bo->size is redundant because the base GEM object already has a size
field with the same value. Drop bo->size and use the base GEM object’s
size instead. While at it, introduce xe_bo_size() to abstract the BO
size.
v2:
- Fix typo in kernel doc (Ashutosh)
- Fix kunit (CI)
- Fix line wrap (Checkpatch)
v3:
- Fix sriov build (CI)
v4:
- Fix display build (CI)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://lore.kernel.org/r/20250625144128.2827577-1-matthew.brost@intel.com
Move the force_wake_get to the beginning of the function so the gpu
timestamp can be closer to the sampled value for exec queues. This
avoids additional delays waiting for force wake ack which can make the
proportion between cycles/total_cycles fluctuate around the real value.
For a gputop-like application getting 2 samples to calculate the
utilization:
sample 0:
read_exec_queue_timestamp
<<<< (A)
read_gpu_timestamp
sample 1:
read_exec_queue_timestamp
<<<<< (B)
read_gpu_timestamp
In the above case, utilization can be bigger or smaller than it should
be, depending on if (A) or (B) receives additional delay, respectively.
With this a LNL system that was failing on
`xe_drm_fdinfo --r utilization-single-full-load` after ~60 iterations,
get to run to 100 without a failure. This is still not perfect, and it's
easy to introduce errors by just loading the CPU with `stress --cpu
$(nproc)` - the same igt test in this case fails after 2 or 3
iterations. That will be dealt with in the test itself, using a longer
sampling period.
v2: Rename function and add another to get "any engine", preparing for
caching the hwe in future (Umesh / Jonathan)
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20241108053318.3483678-3-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
When an exec queue is killed it triggers an async process of asking the
GuC to schedule the context out. The timestamp in the context image is
only updated when this process completes. In case a userspace process
kills an exec and tries to read the timestamp, it may not get an updated
runtime.
Add synchronization between the process reading the fdinfo and the exec
queue being killed. After reading all the timestamps, wait on exec
queues in the process of being killed. When that wait is over,
xe_exec_queue_fini() was already called and updated the timestamps.
v2: Do not update pending_removal before validating user args
(Matthew Auld)
v3: Move wait on pending to be done before getting any timestamp
so it's more likely for the gpu and exec queue timestamps to
be closer together
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/2667
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20241108053318.3483678-2-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
xe_force_wake_get() now returns the reference count-incremented domain
mask. If it fails for individual domains, the return value will always
be 0. However, for XE_FORCEWAKE_ALL, it may return a non-zero value even
in the event of failure. Use helper xe_force_wake_ref_has_domain to
verify all domains are initialized or not. Update the return handling of
xe_force_wake_get() to reflect this behavior, and ensure that the return
value is passed as input to xe_force_wake_put().
v3
- return xe_wakeref_t instead of int in xe_force_wake_get()
- xe_force_wake_put() error doesn't need to be checked. It internally
WARNS on domain ack failure.
v5
- return unsigned int from xe_force_wake_get()
v6
- use xe_force_wake_ref_has_domain()
v7
- Fix commit message
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Badal Nilawar <badal.nilawar@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20241014075601.2324382-15-himal.prasad.ghimiray@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Both the VM / exec queue file lock protect the lookup and reference to
the object, nothing more. These locks are not intended anything else
underneath them. XA have their own locking too, so no need to take the
VM / exec queue file lock aside from when doing a lookup and reference
get.
Add some kernel doc to make this clear and cleanup a few typos too.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240921011712.2681510-1-matthew.brost@intel.com
xe_file_close triggers an asynchronous queue cleanup and then frees up
the xef object. Since queue cleanup flushes all pending jobs and the KMD
stores client usage stats into the xef object after jobs are flushed, we
see a use-after-free for the xef object. Resolve this by taking a
reference to xef from xe_exec_queue.
While at it, revert an earlier change that contained a partial work
around for this issue.
v2:
- Take a ref to xef even for the VM bind queue (Matt)
- Squash patches relevant to that fix and work around (Lucas)
v3: Fix typo (Lucas)
Fixes: ce62827bc2 ("drm/xe: Do not access xe file when updating exec queue run_ticks")
Fixes: 6109f24f87 ("drm/xe: Add helper to accumulate exec queue runtime")
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1908
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240718210548.3580382-5-umesh.nerlige.ramappa@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Per client engine utilization uses RING_TIMESTAMP to return
drm-total-cycles to the user. Current code uses XE_FW_GT to read this
register on the first available engine in a GT. When testing on DG2, it
is observed that this value is 0 when running test on some engines. To
resolve that, get the hwe domain specific FW for reading the engine
timestamp.
v2:
- update commit message
- use domain specific FW (Matt)
v3:
- Drop check for hwe in the helper (Matt, Michal)
v4:
- checkpatch fixes
v5: Rebase
Fixes: 188ced1e0f ("drm/xe/client: Print runtime to fdinfo")
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240627235105.2631135-1-umesh.nerlige.ramappa@intel.com
The current code is running into a use after free case where xe file is
closed before the exec queue run_ticks can be updated. This is occurring
in the xe_file_close path. To fix that, do not access xe file when
updating the exec queue run_ticks. Instead store the exec queue run_ticks
locally in the exec queue object and accumulate it when the user dumps
the drm client stats. We know that the xe file is valid when user is
dumping the run_ticks for the drm client, so this effectively
removes the dependency on xe file object in
xe_exec_queue_update_run_ticks().
v2:
- Fix the accumulation of q->run_ticks delta into xe file run_ticks
- s/runtime/run_ticks/ (Rodrigo)
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/1908
Fixes: 6109f24f87 ("drm/xe: Add helper to accumulate exec queue runtime")
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240524234744.1352543-2-umesh.nerlige.ramappa@intel.com
Print the accumulated runtime for client when printing fdinfo.
Each time a query is done it first does 2 things:
1) loop through all the exec queues for the current client and
accumulate the runtime, per engine class. CTX_TIMESTAMP is used for
that, being read from the context image.
2) Read a "GPU timestamp" that can be used for considering "how much GPU
time has passed" and that has the same unit/refclock as the one
recording the runtime. RING_TIMESTAMP is used for that via MMIO.
Since for all current platforms RING_TIMESTAMP follows the same
refclock, just read it once, using any first engine available.
This is exported to userspace as 2 numbers in fdinfo:
drm-cycles-<class>: <RUNTIME>
drm-total-cycles-<class>: <TIMESTAMP>
Userspace is expected to collect at least 2 samples, which allows to
know the client engine busyness as per:
RUNTIME1 - RUNTIME0
busyness = ---------------------
T1 - T0
Since drm-cycles-<class> always starts at 0, it's also possible to know
if and engine was ever used by a client.
It's expected that userspace will read any 2 samples every few seconds.
Given the update frequency of the counters involved and that
CTX_TIMESTAMP is 32-bits, the counter for each exec_queue can wrap
around (assuming 100% utilization) after ~200s. The wraparound is not
perceived by userspace since it's just accumulated for all the
exec_queues in a 64-bit counter) but the measurement will not be
accurate if the samples are too far apart.
This could be mitigated by adding a workqueue to accumulate the counters
every so often, but it's additional complexity for something that is
done already by userspace every few seconds in tools like gputop (from
igt), htop, nvtop, etc, with none of them really defaulting to 1 sample
per minute or more.
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240517204310.88854-9-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
drm/xe feature pull for v6.9:
UAPI Changes:
- New query to the GuC firmware submission version. (José Roberto de Souza)
- Remove unused persistent exec_queues (Thomas Hellström)
- Add vram frequency sysfs attributes (Sujaritha Sundaresan, Rodrigo Vivi)
- Add the flag XE_VM_BIND_FLAG_DUMPABLE to notify devcoredump that mapping
should be dumped (Maarten Lankhorst)
Cross-drivers Changes:
- Make sure intel_wakeref_t is treated as opaque type on i915-display
and fix its type on xe
Driver Changes:
- Drop pre-production workarounds (Matt Roper)
- Drop kunit tests for unsuported platforms: PVC and pre-production DG2 (Lucas De Marchi)
- Start pumbling SR-IOV support with memory based interrupts
for VF (Michal Wajdeczko)
- Allow to map BO in GGTT with PAT index corresponding to
XE_CACHE_UC to work with memory based interrupts (Michal Wajdeczko)
- Improve logging with GT-oriented drm_printers (Michal Wajdeczko)
- Add GuC Doorbells Manager as prep work SR-IOV during
VF provisioning ((Michal Wajdeczko)
- Refactor fake device handling in kunit integration ((Michal Wajdeczko)
- Implement additional workarounds for xe2 and MTL (Tejas Upadhyay,
Lucas De Marchi, Shekhar Chauhan, Karthik Poosa)
- Program a few registers according to perfomance guide spec for Xe2 (Shekhar Chauhan)
- Add error handling for non-blocking communication with GuC (Daniele Ceraolo Spurio)
- Fix remaining 32b build issues and enable it back (Lucas De Marchi)
- Fix build with CONFIG_DEBUG_FS=n (Jani Nikula)
- Fix warnings from GuC ABI headers (Matthew Brost)
- Introduce Relay Communication for SR-IOV for VF <-> GuC <-> PF (Michal Wajdeczko)
- Add mocs reset kunit (Ruthuvikas Ravikumar)
- Fix spellings (Colin Ian King)
- Disable mid-thread preemption when not properly supported by hardware (Nirmoy Das)
- Release mmap mappings on rpm suspend (Badal Nilawar)
- Fix BUG_ON on xe_exec by moving fence reservation to the validate stage (Matthew Auld)
- Fix xe_exec by reserving extra fence slot for CPU bind (Matthew Brost)
- Fix xe_exec with full long running exec queue, now returning
-EWOULDBLOCK to userspace (Matthew Brost)
- Fix CT irq handler when CT is disabled (Matthew Brost)
- Fix VM_BIND_OP_UNMAP_ALL without any bound vmas (Thomas Hellström)
- Fix missing __iomem annotations (Thomas Hellström)
- Fix exec queue priority handling with GuC (Brian Welty)
- Fix setting SLPC flag to GuC when it's not supported (Vinay Belgaumkar)
- Fix C6 disabling without SLPC (Matt Roper)
- Drop -Wstringop-overflow to fix build with GCC11 (Paul E. McKenney)
- Circumvent bogus -Wstringop-overflow in one case (Arnd Bergmann)
- Refactor exec_queue user extensions handling and fix USM attributes
being applied too late (Brian Welty)
- Use circ_buf head/tail convention (Matthew Brost)
- Fail build if circ_buf-related defines are modified with incompatible values
(Matthew Brost)
- Fix several error paths (Dan Carpenter)
- Fix CCS copy for small VRAM copy chunks (Thomas Hellström)
- Rework driver initialization order and paths to account for driver running
in VF mode (Michal Wajdeczko)
- Initialize GuC earlier during probe to handle driver in VF mode (Michał Winiarski)
- Fix migration use of MI_STORE_DATA_IMM to write PTEs (Matt Roper)
- Fix bounds checking in __xe_bo_placement_for_flags (Brian Welty)
- Drop display dependency on CONFIG_EXPERT (Jani Nikula)
- Do not hand-roll kstrdup when creating snapshot (Michal Wajdeczko)
- Stop creating one kunit module per kunit suite (Lucas De Marchi)
- Reduce scope and constify variables (Thomas Hellström, Jani Nikula, Michal Wajdeczko)
- Improve and document xe_guc_ct_send_recv() (Michal Wajdeczko)
- Add proxy communication between CSME and GSC uC (Daniele Ceraolo Spurio)
- Fix size calculation when writing pgtable (Fei Yang)
- Make sure cfb is page size aligned in stolen memory (Vinod Govindapillai)
- Stop printing guc log to dmesg when waiting for GuC fails (Rodrigo Vivi)
- Use XE_CACHE_WB instead of XE_CACHE_NONE for cpu coherency on migration
(Himal Prasad Ghimiray)
- Fix error path in xe_vm_create (Moti Haimovski)
- Fix warnings in doc generation (Thomas Hellström, Badal Nilawar)
- Improve devcoredump content for mesa debugging (José Roberto de Souza)
- Fix crash in trace_dma_fence_init() (José Roberto de Souza)
- Improve CT state change handling (Matthew Brost)
- Toggle USM support for Xe2 (Lucas De Marchi)
- Reduces code duplication to emit PIPE_CONTROL (José Roberto de Souza)
- Canonicalize addresses where needed for Xe2 and add to devcoredump
(José Roberto de Souza)
- Only allow 1 ufence per exec / bind IOCTL (Matthew Brost)
- Move all display code to display/ (Jani Nikula)
- Fix sparse warnings by correctly using annotations (Thomas Hellström)
- Warn on job timeouts instead of using asserts (Matt Roper)
- Prefix macros to avoid clashes with sparc (Matthew Brost)
- Fix -Walloc-size by subclassing instead of allocating size smaller than struct (Thomas Hellström)
- Add status check during gsc header readout (Suraj Kandpal)
- Fix infinite loop in vm_bind_ioctl_ops_unwind() (Matthew Brost)
- Fix fence refcounting (Matthew Brost)
- Fix picking incorrect userptr VMA (Matthew Brost)
- Fix USM on integrated by mapping both mem.kernel_bb_pool and usm.bb_pool (Matthew Brost)
- Fix double initialization of display power domains (Xiaoming Wang)
- Check expected uC versions by major.minor.patch instead of just major.minor (John Harrison)
- Bump minimum GuC version to 70.19.2 for all platforms under force-probe
(John Harrison)
- Add GuC firmware loading for Lunar Lake (John Harrison)
- Use kzalloc() instead of hand-rolled alloc + memset (Nirmoy Das)
- Fix max page size of VMA during a REMAP (Matthew Brost)
- Don't ignore error when pinning pages in kthread (Matthew Auld)
- Refactor xe hwmon (Karthik Poosa)
- Add debug logs for D3cold (Riana Tauro)
- Remove broken TEST_VM_ASYNC_OPS_ERROR (Matthew Brost)
- Always allow to override firmware blob with module param and improve
log when no firmware is found (Lucas De Marchi)
- Fix shift-out-of-bounds due to xe_vm_prepare_vma() accepting zero fences (Thomas Hellström)
- Fix shift-out-of-bounds by distinguishing xe_pt/xe_pt_dir subclass (Thomas Hellström)
- Fail driver bind if platform supports MSIX, but fails to allocate all of them (Dani Liberman)
- Fix intel_fbdev thinking memory is backed by shmem (Matthew Auld)
- Prefer drm_dbg() over dev_dbg() (Jani Nikula)
- Avoid function cast warnings with clang-16 (Arnd Bergmann)
- Enhance xe_bo_move trace (Priyanka Dandamudi)
- Fix xe_vma_set_pte_size() not setting the right gpuva.flags for 4K size (Matthew Brost)
- Add XE_VMA_PTE_64K VMA flag (Matthew Brost)
- Return 2MB page size for compact 64k PTEs (Matthew Brost)
- Remove usage of the deprecated ida_simple_xx() API (Christophe JAILLET)
- Fix modpost warning on xe_mocs live kunit module (Ashutosh Dixit)
- Drop extra newline in from sysfs files (Ashutosh Dixit)
- Implement VM snapshot support for BO's and userptr (Maarten Lankhorst)
- Add debug logs when skipping rebinds (Matthew Brost)
- Fix code generation when mixing build directories (Dafna Hirschfeld)
- Prefer struct_size over open coded arithmetic (Erick Archer)
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/dbdkrwmcoqqlwftuc3olbauazc3pbamj26wa34puztowsnauoh@i3zms7ut4yuw
Use the newly added drm_print_memory_stats helper to show memory
utilisation of our objects in drm/driver specific fdinfo output.
To collect the stats we walk the per memory regions object lists
and accumulate object size into the respective drm_memory_stats
categories.
Objects with multiple possible placements are reported in multiple
regions for total and shared sizes, while other categories are
counted only for the currently active region.
V4:
- Remove rcu lock - Auld/Thomas
- take refcnt only if its non-zero - Auld
- DMA_RESV_USAGE_BOOKKEEP covers all fences - Auld
- covert to xe_bo for public objects
V3:
- dont use xe_bo_get/put, not needed
- use designated initializer - Jani
- use list_for_each_entry_rcu
- Fix Checkpatch err - CI
V2:
- Use static initializer for mem_type - Himal/Jani
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
In order to show per client memory consumption, we
need tracking support APIs to add at every bo consumption
and removal. Adding APIs here to add tracking calls at
places wherever it is applicable.
V5:
- Rebase
V4:
- remove client bo before vm_put
- spin_lock_irqsave not required - Auld
V3:
- update .h to return xe_drm_client_remove_bo void
- protect xe_drm_client_remove_bo under CONFIG_PROC_FS check - Himal
- Fixed Checkpatch error - CI
V2:
- make xe_drm_client_remove_bo return void - Himal
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
