An xe file can outlive the associated process as the GPU cleanup is just
triggered upon file close (process kill) and completes sometime later.
If the file close triggers error conditions (GPU hangs) the process
cannot be safely referenced to retrieve the name and pid for debug
information. Store the process name and pid directly in the xe file to
be safe.
v2:
- Access file->pid via rcu_access_pointer (Matthew Auld)
Fixes: b10d0c5e9d ("drm/xe: Add process name to devcoredump")
Fixes: f6ca930d97 ("drm/xe: Add process name and PID to job timedout message")
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240723151045.1725417-1-matthew.brost@intel.com
Wedge the entire device, not just GT which may have triggered the wedge.
To implement this, cleanup the layering so xe_device_declare_wedged()
calls into the lower layers (GT) to ensure entire device is wedged.
While we are here, also signal any pending GT TLB invalidations upon
wedging device.
Lastly, short circuit reset wait if device is wedged.
v2:
- Short circuit reset wait if device is wedged (Local testing)
Fixes: 8ed9aaae39 ("drm/xe: Force wedged state and block GT reset upon any GPU hang")
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240716063902.1390130-1-matthew.brost@intel.com
In Xe, the perf layer allows capture of HW counter streams. These HW
counters are generally performance related but don't have to be necessarily
so. Also, the name "perf" is a carryover from i915 and is not preferred.
Here we propose the name "observation" for this common layer which allows
capture of different types of these counter streams.
v2: Rename observability layer to observation layer (Lucas/Rodrigo)
v3: Rename sysctl file to "observation_paranoid" (Jose)
Fixes: 52c2e956dc ("drm/xe/perf/uapi: "Perf" layer to support multiple perf counter stream types")
Fixes: fe8929bdf8 ("drm/xe/perf/uapi: Add perf_stream_paranoid sysctl")
Acked-by: Lucas De Marchi <lucas.demarchi@intel.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Acked-by: José Roberto de Souza <jose.souza@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240703164801.2561423-1-ashutosh.dixit@intel.com
The cleanup is done by devres in irq_uninstall.
Commit bbc9651fe9 ("drm/xe/irq: move irq_uninstall over to devm")
resolved the ordering issue where irq_uninstall (registered with drmm)
was called after pci_free_irq_vectors (registered with devm upon calling
pci_alloc_irq_vectors). This happened because drmm action list is
registered with devm very early in the init flow - before
pci_alloc_irq_vectors.
Now that irq_uninstall is registered with devm, it will be called before
pci_free_irq_vectors and we can remove xe_irq_shutdown.
Signed-off-by: Ilia Levi <illevi@habana.ai>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240606124705.822451-1-illevi@habana.ai
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
This WA requires us to limit media GT frequency requests to a certain
cap value during driver load. Freq limits are restored after load
completes, so perf will not be affected during normal operations.
During normal driver operation, this WA requires dummy writes to media
offset 0x380D8C after every ~63 GGTT writes. This will ensure completion
of the LMEM writes originating from Gunit.
During driver unload(before FLR), the WA requires that we set requested
frequency to the cap value again.
v3: Do not use WA number in function name. Call WA wrapper from xe_device.
Rename some variables, check for locks in the correct function (Rodrigo).
Ensure reset path is also covered for this WA.
v4: Fix BAT failure
v5: Add a function pointer for ggtt_ops (Michal W)
v6: Fix name collision and use static function (Rodrigo)
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240620224928.3986377-2-vinay.belgaumkar@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Need to sync some header include that propagated through
drm-intel-next.
v2: After some changes in drm/drm-next
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Introduce add/remove config perf ops for OA. OA configurations consist of a
set of event/counter select register address/value pairs. The add_config
perf op validates and stores such configurations and also exposes them in
the metrics sysfs. These configurations will be programmed to OA unit HW
when an OA stream using a configuration is opened. The OA stream can also
switch to other stored configurations.
v2: Start config id's from 1 and other minor review comments (Umesh)
v3: Add 32 bit build
v4: Add kernel doc for non-static functions (Michal)
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Acked-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240618014609.3233427-6-ashutosh.dixit@intel.com
In Xe, the plan is to support multiple types of perf counter streams (OA is
only one type of these streams). Rather than introduce NxM ioctls for
these (N perf streams with M ioctl's per perf stream), we decide to
multiplex these (N different stream types and the M ops for each of these
stream types) through a single PERF ioctl. This multiplexing is the purpose
of the PERF layer.
In addition to PERF DRM ioctl's, another set of ioctl's on the PERF fd are
defined. These are expected to be common to different PERF stream types and
therefore defined at the PERF layer itself.
v2: Add param_size to 'struct drm_xe_perf_param' (Umesh)
v3: Rename 'enum drm_xe_perf_ops' to
'enum drm_xe_perf_ioctls' (Guy Zadicario)
Add DRM_ prefix to ioctl names to indicate uapi names
v4: Add 'enum drm_xe_perf_op' previously missed out (Guy Zadicario)
v5: Squash the ops and PERF layer patches into a single patch (Umesh)
Remove param_size from struct 'drm_xe_perf_param' (Umesh)
v6: Add DRM_XE_PERF_IOCTL_STATUS
v7: Add DRM_XE_PERF_IOCTL_INFO
v8: Fix Copyright years, fix DRM_XE_PERF_TYPE_MAX, move '#include
"xe_perf.h"' to xe_perf.c, add kernel doc (Michal)
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Acked-by: Guy Zadicario <gzadicario@habana.ai>
Acked-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240618014609.3233427-2-ashutosh.dixit@intel.com
UAPI Changes:
- Expose the L3 bank mask (Francois)
Cross-subsystem Changes:
- Update Xe driver maintainers (Oded)
Display (i915):
- Add missing include to intel_vga.c (Michal Wajdeczko)
Driver Changes:
- Fix Display (xe-only) detection for ADL-N (Lucas)
- Runtime PM fixes that enabled PC-10 and D3Cold (Francois, Rodrigo)
- Fix unexpected silent drm backmerge issues (Thomas)
- More (a lot more) preparation for SR-IOV support (Michal Wajdeczko)
- Devcoredump fixes and improvements (Jose, Tejas, Matt Brost)
- Introduce device 'wedged' state (Rodrigo)
- Improve debug and info messages (Michal Wajdeczko, Rodrigo, Nirmoy)
- Adding or fixing workarounds (Tejas, Shekhar, Lucas, Bommu)
- Check result of drmm_mutex_init (Michal Wajdeczko)
- Enlarge the critical dma fence area for preempt fences (Matt Auld)
- Prevent UAF in VM's rebind work (Matt Auld)
- GuC submit related clean-ups and fixes (Matt Brost, Himal, Jonathan, Niranjana)
- Prefer local helpers to perform dma reservation locking (Himal)
- Spelling and typo fixes (Colin, Francois)
- Prep patches for 1 job per VM bind IOCTL (no uapi change yet) (Matt Brost)
- Remove uninitialized end var from xe_gt_tlb_invalidation_range (Nirmoy)
- GSC related changes targeting LNL support (Daniele)
- Fix assert in L3 bank mask generation (Francois)
- Perform dma_map when moving system buffer objects to TT (Thomas)
- Add helpers for manipulating macro arguments (Michal Wajdeczko)
- Refactor default device atomic settings (Nirmoy)
- Add debugfs node to dump mocs (Janga)
- Use ordered WQ for G2H handler (Matt Brost)
- Clean up and fixes in header includes (Michal Wajdeczko)
- Prefer flexible-array over deprecated zero-lenght ones (Lucas)
- Add Indirect Ring State support (Niranjana)
- Fix UBSAN shift-out-of-bounds failure (Shuicheng)
- HWMon fixes and additions (Karthik)
- Clean-up refactor around probe init functions (Lucas, Michal Wajdeczko)
- Fix PCODE init function (Himal)
- Only use reserved BCS instances for usm migrate exec queue (Matt Brost)
- Only zap PTEs as needed (Matt Brost)
- Per client usage info (Lucas)
- Core hotunplug improvements converting stuff towards devm (Matt Auld)
- Don't emit false error if running in execlist mode (Michal Wajdeczko)
- Remove unused struct (Dr. David)
- Support/debug for slow GuC loads (John Harrison)
- Decouple job seqno and lrc seqno (Matt Brost)
- Allow migrate vm gpu submissions from reclaim context (Thomas)
- Rename drm-client running time to run_ticks and fix a UAF (Umesh)
- Check empty pinned BO list with lock held (Nirmoy)
- Drop undesired prefix from the platform name (Michal Wajdeczko)
- Remove unwanted mutex locking on xe file close (Niranjana)
- Replace format-less snprintf() with strscpy() (Arnd)
- Other general clean-ups on registers definitions and function names (Michal Wajdeczko)
- Add kernel-doc to some xe_lrc interfaces (Niranajana)
- Use missing lock in relay_needs_worker (Nirmoy)
- Drop redundant W=1 warnings from Makefile (Jani)
- Simplify if condition in preempt fences code (Thorsten)
- Flush engine buffers before signalling user fence on all engines (Andrzej)
- Don't overmap identity VRAM mapping (Matt Brost)
- Do not dereference NULL job->fence in trace points (Matt Brost)
- Add synchronous gt reset debugfs (Jonathan)
- Xe gt_idle fixes (Riana)
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/ZmItmuf7vq_xvRjJ@intel.com
This reverts commit cd506a33b0.
The gt_remove function was explicitly added as part of the remove flow
instead of using drmm/devm automatic cleanup due to it being illegal
to remove a component after the driver has been detached from the pci
device; the GSC proxy component is removed as part of gt_remove, so we
need to do it in the pci cleanup flow. The function already has a
comment above it to explain this.
Note that the change to use the devm also caused an invalid pointer
deref in the gsc_proxy unbind function, but I didn't bother to debug
which pointer was bad since we shouldn't be calling the unbind that
late anyway and this revert fixes it.
Both issue were not seen in CI because the GSC loading is temporarily
disabled due to a critical bug, which means we're not binding the
component.
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Andrzej Hajda <andrzej.hajda@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240528182354.1200424-1-daniele.ceraolospurio@intel.com
This is quite broken since we are nuking the pdev link to the private
driver struct, but note here that driver_release is called when the
drm_device is released (poor mans drmm), which can be long after the
device has been removed. So here what we are actually doing is nuking
the pdev link for what is potentially bound to a different drm_device.
If that happens before our pci remove callback is triggered (for the new
drm_device) we silently exit and skip some important cleanup steps,
resulting in hilarity.
There should be no reason to implement driver_release, when we already
have nicer stuff like drmm, so just remove completely. The actual pdev
link is already nuked when removing the device.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Andrzej Hajda <andrzej.hajda@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240522102143.128069-19-matthew.auld@intel.com
Unregister all in-kernel clients before unloading the i915 driver. For
other drivers, drm_dev_unregister() does this automatically. As i915 and
xe do not use this helper, they have to perform the call by themselves.
Note that there are currently no in-kernel clients in i915 or xe. The
patch prepares the drivers for a related update of their fbdev support.
v8:
- unregister clients in intel_display_driver_unregister() (Jani)
- mention xe in commit message (Rodrigo, Jani)
v7:
- update xe driver
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Acked-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240409081029.17843-5-tzimmermann@suse.de
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
In many validation situations when debugging GPU Hangs,
it is useful to preserve the GT situation from the moment
that the timeout occurred.
This patch introduces a module parameter that could be used
on situations like this.
If xe.wedged module parameter is set to 2, Xe will be declared
wedged on every single execution timeout (a.k.a. GPU hang) right
after devcoredump snapshot capture and without attempting any
kind of GT reset and blocking entirely any kind of execution.
v2: Really block gt_reset from guc side. (Lucas)
s/wedged/busted (Lucas)
v3: - s/busted/wedged
- Really use global_flags (Dafna)
- More robust timeout handling when wedging it.
v4: A really robust clean exit done by Matt Brost.
No more kernel warns on unbind.
v5: Simplify error message (Lucas)
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Dafna Hirschfeld <dhirschfeld@habana.ai>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Cc: Himanshu Somaiya <himanshu.somaiya@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240423221817.1285081-3-rodrigo.vivi@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Introduce a very simple 'wedged' state where any attempt
to access the GPU is entirely blocked.
On some critical cases, like on gt_reset failure, we need to
block any other attempt to use the GPU. Otherwise we are at
a risk of reaching cases that would force us to reboot the machine.
So, when this cases are identified we corner and block any GPU
access. No IOCTL and not even another GT reset should be attempted.
The 'wedged' state in Xe is an end state with no way back.
Only a device "re-probe" (unbind + bind) can restore the GPU access.
v2: - s/wedged/busted (Lucas)
- use unbind+bind instead of module reload (Lucas)
- added more info on unbind operations and instruction on bug report
- only print the message once.
v3: - s/busted/wedged (Ashutosh, Tvrtko, Thomas)
- don't assume user has sudo and tee available (Lucas)
v4: - remove unnecessary cases around ct communication or migration.
Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Cc: Tvrtko Ursulin <tursulin@ursulin.net>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Anshuman Gupta <anshuman.gupta@intel.com>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> #v2
Link: https://patchwork.freedesktop.org/patch/msgid/20240423221817.1285081-1-rodrigo.vivi@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Let's simply convert all the current callers towards direct
xe_pm_runtime access and remove this extra layer of indirection.
No functional change is expected with this patch since
xe_mem_access_get was already using the xe_pm_runtime_get_noresume
at this point.
v2: Convert all the current callers instead of a big refactor
at once.
v3: - Rebased
- Squashed the GSC/HDCP
- Added a new case: sriov_pf_policy
- Improved commit message to highlight that
there's no functional change in this patch.
Reviewed-by: Matthew Auld <matthew.auld@intel.com> #v2
Cc: Suraj Kandpal <suraj.kandpal@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240418143049.43231-1-rodrigo.vivi@intel.com
The mem_access itself is not holding any lock, but attempting
to train lockdep with possible scarring locks happening during
runtime pm. We are going soon to kill the mem_access get and put
helpers in favor of direct xe_pm_runtime calls, so let's just
move this lock around to where it now belongs.
v2: s/lockdep_training/lockdep_prime (Matt Auld)
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240417203952.25503-4-rodrigo.vivi@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Lmem init check should be done only after pcode initialization
status is complete. Move lmem init check after pcode status
check. Also wait for a short while after pcode status check
to allow completion of the task.
Failing to do so, can lead to aborting the module load
leaving the system unusable. Wait until the lmem initialization
is complete within a timeout (60s) or till the user aborts.
v2: use bool as return type
re-order the code comment (Rodrigo)
add comment for deferring probe (Himal)
v3: rebase
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240410085005.1126343-3-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
