| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux into xfs-6.8-mergeA
xfs: fix realtime geometry integer overflows
While reading through the realtime geometry support code in xfsprogs, I
noticed a discrepancy between the sb_rextslog computation used when
writing out the superblock during mkfs and the validation code used in
xfs_repair. This discrepancy would lead to system failure for a runt rt
volume having more than 1 rt block but zero rt extents in length. Most
people aren't going to configure a 1M extent size for their 360k rt
floppy disk volume, but I did!
In the process of studying that code, it occurred to me that there is a
second bug in the computation -- the use of highbit32 for a 64-bit
value means that the upper 32 bits are not considered in the search for
a high bit. This causes the creation of a realtime summary file that is
the wrong length. If rextents is a multiple of U32_MAX then this will
appear to work fine because highbit32 returns -1 for an input of 0; but
for all other cases the rt summary is undersized, leading to failures.
Fix the first problem by standardizing the computation with a helper in
libxfs; and the second problem by correcting the computation. This will
cause any existing rt volumes larger than 2^32 blocks to fail validation
but they probably were already crashing anyway.
This has been lightly tested with fstests. Enjoy!
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandanbabu@kernel.org>
* tag 'fix-rtmount-overflows-6.8_2023-12-06' of https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux:
xfs: don't allow overly small or large realtime volumes
xfs: fix 32-bit truncation in xfs_compute_rextslog
xfs: make rextslog computation consistent with mkfs
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux into xfs-6.8-mergeA
xfs: continue removing defer item boilerplate
Now that we've restructured log intent item recovery to reconstruct the
incore deferred work state, apply further cleanups to that code to
remove boilerplate that is duplicated across all the _item.c files.
Having done that, collapse a bunch of trivial helpers to reduce the
overall call chain. That enables us to refactor the relog code so that
the ->relog_item implementations only have to know how to format the
implementation-specific data encoded in an intent item and don't
themselves have to handle the log item juggling.
This has been lightly tested with fstests. Enjoy!
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandanbabu@kernel.org>
* tag 'reconstruct-defer-cleanups-6.8_2023-12-06' of https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux:
xfs: move ->iop_relog to struct xfs_defer_op_type
xfs: collapse the ->create_done functions
xfs: hoist xfs_trans_add_item calls to defer ops functions
xfs: clean out XFS_LI_DIRTY setting boilerplate from ->iop_relog
xfs: use xfs_defer_create_done for the relogging operation
xfs: hoist ->create_intent boilerplate to its callsite
xfs: collapse the ->finish_item helpers
xfs: hoist intent done flag setting to ->finish_item callsite
xfs: don't set XFS_TRANS_HAS_INTENT_DONE when there's no ATTRD log item
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux into xfs-6.8-mergeA
xfs: log intent item recovery should reconstruct defer work state
Long Li reported a KASAN report from a UAF when intent recovery fails:
==================================================================
BUG: KASAN: slab-use-after-free in xfs_cui_release+0xb7/0xc0
Read of size 4 at addr ffff888012575e60 by task kworker/u8:3/103
CPU: 3 PID: 103 Comm: kworker/u8:3 Not tainted 6.4.0-rc7-next-20230619-00003-g94543a53f9a4-dirty #166
Workqueue: xfs-cil/sda xlog_cil_push_work
Call Trace:
<TASK>
dump_stack_lvl+0x50/0x70
print_report+0xc2/0x600
kasan_report+0xb6/0xe0
xfs_cui_release+0xb7/0xc0
xfs_cud_item_release+0x3c/0x90
xfs_trans_committed_bulk+0x2d5/0x7f0
xlog_cil_committed+0xaba/0xf20
xlog_cil_push_work+0x1a60/0x2360
process_one_work+0x78e/0x1140
worker_thread+0x58b/0xf60
kthread+0x2cd/0x3c0
ret_from_fork+0x1f/0x30
</TASK>
Allocated by task 531:
kasan_save_stack+0x22/0x40
kasan_set_track+0x25/0x30
__kasan_slab_alloc+0x55/0x60
kmem_cache_alloc+0x195/0x5f0
xfs_cui_init+0x198/0x1d0
xlog_recover_cui_commit_pass2+0x133/0x5f0
xlog_recover_items_pass2+0x107/0x230
xlog_recover_commit_trans+0x3e7/0x9c0
xlog_recovery_process_trans+0x140/0x1d0
xlog_recover_process_ophdr+0x1a0/0x3d0
xlog_recover_process_data+0x108/0x2d0
xlog_recover_process+0x1f6/0x280
xlog_do_recovery_pass+0x609/0xdb0
xlog_do_log_recovery+0x84/0xe0
xlog_do_recover+0x7d/0x470
xlog_recover+0x25f/0x490
xfs_log_mount+0x2dd/0x6f0
xfs_mountfs+0x11ce/0x1e70
xfs_fs_fill_super+0x10ec/0x1b20
get_tree_bdev+0x3c8/0x730
vfs_get_tree+0x89/0x2c0
path_mount+0xecf/0x1800
do_mount+0xf3/0x110
__x64_sys_mount+0x154/0x1f0
do_syscall_64+0x39/0x80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
Freed by task 531:
kasan_save_stack+0x22/0x40
kasan_set_track+0x25/0x30
kasan_save_free_info+0x2b/0x40
__kasan_slab_free+0x114/0x1b0
kmem_cache_free+0xf8/0x510
xfs_cui_item_free+0x95/0xb0
xfs_cui_release+0x86/0xc0
xlog_recover_cancel_intents.isra.0+0xf8/0x210
xlog_recover_finish+0x7e7/0x980
xfs_log_mount_finish+0x2bb/0x4a0
xfs_mountfs+0x14bf/0x1e70
xfs_fs_fill_super+0x10ec/0x1b20
get_tree_bdev+0x3c8/0x730
vfs_get_tree+0x89/0x2c0
path_mount+0xecf/0x1800
do_mount+0xf3/0x110
__x64_sys_mount+0x154/0x1f0
do_syscall_64+0x39/0x80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
The buggy address belongs to the object at ffff888012575dc8
which belongs to the cache xfs_cui_item of size 432
The buggy address is located 152 bytes inside of
freed 432-byte region [ffff888012575dc8, ffff888012575f78)
The buggy address belongs to the physical page:
page:ffffea0000495d00 refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff888012576208 pfn:0x12574
head:ffffea0000495d00 order:2 entire_mapcount:0 nr_pages_mapped:0 pincount:0
flags: 0x1fffff80010200(slab|head|node=0|zone=1|lastcpupid=0x1fffff)
page_type: 0xffffffff()
raw: 001fffff80010200 ffff888012092f40 ffff888014570150 ffff888014570150
raw: ffff888012576208 00000000001e0010 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected
Memory state around the buggy address:
ffff888012575d00: fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc fc
ffff888012575d80: fc fc fc fc fc fc fc fc fc fa fb fb fb fb fb fb
>ffff888012575e00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
^
ffff888012575e80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
ffff888012575f00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fc
==================================================================
"If process intents fails, intent items left in AIL will be delete
from AIL and freed in error handling, even intent items that have been
recovered and created done items. After this, uaf will be triggered when
done item committed, because at this point the released intent item will
be accessed.
xlog_recover_finish xlog_cil_push_work
---------------------------- ---------------------------
xlog_recover_process_intents
xfs_cui_item_recover//cui_refcount == 1
xfs_trans_get_cud
xfs_trans_commit
<add cud item to cil>
xfs_cui_item_recover
<error occurred and return>
xlog_recover_cancel_intents
xfs_cui_release //cui_refcount == 0
xfs_cui_item_free //free cui
<release other intent items>
xlog_force_shutdown //shutdown
<...>
<push items in cil>
xlog_cil_committed
xfs_cud_item_release
xfs_cui_release // UAF
"Intent log items are created with a reference count of 2, one for the
creator, and one for the intent done object. Log recovery explicitly
drops the creator reference after it is inserted into the AIL, but it
then processes the log item as if it also owns the intent-done reference.
"The code in ->iop_recovery should assume that it passes the reference
to the done intent, we can remove the intent item from the AIL after
creating the done-intent, but if that code fails before creating the
done-intent then it needs to release the intent reference by log recovery
itself.
"That way when we go to cancel the intent, the only intents we find in
the AIL are the ones we know have not been processed yet and hence we
can safely drop both the creator and the intent done reference from
xlog_recover_cancel_intents().
"Hence if we remove the intent from the list of intents that need to
be recovered after we have done the initial recovery, we acheive two
things:
"1. the tail of the log can be moved forward with the commit of the
done intent or new intent to continue the operation, and
"2. We avoid the problem of trying to determine how many reference
counts we need to drop from intent recovery cancelling because we
never come across intents we've actually attempted recovery on."
Restated: The cause of the UAF is that xlog_recover_cancel_intents
thinks that it owns the refcount on any intent item in the AIL, and that
it's always safe to release these intent items. This is not true after
the recovery function creates an log intent done item and points it at
the log intent item because releasing the done item always releases the
intent item.
The runtime defer ops code avoids all this by tracking both the log
intent and the intent done items, and releasing only the intent done
item if both have been created. Long Li proposed fixing this by adding
state flags, but I have a more comprehensive fix.
First, observe that the latter half of the intent _recover functions are
nearly open-coded versions of the corresponding _finish_one function
that uses an onstack deferred work item to single-step through the item.
Second, notice that the recover function is not an exact match because
of the odd behavior that unfinished recovered work items are relogged
with separate log intent items instead of a single new log intent item,
which is what the defer ops machinery does.
Dave and I have long suspected that recovery should be reconstructing
the defer work state from what's in the recovered intent item. Now we
finally have an excuse to refactor the code to do that.
This series starts by fixing a resource leak in LARP recovery. We fix
the bug that Long Li reported by switching the intent recovery code to
construct chains of xfs_defer_pending objects and then using the defer
pending objects to track the intent/done item ownership. Finally, we
clean up the code to reconstruct the exact incore state, which means we
can remove all the opencoded _recover code, which makes maintaining log
items much easier.
v2: minor changes per review comments
v3: pick up more rvb tags, fix build errors
This has been lightly tested with fstests. Enjoy!
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandanbabu@kernel.org>
* tag 'reconstruct-defer-work-6.8_2023-12-06' of https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux:
xfs: move ->iop_recover to xfs_defer_op_type
xfs: use xfs_defer_finish_one to finish recovered work items
xfs: dump the recovered xattri log item if corruption happens
xfs: recreate work items when recovering intent items
xfs: transfer recovered intent item ownership in ->iop_recover
xfs: pass the xfs_defer_pending object to iop_recover
xfs: use xfs_defer_pending objects to recover intent items
xfs: don't leak recovered attri intent items
|
|
Don't allow realtime volumes that are less than one rt extent long.
This has been broken across 4 LTS kernels with nobody noticing, so let's
just disable it.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
The only log items that need relogging are the ones created for deferred
work operations, and the only part of the code base that relogs log
items is the deferred work machinery. Move the function pointers.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
It's quite reasonable that some customer somewhere will want to
configure a realtime volume with more than 2^32 extents. If they try to
do this, the highbit32() call will truncate the upper bits of the
xfs_rtbxlen_t and produce the wrong value for rextslog. This in turn
causes the rsumlevels to be wrong, which results in a realtime summary
file that is the wrong length. Fix that.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
There's a weird discrepancy in xfsprogs dating back to the creation of
the Linux port -- if there are zero rt extents, mkfs will set
sb_rextents and sb_rextslog both to zero:
sbp->sb_rextslog =
(uint8_t)(rtextents ?
libxfs_highbit32((unsigned int)rtextents) : 0);
However, that's not the check that xfs_repair uses for nonzero rtblocks:
if (sb->sb_rextslog !=
libxfs_highbit32((unsigned int)sb->sb_rextents))
The difference here is that xfs_highbit32 returns -1 if its argument is
zero. Unfortunately, this means that in the weird corner case of a
realtime volume shorter than 1 rt extent, xfs_repair will immediately
flag a freshly formatted filesystem as corrupt. Because mkfs has been
writing ondisk artifacts like this for decades, we have to accept that
as "correct". TBH, zero rextslog for zero rtextents makes more sense to
me anyway.
Regrettably, the superblock verifier checks created in commit copied
xfs_repair even though mkfs has been writing out such filesystems for
ages. Fix the superblock verifier to accept what mkfs spits out; the
userspace version of this patch will have to fix xfs_repair as well.
Note that the new helper leaves the zeroday bug where the upper 32 bits
of sb_rextents is ripped off and fed to highbit32. This leads to a
seriously undersized rt summary file, which immediately breaks mkfs:
$ hugedisk.sh foo /dev/sdc $(( 0x100000080 * 4096))B
$ /sbin/mkfs.xfs -f /dev/sda -m rmapbt=0,reflink=0 -r rtdev=/dev/mapper/foo
meta-data=/dev/sda isize=512 agcount=4, agsize=1298176 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=0 bigtime=1 inobtcount=1 nrext64=1
data = bsize=4096 blocks=5192704, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=16384, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =/dev/mapper/foo extsz=4096 blocks=4294967424, rtextents=4294967424
Discarding blocks...Done.
mkfs.xfs: Error initializing the realtime space [117 - Structure needs cleaning]
The next patch will drop support for rt volumes with fewer than 1 or
more than 2^32-1 rt extents, since they've clearly been broken forever.
Fixes: f8e566c0f5e1f ("xfs: validate the realtime geometry in xfs_validate_sb_common")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
Move the meat of the ->create_done function helpers into ->create_done
to reduce the amount of boilerplate.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
Remove even more repeated boilerplate.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
Hoist this dirty flag setting to the ->iop_relog callsite to reduce
boilerplate.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
Now that we have a helper to handle creating a log intent done item and
updating all the necessary state flags, use it to reduce boilerplate in
the ->iop_relog implementations.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
Hoist the dirty flag setting code out of each ->create_intent
implementation up to the callsite to reduce boilerplate further.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
Each log item's ->finish_item function sets up a small amount of state
and calls another function to do the work. Collapse that other function
into ->finish_item to reduce the call stack height.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
Finish off the series by moving the intent item recovery function
pointer to the xfs_defer_op_type struct, since this is really a deferred
work function now.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
Each log intent item's ->finish_item call chain inevitably includes some
code to set the dirty flag of the transaction. If there's an associated
log intent done item, it also sets the item's dirty flag and the
transaction's INTENT_DONE flag. This is repeated throughout the
codebase.
Reduce the LOC by moving all that to xfs_defer_finish_one.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
Get rid of the open-coded calls to xfs_defer_finish_one. This also
means that the recovery transaction takes care of cleaning up the dfp,
and we have solved (I hope) all the ownership issues in recovery.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
XFS_TRANS_HAS_INTENT_DONE is a flag to the CIL that we've added a log
intent done item to the transaction. This enables an optimization
wherein we avoid writing out log intent and log intent done items if
they would have ended up in the same checkpoint. This reduces writes to
the ondisk log and speeds up recovery as a result.
However, callers can use the defer ops machinery to modify xattrs
without using the log items. In this situation, there won't be an
intent done item, so we do not need to set the flag.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
If xfs_attri_item_recover receives a corruption error when it tries to
finish a recovered log intent item, it should dump the log item for
debugging, just like all the other log intent items.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
Recreate work items for each xfs_defer_pending object when we are
recovering intent items.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
Now that we pass the xfs_defer_pending object into the intent item
recovery functions, we know exactly when ownership of the sole refcount
passes from the recovery context to the intent done item. At that
point, we need to null out dfp_intent so that the recovery mechanism
won't release it. This should fix the UAF problem reported by Long Li.
Note that we still want to recreate the full deferred work state. That
will be addressed in the next patches.
Fixes: 2e76f188fd90 ("xfs: cancel intents immediately if process_intents fails")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
Now that log intent item recovery recreates the xfs_defer_pending state,
we should pass that into the ->iop_recover routines so that the intent
item can finish the recreation work.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
One thing I never quite got around to doing is porting the log intent
item recovery code to reconstruct the deferred pending work state. As a
result, each intent item open codes xfs_defer_finish_one in its recovery
method, because that's what the EFI code did before xfs_defer.c even
existed.
This is a gross thing to have left unfixed -- if an EFI cannot proceed
due to busy extents, we end up creating separate new EFIs for each
unfinished work item, which is a change in behavior from what runtime
would have done.
Worse yet, Long Li pointed out that there's a UAF in the recovery code.
The ->commit_pass2 function adds the intent item to the AIL and drops
the refcount. The one remaining refcount is now owned by the recovery
mechanism (aka the log intent items in the AIL) with the intent of
giving the refcount to the intent done item in the ->iop_recover
function.
However, if something fails later in recovery, xlog_recover_finish will
walk the recovered intent items in the AIL and release them. If the CIL
hasn't been pushed before that point (which is possible since we don't
force the log until later) then the intent done release will try to free
its associated intent, which has already been freed.
This patch starts to address this mess by having the ->commit_pass2
functions recreate the xfs_defer_pending state. The next few patches
will fix the recovery functions.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
If recovery finds an xattr log intent item calling for the removal of an
attribute and the file doesn't even have an attr fork, we know that the
removal is trivially complete. However, we can't just exit the recovery
function without doing something about the recovered log intent item --
it's still on the AIL, and not logging an attrd item means it stays
there forever.
This has likely not been seen in practice because few people use LARP
and the runtime code won't log the attri for a no-attrfork removexattr
operation. But let's fix this anyway.
Also we shouldn't really be testing the attr fork presence until we've
taken the ILOCK, though this doesn't matter much in recovery, which is
single threaded.
Fixes: fdaf1bb3cafc ("xfs: ATTR_REPLACE algorithm with LARP enabled needs rework")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
|
|
Pull smb client fixes from Steve French:
- Two fallocate fixes
- Fix warnings from new gcc
- Two symlink fixes
* tag 'v6.7-rc3-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
smb: client, common: fix fortify warnings
cifs: Fix FALLOC_FL_INSERT_RANGE by setting i_size after EOF moved
cifs: Fix FALLOC_FL_ZERO_RANGE by setting i_size if EOF moved
smb: client: report correct st_size for SMB and NFS symlinks
smb: client: fix missing mode bits for SMB symlinks
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394
Pull firewire fix from Takashi Sakamoto:
"A single patch to fix long-standing issue of memory leak at failure of
device registration for fw_unit. We rarely encounter the issue, but it
should be applied to stable releases, since it fixes inappropriate API
usage"
* tag 'firewire-fixes-6.7-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394:
firewire: core: fix possible memory leak in create_units()
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc fixes from Michael Ellerman:
- Fix corruption of f0/vs0 during FP/Vector save, seen as userspace
crashes when using io-uring workers (in particular with MariaDB)
- Fix KVM_RUN potentially clobbering all host userspace FP/Vector
registers
Thanks to Timothy Pearson, Jens Axboe, and Nicholas Piggin.
* tag 'powerpc-6.7-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
KVM: PPC: Book3S HV: Fix KVM_RUN clobbering FP/VEC user registers
powerpc: Don't clobber f0/vs0 during fp|altivec register save
|
|
Pull vfio fixes from Alex Williamson:
- Fix the lifecycle of a mutex in the pds variant driver such that a
reset prior to opening the device won't find it uninitialized.
Implement the release path to symmetrically destroy the mutex. Also
switch a different lock from spinlock to mutex as the code path has
the potential to sleep and doesn't need the spinlock context
otherwise (Brett Creeley)
- Fix an issue detected via randconfig where KVM tries to symbol_get an
undeclared function. The symbol is temporarily declared
unconditionally here, which resolves the problem and avoids churn
relative to a series pending for the next merge window which resolves
some of this symbol ugliness, but also fixes Kconfig dependencies
(Sean Christopherson)
* tag 'vfio-v6.7-rc4' of https://github.com/awilliam/linux-vfio:
vfio: Drop vfio_file_iommu_group() stub to fudge around a KVM wart
vfio/pds: Fix possible sleep while in atomic context
vfio/pds: Fix mutex lock->magic != lock warning
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip
Pull xen fixes from Juergen Gross:
- A fix for the Xen event driver setting the correct return value when
experiencing an allocation failure
- A fix for allocating space for a struct in the percpu area to not
cross page boundaries (this one is for x86, a similar one for Arm was
already in the pull request for rc3)
* tag 'for-linus-6.7a-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
xen/events: fix error code in xen_bind_pirq_msi_to_irq()
x86/xen: fix percpu vcpu_info allocation
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull probes fixes from Masami Hiramatsu:
- objpool: Fix objpool overrun case on memory/cache access delay
especially on the big.LITTLE SoC. The objpool uses a copy of object
slot index internal loop, but the slot index can be changed on
another processor in parallel. In that case, the difference of 'head'
local copy and the 'slot->last' index will be bigger than local slot
size. In that case, we need to re-read the slot::head to update it.
- kretprobe: Fix to use appropriate rcu API for kretprobe holder. Since
kretprobe_holder::rp is RCU managed, it should use
rcu_assign_pointer() and rcu_dereference_check() correctly. Also
adding __rcu tag for finding wrong usage by sparse.
- rethook: Fix to use appropriate rcu API for rethook::handler. The
same as kretprobe, rethook::handler is RCU managed and it should use
rcu_assign_pointer() and rcu_dereference_check(). This also adds
__rcu tag for finding wrong usage by sparse.
* tag 'probes-fixes-v6.7-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
rethook: Use __rcu pointer for rethook::handler
kprobes: consistent rcu api usage for kretprobe holder
lib: objpool: fix head overrun on RK3588 SBC
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull power management fixes from Rafael Wysocki:
"These fix issues in two cpufreq drivers, in the AMD P-state driver and
in the power-capping DTPM framework.
Specifics:
- Fix the AMD P-state driver's EPP sysfs interface in the cases when
the performance governor is in use (Ayush Jain)
- Make the ->fast_switch() callback in the AMD P-state driver return
the target frequency as expected (Gautham R. Shenoy)
- Allow user space to control the range of frequencies to use via
scaling_min_freq and scaling_max_freq when AMD P-state driver is in
use (Wyes Karny)
- Prevent power domains needed for wakeup signaling from being turned
off during system suspend on Qualcomm systems and prevent
performance states votes from runtime-suspended devices from being
lost across a system suspend-resume cycle in qcom-cpufreq-nvmem
(Stephan Gerhold)
- Fix disabling the 792 Mhz OPP in the imx6q cpufreq driver for the
i.MX6ULL types that can run at that frequency (Christoph
Niedermaier)
- Eliminate unnecessary and harmful conversions to uW from the DTPM
(dynamic thermal and power management) framework (Lukasz Luba)"
* tag 'pm-6.7-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
cpufreq/amd-pstate: Only print supported EPP values for performance governor
cpufreq/amd-pstate: Fix scaling_min_freq and scaling_max_freq update
powercap: DTPM: Fix unneeded conversions to micro-Watts
cpufreq/amd-pstate: Fix the return value of amd_pstate_fast_switch()
pmdomain: qcom: rpmpd: Set GENPD_FLAG_ACTIVE_WAKEUP
cpufreq: qcom-nvmem: Preserve PM domain votes in system suspend
cpufreq: qcom-nvmem: Enable virtual power domain devices
cpufreq: imx6q: Don't disable 792 Mhz OPP unnecessarily
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull ACPI fixes from Rafael Wysocki:
"This fixes a recently introduced build issue on ARM32 and a NULL
pointer dereference in the ACPI backlight driver due to a design issue
exposed by a recent change in the ACPI bus type code.
Specifics:
- Fix a recently introduced build issue on ARM32 platforms caused by
an inadvertent header file breakage (Dave Jiang)
- Eliminate questionable usage of acpi_driver_data() in the ACPI
backlight cooling device code that leads to NULL pointer
dereferences after recent ACPI core changes (Hans de Goede)"
* tag 'acpi-6.7-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
ACPI: video: Use acpi_video_device for cooling-dev driver data
ACPI: Fix ARM32 platforms compile issue introduced by fw_table changes
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fix from Catalin Marinas:
"Fix a regression where the arm64 KPTI ends up enabled even on systems
that don't need it"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: Avoid enabling KPTI unnecessarily
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
Pull iommu fixes from Joerg Roedel:
- Fix race conditions in device probe path
- Handle ERR_PTR() returns in __iommu_domain_alloc() path
- Update MAINTAINERS entry for Qualcom IOMMUs
- Printk argument fix in device tree specific code
- Several Intel VT-d fixes from Lu Baolu:
- Do not support enforcing cache coherency for non-empty domains
- Avoid devTLB invalidation if iommu is off
- Disable PCI ATS in legacy passthrough mode
- Support non-PCI devices when clearing context
- Fix incorrect cache invalidation for mm notification
- Add MTL to quirk list to skip TE disabling
- Set variable intel_dirty_ops to static
* tag 'iommu-fixes-v6.7-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
iommu: Fix printk arg in of_iommu_get_resv_regions()
iommu/vt-d: Set variable intel_dirty_ops to static
iommu/vt-d: Fix incorrect cache invalidation for mm notification
iommu/vt-d: Add MTL to quirk list to skip TE disabling
iommu/vt-d: Make context clearing consistent with context mapping
iommu/vt-d: Disable PCI ATS in legacy passthrough mode
iommu/vt-d: Omit devTLB invalidation requests when TES=0
iommu/vt-d: Support enforce_cache_coherency only for empty domains
iommu: Avoid more races around device probe
MAINTAINERS: list all Qualcomm IOMMU drivers in the QUALCOMM IOMMU entry
iommu: Flow ERR_PTR out from __iommu_domain_alloc()
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"No surprise here, including only a collection of HD-audio
device-specific small fixes"
* tag 'sound-6.7-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: hda: Disable power-save on KONTRON SinglePC
ALSA: hda/realtek: Add supported ALC257 for ChromeOS
ALSA: hda/realtek: Headset Mic VREF to 100%
ALSA: hda: intel-nhlt: Ignore vbps when looking for DMIC 32 bps format
ALSA: hda: cs35l56: Enable low-power hibernation mode on SPI
ALSA: cs35l41: Fix for old systems which do not support command
ALSA: hda: cs35l41: Remove unnecessary boolean state variable firmware_running
ALSA: hda - Fix speaker and headset mic pin config for CHUWI CoreBook XPro
|
|
Pull drm fixes from Dave Airlie:
"Weekly fixes, mostly amdgpu fixes with a scattering of nouveau, i915,
and a couple of reverts. Hopefully it will quieten down in coming
weeks.
drm:
- Revert unexport of prime helpers for fd/handle conversion
dma_resv:
- Do not double add fences in dma_resv_add_fence.
gpuvm:
- Fix GPUVM license identifier.
i915:
- Mark internal GSC engine with reserved uabi class
- Take VGA converters into account in eDP probe
- Fix intel_pre_plane_updates() call to ensure workarounds get applied
panel:
- Revert panel fixes as they require exporting device_is_dependent.
nouveau:
- fix oversized allocations in new vm path
- fix zero-length array
- remove a stray lock
nt36523:
- Fix error check for nt36523.
amdgpu:
- DMUB fix
- DCN 3.5 fixes
- XGMI fix
- DCN 3.2 fixes
- Vangogh suspend fix
- NBIO 7.9 fix
- GFX11 golden register fix
- Backlight fix
- NBIO 7.11 fix
- IB test overflow fix
- DCN 3.1.4 fixes
- fix a runtime pm ref count
- Retimer fix
- ABM fix
- DCN 3.1.5 fix
- Fix AGP addressing
- Fix possible memory leak in SMU error path
- Make sure PME is enabled in D3
- Fix possible NULL pointer dereference in debugfs
- EEPROM fix
- GC 9.4.3 fix
amdkfd:
- IP version check fix
- Fix memory leak in pqm_uninit()"
* tag 'drm-fixes-2023-12-01' of git://anongit.freedesktop.org/drm/drm: (53 commits)
Revert "drm/prime: Unexport helpers for fd/handle conversion"
drm/amdgpu: Use another offset for GC 9.4.3 remap
drm/amd/display: Fix some HostVM parameters in DML
drm/amdkfd: Free gang_ctx_bo and wptr_bo in pqm_uninit
drm/amdgpu: Update EEPROM I2C address for smu v13_0_0
drm/amd/display: Allow DTBCLK disable for DCN35
drm/amdgpu: Fix cat debugfs amdgpu_regs_didt causes kernel null pointer
drm/amd: Enable PCIe PME from D3
drm/amd/pm: fix a memleak in aldebaran_tables_init
drm/amdgpu: fix AGP addressing when GART is not at 0
drm/amd/display: update dcn315 lpddr pstate latency
drm/amd/display: fix ABM disablement
drm/amd/display: Fix black screen on video playback with embedded panel
drm/amd/display: Fix conversions between bytes and KB
drm/amdkfd: Use common function for IP version check
drm/amd/display: Remove config update
drm/amd/display: Update DCN35 clock table policy
drm/amd/display: force toggle rate wa for first link training for a retimer
drm/amdgpu: correct the amdgpu runtime dereference usage count
drm/amd/display: Update min Z8 residency time to 2100 for DCN314
...
|
|
Pull io_uring fixes from Jens Axboe:
- Fix an issue with discontig page checking for IORING_SETUP_NO_MMAP
- Fix an issue with not allowing IORING_SETUP_NO_MMAP also disallowing
mmap'ed buffer rings
- Fix an issue with deferred release of memory mapped pages
- Fix a lockdep issue with IORING_SETUP_NO_MMAP
- Use fget/fput consistently, even from our sync system calls. No real
issue here, but if we were ever to allow closing io_uring descriptors
it would be required. Let's play it safe and just use the full ref
counted versions upfront. Most uses of io_uring are threaded anyway,
and hence already doing the full version underneath.
* tag 'io_uring-6.7-2023-11-30' of git://git.kernel.dk/linux:
io_uring: use fget/fput consistently
io_uring: free io_buffer_list entries via RCU
io_uring/kbuf: prune deferred locked cache when tearing down
io_uring/kbuf: recycle freed mapped buffer ring entries
io_uring/kbuf: defer release of mapped buffer rings
io_uring: enable io_mem_alloc/free to be used in other parts
io_uring: don't guard IORING_OFF_PBUF_RING with SETUP_NO_MMAP
io_uring: don't allow discontig pages for IORING_SETUP_NO_MMAP
|
|
Pull block fixes from Jens Axboe:
- NVMe pull request via Keith:
- Invalid namespace identification error handling (Marizio Ewan,
Keith)
- Fabrics keep-alive tuning (Mark)
- Fix for a bad error check regression in bcache (Markus)
- Fix for a performance regression with O_DIRECT (Ming)
- Fix for a flush related deadlock (Ming)
- Make the read-only warn on per-partition (Yu)
* tag 'block-6.7-2023-12-01' of git://git.kernel.dk/linux:
nvme-core: check for too small lba shift
blk-mq: don't count completed flush data request as inflight in case of quiesce
block: Document the role of the two attribute groups
block: warn once for each partition in bio_check_ro()
block: move .bd_inode into 1st cacheline of block_device
nvme: check for valid nvme_identify_ns() before using it
nvme-core: fix a memory leak in nvme_ns_info_from_identify()
nvme: fine-tune sending of first keep-alive
bcache: revert replacing IS_ERR_OR_NULL with IS_ERR
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm
Pull device mapper fixes from Mike Snitzer:
- Fix DM verity target's FEC support to always initialize IO before it
frees it. Also fix alignment of struct dm_verity_fec_io within the
per-bio-data
- Fix DM verity target to not FEC failed readahead IO
- Update DM flakey target to use MAX_ORDER rather than MAX_ORDER - 1
* tag 'dm-6.7/dm-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
dm-flakey: start allocating with MAX_ORDER
dm-verity: align struct dm_verity_fec_io properly
dm verity: don't perform FEC for failed readahead IO
dm verity: initialize fec io before freeing it
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI fixes from James Bottomley:
"Three small fixes, one in drivers.
The core changes are to the internal representation of flags in
scsi_devices which removes space wasting bools in favour of single bit
flags and to add a flag to force a runtime resume which is used by ATA
devices"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: sd: Fix system start for ATA devices
scsi: Change SCSI device boolean fields to single bit flags
scsi: ufs: core: Clear cmd if abort succeeds in MCQ mode
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs
Pull ext2 fix from Jan Kara:
"Fix an ext2 bug introduced by changes in ext2 & iomap stepping on each
other toes (apparently ext2 driver does not get much testing in
linux-next)"
* tag 'fs_for_v6.7-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
ext2: Fix ki_pos update for DIO buffered-io fallback case
|
|
Pull more bcachefs bugfixes from Kent Overstreet:
- bcache & bcachefs were broken with CFI enabled; patch for closures to
fix type punning
- mark erasure coding as extra-experimental; there are incompatible
disk space accounting changes coming for erasure coding, and I'm
still seeing checksum errors in some tests
- several fixes for durability-related issues (durability is a device
specific setting where we can tell bcachefs that data on a given
device should be counted as replicated x times)
- a fix for a rare livelock when a btree node merge then updates a
parent node that is almost full
- fix a race in the device removal path, where dropping a pointer in a
btree node to a device would be clobbered by an in flight btree write
updating the btree node key on completion
- fix one SRCU lock hold time warning in the btree gc code - ther's
still a bunch more of these to fix
- fix a rare race where we'd start copygc before initializing the "are
we rw" percpu refcount; copygc would think we were already ro and die
immediately
* tag 'bcachefs-2023-11-29' of https://evilpiepirate.org/git/bcachefs: (23 commits)
bcachefs: Extra kthread_should_stop() calls for copygc
bcachefs: Convert gc_alloc_start() to for_each_btree_key2()
bcachefs: Fix race between btree writes and metadata drop
bcachefs: move journal seq assertion
bcachefs: -EROFS doesn't count as move_extent_start_fail
bcachefs: trace_move_extent_start_fail() now includes errcode
bcachefs: Fix split_race livelock
bcachefs: Fix bucket data type for stripe buckets
bcachefs: Add missing validation for jset_entry_data_usage
bcachefs: Fix zstd compress workspace size
bcachefs: bpos is misaligned on big endian
bcachefs: Fix ec + durability calculation
bcachefs: Data update path won't accidentaly grow replicas
bcachefs: deallocate_extra_replicas()
bcachefs: Proper refcounting for journal_keys
bcachefs: preserve device path as device name
bcachefs: Fix an endianness conversion
bcachefs: Start gc, copygc, rebalance threads after initing writes ref
bcachefs: Don't stop copygc thread on device resize
bcachefs: Make sure bch2_move_ratelimit() also waits for move_ops
...
|
|
Merge a fix for a recently introduced build issue on ARM32 platforms
caused by an inadvertent header file breakage (Dave Jiang).
* acpi-tables:
ACPI: Fix ARM32 platforms compile issue introduced by fw_table changes
|
|
Merge a power capping fix for 6.7-rc4 which eliminates unnecessary
and harmful conversions to uW from the DTPM (dynamic thermal and power
management) framework (Lukasz Luba).
* powercap:
powercap: DTPM: Fix unneeded conversions to micro-Watts
|
|
Pull NVMe fixes from Keith:
"nvme fixes for Linux 6.7
- Invalid namespace identification error handling (Marizio Ewan, Keith)
- Fabrics keep-alive tuning (Mark)"
* tag 'nvme-6.7-2023-12-01' of git://git.infradead.org/nvme:
nvme-core: check for too small lba shift
nvme: check for valid nvme_identify_ns() before using it
nvme-core: fix a memory leak in nvme_ns_info_from_identify()
nvme: fine-tune sending of first keep-alive
|
|
The block layer doesn't support logical block sizes smaller than 512
bytes. The nvme spec doesn't support that small either, but the driver
isn't checking to make sure the device responded with usable data.
Failing to catch this will result in a kernel bug, either from a
division by zero when stacking, or a zero length bio.
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
Request queue quiesce may interrupt flush sequence, and the original request
may have been marked as COMPLETE, but can't get finished because of
queue quiesce.
This way is fine from driver viewpoint, because flush sequence is block
layer concept, and it isn't related with driver.
However, driver(such as dm-rq) can call blk_mq_queue_inflight() to count &
drain inflight requests, then the wait & drain never gets done because
the completed & not-finished flush request is counted as inflight.
Fix this issue by not counting completed flush data request as inflight in
case of quiesce.
Cc: Mike Snitzer <snitzer@kernel.org>
Cc: David Jeffery <djeffery@redhat.com>
Cc: John Pittman <jpittman@redhat.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20231201085605.577730-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
The variable phys is defined as (struct resource *) which aligns with
the printk format specifier %pr. Taking the address of it results in a
value of type (struct resource **) which is incompatible with the format
specifier %pr. Therefore, remove the address of operator (&).
Fixes: a5bf3cfce8cb ("iommu: Implement of_iommu_get_resv_regions()")
Signed-off-by: Daniel Mentz <danielmentz@google.com>
Acked-by: Thierry Reding <treding@nvidia.com>
Link: https://lore.kernel.org/r/20231108062226.928985-1-danielmentz@google.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
|
|
Since the rethook::handler is an RCU-maganged pointer so that it will
notice readers the rethook is stopped (unregistered) or not, it should
be an __rcu pointer and use appropriate functions to be accessed. This
will use appropriate memory barrier when accessing it. OTOH,
rethook::data is never changed, so we don't need to check it in
get_kretprobe().
NOTE: To avoid sparse warning, rethook::handler is defined by a raw
function pointer type with __rcu instead of rethook_handler_t.
Link: https://lore.kernel.org/all/170126066201.398836.837498688669005979.stgit@devnote2/
Fixes: 54ecbe6f1ed5 ("rethook: Add a generic return hook")
Cc: stable@vger.kernel.org
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202311241808.rv9ceuAh-lkp@intel.com/
Tested-by: JP Kobryn <inwardvessel@gmail.com>
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
|
|
It seems that the pointer-to-kretprobe "rp" within the kretprobe_holder is
RCU-managed, based on the (non-rethook) implementation of get_kretprobe().
The thought behind this patch is to make use of the RCU API where possible
when accessing this pointer so that the needed barriers are always in place
and to self-document the code.
The __rcu annotation to "rp" allows for sparse RCU checking. Plain writes
done to the "rp" pointer are changed to make use of the RCU macro for
assignment. For the single read, the implementation of get_kretprobe()
is simplified by making use of an RCU macro which accomplishes the same,
but note that the log warning text will be more generic.
I did find that there is a difference in assembly generated between the
usage of the RCU macros vs without. For example, on arm64, when using
rcu_assign_pointer(), the corresponding store instruction is a
store-release (STLR) which has an implicit barrier. When normal assignment
is done, a regular store (STR) is found. In the macro case, this seems to
be a result of rcu_assign_pointer() using smp_store_release() when the
value to write is not NULL.
Link: https://lore.kernel.org/all/20231122132058.3359-1-inwardvessel@gmail.com/
Fixes: d741bf41d7c7 ("kprobes: Remove kretprobe hash")
Cc: stable@vger.kernel.org
Signed-off-by: JP Kobryn <inwardvessel@gmail.com>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
|
