Pavel Skripkin [Tue, 27 Jul 2021 16:59:57 +0000 (19:59 +0300)]
can: usb_8dev: fix memory leak
In usb_8dev_start() MAX_RX_URBS coherent buffers are allocated and
there is nothing, that frees them:
1) In callback function the urb is resubmitted and that's all
2) In disconnect function urbs are simply killed, but URB_FREE_BUFFER
is not set (see usb_8dev_start) and this flag cannot be used with
coherent buffers.
So, all allocated buffers should be freed with usb_free_coherent()
explicitly.
Side note: This code looks like a copy-paste of other can drivers. The
same patch was applied to mcba_usb driver and it works nice with real
hardware. There is no change in functionality, only clean-up code for
coherent buffers.
Yasushi reported, that his Microchip CAN Analyzer stopped working
since commit 91c02557174b ("can: mcba_usb: fix memory leak in
mcba_usb"). The problem was in missing urb->transfer_dma
initialization.
In my previous patch to this driver I refactored mcba_usb_start() code
to avoid leaking usb coherent buffers. To archive it, I passed local
stack variable to usb_alloc_coherent() and then saved it to private
array to correctly free all coherent buffers on ->close() call. But I
forgot to initialize urb->transfer_dma with variable passed to
usb_alloc_coherent().
All of this was causing device to not work, since dma addr 0 is not
valid and following log can be found on bug report page, which points
exactly to problem described above.
| DMAR: [DMA Write] Request device [00:14.0] PASID ffffffff fault addr 0 [fault reason 05] PTE Write access is not set
Dan Carpenter [Thu, 29 Jul 2021 14:12:46 +0000 (17:12 +0300)]
can: hi311x: fix a signedness bug in hi3110_cmd()
The hi3110_cmd() is supposed to return zero on success and negative
error codes on failure, but it was accidentally declared as a u8 when
it needs to be an int type.
Merge tag 'drm-fixes-2021-07-30' of git://anongit.freedesktop.org/drm/drm
Pull drm fixes from Dave Airlie:
"Regular drm fixes pull, seems about the right size, lots of small
fixes across the board, mostly amdgpu, but msm and i915 are in there
along with panel and ttm.
amdgpu:
- Fix resource leak in an error path
- Avoid stack contents exposure in error path
- pmops check fix for S0ix vs S3
- DCN 2.1 display fixes
- DCN 2.0 display fix
- Backlight control fix for laptops with HDR panels
- Maintainers updates
i915:
- Fix vbt port mask
- Fix around reading the right DSC disable fuse in display_ver 10
- Split display version 9 and 10 in intel_setup_outputs
* tag 'drm-fixes-2021-07-30' of git://anongit.freedesktop.org/drm/drm:
maintainers: add bugs and chat URLs for amdgpu
drm/amdgpu/display: only enable aux backlight control for OLED panels
drm/amd/display: ensure dentist display clock update finished in DCN20
drm/amd/display: Add missing DCN21 IP parameter
drm/amd/display: Guard DST_Y_PREFETCH register overflow in DCN21
drm/amdgpu: Check pmops for desired suspend state
drm/msm/dp: Initialize dp->aux->drm_dev before registration
drm/msm/dp: signal audio plugged change at dp_pm_resume
drm/msm/dp: Initialize the INTF_CONFIG register
drm/msm/dp: use dp_ctrl_off_link_stream during PHY compliance test run
drm/msm: Fix display fault handling
drm/msm/dpu: Fix sm8250_mdp register length
drm/amdgpu: Avoid printing of stack contents on firmware load error
drm/amdgpu: Fix resource leak on probe error path
drm/i915/display: split DISPLAY_VER 9 and 10 in intel_setup_outputs()
drm/i915: fix not reading DSC disable fuse in GLK
drm/i915/bios: Fix ports mask
drm/panel: panel-simple: Fix proper bpc for ytc700tlag_05_201c
drm/ttm: Initialize debugfs from ttm_global_init()
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mattst88/alpha
Pull alpha updates from Matt Turner:
"They're mostly small janitorial fixes but there's also more important
ones:
- drop the alpha-specific x86 binary loader (David Hildenbrand)
- regression fix for at least Marvel platforms (Mike Rapoport)
- fix for a scary-looking typo (Zheng Yongjun)"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mattst88/alpha:
alpha: register early reserved memory in memblock
alpha: fix spelling mistakes
alpha: Remove space between * and parameter name
alpha: fp_emul: avoid init/cleanup_module names
alpha: Add syscall_get_return_value()
binfmt: remove support for em86 (alpha only)
alpha: fix typos in a comment
alpha: defconfig: add necessary configs for boot testing
alpha: Send stop IPI to send to online CPUs
alpha: convert comma to semicolon
alpha: remove undef inline in compiler.h
alpha: Kconfig: Replace HTTP links with HTTPS ones
alpha: __udiv_qrnnd should be exported
scsi: acornscsi: Fix fall-through warning for clang
Fix the following fallthrough warning (on ARM):
drivers/scsi/arm/acornscsi.c:2651:2: warning: unannotated fall-through between switch labels [-Wimplicit-fallthrough]
case res_success:
^
drivers/scsi/arm/acornscsi.c:2651:2: note: insert '__attribute__((fallthrough));' to silence this warning
case res_success:
^
__attribute__((fallthrough));
drivers/scsi/arm/acornscsi.c:2651:2: note: insert 'break;' to avoid fall-through
case res_success:
^
break; Reported-by: kernel test robot <[email protected]> Link: https://lore.kernel.org/lkml/[email protected]/ Signed-off-by: Gustavo A. R. Silva <[email protected]>
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull kvm fixes from Paolo Bonzini:
"ARM:
- Fix MTE shared page detection
- Enable selftest's use of PMU registers when asked to
s390:
- restore 5.13 debugfs names
x86:
- fix sizes for vcpu-id indexed arrays
- fixes for AMD virtualized LAPIC (AVIC)
- other small bugfixes
Generic:
- access tracking performance test
- dirty_log_perf_test command line parsing fix
- Fix selftest use of obsolete pthread_yield() in favour of
sched_yield()
- use cpu_relax when halt polling
- fixed missing KVM_CLEAR_DIRTY_LOG compat ioctl"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: add missing compat KVM_CLEAR_DIRTY_LOG
KVM: use cpu_relax when halt polling
KVM: SVM: use vmcb01 in svm_refresh_apicv_exec_ctrl
KVM: SVM: tweak warning about enabled AVIC on nested entry
KVM: SVM: svm_set_vintr don't warn if AVIC is active but is about to be deactivated
KVM: s390: restore old debugfs names
KVM: SVM: delay svm_vcpu_init_msrpm after svm->vmcb is initialized
KVM: selftests: Introduce access_tracking_perf_test
KVM: selftests: Fix missing break in dirty_log_perf_test arg parsing
x86/kvm: fix vcpu-id indexed array sizes
KVM: x86: Check the right feature bit for MSR_KVM_ASYNC_PF_ACK access
docs: virt: kvm: api.rst: replace some characters
KVM: Documentation: Fix KVM_CAP_ENFORCE_PV_FEATURE_CPUID name
KVM: nSVM: Swap the parameter order for svm_copy_vmrun_state()/svm_copy_vmloadsave_state()
KVM: nSVM: Rename nested_svm_vmloadsave() to svm_copy_vmloadsave_state()
KVM: arm64: selftests: get-reg-list: actually enable pmu regs in pmu sublist
KVM: selftests: change pthread_yield to sched_yield
KVM: arm64: Fix detection of shared VMAs on guest fault
Darrick J. Wong [Mon, 26 Jul 2021 23:43:17 +0000 (16:43 -0700)]
xfs: prevent spoofing of rtbitmap blocks when recovering buffers
While reviewing the buffer item recovery code, the thought occurred to
me: in V5 filesystems we use log sequence number (LSN) tracking to avoid
replaying older metadata updates against newer log items. However, we
use the magic number of the ondisk buffer to find the LSN of the ondisk
metadata, which means that if an attacker can control the layout of the
realtime device precisely enough that the start of an rt bitmap block
matches the magic and UUID of some other kind of block, they can control
the purported LSN of that spoofed block and thereby break log replay.
Since realtime bitmap and summary blocks don't have headers at all, we
have no way to tell if a block really should be replayed. The best we
can do is replay unconditionally and hope for the best.
Two checkpoints in a single iclog. One is complete, the other just
contains the start record and overruns into a new iclog.
Timeline:
Before S1: Cache flush, log tail = X
At S1: Metadata stable, write start record and checkpoint
At C1: Write commit record, set NEED_FUA
Single iclog checkpoint, so no need for NEED_FLUSH
Log tail still = X, so no need for NEED_FLUSH
After C1,
Before S2: Cache flush, log tail = X
At S2: Metadata stable, write start record and checkpoint
After S2: Log tail moves to X+1
At EOIC: End of iclog, more journal data to write
Releases iclog
Not a commit iclog, so no need for NEED_FLUSH
Writes log tail X+1 into iclog.
At this point, the iclog has tail X+1 and NEED_FUA set. There has
been no cache flush for the metadata between X and X+1, and the
iclog writes the new tail permanently to the log. THis is sufficient
to violate on disk metadata/journal ordering.
We have two options here. The first is to detect this case in some
manner and ensure that the partial checkpoint write sets NEED_FLUSH
when the iclog is already marked NEED_FUA and the log tail changes.
This seems somewhat fragile and quite complex to get right, and it
doesn't actually make it obvious what underlying problem it is
actually addressing from reading the code.
The second option seems much cleaner to me, because it is derived
directly from the requirements of the C1 commit record in the iclog.
That is, when we write this commit record to the iclog, we've
guaranteed that the metadata/data ordering is correct for tail
update purposes. Hence if we only write the log tail into the iclog
for the *first* commit record rather than the log tail at the last
release, we guarantee that the log tail does not move past where the
the first commit record in the log expects it to be.
IOWs, taking the first option means that replay of C1 becomes
dependent on future operations doing the right thing, not just the
C1 checkpoint itself doing the right thing. This makes log recovery
almost impossible to reason about because now we have to take into
account what might or might not have happened in the future when
looking at checkpoints in the log rather than just having to
reconstruct the past...
Dave Chinner [Tue, 27 Jul 2021 23:23:50 +0000 (16:23 -0700)]
xfs: Enforce attr3 buffer recovery order
From the department of "WTAF? How did we miss that!?"...
When we are recovering a buffer, the first thing we do is check the
buffer magic number and extract the LSN from the buffer. If the LSN
is older than the current LSN, we replay the modification to it. If
the metadata on disk is newer than the transaction in the log, we
skip it. This is a fundamental v5 filesystem metadata recovery
behaviour.
generic/482 failed with an attribute writeback failure during log
recovery. The write verifier caught the corruption before it got
written to disk, and the attr buffer dump looked like:
The highlighted bytes are the LSN that was replayed into the
buffer: 0x100000538. This is cycle 1, block 0x538. Prior to replay,
that block on disk looks like this:
Note the LSN stamped into the buffer on disk: 1/0x900. The version
on disk is much newer than the log transaction that was being
replayed. That's a bug, and should -never- happen.
So I immediately went to look at xlog_recover_get_buf_lsn() to check
that we handled the LSN correctly. I was wondering if there was a
similar "two commits with the same start LSN skips the second
replay" problem with buffers. I didn't get that far, because I found
a much more basic, rudimentary bug: xlog_recover_get_buf_lsn()
doesn't recognise buffers with XFS_ATTR3_LEAF_MAGIC set in them!!!
IOWs, attr3 leaf buffers fall through the magic number checks
unrecognised, so trigger the "recover immediately" behaviour instead
of undergoing an LSN check. IOWs, we incorrectly replay ATTR3 leaf
buffers and that causes silent on disk corruption of inode attribute
forks and potentially other things....
Git history shows this is *another* zero day bug, this time
introduced in commit 50d5c8d8e938 ("xfs: check LSN ordering for v5
superblocks during recovery") which failed to handle the attr3 leaf
buffers in recovery. And we've failed to handle them ever since...
to format the log inode. It writes the LSN from the inode item into
the log inode, and if recovery decides the inode item needs to be
replayed, it recovers the log inode LSN field and writes it into the
on disk inode LSN field.
Now this might seem like a reasonable thing to do, but it is wrong
on multiple levels. Firstly, if the item is not yet in the AIL,
item->li_lsn is zero. i.e. the first time the inode it is logged and
formatted, the LSN we write into the log inode will be zero. If we
only log it once, recovery will run and can write this zero LSN into
the inode.
This means that the next time the inode is logged and log recovery
runs, it will *always* replay changes to the inode regardless of
whether the inode is newer on disk than the version in the log and
that violates the entire purpose of recording the LSN in the inode
at writeback time (i.e. to stop it going backwards in time on disk
during recovery).
Secondly, if we commit the CIL to the journal so the inode item
moves to the AIL, and then relog the inode, the LSN that gets
stamped into the log inode will be the LSN of the inode's current
location in the AIL, not it's age on disk. And it's not the LSN that
will be associated with the current change. That means when log
recovery replays this inode item, the LSN that ends up on disk is
the LSN for the previous changes in the log, not the current
changes being replayed. IOWs, after recovery the LSN on disk is not
in sync with the LSN of the modifications that were replayed into
the inode. This, again, violates the recovery ordering semantics
that on-disk writeback LSNs provide.
Hence the inode LSN in the log dinode is -always- invalid.
Thirdly, recovery actually has the LSN of the log transaction it is
replaying right at hand - it uses it to determine if it should
replay the inode by comparing it to the on-disk inode's LSN. But it
doesn't use that LSN to stamp the LSN into the inode which will be
written back when the transaction is fully replayed. It uses the one
in the log dinode, which we know is always going to be incorrect.
Looking back at the change history, the inode logging was broken by
commit 93f958f9c41f ("xfs: cull unnecessary icdinode fields") way
back in 2016 by a stupid idiot who thought he knew how this code
worked. i.e. me. That commit replaced an in memory di_lsn field that
was updated only at inode writeback time from the inode item.li_lsn
value - and hence always contained the same LSN that appeared in the
on-disk inode - with a read of the inode item LSN at inode format
time. CLearly these are not the same thing.
Before 93f958f9c41f, the log recovery behaviour was irrelevant,
because the LSN in the log inode always matched the on-disk LSN at
the time the inode was logged, hence recovery of the transaction
would never make the on-disk LSN in the inode go backwards or get
out of sync.
A symptom of the problem is this, caught from a failure of
generic/482. Before log recovery, the inode has been allocated but
never used:
You can see that the LSN of the on-disk inode is 0, even though it
clearly has been written to disk. I point out this inode, because
the generic/482 failure occurred because several adjacent inodes in
this specific inode cluster were not replayed correctly and still
appeared to be zero on disk when all the other metadata (inobt,
finobt, directories, etc) indicated they should be allocated and
written back.
The fix for this is two-fold. The first is that we need to either
revert the LSN changes in 93f958f9c41f or stop logging the inode LSN
altogether. If we do the former, log recovery does not need to
change but we add 8 bytes of memory per inode to store what is
largely a write-only inode field. If we do the latter, log recovery
needs to stamp the on-disk inode in the same manner that inode
writeback does.
I prefer the latter, because we shouldn't really be trying to log
and replay changes to the on disk LSN as the on-disk value is the
canonical source of the on-disk version of the inode. It also
matches the way we recover buffer items - we create a buf_log_item
that carries the current recovery transaction LSN that gets stamped
into the buffer by the write verifier when it gets written back
when the transaction is fully recovered.
However, this might break log recovery on older kernels even more,
so I'm going to simply ignore the logged value in recovery and stamp
the on-disk inode with the LSN of the transaction being recovered
that will trigger writeback on transaction recovery completion. This
will ensure that the on-disk inode LSN always reflects the LSN of
the last change that was written to disk, regardless of whether it
comes from log recovery or runtime writeback.
Fixes: 93f958f9c41f ("xfs: cull unnecessary icdinode fields") Signed-off-by: Dave Chinner <[email protected]> Reviewed-by: Darrick J. Wong <[email protected]> Signed-off-by: Darrick J. Wong <[email protected]>
Dave Chinner [Tue, 27 Jul 2021 23:23:49 +0000 (16:23 -0700)]
xfs: avoid unnecessary waits in xfs_log_force_lsn()
Before waiting on a iclog in xfs_log_force_lsn(), we don't check to
see if the iclog has already been completed and the contents on
stable storage. We check for completed iclogs in xfs_log_force(), so
we should do the same thing for xfs_log_force_lsn().
This fixed some random up-to-30s pauses seen in unmounting
filesystems in some tests. A log force ends up waiting on completed
iclog, and that doesn't then get flushed (and hence the log force
get completed) until the background log worker issues a log force
that flushes the iclog in question. Then the unmount unblocks and
continues.
Dave Chinner [Tue, 27 Jul 2021 23:23:49 +0000 (16:23 -0700)]
xfs: log forces imply data device cache flushes
After fixing the tail_lsn vs cache flush race, generic/482 continued
to fail in a similar way where cache flushes were missing before
iclog FUA writes. Tracing of iclog state changes during the fsstress
workload portion of the test (via xlog_iclog* events) indicated that
iclog writes were coming from two sources - CIL pushes and log
forces (due to fsync/O_SYNC operations). All of the cases where a
recovery problem was triggered indicated that the log force was the
source of the iclog write that was not preceeded by a cache flush.
This was an oversight in the modifications made in commit eef983ffeae7 ("xfs: journal IO cache flush reductions"). Log forces
for fsync imply a data device cache flush has been issued if an
iclog was flushed to disk and is indicated to the caller via the
log_flushed parameter so they can elide the device cache flush if
the journal issued one.
The change in eef983ffeae7 results in iclogs only issuing a cache
flush if XLOG_ICL_NEED_FLUSH is set on the iclog, but this was not
added to the iclogs that the log force code flushes to disk. Hence
log forces are no longer guaranteeing that a cache flush is issued,
hence opening up a potential on-disk ordering failure.
Log forces should also set XLOG_ICL_NEED_FUA as well to ensure that
the actual iclogs it forces to the journal are also on stable
storage before it returns to the caller.
This patch introduces the xlog_force_iclog() helper function to
encapsulate the process of taking a reference to an iclog, switching
its state if WANT_SYNC and flushing it to stable storage correctly.
Both xfs_log_force() and xfs_log_force_lsn() are converted to use
it, as is xlog_unmount_write() which has an elaborate method of
doing exactly the same "write this iclog to stable storage"
operation.
Further, if the log force code needs to wait on a iclog in the
WANT_SYNC state, it needs to ensure that iclog also results in a
cache flush being issued. This covers the case where the iclog
contains the commit record of the CIL flush that the log force
triggered, but it hasn't been written yet because there is still an
active reference to the iclog.
Note: this whole cache flush whack-a-mole patch is a result of log
forces still being iclog state centric rather than being CIL
sequence centric. Most of this nasty code will go away in future
when log forces are converted to wait on CIL sequence push
completion rather than iclog completion. With the CIL push algorithm
guaranteeing that the CIL checkpoint is fully on stable storage when
it completes, we no longer need to iterate iclogs and push them to
ensure a CIL sequence push has completed and so all this nasty iclog
iteration and flushing code will go away.
Dave Chinner [Tue, 27 Jul 2021 23:23:48 +0000 (16:23 -0700)]
xfs: factor out forced iclog flushes
We force iclogs in several places - we need them all to have the
same cache flush semantics, so start by factoring out the iclog
force into a common helper.
Dave Chinner [Tue, 27 Jul 2021 23:23:48 +0000 (16:23 -0700)]
xfs: fix ordering violation between cache flushes and tail updates
There is a race between the new CIL async data device metadata IO
completion cache flush and the log tail in the iclog the flush
covers being updated. This can be seen by repeating generic/482 in a
loop and eventually log recovery fails with a failures such as this:
Analysis of the logwrite replay shows that there were no writes to
the data device between the FUA @ write 124 and the FUA at write @
125, but log recovery @ 125 failed. The difference was the one log
write @ 125 moved the tail of the log forwards from (1,8) to (1,32)
and so the inode create intent in (1,8) was not replayed and so the
inode cluster was zero on disk when replay of the first inode item
in (1,32) was attempted.
What this meant was that the journal write that occurred at @ 125
did not ensure that metadata completed before the iclog was written
was correctly on stable storage. The tail of the log moved forward,
so IO must have been completed between the two iclog writes. This
means that there is a race condition between the unconditional async
cache flush in the CIL push work and the tail LSN that is written to
the iclog. This happens like so:
CIL push work AIL push work
------------- -------------
Add to committing list
start async data dev cache flush
.....
<flush completes>
<all writes to old tail lsn are stable>
xlog_write
.... push inode create buffer
<start IO>
.....
xlog_write(commit record)
.... <IO completes>
log tail moves
xlog_assign_tail_lsn()
start_lsn == commit_lsn
<no iclog preflush!>
xlog_state_release_iclog
__xlog_state_release_iclog()
<writes *new* tail_lsn into iclog>
xlog_sync()
....
submit_bio()
<tail in log moves forward without flushing written metadata>
Essentially, this can only occur if the commit iclog is issued
without a cache flush. If the iclog bio is submitted with
REQ_PREFLUSH, then it will guarantee that all the completed IO is
one stable storage before the iclog bio with the new tail LSN in it
is written to the log.
IOWs, the tail lsn that is written to the iclog needs to be sampled
*before* we issue the cache flush that guarantees all IO up to that
LSN has been completed.
To fix this without giving up the performance advantage of the
flush/FUA optimisations (e.g. g/482 runtime halves with 5.14-rc1
compared to 5.13), we need to ensure that we always issue a cache
flush if the tail LSN changes between the initial async flush and
the commit record being written. THis requires sampling the tail_lsn
before we start the flush, and then passing the sampled tail LSN to
xlog_state_release_iclog() so it can determine if the the tail LSN
has changed while writing the checkpoint. If the tail LSN has
changed, then it needs to set the NEED_FLUSH flag on the iclog and
we'll issue another cache flush before writing the iclog.
Dave Chinner [Tue, 27 Jul 2021 23:23:47 +0000 (16:23 -0700)]
xfs: external logs need to flush data device
The recent journal flush/FUA changes replaced the flushing of the
data device on every iclog write with an up-front async data device
cache flush. Unfortunately, the assumption of which this was based
on has been proven incorrect by the flush vs log tail update
ordering issue. As the fix for that issue uses the
XLOG_ICL_NEED_FLUSH flag to indicate that data device needs a cache
flush, we now need to (once again) ensure that an iclog write to
external logs that need a cache flush to be issued actually issue a
cache flush to the data device as well as the log device.
Michael Ellerman [Thu, 29 Jul 2021 12:56:36 +0000 (22:56 +1000)]
powerpc/vdso: Don't use r30 to avoid breaking Go lang
The Go runtime uses r30 for some special value called 'g'. It assumes
that value will remain unchanged even when calling VDSO functions.
Although r30 is non-volatile across function calls, the callee is free
to use it, as long as the callee saves the value and restores it before
returning.
It used to be true by accident that the VDSO didn't use r30, because the
VDSO was hand-written asm. When we switched to building the VDSO from C
the compiler started using r30, at least in some builds, leading to
crashes in Go. eg:
~/go/src$ ./all.bash
Building Go cmd/dist using /usr/lib/go-1.16. (go1.16.2 linux/ppc64le)
Building Go toolchain1 using /usr/lib/go-1.16.
go build os/exec: /usr/lib/go-1.16/pkg/tool/linux_ppc64le/compile: signal: segmentation fault
go build reflect: /usr/lib/go-1.16/pkg/tool/linux_ppc64le/compile: signal: segmentation fault
go tool dist: FAILED: /usr/lib/go-1.16/bin/go install -gcflags=-l -tags=math_big_pure_go compiler_bootstrap bootstrap/cmd/...: exit status 1
There are patches in flight to fix Go[1], but until they are released
and widely deployed we can workaround it in the VDSO by avoiding use of
r30.
Note this only works with GCC, clang does not support -ffixed-rN.
powerpc/pseries: Fix regression while building external modules
With commit c9f3401313a5 ("powerpc: Always enable queued spinlocks for
64s, disable for others") CONFIG_PPC_QUEUED_SPINLOCKS is always
enabled on ppc64le, external modules that use spinlock APIs are
failing.
ERROR: modpost: GPL-incompatible module XXX.ko uses GPL-only symbol 'shared_processor'
Before the above commit, modules were able to build without any
issues. Also this problem is not seen on other architectures. This
problem can be workaround if CONFIG_UNINLINE_SPIN_UNLOCK is enabled in
the config. However CONFIG_UNINLINE_SPIN_UNLOCK is not enabled by
default and only enabled in certain conditions like
CONFIG_DEBUG_SPINLOCKS is set in the kernel config.
Given that spin locks are one of the basic facilities for module code,
this effectively makes it impossible to build/load almost any non GPL
modules on ppc64le.
This was first reported at https://github.com/openzfs/zfs/issues/11172
Currently shared_processor is exported as GPL only symbol.
Fix this for parity with other architectures by exposing
shared_processor to non-GPL modules too.
Michael Zaidman [Thu, 29 Jul 2021 10:26:03 +0000 (13:26 +0300)]
HID: ft260: fix device removal due to USB disconnect
This commit fixes a functional regression introduced by the commit 82f09a637dd3
("HID: ft260: improve error handling of ft260_hid_feature_report_get()")
when upon USB disconnect, the FTDI FT260 i2c device is still available within
the /dev folder.
In my company's product, where the host USB to FT260 USB connection is
hard-wired in the PCB, the issue is not reproducible. To reproduce it, I used
the VirtualBox Ubuntu 20.04 VM and the UMFT260EV1A development module for the
FTDI FT260 chip:
Plug the UMFT260EV1A module into a USB port and attach it to VM.
The VM shows 2 i2c devices under the /dev:
michael@michael-VirtualBox:~$ ls /dev/i2c-*
/dev/i2c-0 /dev/i2c-1
The i2c-0 is not related to the FTDI FT260:
michael@michael-VirtualBox:~$ cat /sys/bus/i2c/devices/i2c-0/name
SMBus PIIX4 adapter at 4100
The i2c-1 is created by hid-ft260.ko:
michael@michael-VirtualBox:~$ cat /sys/bus/i2c/devices/i2c-1/name
FT260 usb-i2c bridge on hidraw1
Now, detach the FTDI FT260 USB device from VM. We expect the /dev/i2c-1
to disappear, but it's still here:
michael@michael-VirtualBox:~$ ls /dev/i2c-*
/dev/i2c-0 /dev/i2c-1
And the kernel log shows:
[ +0.001202] usb 2-2: USB disconnect, device number 3
[ +0.000109] ft260 0003:0403:6030.0002: failed to retrieve system status
[ +0.000316] ft260 0003:0403:6030.0003: failed to retrieve system status
It happens because the commit 82f09a637dd3 changed the ft260_get_system_config()
return logic. This caused the ft260_is_interface_enabled() to exit with error
upon the FT260 device USB disconnect, which in turn, aborted the ft260_remove()
before deleting the FT260 i2c device and cleaning its sysfs stuff.
This commit restores the FT260 USB removal functionality and improves the
ft260_is_interface_enabled() code to handle correctly all chip modes defined
by the device interface configuration pins DCNF0 and DCNF1.
Mike Rapoport [Tue, 27 Jul 2021 20:38:24 +0000 (23:38 +0300)]
alpha: register early reserved memory in memblock
The memory reserved by console/PALcode or non-volatile memory is not added
to memblock.memory.
Since commit fa3354e4ea39 (mm: free_area_init: use maximal zone PFNs rather
than zone sizes) the initialization of the memory map relies on the
accuracy of memblock.memory to properly calculate zone sizes. The holes in
memblock.memory caused by absent regions reserved by the firmware cause
incorrect initialization of struct pages which leads to BUG() during the
initial page freeing:
Dave Airlie [Thu, 29 Jul 2021 02:13:55 +0000 (12:13 +1000)]
Merge tag 'drm-intel-fixes-2021-07-28' of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes
Display related fixes:
- Fix vbt port mask
- Fix around reading the right DSC disable fuse in display_ver 10
- Split display version 9 and 10 in intel_setup_outputs
Dave Airlie [Thu, 29 Jul 2021 01:28:06 +0000 (11:28 +1000)]
Merge tag 'drm-msm-fixes-2021-07-27' of https://gitlab.freedesktop.org/drm/msm into drm-fixes
A few fixes for v5.14, including a fix for a crash if display triggers
an iommu fault (which tends to happen at probe time on devices with
bootloader fw that leaves display enabled as kernel starts)
Daniel Borkmann [Tue, 13 Jul 2021 08:18:31 +0000 (08:18 +0000)]
bpf: Fix leakage due to insufficient speculative store bypass mitigation
Spectre v4 gadgets make use of memory disambiguation, which is a set of
techniques that execute memory access instructions, that is, loads and
stores, out of program order; Intel's optimization manual, section 2.4.4.5:
A load instruction micro-op may depend on a preceding store. Many
microarchitectures block loads until all preceding store addresses are
known. The memory disambiguator predicts which loads will not depend on
any previous stores. When the disambiguator predicts that a load does
not have such a dependency, the load takes its data from the L1 data
cache. Eventually, the prediction is verified. If an actual conflict is
detected, the load and all succeeding instructions are re-executed.
af86ca4e3088 ("bpf: Prevent memory disambiguation attack") tried to mitigate
this attack by sanitizing the memory locations through preemptive "fast"
(low latency) stores of zero prior to the actual "slow" (high latency) store
of a pointer value such that upon dependency misprediction the CPU then
speculatively executes the load of the pointer value and retrieves the zero
value instead of the attacker controlled scalar value previously stored at
that location, meaning, subsequent access in the speculative domain is then
redirected to the "zero page".
The sanitized preemptive store of zero prior to the actual "slow" store is
done through a simple ST instruction based on r10 (frame pointer) with
relative offset to the stack location that the verifier has been tracking
on the original used register for STX, which does not have to be r10. Thus,
there are no memory dependencies for this store, since it's only using r10
and immediate constant of zero; hence af86ca4e3088 /assumed/ a low latency
operation.
However, a recent attack demonstrated that this mitigation is not sufficient
since the preemptive store of zero could also be turned into a "slow" store
and is thus bypassed as well:
[...]
// r2 = oob address (e.g. scalar)
// r7 = pointer to map value
31: (7b) *(u64 *)(r10 -16) = r2
// r9 will remain "fast" register, r10 will become "slow" register below
32: (bf) r9 = r10
// JIT maps BPF reg to x86 reg:
// r9 -> r15 (callee saved)
// r10 -> rbp
// train store forward prediction to break dependency link between both r9
// and r10 by evicting them from the predictor's LRU table.
33: (61) r0 = *(u32 *)(r7 +24576)
34: (63) *(u32 *)(r7 +29696) = r0
35: (61) r0 = *(u32 *)(r7 +24580)
36: (63) *(u32 *)(r7 +29700) = r0
37: (61) r0 = *(u32 *)(r7 +24584)
38: (63) *(u32 *)(r7 +29704) = r0
39: (61) r0 = *(u32 *)(r7 +24588)
40: (63) *(u32 *)(r7 +29708) = r0
[...]
543: (61) r0 = *(u32 *)(r7 +25596)
544: (63) *(u32 *)(r7 +30716) = r0
// prepare call to bpf_ringbuf_output() helper. the latter will cause rbp
// to spill to stack memory while r13/r14/r15 (all callee saved regs) remain
// in hardware registers. rbp becomes slow due to push/pop latency. below is
// disasm of bpf_ringbuf_output() helper for better visual context:
//
// ffffffff8117ee20: 41 54 push r12
// ffffffff8117ee22: 55 push rbp
// ffffffff8117ee23: 53 push rbx
// ffffffff8117ee24: 48 f7 c1 fc ff ff ff test rcx,0xfffffffffffffffc
// ffffffff8117ee2b: 0f 85 af 00 00 00 jne ffffffff8117eee0 <-- jump taken
// [...]
// ffffffff8117eee0: 49 c7 c4 ea ff ff ff mov r12,0xffffffffffffffea
// ffffffff8117eee7: 5b pop rbx
// ffffffff8117eee8: 5d pop rbp
// ffffffff8117eee9: 4c 89 e0 mov rax,r12
// ffffffff8117eeec: 41 5c pop r12
// ffffffff8117eeee: c3 ret
545: (18) r1 = map[id:4]
547: (bf) r2 = r7
548: (b7) r3 = 0
549: (b7) r4 = 4
550: (85) call bpf_ringbuf_output#194288
// instruction 551 inserted by verifier \
551: (7a) *(u64 *)(r10 -16) = 0 | /both/ are now slow stores here
// storing map value pointer r7 at fp-16 | since value of r10 is "slow".
552: (7b) *(u64 *)(r10 -16) = r7 /
// following "fast" read to the same memory location, but due to dependency
// misprediction it will speculatively execute before insn 551/552 completes.
553: (79) r2 = *(u64 *)(r9 -16)
// in speculative domain contains attacker controlled r2. in non-speculative
// domain this contains r7, and thus accesses r7 +0 below.
554: (71) r3 = *(u8 *)(r2 +0)
// leak r3
As can be seen, the current speculative store bypass mitigation which the
verifier inserts at line 551 is insufficient since /both/, the write of
the zero sanitation as well as the map value pointer are a high latency
instruction due to prior memory access via push/pop of r10 (rbp) in contrast
to the low latency read in line 553 as r9 (r15) which stays in hardware
registers. Thus, architecturally, fp-16 is r7, however, microarchitecturally,
fp-16 can still be r2.
Initial thoughts to address this issue was to track spilled pointer loads
from stack and enforce their load via LDX through r10 as well so that /both/
the preemptive store of zero /as well as/ the load use the /same/ register
such that a dependency is created between the store and load. However, this
option is not sufficient either since it can be bypassed as well under
speculation. An updated attack with pointer spill/fills now _all_ based on
r10 would look as follows:
[...]
// r2 = oob address (e.g. scalar)
// r7 = pointer to map value
[...]
// longer store forward prediction training sequence than before.
2062: (61) r0 = *(u32 *)(r7 +25588)
2063: (63) *(u32 *)(r7 +30708) = r0
2064: (61) r0 = *(u32 *)(r7 +25592)
2065: (63) *(u32 *)(r7 +30712) = r0
2066: (61) r0 = *(u32 *)(r7 +25596)
2067: (63) *(u32 *)(r7 +30716) = r0
// store the speculative load address (scalar) this time after the store
// forward prediction training.
2068: (7b) *(u64 *)(r10 -16) = r2
// preoccupy the CPU store port by running sequence of dummy stores.
2069: (63) *(u32 *)(r7 +29696) = r0
2070: (63) *(u32 *)(r7 +29700) = r0
2071: (63) *(u32 *)(r7 +29704) = r0
2072: (63) *(u32 *)(r7 +29708) = r0
2073: (63) *(u32 *)(r7 +29712) = r0
2074: (63) *(u32 *)(r7 +29716) = r0
2075: (63) *(u32 *)(r7 +29720) = r0
2076: (63) *(u32 *)(r7 +29724) = r0
2077: (63) *(u32 *)(r7 +29728) = r0
2078: (63) *(u32 *)(r7 +29732) = r0
2079: (63) *(u32 *)(r7 +29736) = r0
2080: (63) *(u32 *)(r7 +29740) = r0
2081: (63) *(u32 *)(r7 +29744) = r0
2082: (63) *(u32 *)(r7 +29748) = r0
2083: (63) *(u32 *)(r7 +29752) = r0
2084: (63) *(u32 *)(r7 +29756) = r0
2085: (63) *(u32 *)(r7 +29760) = r0
2086: (63) *(u32 *)(r7 +29764) = r0
2087: (63) *(u32 *)(r7 +29768) = r0
2088: (63) *(u32 *)(r7 +29772) = r0
2089: (63) *(u32 *)(r7 +29776) = r0
2090: (63) *(u32 *)(r7 +29780) = r0
2091: (63) *(u32 *)(r7 +29784) = r0
2092: (63) *(u32 *)(r7 +29788) = r0
2093: (63) *(u32 *)(r7 +29792) = r0
2094: (63) *(u32 *)(r7 +29796) = r0
2095: (63) *(u32 *)(r7 +29800) = r0
2096: (63) *(u32 *)(r7 +29804) = r0
2097: (63) *(u32 *)(r7 +29808) = r0
2098: (63) *(u32 *)(r7 +29812) = r0
// overwrite scalar with dummy pointer; same as before, also including the
// sanitation store with 0 from the current mitigation by the verifier.
2099: (7a) *(u64 *)(r10 -16) = 0 | /both/ are now slow stores here
2100: (7b) *(u64 *)(r10 -16) = r7 | since store unit is still busy.
// load from stack intended to bypass stores.
2101: (79) r2 = *(u64 *)(r10 -16)
2102: (71) r3 = *(u8 *)(r2 +0)
// leak r3
[...]
Looking at the CPU microarchitecture, the scheduler might issue loads (such
as seen in line 2101) before stores (line 2099,2100) because the load execution
units become available while the store execution unit is still busy with the
sequence of dummy stores (line 2069-2098). And so the load may use the prior
stored scalar from r2 at address r10 -16 for speculation. The updated attack
may work less reliable on CPU microarchitectures where loads and stores share
execution resources.
This concludes that the sanitizing with zero stores from af86ca4e3088 ("bpf:
Prevent memory disambiguation attack") is insufficient. Moreover, the detection
of stack reuse from af86ca4e3088 where previously data (STACK_MISC) has been
written to a given stack slot where a pointer value is now to be stored does
not have sufficient coverage as precondition for the mitigation either; for
several reasons outlined as follows:
1) Stack content from prior program runs could still be preserved and is
therefore not "random", best example is to split a speculative store
bypass attack between tail calls, program A would prepare and store the
oob address at a given stack slot and then tail call into program B which
does the "slow" store of a pointer to the stack with subsequent "fast"
read. From program B PoV such stack slot type is STACK_INVALID, and
therefore also must be subject to mitigation.
2) The STACK_SPILL must not be coupled to register_is_const(&stack->spilled_ptr)
condition, for example, the previous content of that memory location could
also be a pointer to map or map value. Without the fix, a speculative
store bypass is not mitigated in such precondition and can then lead to
a type confusion in the speculative domain leaking kernel memory near
these pointer types.
While brainstorming on various alternative mitigation possibilities, we also
stumbled upon a retrospective from Chrome developers [0]:
[...] For variant 4, we implemented a mitigation to zero the unused memory
of the heap prior to allocation, which cost about 1% when done concurrently
and 4% for scavenging. Variant 4 defeats everything we could think of. We
explored more mitigations for variant 4 but the threat proved to be more
pervasive and dangerous than we anticipated. For example, stack slots used
by the register allocator in the optimizing compiler could be subject to
type confusion, leading to pointer crafting. Mitigating type confusion for
stack slots alone would have required a complete redesign of the backend of
the optimizing compiler, perhaps man years of work, without a guarantee of
completeness. [...]
From BPF side, the problem space is reduced, however, options are rather
limited. One idea that has been explored was to xor-obfuscate pointer spills
to the BPF stack:
[...]
// preoccupy the CPU store port by running sequence of dummy stores.
[...]
2106: (63) *(u32 *)(r7 +29796) = r0
2107: (63) *(u32 *)(r7 +29800) = r0
2108: (63) *(u32 *)(r7 +29804) = r0
2109: (63) *(u32 *)(r7 +29808) = r0
2110: (63) *(u32 *)(r7 +29812) = r0
// overwrite scalar with dummy pointer; xored with random 'secret' value
// of 943576462 before store ...
2111: (b4) w11 = 943576462
2112: (af) r11 ^= r7
2113: (7b) *(u64 *)(r10 -16) = r11
2114: (79) r11 = *(u64 *)(r10 -16)
2115: (b4) w2 = 943576462
2116: (af) r2 ^= r11
// ... and restored with the same 'secret' value with the help of AX reg.
2117: (71) r3 = *(u8 *)(r2 +0)
[...]
While the above would not prevent speculation, it would make data leakage
infeasible by directing it to random locations. In order to be effective
and prevent type confusion under speculation, such random secret would have
to be regenerated for each store. The additional complexity involved for a
tracking mechanism that prevents jumps such that restoring spilled pointers
would not get corrupted is not worth the gain for unprivileged. Hence, the
fix in here eventually opted for emitting a non-public BPF_ST | BPF_NOSPEC
instruction which the x86 JIT translates into a lfence opcode. Inserting the
latter in between the store and load instruction is one of the mitigations
options [1]. The x86 instruction manual notes:
[...] An LFENCE that follows an instruction that stores to memory might
complete before the data being stored have become globally visible. [...]
The latter meaning that the preceding store instruction finished execution
and the store is at minimum guaranteed to be in the CPU's store queue, but
it's not guaranteed to be in that CPU's L1 cache at that point (globally
visible). The latter would only be guaranteed via sfence. So the load which
is guaranteed to execute after the lfence for that local CPU would have to
rely on store-to-load forwarding. [2], in section 2.3 on store buffers says:
[...] For every store operation that is added to the ROB, an entry is
allocated in the store buffer. This entry requires both the virtual and
physical address of the target. Only if there is no free entry in the store
buffer, the frontend stalls until there is an empty slot available in the
store buffer again. Otherwise, the CPU can immediately continue adding
subsequent instructions to the ROB and execute them out of order. On Intel
CPUs, the store buffer has up to 56 entries. [...]
One small upside on the fix is that it lifts constraints from af86ca4e3088
where the sanitize_stack_off relative to r10 must be the same when coming
from different paths. The BPF_ST | BPF_NOSPEC gets emitted after a BPF_STX
or BPF_ST instruction. This happens either when we store a pointer or data
value to the BPF stack for the first time, or upon later pointer spills.
The former needs to be enforced since otherwise stale stack data could be
leaked under speculation as outlined earlier. For non-x86 JITs the BPF_ST |
BPF_NOSPEC mapping is currently optimized away, but others could emit a
speculation barrier as well if necessary. For real-world unprivileged
programs e.g. generated by LLVM, pointer spill/fill is only generated upon
register pressure and LLVM only tries to do that for pointers which are not
used often. The program main impact will be the initial BPF_ST | BPF_NOSPEC
sanitation for the STACK_INVALID case when the first write to a stack slot
occurs e.g. upon map lookup. In future we might refine ways to mitigate
the latter cost.
Daniel Borkmann [Tue, 13 Jul 2021 08:18:31 +0000 (08:18 +0000)]
bpf: Introduce BPF nospec instruction for mitigating Spectre v4
In case of JITs, each of the JIT backends compiles the BPF nospec instruction
/either/ to a machine instruction which emits a speculation barrier /or/ to
/no/ machine instruction in case the underlying architecture is not affected
by Speculative Store Bypass or has different mitigations in place already.
This covers both x86 and (implicitly) arm64: In case of x86, we use 'lfence'
instruction for mitigation. In case of arm64, we rely on the firmware mitigation
as controlled via the ssbd kernel parameter. Whenever the mitigation is enabled,
it works for all of the kernel code with no need to provide any additional
instructions here (hence only comment in arm64 JIT). Other archs can follow
as needed. The BPF nospec instruction is specifically targeting Spectre v4
since i) we don't use a serialization barrier for the Spectre v1 case, and
ii) mitigation instructions for v1 and v4 might be different on some archs.
The BPF nospec is required for a future commit, where the BPF verifier does
annotate intermediate BPF programs with speculation barriers.
Merge tag 'fixes_for_v5.14-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs
Pull ext2 and reiserfs fixes from Jan Kara:
"A fix for the ext2 conversion to kmap_local() and two reiserfs
hardening fixes"
* tag 'fixes_for_v5.14-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
reiserfs: check directory items on read from disk
fs/ext2: Avoid page_address on pages returned by ext2_get_page
reiserfs: add check for root_inode in reiserfs_fill_super
Merge tag 'platform-drivers-x86-v5.14-2' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86
Pull x86 platform driver fixes from Hans de Goede:
"A set of bug-fixes and new hardware ids.
Highlights:
- amd-pmc fixes
- think-lmi fixes
- various new hardware-ids"
* tag 'platform-drivers-x86-v5.14-2' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86:
platform/x86: gigabyte-wmi: add support for B550 Aorus Elite V2
platform/x86: intel-hid: add Alder Lake ACPI device ID
platform/x86: think-lmi: Fix possible mem-leaks on tlmi_analyze() error-exit
platform/x86: think-lmi: Split kobject_init() and kobject_add() calls
platform/x86: think-lmi: Move pending_reboot_attr to the attributes sysfs dir
platform/x86: amd-pmc: Fix undefined reference to __udivdi3
platform/x86: amd-pmc: Fix missing unlock on error in amd_pmc_send_cmd()
platform/x86: wireless-hotkey: remove hardcoded "hp" from the error message
platform/x86: amd-pmc: Use return code on suspend
platform/x86: amd-pmc: Add new acpi id for future PMC controllers
platform/x86: amd-pmc: Add support for ACPI ID AMDI0006
platform/x86: amd-pmc: Add support for logging s0ix counters
platform/x86: amd-pmc: Add support for logging SMU metrics
platform/x86: amd-pmc: call dump registers only once
platform/x86: amd-pmc: Fix SMU firmware reporting mechanism
platform/x86: amd-pmc: Fix command completion code
platform/x86: think-lmi: Add pending_reboot support
unix_gc() assumes that candidate sockets can never gain an external
reference (i.e. be installed into an fd) while the unix_gc_lock is
held. Except for MSG_PEEK this is guaranteed by modifying inflight
count under the unix_gc_lock.
MSG_PEEK does not touch any variable protected by unix_gc_lock (file
count is not), yet it needs to be serialized with garbage collection.
Do this by locking/unlocking unix_gc_lock:
1) increment file count
2) lock/unlock barrier to make sure incremented file count is visible
to garbage collection
3) install file into fd
This is a lock barrier (unlike smp_mb()) that ensures that garbage
collection is run completely before or completely after the barrier.
As a note, prior to commit cf89af146b7e ("btrfs: dev-replace: fail
mount if we don't have replace item with target device"), rw_devices
was decremented on removing a writable device in
__btrfs_free_extra_devids only if the BTRFS_DEV_STATE_REPLACE_TGT bit
was not set for the device. However, this check does not need to be
reinstated as it is now redundant and incorrect.
In __btrfs_free_extra_devids, we skip removing the device if it is the
target for replacement. This is done by checking whether device->devid
== BTRFS_DEV_REPLACE_DEVID. Since BTRFS_DEV_STATE_REPLACE_TGT is set
only on the device with devid BTRFS_DEV_REPLACE_DEVID, no devices
should have the BTRFS_DEV_STATE_REPLACE_TGT bit set after the check,
and so it's redundant to test for that bit.
Additionally, following commit 82372bc816d7 ("Btrfs: make
the logic of source device removing more clear"), rw_devices is
incremented whenever a writeable device is added to the alloc
list (including the target device in btrfs_dev_replace_finishing), so
all removals of writable devices from the alloc list should also be
accompanied by a decrement to rw_devices.
Filipe Manana [Tue, 27 Jul 2021 10:24:43 +0000 (11:24 +0100)]
btrfs: fix lost inode on log replay after mix of fsync, rename and inode eviction
When checking if we need to log the new name of a renamed inode, we are
checking if the inode and its parent inode have been logged before, and if
not we don't log the new name. The check however is buggy, as it directly
compares the logged_trans field of the inodes versus the ID of the current
transaction. The problem is that logged_trans is a transient field, only
stored in memory and never persisted in the inode item, so if an inode
was logged before, evicted and reloaded, its logged_trans field is set to
a value of 0, meaning the check will return false and the new name of the
renamed inode is not logged. If the old parent directory was previously
fsynced and we deleted the logged directory entries corresponding to the
old name, we end up with a log that when replayed will delete the renamed
inode.
# Add some new file to A and fsync directory A.
$ touch /mnt/A/bar
$ xfs_io -c "fsync" /mnt/A
# Now trigger inode eviction. We are only interested in triggering
# eviction for the inode of directory A.
$ echo 2 > /proc/sys/vm/drop_caches
# Move foo from directory A to directory B.
# This deletes the directory entries for foo in A from the log, and
# does not add the new name for foo in directory B to the log, because
# logged_trans of A is 0, which is less than the current transaction ID.
$ mv /mnt/A/foo /mnt/B/foo
# Now make an fsync to anything except A, B or any file inside them,
# like for example create a file at the root directory and fsync this
# new file. This syncs the log that contains all the changes done by
# previous rename operation.
$ touch /mnt/baz
$ xfs_io -c "fsync" /mnt/baz
<power fail>
# Mount the filesystem and replay the log.
$ mount /dev/sdc /mnt
# Check the filesystem content.
$ ls -1R /mnt
/mnt/:
A
B
baz
/mnt/A:
bar
/mnt/B:
$
# File foo is gone, it's neither in A/ nor in B/.
Fix this by using the inode_logged() helper at btrfs_log_new_name(), which
safely checks if an inode was logged before in the current transaction.
btrfs: mark compressed range uptodate only if all bio succeed
In compression write endio sequence, the range which the compressed_bio
writes is marked as uptodate if the last bio of the compressed (sub)bios
is completed successfully. There could be previous bio which may
have failed which is recorded in cb->errors.
Set the writeback range as uptodate only if cb->errors is zero, as opposed
to checking only the last bio's status.
Backporting notes: in all versions up to 4.4 the last argument is always
replaced by "!cb->errors".
The current assumption that methods to read PCH FIVR attributes will
return integer, is not correct. There is no good way to return integer
as negative numbers are also valid.
These read methods return a package of integers. The first integer returns
status, which is 0 on success and any other value for failure. When the
returned status is zero, then the second integer returns the actual value.
This change fixes this issue by replacing acpi_evaluate_integer() with
acpi_evaluate_object() and use acpi_extract_package() to extract results.
Hui Wang [Wed, 28 Jul 2021 15:19:58 +0000 (23:19 +0800)]
Revert "ACPI: resources: Add checks for ACPI IRQ override"
The commit 0ec4e55e9f57 ("ACPI: resources: Add checks for ACPI IRQ
override") introduces regression on some platforms, at least it makes
the UART can't get correct irq setting on two different platforms,
and it makes the kernel can't bootup on these two platforms.
io_uring: fix poll requests leaking second poll entries
For pure poll requests, it doesn't remove the second poll wait entry
when it's done, neither after vfs_poll() or in the poll completion
handler. We should remove the second poll wait entry.
And we use io_poll_remove_double() rather than io_poll_remove_waitqs()
since the latter has some redundant logic.
io_uring: don't block level reissue off completion path
Some setups, like SCSI, can throw spurious -EAGAIN off the softirq
completion path. Normally we expect this to happen inline as part
of submission, but apparently SCSI has a weird corner case where it
can happen as part of normal completions.
This should be solved by having the -EAGAIN bubble back up the stack
as part of submission, but previous attempts at this failed and we're
not just quite there yet. Instead we currently use REQ_F_REISSUE to
handle this case.
For now, catch it in io_rw_should_reissue() and prevent a reissue
from a bogus path.
Jason Gerecke [Mon, 19 Jul 2021 20:55:31 +0000 (13:55 -0700)]
HID: wacom: Skip processing of touches with negative slot values
The `input_mt_get_slot_by_key` function may return a negative value
if an error occurs (e.g. running out of slots). If this occurs we
should really avoid reporting any data for the slot.
Jason Gerecke [Mon, 19 Jul 2021 20:55:28 +0000 (13:55 -0700)]
HID: wacom: Re-enable touch by default for Cintiq 24HDT / 27QHDT
Commit 670e90924bfe ("HID: wacom: support named keys on older devices")
added support for sending named events from the soft buttons on the
24HDT and 27QHDT. In the process, however, it inadvertantly disabled the
touchscreen of the 24HDT and 27QHDT by default. The
`wacom_set_shared_values` function would normally enable touch by default
but because it checks the state of the non-shared `has_mute_touch_switch`
flag and `wacom_setup_touch_input_capabilities` sets the state of the
/shared/ version, touch ends up being disabled by default.
This patch sets the non-shared flag, letting `wacom_set_shared_values`
take care of copying the value over to the shared version and setting
the default touch state to "on".
HID: apple: Add support for Keychron K1 wireless keyboard
The Keychron K1 wireless keyboard has a set of Apple-like function keys
and an Fn key that works like on an Apple bluetooth keyboard. It
identifies as an Apple Alu RevB ANSI keyboard (05ac:024f) over USB and
BT. Use hid-apple for it so the Fn key and function keys work correctly.
nfc: nfcsim: fix use after free during module unload
There is a use after free memory corruption during module exit:
- nfcsim_exit()
- nfcsim_device_free(dev0)
- nfc_digital_unregister_device()
This iterates over command queue and frees all commands,
- dev->up = false
- nfcsim_link_shutdown()
- nfcsim_link_recv_wake()
This wakes the sleeping thread nfcsim_link_recv_skb().
- nfcsim_link_recv_skb()
Wake from wait_event_interruptible_timeout(),
call directly the deb->cb callback even though (dev->up == false),
- digital_send_cmd_complete()
Dereference of "struct digital_cmd" cmd which was freed earlier by
nfc_digital_unregister_device().
This causes memory corruption shortly after (with unrelated stack
trace):
nfc nfc0: NFC: nfcsim_recv_wq: Device is down
llcp: nfc_llcp_recv: err -19
nfc nfc1: NFC: nfcsim_recv_wq: Device is down
BUG: unable to handle page fault for address: ffffffffffffffed
Call Trace:
fsnotify+0x54b/0x5c0
__fsnotify_parent+0x1fe/0x300
? vfs_write+0x27c/0x390
vfs_write+0x27c/0x390
ksys_write+0x63/0xe0
do_syscall_64+0x3b/0x90
entry_SYSCALL_64_after_hwframe+0x44/0xae
This flow of calling digital_send_cmd_complete() callback on driver exit
is specific to nfcsim which implements reading and sending work queues.
Since the NFC digital device was unregistered, the callback should not
be called.
blkdev_get_no_open acquires a reference to the block_device through
the block device inode and then tries to acquire a device model
reference to the gendisk. But at this point the disk migh already
be freed (although the race is free). Fix this by only freeing the
gendisk from the whole device bdevs ->free_inode callback as well.
blk-iocost: fix operation ordering in iocg_wake_fn()
iocg_wake_fn() open-codes wait_queue_entry removal and wakeup because it
wants the wq_entry to be always removed whether it ended up waking the
task or not. finish_wait() tests whether wq_entry needs removal without
grabbing the wait_queue lock and expects the waker to use
list_del_init_careful() after all waking operations are complete, which
iocg_wake_fn() didn't do. The operation order was wrong and the regular
list_del_init() was used.
The result is that if a waiter wakes up racing the waker, it can free pop
the wq_entry off stack before the waker is still looking at it, which can
lead to a backtrace like the following.
net/mlx5e: Fix nullptr in mlx5e_hairpin_get_mdev()
The result of __dev_get_by_index() is not checked for NULL and then gets
dereferenced immediately.
Also, __dev_get_by_index() must be called while holding either RTNL lock
or @dev_base_lock, which isn't satisfied by mlx5e_hairpin_get_mdev() or
its callers. This makes the underlying hlist_for_each_entry() loop not
safe, and can have adverse effects in itself.
Fix by using dev_get_by_index() and handling nullptr return value when
ifindex device is not found. Update mlx5e_hairpin_get_mdev() callers to
check for possible PTR_ERR() result.
Aya Levin [Wed, 16 Jun 2021 16:11:03 +0000 (19:11 +0300)]
net/mlx5: Unload device upon firmware fatal error
When fw_fatal reporter reports an error, the firmware in not responding.
Unload the device to ensure that the driver closes all its resources,
even if recovery is not due (user disabled auto-recovery or reporter is
in grace period). On successful recovery the device is loaded back up.
Fixes: b3bd076f7501 ("net/mlx5: Report devlink health on FW fatal issues") Signed-off-by: Aya Levin <[email protected]> Reviewed-by: Moshe Shemesh <[email protected]> Signed-off-by: Saeed Mahameed <[email protected]>
net/mlx5e: Add NETIF_F_HW_TC to hw_features when HTB offload is available
If a feature flag is only present in features, but not in hw_features,
the user can't reset it. Although hw_features may contain NETIF_F_HW_TC
by the point where the driver checks whether HTB offload is supported,
this flag is controlled by another condition that may not hold. Set it
explicitly to make sure the user can disable it.
Tariq Toukan [Wed, 30 Jun 2021 10:45:05 +0000 (13:45 +0300)]
net/mlx5e: RX, Avoid possible data corruption when relaxed ordering and LRO combined
When HW aggregates packets for an LRO session, it writes the payload
of two consecutive packets of a flow contiguously, so that they usually
share a cacheline.
The first byte of a packet's payload is written immediately after
the last byte of the preceding packet.
In this flow, there are two consecutive write requests to the shared
cacheline:
1. Regular write for the earlier packet.
2. Read-modify-write for the following packet.
In case of relaxed-ordering on, these two writes might be re-ordered.
Using the end padding optimization (to avoid partial write for the last
cacheline of a packet) becomes problematic if the two writes occur
out-of-order, as the padding would overwrite payload that belongs to
the following packet, causing data corruption.
Avoid this by disabling the end padding optimization when both
LRO and relaxed-ordering are enabled.
Roi Dayan [Wed, 2 Jun 2021 11:17:07 +0000 (14:17 +0300)]
net/mlx5: E-Switch, handle devcom events only for ports on the same device
This is the same check as LAG mode checks if to enable lag.
This will fix adding peer miss rules if lag is not supported
and even an incorrect rules in socket direct mode.
Also fix the incorrect comment on mlx5_get_next_phys_dev() as flow #1
doesn't exists.
Maor Dickman [Tue, 22 Jun 2021 14:07:02 +0000 (17:07 +0300)]
net/mlx5: E-Switch, Set destination vport vhca id only when merged eswitch is supported
Destination vport vhca id is valid flag is set only merged eswitch isn't supported.
Change destination vport vhca id value to be set also only when merged eswitch
is supported.
Fixes: e4ad91f23f10 ("net/mlx5e: Split offloaded eswitch TC rules for port mirroring") Signed-off-by: Maor Dickman <[email protected]> Reviewed-by: Roi Dayan <[email protected]> Signed-off-by: Saeed Mahameed <[email protected]>
Maor Gottlieb [Mon, 26 Jul 2021 06:20:14 +0000 (09:20 +0300)]
net/mlx5: Fix flow table chaining
Fix a bug when flow table is created in priority that already
has other flow tables as shown in the below diagram.
If the new flow table (FT-B) has the lowest level in the priority,
we need to connect the flow tables from the previous priority (p0)
to this new table. In addition when this flow table is destroyed
(FT-B), we need to connect the flow tables from the previous
priority (p0) to the next level flow table (FT-C) in the same
priority of the destroyed table (if exists).
Merge branch 'sockmap fixes picked up by stress tests'
John Fastabend says:
====================
Running stress tests with recent patch to remove an extra lock in sockmap
resulted in a couple new issues popping up. It seems only one of them
is actually related to the patch:
799aa7f98d53 ("skmsg: Avoid lock_sock() in sk_psock_backlog()")
The other two issues had existed long before, but I guess the timing
with the serialization we had before was too tight to get any of
our tests or deployments to hit it.
With attached series stress testing sockmap+TCP with workloads that
create lots of short-lived connections no more splats like below were
seen on upstream bpf branch.
John Fastabend [Tue, 27 Jul 2021 16:05:00 +0000 (09:05 -0700)]
bpf, sockmap: Fix memleak on ingress msg enqueue
If backlog handler is running during a tear down operation we may enqueue
data on the ingress msg queue while tear down is trying to free it.
sk_psock_backlog()
sk_psock_handle_skb()
skb_psock_skb_ingress()
sk_psock_skb_ingress_enqueue()
sk_psock_queue_msg(psock,msg)
spin_lock(ingress_lock)
sk_psock_zap_ingress()
_sk_psock_purge_ingerss_msg()
_sk_psock_purge_ingress_msg()
-- free ingress_msg list --
spin_unlock(ingress_lock)
spin_lock(ingress_lock)
list_add_tail(msg,ingress_msg) <- entry on list with no one
left to free it.
spin_unlock(ingress_lock)
To fix we only enqueue from backlog if the ENABLED bit is set. The tear
down logic clears the bit with ingress_lock set so we wont enqueue the
msg in the last step.
John Fastabend [Tue, 27 Jul 2021 16:04:59 +0000 (09:04 -0700)]
bpf, sockmap: On cleanup we additionally need to remove cached skb
Its possible if a socket is closed and the receive thread is under memory
pressure it may have cached a skb. We need to ensure these skbs are
free'd along with the normal ingress_skb queue.
Before 799aa7f98d53 ("skmsg: Avoid lock_sock() in sk_psock_backlog()") tear
down and backlog processing both had sock_lock for the common case of
socket close or unhash. So it was not possible to have both running in
parrallel so all we would need is the kfree in those kernels.
But, latest kernels include the commit 799aa7f98d5e and this requires a
bit more work. Without the ingress_lock guarding reading/writing the
state->skb case its possible the tear down could run before the state
update causing it to leak memory or worse when the backlog reads the state
it could potentially run interleaved with the tear down and we might end up
free'ing the state->skb from tear down side but already have the reference
from backlog side. To resolve such races we wrap accesses in ingress_lock
on both sides serializing tear down and backlog case. In both cases this
only happens after an EAGAIN error case so having an extra lock in place
is likely fine. The normal path will skip the locks.
Note, we check state->skb before grabbing lock. This works because
we can only enqueue with the mutex we hold already. Avoiding a race
on adding state->skb after the check. And if tear down path is running
that is also fine if the tear down path then removes state->skb we
will simply set skb=NULL and the subsequent goto is skipped. This
slight complication avoids locking in normal case.
With this fix we no longer see this warning splat from tcp side on
socket close when we hit the above case with redirect to ingress self.
John Fastabend [Tue, 27 Jul 2021 16:04:58 +0000 (09:04 -0700)]
bpf, sockmap: Zap ingress queues after stopping strparser
We don't want strparser to run and pass skbs into skmsg handlers when
the psock is null. We just sk_drop them in this case. When removing
a live socket from map it means extra drops that we do not need to
incur. Move the zap below strparser close to avoid this condition.
This way we stop the stream parser first stopping it from processing
packets and then delete the psock.
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma
Pull rdma fixes from Jason Gunthorpe:
"Nothing very exciting here, mainly just a bunch of irdma fixes. irdma
is a new driver this cycle so it to be expected.
- Many more irdma fixups from bots/etc
- bnxt_re regression in their counters from a FW upgrade
- User triggerable memory leak in rxe"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
RDMA/irdma: Change returned type of irdma_setup_virt_qp to void
RDMA/irdma: Change the returned type of irdma_set_hw_rsrc to void
RDMA/irdma: change the returned type of irdma_sc_repost_aeq_entries to void
RDMA/irdma: Check vsi pointer before using it
RDMA/rxe: Fix memory leak in error path code
RDMA/irdma: Change the returned type to void
RDMA/irdma: Make spdxcheck.py happy
RDMA/irdma: Fix unused variable total_size warning
RDMA/bnxt_re: Fix stats counters
Paolo Bonzini [Tue, 27 Jul 2021 12:43:10 +0000 (08:43 -0400)]
KVM: add missing compat KVM_CLEAR_DIRTY_LOG
The arguments to the KVM_CLEAR_DIRTY_LOG ioctl include a pointer,
therefore it needs a compat ioctl implementation. Otherwise,
32-bit userspace fails to invoke it on 64-bit kernels; for x86
it might work fine by chance if the padding is zero, but not
on big-endian architectures.
Reported-by: Thomas Sattler Cc: [email protected] Fixes: 2a31b9db1535 ("kvm: introduce manual dirty log reprotect") Reviewed-by: Peter Xu <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>
Li RongQing [Tue, 27 Jul 2021 11:12:47 +0000 (19:12 +0800)]
KVM: use cpu_relax when halt polling
SMT siblings share caches and other hardware, and busy halt polling
will degrade its sibling performance if its sibling is working
Sean Christopherson suggested as below:
"Rather than disallowing halt-polling entirely, on x86 it should be
sufficient to simply have the hardware thread yield to its sibling(s)
via PAUSE. It probably won't get back all performance, but I would
expect it to be close.
This compiles on all KVM architectures, and AFAICT the intended usage
of cpu_relax() is identical for all architectures."
KVM: SVM: use vmcb01 in svm_refresh_apicv_exec_ctrl
Currently when SVM is enabled in guest CPUID, AVIC is inhibited as soon
as the guest CPUID is set.
AVIC happens to be fully disabled on all vCPUs by the time any guest
entry starts (if after migration the entry can be nested).
The reason is that currently we disable avic right away on vCPU from which
the kvm_request_apicv_update was called and for this case, it happens to be
called on all vCPUs (by svm_vcpu_after_set_cpuid).
After we stop doing this, AVIC will end up being disabled only when
KVM_REQ_APICV_UPDATE is processed which is after we done switching to the
nested guest.
Fix this by just using vmcb01 in svm_refresh_apicv_exec_ctrl for avic
(which is a right thing to do anyway).
KVM: SVM: svm_set_vintr don't warn if AVIC is active but is about to be deactivated
It is possible for AVIC inhibit and AVIC active state to be mismatched.
Currently we disable AVIC right away on vCPU which started the AVIC inhibit
request thus this warning doesn't trigger but at least in theory,
if svm_set_vintr is called at the same time on multiple vCPUs,
the warning can happen.
commit bc9e9e672df9 ("KVM: debugfs: Reuse binary stats descriptors")
did replace the old definitions with the binary ones. While doing that
it missed that some files are names different than the counters. This
is especially important for kvm_stat which does have special handling
for counters named instruction_*.
Paolo Bonzini [Mon, 26 Jul 2021 16:39:01 +0000 (12:39 -0400)]
KVM: SVM: delay svm_vcpu_init_msrpm after svm->vmcb is initialized
Right now, svm_hv_vmcb_dirty_nested_enlightenments has an incorrect
dereference of vmcb->control.reserved_sw before the vmcb is checked
for being non-NULL. The compiler is usually sinking the dereference
after the check; instead of doing this ourselves in the source,
ensure that svm_hv_vmcb_dirty_nested_enlightenments is only called
with a non-NULL VMCB.
This test measures the performance effects of KVM's access tracking.
Access tracking is driven by the MMU notifiers test_young, clear_young,
and clear_flush_young. These notifiers do not have a direct userspace
API, however the clear_young notifier can be triggered by marking a
pages as idle in /sys/kernel/mm/page_idle/bitmap. This test leverages
that mechanism to enable access tracking on guest memory.
To measure performance this test runs a VM with a configurable number of
vCPUs that each touch every page in disjoint regions of memory.
Performance is measured in the time it takes all vCPUs to finish
touching their predefined region.
Populating memory : 1.337752570s
Writing to populated memory : 0.010177640s
Reading from populated memory : 0.009548239s
Mark memory idle : 23.973131748s
Writing to idle memory : 0.063584496s
Mark memory idle : 24.924652964s
Reading from idle memory : 0.062042814s
Breaking down the results:
* "Populating memory": The time it takes for all vCPUs to perform the
first write to every page in their region.
* "Writing to populated memory" / "Reading from populated memory": The
time it takes for all vCPUs to write and read to every page in their
region after it has been populated. This serves as a control for the
later results.
* "Mark memory idle": The time it takes for every vCPU to mark every
page in their region as idle through page_idle.
* "Writing to idle memory" / "Reading from idle memory": The time it
takes for all vCPUs to write and read to every page in their region
after it has been marked idle.
This test should be portable across architectures but it is only enabled
for x86_64 since that's all I have tested.
David Matlack [Tue, 13 Jul 2021 22:09:56 +0000 (22:09 +0000)]
KVM: selftests: Fix missing break in dirty_log_perf_test arg parsing
There is a missing break statement which causes a fallthrough to the
next statement where optarg will be null and a segmentation fault will
be generated.
KVM_MAX_VCPU_ID is the maximum vcpu-id of a guest, and not the number
of vcpu-ids. Fix array indexed by vcpu-id to have KVM_MAX_VCPU_ID+1
elements.
Note that this is currently no real problem, as KVM_MAX_VCPU_ID is
an odd number, resulting in always enough padding being available at
the end of those arrays.
Nevertheless this should be fixed in order to avoid rare problems in
case someone is using an even number for KVM_MAX_VCPU_ID.
Yufeng Mo [Tue, 27 Jul 2021 14:03:50 +0000 (22:03 +0800)]
net: hns3: change the method of obtaining default ptp cycle
The ptp cycle is related to the hardware, so it may cause compatibility
issues if a fixed value is used in driver. Therefore, the method of
obtaining this value is changed to read from the register rather than
use a fixed value in driver.
As a safeguard, if we're going to queue async work, do it from task_work
from the original task. This ensures that we can always sanely create
threads, regards of what the reissue context may be.
drm/amd/display: Only set default brightness for OLED
[Why]
We used to unconditionally set backlight path as AUX for panels capable
of backlight adjustment via DPCD in set default brightness.
[How]
This should be limited to OLED panel only since we control backlight via
PWM path for SDR mode in LCD HDR panel.
Reviewed-by: Krunoslav Kovac <[email protected]> Acked-by: Rodrigo Siqueira <[email protected]> Signed-off-by: Camille Cho <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
Changes some other code to only use aux for backlight control on
OLED panels. The commit message seems to indicate that PWM should
be used for SDR mode on HDR panels. Do something similar for
backlight control in general. This may need to be revisited if and
when HDR started to get used.
Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1438
Bug: https://bugzilla.kernel.org/show_bug.cgi?id=213715 Reviewed-by: Roman Li <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
drm/amd/display: ensure dentist display clock update finished in DCN20
[Why]
We don't check DENTIST_DISPCLK_CHG_DONE to ensure dentist
display clockis updated to target value. In some scenarios with large
display clock margin, it will deliver unfinished display clock and cause
issues like display black screen.
[How]
Checking DENTIST_DISPCLK_CHG_DONE to ensure display clock
has been update to target value before driver do other clock related
actions.
projects / linux.git / log