We can do the same cleanup on laundromat.
On invalidate_all_cached_dirs(), run laundromat worker with 0 timeout
and flush it for immediate + sync cleanup.
Signed-off-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
Remove redudant assignment of @rc as it will be overwritten by the
following cifs_file_flush() call.
Reported-by: Steve French <stfrench@microsoft.com>
Addresses-Coverity: 1665925
Fixes: 210627b0aca9 ("smb: client: fix missing timestamp updates with O_TRUNC")
Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.org>
Cc: linux-cifs@vger.kernel.org
Signed-off-by: Steve French <stfrench@microsoft.com>
AIO+DIO may extend the file size, hence we need to make sure ->i_size
is stable across the entire fallocate(2) operation, otherwise it would
become a truncate and then inode size reduced back down when it
finishes.
Fix this by calling netfs_wait_for_outstanding_io() right after
acquiring ->i_rwsem exclusively in cifs_fallocate() and then guarantee
a stable ->i_size across fallocate(2).
Also call netfs_wait_for_outstanding_io() after truncating pagecache
to avoid any potential races with writeback.
Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.org>
Reviewed-by: David Howells <dhowells@redhat.com>
Fixes: 210627b0aca9 ("smb: client: fix missing timestamp updates with O_TRUNC")
Cc: Frank Sorenson <sorenson@redhat.com>
Cc: linux-cifs@vger.kernel.org
Signed-off-by: Steve French <stfrench@microsoft.com>
Don't reuse open handle when changing timestamps to prevent the server
from disabling automatic timestamp updates as per MS-FSA 2.1.4.17.
---8<---
import os
import time
filename = '/mnt/foo'
def print_stat(prefix):
st = os.stat(filename)
print(prefix, ': ', time.ctime(st.st_atime), time.ctime(st.st_ctime))
fd = os.open(filename, os.O_CREAT|os.O_TRUNC|os.O_WRONLY, 0o644)
print_stat('old')
os.utime(fd, None)
time.sleep(2)
os.write(fd, b'foo')
os.close(fd)
time.sleep(2)
print_stat('new')
---8<---
Before patch:
$ mount.cifs //srv/share /mnt -o ...
$ python3 run.py
old : Fri Oct 3 14:01:21 2025 Fri Oct 3 14:01:21 2025
new : Fri Oct 3 14:01:21 2025 Fri Oct 3 14:01:21 2025
After patch:
$ mount.cifs //srv/share /mnt -o ...
$ python3 run.py
old : Fri Oct 3 17:03:34 2025 Fri Oct 3 17:03:34 2025
new : Fri Oct 3 17:03:36 2025 Fri Oct 3 17:03:36 2025
Fixes: b6f2a0f89d ("cifs: for compound requests, use open handle if possible")
Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.org>
Cc: Frank Sorenson <sorenson@redhat.com>
Reviewed-by: David Howells <dhowells@redhat.com>
Cc: linux-cifs@vger.kernel.org
Signed-off-by: Steve French <stfrench@microsoft.com>
Mask off ATTR_MTIME|ATTR_CTIME bits on ATTR_SIZE (e.g. ftruncate(2))
to prevent the client from sending set info calls and then disabling
automatic timestamp updates on server side as per MS-FSA 2.1.4.17.
---8<---
import os
import time
filename = '/mnt/foo'
def print_stat(prefix):
st = os.stat(filename)
print(prefix, ': ', time.ctime(st.st_atime), time.ctime(st.st_ctime))
fd = os.open(filename, os.O_CREAT|os.O_TRUNC|os.O_WRONLY, 0o644)
print_stat('old')
os.ftruncate(fd, 10)
time.sleep(2)
os.write(fd, b'foo')
os.close(fd)
time.sleep(2)
print_stat('new')
---8<---
Before patch:
$ mount.cifs //srv/share /mnt -o ...
$ python3 run.py
old : Fri Oct 3 13:47:03 2025 Fri Oct 3 13:47:03 2025
new : Fri Oct 3 13:47:00 2025 Fri Oct 3 13:47:03 2025
After patch:
$ mount.cifs //srv/share /mnt -o ...
$ python3 run.py
old : Fri Oct 3 13:48:39 2025 Fri Oct 3 13:48:39 2025
new : Fri Oct 3 13:48:41 2025 Fri Oct 3 13:48:41 2025
Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.org>
Cc: Frank Sorenson <sorenson@redhat.com>
Reviewed-by: David Howells <dhowells@redhat.com>
Cc: linux-cifs@vger.kernel.org
Signed-off-by: Steve French <stfrench@microsoft.com>
Don't call ->set_file_info() on open handle to prevent the server from
stopping [cm]time updates automatically as per MS-FSA 2.1.4.17.
Fix this by checking for ATTR_OPEN bit earlier in cifs_setattr() to
prevent ->set_file_info() from being called when opening a file with
O_TRUNC. Do the truncation in ->open() instead.
This also saves two roundtrips when opening a file with O_TRUNC and
there are currently no open handles to be reused.
Before patch:
$ mount.cifs //srv/share /mnt -o ...
$ cd /mnt
$ exec 3>foo; stat -c 'old: %z %y' foo; sleep 2; echo test >&3; exec 3>&-; sleep 2; stat -c 'new: %z %y' foo
old: 2025-10-03 13:26:23.151030500 -0300 2025-10-03 13:26:23.151030500 -0300
new: 2025-10-03 13:26:23.151030500 -0300 2025-10-03 13:26:23.151030500 -0300
After patch:
$ mount.cifs //srv/share /mnt -o ...
$ cd /mnt
$ exec 3>foo; stat -c 'old: %z %y' foo; sleep 2; echo test >&3; exec 3>&-; sleep 2; stat -c 'new: %z %y' foo
$ exec 3>foo; stat -c 'old: %z %y' foo; sleep 2; echo test >&3; exec 3>&-; sleep 2; stat -c 'new: %z %y' foo
old: 2025-10-03 13:28:13.911933800 -0300 2025-10-03 13:28:13.911933800 -0300
new: 2025-10-03 13:28:26.647492700 -0300 2025-10-03 13:28:26.647492700 -0300
Reported-by: Frank Sorenson <sorenson@redhat.com>
Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.org>
Reviewed-by: David Howells <dhowells@redhat.com>
Cc: linux-cifs@vger.kernel.org
Signed-off-by: Steve French <stfrench@microsoft.com>
The return value of copy_to_iter() function will never be negative,
it is the number of bytes copied, or zero if nothing was copied.
Update the check to treat 0 as an error, and return -1 in that case.
Fixes: d08089f649 ("cifs: Change the I/O paths to use an iterator rather than a page list")
Acked-by: Tom Talpey <tom@talpey.com>
Reviewed-by: David Howells <dhowells@redhat.com>
Signed-off-by: Fushuai Wang <wangfushuai@baidu.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
smb2_copychunk_range() used to send a single SRV_COPYCHUNK per
SRV_COPYCHUNK_COPY IOCTL.
Implement variable Chunks[] array in struct copychunk_ioctl and fill it
with struct copychunk (MS-SMB2 2.2.31.1.1), bounded by server-advertised
limits.
This reduces the number of IOCTL requests for large copies.
While we are at it, rename a couple variables to follow the terminology
used in the specification.
Signed-off-by: Henrique Carvalho <henrique.carvalho@suse.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Statements from an if branch and the end of this function implementation
were equivalent.
Thus delete duplicate source code.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Steve French <stfrench@microsoft.com>
For queue-depth I/O policy, this patch fixes unbalanced I/Os across
nvme multipaths.
Issue Description:
The RETRY disposition incorrectly increments ns->ctrl->nr_active
counter and reinitializes iostat start-time. In such cases nr_active
counter never goes back to zero until that path disconnects and
reconnects.
Such a path is not chosen for new I/Os if multiple RETRY cases on a given
a path cause its queue-depth counter to be artificially higher compared
to other paths. This leads to unbalanced I/Os across paths.
The patch skips incrementing nr_active if NVME_MPATH_CNT_ACTIVE is already
set. And it skips restarting io stats if NVME_MPATH_IO_STATS is already set.
base-commit: e989a3da2d371a4b6597ee8dee5c72e407b4db7a
Fixes: d4d957b53d ("nvme-multipath: support io stats on the mpath device")
Signed-off-by: Amit Chaudhary <achaudhary@purestorage.com>
Reviewed-by: Randy Jennings <randyj@purestorage.com>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Drawing from commit d2624d90a0 ("drm/panthor: assign unique names to
queues"), give scheduler queues proper names that reflect the function
of their JM slot, so that this will be shown when gathering DRM
scheduler tracepoints.
Signed-off-by: Adrián Larumbe <adrian.larumbe@collabora.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20251009114313.1374948-1-adrian.larumbe@collabora.com
The previous timeout of 500us seems to be too small; panning the map in
the Roll20 VTT in Firefox on a KDE/Wayland desktop reliably triggered
timeouts within a few seconds of usage, causing the monitor to freeze
and the following to be printed to dmesg:
[Jul30 13:44] xe 0000:03:00.0: [drm] *ERROR* GT0: Global invalidation timeout
[Jul30 13:48] xe 0000:03:00.0: [drm] *ERROR* [CRTC:82:pipe A] flip_done timed out
I haven't hit a single timeout since increasing it to 1000us even after
several multi-hour testing sessions.
Fixes: 0dd2dd0182 ("drm/xe: Move DSB l2 flush to a more sensible place")
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/5710
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: stable@vger.kernel.org
Cc: Maarten Lankhorst <dev@lankhorst.se>
Reviewed-by: Shuicheng Lin <shuicheng.lin@intel.com>
Link: https://lore.kernel.org/r/20250912223254.147940-1-kenneth@whitecape.org
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Some functions in drm multiply hdisplay and vdisplay with a third
factor, which can result in a sign extension according to static
analysis due to an implicit s32 promotion. Use a cast to u32 to
prevent this.
Suggested-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Krzystof Karas <krzysztof.karas@intel.com>
Signed-off-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Andi Shyti <andi.shyti@intel.com>
Link: https://lore.kernel.org/r/20251007153645.90920-2-jonathan.cavitt@intel.com
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
drm_crtc_accurate_vblank_count takes a spinlock, which we should avoid
in tracepoints and debug functions.
This also prevents taking the spinlock 2x during the critical
section of pipe updates for DSI updates.
Acked-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250829131701.15607-2-dev@lankhorst.se
Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
Luca Abeni reported this:
| BUG: scheduling while atomic: kworker/u8:2/15203/0x00000003
| CPU: 1 PID: 15203 Comm: kworker/u8:2 Not tainted 4.19.1-rt3 #10
| Call Trace:
| rt_spin_lock+0x3f/0x50
| gen6_read32+0x45/0x1d0 [i915]
| g4x_get_vblank_counter+0x36/0x40 [i915]
| trace_event_raw_event_i915_pipe_update_start+0x7d/0xf0 [i915]
The tracing events use trace_intel_pipe_update_start() among other events
use functions acquire spinlock_t locks which are transformed into
sleeping locks on PREEMPT_RT. A few trace points use
intel_get_crtc_scanline(), others use ->get_vblank_counter() wich also
might acquire a sleeping locks on PREEMPT_RT.
At the time the arguments are evaluated within trace point, preemption
is disabled and so the locks must not be acquired on PREEMPT_RT.
Based on this I don't see any other way than disable trace points on
PREMPT_RT.
[mlankhorst]
The original patch was insufficient, and since the tracing
infrastructure does not allow for partial disabling of tracepoints.
Completely disable tracing for the entire i915 driver in PREEMPT_RT,
a separate fix for display tracepoints on xe is added to make those
work.
Cc: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Reported-by: Luca Abeni <lucabe72@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Co-developed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Link: https://lore.kernel.org/r/20250828090944.101069-1-dev@lankhorst.se
Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
The regulator is optional, skip the setup instead of returning an
error if it is not present
Signed-off-by: Rain Yang <jiyu.yang@nxp.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250928090334.35389-2-jiyu.yang@oss.nxp.com
Not all platforms require the mali-supply regulator. This change removes
it from the required list in the binding schema, and make mali-supply
required for rk3588 only.
Signed-off-by: Rain Yang <jiyu.yang@nxp.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250928090334.35389-1-jiyu.yang@oss.nxp.com
Currently the Panthor driver needs the GPU to be powered down
between suspend and resume. If this is not done, then the
MCU_CONTROL register will be preserved as AUTO, which again will
cause a premature FW boot on resume. The FW will go directly into
fatal state in this case.
This case needs to be handled as there is no guarantee that the
GPU will be powered down after the suspend callback on all platforms.
The fix is to call panthor_fw_stop() in "pre-reset" path to ensure
the MCU_CONTROL register is cleared (set DISABLE). This matches
well with the already existing call to panthor_fw_start() from the
"post-reset" path.
Signed-off-by: Ketil Johnsen <ketil.johnsen@arm.com>
Acked-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Fixes: 2718d91816 ("drm/panthor: Add the FW logical block")
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20251008105112.4077015-1-ketil.johnsen@arm.com
Convert the Devicetree binding documentation for hisilicon,hix5hd2-i2c
from plain text to DT binding schema.
Signed-off-by: Kael D'Alcamo <dev@kael-k.io>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Some VF2GUC actions may take longer to process. Increase default timeout
after received BUSY indication to 2sec to cover all worst case scenarios.
Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-35-matthew.brost@intel.com
Rebase the CCS save/restore BB's GGTT addresses during VF post-migration
recovery by setting the software ring tail to zero, the LRC ring head to
zero, and rewriting the jump-to-BB instructions.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-34-matthew.brost@intel.com
It is possible that the media GT's VF post-migration recovery work item
gets scheduled before the primary GT's work item. Since the media GT
depends on the primary GT's work item to complete CCS restore, if the
media GT's work item is scheduled first, detect this condition and
re-queue the media GT's work item for a later time.
v5:
- Adjust debug message (Tomasz)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-33-matthew.brost@intel.com
VF CCS restore is a primary GT operation on which the media GT depends.
Therefore, it doesn't make much sense to run these operations in
parallel. To address this, point the media GT's ordered work queue to
the primary GT's ordered work queue on platforms that require (PTL VFs)
CCS restore as part of VF post-migration recovery.
v7:
- Remove bool from xe_gt_alloc (Lucas)
v9:
- Fix typo (Lucas)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-32-matthew.brost@intel.com
The migrate VM builds the CCS metadata save/restore batch buffer (BB) in
advance and retains it so the GuC can submit it directly when saving a
VM’s state.
When a VM migrates between VFs, the GGTT base can change. Any GGTT-based
addresses embedded in the BB would then have to be parsed and patched.
Use PPGTT addresses in the BB (including for TLB invalidation) so the BB
remains GGTT-agnostic and requires no address fixups during migration.
Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-31-matthew.brost@intel.com
A race condition exists where a paused VF's H2G request can be processed
and subsequently rejected. This rejection results in a FAST_REQ failure
being delivered to the KMD, which then terminates the CT via a dead
worker and triggers a GT reset—an undesirable outcome.
This workaround mitigates the issue by checking if a VF post-migration
recovery is in progress and aborting these adverse actions accordingly.
The GuC firmware will address this bug in an upcoming release. Once that
version is available and VF migration depends on it, this workaround can
be safely removed.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-30-matthew.brost@intel.com
Helpful to manually verify the GuC state machine can correctly replay
the state during a VF post-migration recovery. All replay paths have
been manually verified as triggered and working during testing.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-29-matthew.brost@intel.com
A queue must be in the submission backend's tracking state before the
LRC is created to avoid a race condition where the LRC's GGTT addresses
are not properly fixed up during VF post-migration recovery.
Move the queue initialization—which adds the queue to the submission
backend's tracking state—before LRC creation.
Also wait on pending GGTT fixups before allocating LRCs to avoid racing
with fixups.
v2:
- Wait on VF GGTT fixes before creating LRC (testing)
v5:
- Adjust comment in code (Tomasz)
- Reduce race window
v7:
- Only wakeup waiters in recovery path (CI)
- Wakeup waiters on abort
- Use GT warn on (Michal)
- Fix kernel doc for LRC ring size function (Tomasz)
v8:
- Guard against migration not supported or no memirq (CI)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-28-matthew.brost@intel.com
Fixup GuC submission pause / unpause functions to properly replay any
possible state lost during VF post migration recovery.
v3:
- Add helpers for revert / replay (Tomasz)
- Add comment around WQ NOPs (Tomasz)
v7:
- Only fixup / replay parallel queues once (Testing)
- Skip unpause step on queues created after resfix done (Testing)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-27-matthew.brost@intel.com
If VF post-migration recovery fails, the device is wedged. However,
submission queues still need to be enabled for proper cleanup. In such
cases, call into the GuC submission backend to restart all queues that
were previously paused.
v3:
- s/Avort/Abort (Tomasz)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-26-matthew.brost@intel.com
Before RESFIX_DONE, all CTs stuck in the H2G queue need to be squashed,
as they may contain actions which contain invalid GGTT references or are
unnecessary after HW change.
Starting the CTs clears all H2Gs in the queue. Any lost H2Gs are
resubmitted by the GuC submission state machine.
v3:
- Don't mess with head / tail values (Michal)
v4:
- Don't mess with broke (Michal)
- Add CTB_H2G_BUFFER_OFFSET (Michal)
v5:
- Adjust commit message (Tomasz)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-25-matthew.brost@intel.com
GuC needs to be live for the GuC submission state machine to resubmit
anything lost during VF post-migration recovery. Therefore, move the
kickstart step after `resfix` to ensure proper resubmission.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-23-matthew.brost@intel.com
TLB invalidations requests can be lost during VF post-migration
recovery. Since the VF has migrated, these invalidations are no longer
needed.
Reset the TLB invalidation frontend, which will signal all pending
fences.
v3:
- Move TLB invalidation reset after pausing submission (Tomasz)
- Adjust commit message (Michal)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-22-matthew.brost@intel.com
Flushing CTs (i.e., progressing all pending G2H messages) gives VF
post-migration recovery an accurate view of which H2G messages the GuC
has processed, enabling the GuC submission state machine to correctly
rebuild all state.
Also, stop all CT traffic, as the CT is not live during VF
post-migration recovery.
v3:
- xe_guc_ct_flush_and_stop rename (Michal)
- Drop extra GuC CT WQ wake up (Michal)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-21-matthew.brost@intel.com
The only case where the GuC submission backend cannot reason 100%
correctly is when a GuC context is registered during VF post-migration
recovery. In this scenario, it's possible that the GuC context register
H2G is processed, but the immediately following schedule-enable H2G gets
lost. The schedule-enable G2H "done" response is how the GuC state machine
determines whether context registration has completed.
A double register is harmless when using `GUC_HXG_TYPE_EVENT`, as GuC
simply drops the duplicate H2G. To keep things simple, use
`GUC_HXG_TYPE_EVENT` for all context registrations on VFs.
v5:
- Check for xe_sriov_vf_migration_supported (Tomasz)
v7:
- Add comment about subsequent protocol failures (Tomasz)
- Modify commit message (Michal)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-20-matthew.brost@intel.com
Blocking in work queues on a hardware action that may never occur —
especially when it depends on a software fixup also scheduled on the
a work queue — is a recipe for deadlock. This situation arises with
the preempt rebind worker and VF post-migration recovery. To prevent
potential deadlocks, avoid indefinite blocking in the preempt rebind
worker for VFs that support migration.
v4:
- Use dma_fence_wait_timeout (CI)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-19-matthew.brost@intel.com
If VF post-migration recovery is in progress, the recovery flow will
rebuild all GuC submission state. In this case, exit all waiters to
ensure that submission queue scheduling can also be paused. Avoid taking
any adverse actions after aborting the wait.
As part of waking up the GuC backend, suspend_wait can now return
-EAGAIN indicating the waiter should be retried. If the caller is
running on work item, that work item need to be requeued to avoid a
deadlock for the work item blocking the VF migration recovery work item.
v3:
- Don't block in preempt fence work queue as this can interfere with VF
post-migration work queue scheduling leading to deadlock (Testing)
- Use xe_gt_recovery_inprogress (Michal)
v5:
- Use static function for vf_recovery (Michal)
- Add helper to wake CT waiters (Michal)
- Move some code to following patch (Michal)
- Adjust commit message to explain suspend_wait returning -EAGAIN (Michal)
- Add kernel doc to suspend_wait around returning -EAGAIN
v7:
- Add comment on why a shared wait queue is need on VFs (Michal)
- Guard again suspend_wait signaling early on resfix donw (Tomasz)
v8:
- Fix kernel doc (CI)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-18-matthew.brost@intel.com
With well-behaved software, a GT reset should never occur, nor should it
happen during VF post-migration recovery. If it does, trigger a warning
but suppress the GT reset, as VF post-migration recovery is expected to
bring the VF back to a working state.
v3:
- Better commit message (Tomasz)
v5:
- Use xe_gt_WARN_ON (Michal)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-17-matthew.brost@intel.com
Be cautious and ensure the VF post-migration worker is not running
during driver unload.
v3:
- More teardown later in driver init, use devm (Tomasz)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-16-matthew.brost@intel.com
As multi-GT VF post-migration recovery can run in parallel on different
workqueues, but both GTs point to the same GGTT, only one GT needs to
shift the GGTT. However, both GTs need to know when this step has
completed. To coordinate this, perform the GGTT shift under the GGTT
lock. With shift being done under the lock, storing the shift value
becomes unnecessary.
In addition to above, move the GGTT VF config from the GT to the tile.
v3:
- Update commmit message (Tomasz)
v4:
- Move GGTT values to tile state (Michal)
- Use GGTT lock (Michal)
v5:
- Only take GGTT lock during recovery (CI)
- Drop goto in vf_get_submission_cfg (Michal)
- Add kernel doc around recovery in xe_gt_sriov_vf_query_config (Michal)
v7:
- Drop recovery variable (Michal)
- Use _locked naming (Michal)
- Use guard (Michal)
v9:
- Break LMEM changes into different patch (Michal)
- Fix layering (Michal)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-15-matthew.brost@intel.com
The LMEM VF provision is tile-layer-specific information. Move the LMEM
configuration to the tile layer accordingly.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-14-matthew.brost@intel.com
The GGTT lock is needed very early during GT initialization for a VF;
move the GGTT lock initialization to the allocation phase.
v8:
- Rework function structure (Michal)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-13-matthew.brost@intel.com
VF post migration recovery is the path of dma-fence signaling / reclaim,
avoid memory allocations in this path.
v3:
- s/lrc_wa_bb/scratch (Tomasz)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-12-matthew.brost@intel.com
While VF post-migration recovery is in progress, abort H2G sends with
-ECANCEL. These messages are treated as lost, and TLB invalidation
errors are suppressed. During this phase, the H2G channel is down, and
VF recovery requires the CT lock to proceed.
v3:
- Use xe_gt_recovery_inprogress (Michal)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-11-matthew.brost@intel.com
VF recovery is a per-GT operation, so it makes sense to isolate it to a
per-GT queue. Scheduling this operation on the same worker as the GT
reset and TDR not only aligns with this design but also helps avoid race
conditions, as those operations can also modify the queue state.
v2:
- Fix lockdep splat (Adam)
- Use xe_sriov_vf_migration_supported helper
v3:
- Drop xe_gt_sriov_ prefix for private functions (Michal)
- Drop message in xe_gt_sriov_vf_migration_init_early (Michal)
- Logic rework in vf_post_migration_notify_resfix_done (Michal)
- Rework init sequence layering (Michal)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-10-matthew.brost@intel.com
Add xe_gt_recovery_pending helper.
This helper serves as the singular point to determine whether a GT
recovery is currently in progress. Expected callers include the GuC CT
layer and the GuC submission layer. Atomically visable as soon as vCPU
are unhalted until VF recovery completes.
v3:
- Add GT layer xe_gt_recovery_inprogress (Michal)
- Don't blow up in memirq not enabled (CI)
- Add __memirq_received with clear argument (Michal)
- xe_memirq_sw_int_0_irq_pending rename (Michal)
- Use offset in xe_memirq_sw_int_0_irq_pending (Michal)
v4:
- Refactor xe_gt_recovery_inprogress logic around memirq (Michal)
v5:
- s/inprogress/pending (Michal)
v7:
- Fix typos, adjust comment (Michal)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-9-matthew.brost@intel.com
The LRC W/A currently checks for LRC being iomem in some places, while
in others it checks if the scratch buffer is non-NULL. This
inconsistency causes issues with the VF post-migration recovery code,
which blindly passes in a scratch buffer.
This patch standardizes the check by consistently verifying whether the
LRC is iomem to determine if the scratch buffer should be used.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-8-matthew.brost@intel.com
Now that we save the job's head during submission, it's no longer
necessary to adjust the LRC ring head during resubmission. Instead, a
software-based adjustment of the tail will overwrite the old jobs in
place. For some odd reason, adjusting the LRC ring head didn't work on
parallel queues, which was causing issues in our CI.
v5:
- Add comment in guc_exec_queue_start explaning why the function works
(Auld)
v7:
- Only adjust first state on first unsignaled job (Auld)
v8:
- Break unsignaled job handling to separate patch (Auld)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-7-matthew.brost@intel.com
In all cases where the first pending job helper is called, we only want
to retrieve the first unsignaled pending job, as this helper is used
exclusively in recovery flows. It is possible for signaled jobs to
remain in the pending list as the scheduler is stopped, so those should
be skipped.
Also, add kernel documentation to clarify this behavior.
v8:
- Split out into own patch (Auld)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-6-matthew.brost@intel.com
VF migration requires jobs to remain pending so they can be replayed
after the VF comes back. Previously, LR job fences were intentionally
signaled immediately after submission to avoid the risk of exporting
them, as these fences do not naturally signal in a timely manner and
could break dma-fence contracts. A side effect of this approach was that
LR jobs were never added to the DRM scheduler’s pending list, preventing
them from being tracked for later resubmission.
We now avoid signaling LR job fences and ensure they are never exported;
Xe already guards against exporting these internal fences. With that
guarantee in place, we can safely track LR jobs in the scheduler’s
pending list so they are eligible for resubmission during VF
post-migration recovery (and similar recovery paths).
An added benefit is that LR queues now gain the DRM scheduler’s built-in
flow control over ring usage rather than rejecting new jobs in the exec
IOCTL if the ring is full.
v2:
- Ensure DRM scheduler TDR doesn't run for LR jobs
- Stack variable for killed_or_banned_or_wedged
v4:
- Clarify commit message (Tomasz)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
Link: https://lore.kernel.org/r/20251008214532.3442967-5-matthew.brost@intel.com