpoint_maxp() returns wMaxPacketSize in its raw form. Without taking into consideration that it also contains other bits reserved for isochronous endpoints. This patch fixes one occasion where this is a problem by making sure that we initialize ep->maxpacket only with lower 10 bits of the value returned by usb_endpoint_maxp(). Note that seperate patches will be necessary to audit all call sites of usb_endpoint_maxp() and make sure that usb_endpoint_maxp() only returns lower 10 bits of wMaxPacketSize. Signed-off-by:
Felipe Balbi <felipe.balbi@linux.intel.com> Signed-off-by:
-
Mathias Nyman authored
commit 37be6676 upstream. USB-3 does not have any link state that will avoid negotiating a connection with a plugged-in cable but will signal the host when the cable is unplugged. For USB-3 we used to first set the link to Disabled, then to RxDdetect to be able to detect cable connects or disconnects. But in RxDetect the connected device is detected again and eventually enabled. Instead set the link into U3 and disable remote wakeups for the device. This is what Windows does, and what Alan Stern suggested. Cc: Alan Stern <stern@rowland.harvard.edu> Acked-by:
Mathias Nyman <mathias.nyman@linux.intel.com> Signed-off-by:
-
Nathaniel Quillin authored
commit 30121604 upstream. Add device-id entry for GW Instek AFG-125, which has a byte swapped bInterfaceSubClass (0x20). Signed-off-by:
Nathaniel Quillin <ndq@google.com> Acked-by:
Oliver Neukum <oneukum@suse.com> Signed-off-by:
-
Johan Hovold authored
commit 6774d5f5 upstream. Kill urbs and disable read before returning from open on failure to retrieve the line state. Fixes: 1da177e4 ("Linux-2.6.12-rc2") Signed-off-by:
Johan Hovold <johan@kernel.org> Signed-off-by:
-
Giuseppe Lippolis authored
commit d8a12b71 upstream. Adding registration for 3G modem DWM-158 in usb-serial-option Signed-off-by:
Giuseppe Lippolis <giu.lippolis@gmail.com> Signed-off-by:
-
Daniele Palmas authored
commit 5b09eff0 upstream. This patch adds support for PIDs 0x1040, 0x1041 of Telit LE922A. Since the interface positions are the same than the ones used for other Telit compositions, previous defined blacklists are used. Signed-off-by:
Daniele Palmas <dnlplm@gmail.com> Signed-off-by:
-
Robbie Ko authored
commit 2a7bf53f upstream. If a log tree has a layout like the following: leaf N: ... item 240 key (282 DIR_LOG_ITEM 0) itemoff 8189 itemsize 8 dir log end 1275809046 leaf N + 1: item 0 key (282 DIR_LOG_ITEM 3936149215) itemoff 16275 itemsize 8 dir log end 18446744073709551615 ... When we pass the value 1275809046 + 1 as the parameter start_ret to the function tree-log.c:find_dir_range() (done by replay_dir_deletes()), we end up with path->slots[0] having the value 239 (points to the last item of leaf N, item 240). Because the dir log item in that position has an offset value smaller than *start_ret (1275809046 + 1) we need to move on to the next leaf, however the logic for that is wrong since it compares the current slot to the number of items in the leaf, which is smaller and therefore we don't lookup for the next leaf but instead we set the slot to point to an item that does not exist, at slot 240, and we later operate on that slot which has unexpected content or in the worst case can result in an invalid memory access (accessing beyond the last page of leaf N's extent buffer). So fix the logic that checks when we need to lookup at the next leaf by first incrementing the slot and only after to check if that slot is beyond the last item of the current leaf. Signed-off-by:
Robbie Ko <robbieko@synology.com> Reviewed-by:
Filipe Manana <fdmanana@suse.com> Fixes: e02119d5 (Btrfs: Add a write ahead tree log to optimize synchronous operations) Signed-off-by:
-
Jeff Mahoney authored
This reverts commit 644d1071, upstream commit 6f896054. The original patch for mainline, 6f896054 (Btrfs: don't delay inode ref updates during log replay) lists 1d52c78a (Btrfs: try not to ENOSPC on log replay) as the only pre-3.18 dependency, but it also depends on 67de1176 (Btrfs: introduce the delayed inode ref deletion for the single link inode), which was introduced in 3.14 and isn't in 3.12.y. The -stable commit added the check to btrfs_delayed_update_inode, which may look similar to btrfs_delayed_delete_inode_ref, but it's only superficial. The tops of both functions handle typical delayed node boilerplate. The upshot is that the patch is harmless since the caller already checks to see if we're doing log recovery, so we're not breaking anything. It should be reverted because it makes it appear as if this issue was fixed for users who did backport 67de1176, when it is not. Signed-off-by:
Jeff Mahoney <jeffm@suse.com> Signed-off-by:
-
Michal Hocko authored
commit 777c6e0d upstream. Yu Zhao has noticed that __unregister_cpu_notifier only unregisters its notifiers when HOTPLUG_CPU=y while the registration might succeed even when HOTPLUG_CPU=n if MODULE is enabled. This means that e.g. zswap might keep a stale notifier on the list on the manual clean up during the pool tear down and thus corrupt the list. Resulting in the following [ 144.964346] BUG: unable to handle kernel paging request at ffff880658a2be78 [ 144.971337] IP: [<ffffffffa290b00b>] raw_notifier_chain_register+0x1b/0x40 <snipped> [ 145.122628] Call Trace: [ 145.125086] [<ffffffffa28e5cf8>] __register_cpu_notifier+0x18/0x20 [ 145.131350] [<ffffffffa2a5dd73>] zswap_pool_create+0x273/0x400 [ 145.137268] [<ffffffffa2a5e0fc>] __zswap_param_set+0x1fc/0x300 [ 145.143188] [<ffffffffa2944c1d>] ? trace_hardirqs_on+0xd/0x10 [ 145.149018] [<ffffffffa2908798>] ? kernel_param_lock+0x28/0x30 [ 145.154940] [<ffffffffa2a3e8cf>] ? __might_fault+0x4f/0xa0 [ 145.160511] [<ffffffffa2a5e237>] zswap_compressor_param_set+0x17/0x20 [ 145.167035] [<ffffffffa2908d3c>] param_attr_store+0x5c/0xb0 [ 145.172694] [<ffffffffa290848d>] module_attr_store+0x1d/0x30 [ 145.178443] [<ffffffffa2b2b41f>] sysfs_kf_write+0x4f/0x70 [ 145.183925] [<ffffffffa2b2a5b9>] kernfs_fop_write+0x149/0x180 [ 145.189761] [<ffffffffa2a99248>] __vfs_write+0x18/0x40 [ 145.194982] [<ffffffffa2a9a412>] vfs_write+0xb2/0x1a0 [ 145.200122] [<ffffffffa2a9a732>] SyS_write+0x52/0xa0 [ 145.205177] [<ffffffffa2ff4d97>] entry_SYSCALL_64_fastpath+0x12/0x17 This can be even triggered manually by changing /sys/module/zswap/parameters/compressor multiple times. Fix this issue by making unregister APIs symmetric to the register so there are no surprises. [js] backport to 3.12 Fixes: 47e627bc ("[PATCH] hotplug: Allow modules to use the cpu hotplug notifiers even if !CONFIG_HOTPLUG_CPU") Reported-and-tested-by:
Yu Zhao <yuzhao@google.com> Signed-off-by:
Michal Hocko <mhocko@suse.com> Cc: linux-mm@kvack.org Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Dan Streetman <ddstreet@ieee.org> Link: http://lkml.kernel.org/r/20161207135438.4310-1-mhocko@kernel.orgSigned-off-by:
Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
-
Boris Brezillon authored
commit 7e251bb2 upstream. The current ndelay() macro definition has an extra semi-colon at the end of the line thus leading to a compilation error when ndelay is used in a conditional block without curly braces like this one: if (cond) ndelay(t); else ... which, after the preprocessor pass gives: if (cond) m68k_ndelay(t);; else ... thus leading to the following gcc error: error: 'else' without a previous 'if' Remove this extra semi-colon. Signed-off-by:
Boris Brezillon <boris.brezillon@free-electrons.com> Fixes: c8ee038b ("m68k: Implement ndelay() based on the existing udelay() logic") Signed-off-by:
Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by:
-
추지호 authored
commit b67d0dd7 upstream. Fix for bad memory access while disconnecting. netdev is freed before private data free, and dev is accessed after freeing netdev. This makes a slub problem, and it raise kernel oops with slub debugger config. Signed-off-by:
Jiho Chu <jiho.chu@samsung.com> Signed-off-by:
Marc Kleine-Budde <mkl@pengutronix.de> Signed-off-by:
-
Marc Kleine-Budde authored
commit 332b05ca upstream. This patch adds a check to limit the number of can_filters that can be set via setsockopt on CAN_RAW sockets. Otherwise allocations > MAX_ORDER are not prevented resulting in a warning. Reference: https://lkml.org/lkml/2016/12/2/230Reported-by:
Andrey Konovalov <andreyknvl@google.com> Tested-by:
-
Peter Zijlstra (Intel) authored
commit 7f612a7f upstream. Lukasz reported that perf stat counters overflow handling is broken on KNL/SLM. Both these parts have full_width_write set, and that does indeed have a problem. In order to deal with counter wrap, we must sample the counter at at least half the counter period (see also the sampling theorem) such that we can unambiguously reconstruct the count. However commit: 069e0c3c ("perf/x86/intel: Support full width counting") sets the sampling interval to the full period, not half. Fixing that exposes another issue, in that we must not sign extend the delta value when we shift it right; the counter cannot have decremented after all. With both these issues fixed, counter overflow functions correctly again. Reported-by:
Lukasz Odzioba <lukasz.odzioba@intel.com> Tested-by:
Liang, Kan <kan.liang@intel.com> Tested-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vince Weaver <vincent.weaver@maine.edu> Fixes: 069e0c3c ("perf/x86/intel: Support full width counting") Signed-off-by:
Ingo Molnar <mingo@kernel.org> Signed-off-by:
-
Thomas Gleixner authored
commit 1be5d4fa upstream. While debugging the rtmutex unlock vs. dequeue race Will suggested to use READ_ONCE() in rt_mutex_owner() as it might race against the cmpxchg_release() in unlock_rt_mutex_safe(). Will: "It's a minor thing which will most likely not matter in practice" Careful search did not unearth an actual problem in todays code, but it's better to be safe than surprised. Suggested-by:
Will Deacon <will.deacon@arm.com> Signed-off-by:
-
Thomas Gleixner authored
commit dbb26055 upstream. David reported a futex/rtmutex state corruption. It's caused by the following problem: CPU0 CPU1 CPU2 l->owner=T1 rt_mutex_lock(l) lock(l->wait_lock) l->owner = T1 | HAS_WAITERS; enqueue(T2) boost() unlock(l->wait_lock) schedule() rt_mutex_lock(l) lock(l->wait_lock) l->owner = T1 | HAS_WAITERS; enqueue(T3) boost() unlock(l->wait_lock) schedule() signal(->T2) signal(->T3) lock(l->wait_lock) dequeue(T2) deboost() unlock(l->wait_lock) lock(l->wait_lock) dequeue(T3) ===> wait list is now empty deboost() unlock(l->wait_lock) lock(l->wait_lock) fixup_rt_mutex_waiters() if (wait_list_empty(l)) { owner = l->owner & ~HAS_WAITERS; l->owner = owner ==> l->owner = T1 } lock(l->wait_lock) rt_mutex_unlock(l) fixup_rt_mutex_waiters() if (wait_list_empty(l)) { owner = l->owner & ~HAS_WAITERS; cmpxchg(l->owner, T1, NULL) ===> Success (l->owner = NULL) l->owner = owner ==> l->owner = T1 } That means the problem is caused by fixup_rt_mutex_waiters() which does the RMW to clear the waiters bit unconditionally when there are no waiters in the rtmutexes rbtree. This can be fatal: A concurrent unlock can release the rtmutex in the fastpath because the waiters bit is not set. If the cmpxchg() gets in the middle of the RMW operation then the previous owner, which just unlocked the rtmutex is set as the owner again when the write takes place after the successfull cmpxchg(). The solution is rather trivial: verify that the owner member of the rtmutex has the waiters bit set before clearing it. This does not require a cmpxchg() or other atomic operations because the waiters bit can only be set and cleared with the rtmutex wait_lock held. It's also safe against the fast path unlock attempt. The unlock attempt via cmpxchg() will either see the bit set and take the slowpath or see the bit cleared and release it atomically in the fastpath. It's remarkable that the test program provided by David triggers on ARM64 and MIPS64 really quick, but it refuses to reproduce on x86-64, while the problem exists there as well. That refusal might explain that this got not discovered earlier despite the bug existing from day one of the rtmutex implementation more than 10 years ago. Thanks to David for meticulously instrumenting the code and providing the information which allowed to decode this subtle problem. Reported-by:
David Daney <ddaney@caviumnetworks.com> Tested-by:
David Daney <david.daney@cavium.com> Signed-off-by:
Steven Rostedt <rostedt@goodmis.org> Acked-by:
-
Jan Kara authored
commit 06bd3c36 upstream. Huang has reported that in his powerfail testing he is seeing stale block contents in some of recently allocated blocks although he mounts ext4 in data=ordered mode. After some investigation I have found out that indeed when delayed allocation is used, we don't add inode to transaction's list of inodes needing flushing before commit. Originally we were doing that but commit f3b59291 removed the logic with a flawed argument that it is not needed. The problem is that although for delayed allocated blocks we write their contents immediately after allocating them, there is no guarantee that the IO scheduler or device doesn't reorder things and thus transaction allocating blocks and attaching them to inode can reach stable storage before actual block contents. Actually whenever we attach freshly allocated blocks to inode using a written extent, we should add inode to transaction's ordered inode list to make sure we properly wait for block contents to be written before committing the transaction. So that is what we do in this patch. This also handles other cases where stale data exposure was possible - like filling hole via mmap in data=ordered,nodelalloc mode. The only exception to the above rule are extending direct IO writes where blkdev_direct_IO() waits for IO to complete before increasing i_size and thus stale data exposure is not possible. For now we don't complicate the code with optimizing this special case since the overhead is pretty low. In case this is observed to be a performance problem we can always handle it using a special flag to ext4_map_blocks(). Fixes: f3b59291Reported-by:
"HUANG Weller (CM/ESW12-CN)" <Weller.Huang@cn.bosch.com> Tested-by:
-
- 15 Dec, 2016 2 commits
-
-
Ming Lei authored
commit cebf8fd1 upstream. The global mutex of 'gdp_mutex' is used to serialize creating/querying glue dir and its cleanup. Turns out it isn't a perfect way because part(kobj_kset_leave()) of the actual cleanup action() is done inside the release handler of the glue dir kobject. That means gdp_mutex has to be held before releasing the last reference count of the glue dir kobject. This patch moves glue dir's cleanup after kobject_del() in device_del() for avoiding the race. Cc: Yijing Wang <wangyijing@huawei.com> Reported-by:
Chandra Sekhar Lingutla <clingutla@codeaurora.org> Signed-off-by:
Ming Lei <ming.lei@canonical.com> Signed-off-by:
-
Markus Elfring authored
commit 5f0163a5 upstream. The put_device() function tests whether its argument is NULL and then returns immediately. Thus the test around the call is not needed. This issue was detected by using the Coccinelle software. Signed-off-by:
Markus Elfring <elfring@users.sourceforge.net> Signed-off-by:
-
- 13 Dec, 2016 4 commits
-
-
Jiri Slaby authored
-
Thomas Tai authored
[ Upstream commit 87a349f9 ] A compile warning is introduced by a commit to fix the find_node(). This patch fix the compile warning by moving find_node() into __init section. Because find_node() is only used by memblock_nid_range() which is only used by a __init add_node_ranges(). find_node() and memblock_nid_range() should also be inside __init section. Signed-off-by:
Thomas Tai <thomas.tai@oracle.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
-
Thomas Tai authored
[ Upstream commit 74a5ed5c ] When booting up LDOM, find_node() warns that a physical address doesn't match a NUMA node. WARNING: CPU: 0 PID: 0 at arch/sparc/mm/init_64.c:835 find_node+0xf4/0x120 find_node: A physical address doesn't match a NUMA node rule. Some physical memory will be owned by node 0.Modules linked in: CPU: 0 PID: 0 Comm: swapper Not tainted 4.9.0-rc3 #4 Call Trace: [0000000000468ba0] __warn+0xc0/0xe0 [0000000000468c74] warn_slowpath_fmt+0x34/0x60 [00000000004592f4] find_node+0xf4/0x120 [0000000000dd0774] add_node_ranges+0x38/0xe4 [0000000000dd0b1c] numa_parse_mdesc+0x268/0x2e4 [0000000000dd0e9c] bootmem_init+0xb8/0x160 [0000000000dd174c] paging_init+0x808/0x8fc [0000000000dcb0d0] setup_arch+0x2c8/0x2f0 [0000000000dc68a0] start_kernel+0x48/0x424 [0000000000dcb374] start_early_boot+0x27c/0x28c [0000000000a32c08] tlb_fixup_done+0x4c/0x64 [0000000000027f08] 0x27f08 It is because linux use an internal structure node_masks[] to keep the best memory latency node only. However, LDOM mdesc can contain single latency-group with multiple memory latency nodes. If the address doesn't match the best latency node within node_masks[], it should check for an alternative via mdesc. The warning message should only be printed if the address doesn't match any node_masks[] nor within mdesc. To minimize the impact of searching mdesc every time, the last matched mask and index is stored in a variable. Signed-off-by:
Chris Hyser <chris.hyser@oracle.com> Reviewed-by:
Liam Merwick <liam.merwick@oracle.com> Signed-off-by:
-
Andreas Larsson authored
[ Upstream commit 07b5ab3f ] Signed-off-by:
Andreas Larsson <andreas@gaisler.com> Signed-off-by:
-
