Age | Commit message (Collapse) | Author |
|
Normally scratch page is not allowed when a vm is operate under page
fault mode, i.e., in the existing codes, DRM_XE_VM_CREATE_FLAG_SCRATCH_PAGE
and DRM_XE_VM_CREATE_FLAG_FAULT_MODE are mutual exclusive. The reason
is fault mode relies on recoverable page to work, while scratch page
can mute recoverable page fault.
On xe2 and xe3, out of bound prefetch can cause page fault and further
system hang because xekmd can't resolve such page fault. SYCL and OCL
language runtime requires out of bound prefetch to be silently dropped
without causing any functional problem, thus the existing behavior
doesn't meet language runtime requirement.
At the same time, HW prefetching can cause page fault interrupt. Due to
page fault interrupt overhead (i.e., need Guc and KMD involved to fix
the page fault), HW prefetching can be slowed by many orders of magnitude.
Fix those problems by allowing scratch page under fault mode for xe2 and
xe3. With scratch page in place, HW prefetching could always hit scratch
page instead of causing interrupt.
A side effect is, scratch page could hide application program error.
Application out of bound accesses are hided by scratch page mapping,
instead of get reported to user.
v2: Refine commit message (Thomas)
v3: Move the scratch page flag check to after scratch page wa (Thomas)
v4: drop NEEDS_SCRATCH macro (matt)
Add a comment to DRM_XE_VM_CREATE_FLAG_SCRATCH_PAGE
Signed-off-by: Oak Zeng <oak.zeng@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Link: https://lore.kernel.org/r/20250403165328.2438690-4-oak.zeng@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
|
|
When a vm runs under fault mode, if scratch page is enabled, we need
to clear the scratch page mapping on vm_bind for the vm_bind address
range. Under fault mode, we depend on recoverable page fault to
establish mapping in page table. If scratch page is not cleared, GPU
access of address won't cause page fault because it always hits the
existing scratch page mapping.
When vm_bind with IMMEDIATE flag, there is no need of clearing as
immediate bind can overwrite the scratch page mapping.
So far only is xe2 and xe3 products are allowed to enable scratch page
under fault mode. On other platform we don't allow scratch page under
fault mode, so no need of such clearing.
v2: Rework vm_bind pipeline to clear scratch page mapping. This is similar
to a map operation, with the exception that PTEs are cleared instead of
pointing to valid physical pages. (Matt, Thomas)
TLB invalidation is needed after clear scratch page mapping as larger
scratch page mapping could be backed by physical page and cached in
TLB. (Matt, Thomas)
v3: Fix the case of clearing huge pte (Thomas)
Improve commit message (Thomas)
v4: TLB invalidation on all LR cases, not only the clear on bind
cases (Thomas)
v5: Misc cosmetic changes (Matt)
Drop pt_update_ops.invalidate_on_bind. Directly wire
xe_vma_op.map.invalidata_on_bind to bind_op_prepare/commit (Matt)
v6: checkpatch fix (Matt)
v7: No need to check platform needs_scratch deciding invalidate_on_bind
(Matt)
v8: rebase
v9: rebase
v10: fix an error in xe_pt_stage_bind_entry, introduced in v9 rebase
Signed-off-by: Oak Zeng <oak.zeng@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Link: https://lore.kernel.org/r/20250403165328.2438690-3-oak.zeng@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
|
|
On some platform, scratch page is needed for out of bound prefetch
to work. Introduce a bit in device descriptor to specify whether
this device needs scratch page to work.
v2: introduce a needs_scratch bit in device info (Thomas, Jonathan)
v3: drop NEEDS_SCRATCH macro (Matt)
Signed-off-by: Oak Zeng <oak.zeng@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Link: https://lore.kernel.org/r/20250403165328.2438690-2-oak.zeng@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
|
|
Currently we can run into issues with provisioning VRAM region, due to
requiring contig VRAM BO underneath. We sometimes see that allocation
(multiple GB) can fail even when there is enough free space. We don't
need CPU access to the buffer in the first place, so can forgo pin_map
and therefore also the contig requirement. Keep the same behavior with
save and restore during suspend/resume (which can now be done with
blitter). We also need the VRAM to occupy the same pages so we don't
need to re-program the LMTT, so should still remain pinned (also we
don't want something to try evict it). With that covert over to plain
pinned kernel object.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250403102440.266113-16-matthew.auld@intel.com
|
|
If the kernel bo doesn't care about vmap(), either directly or
indirectly with save/restore then we don't need to force contig for such
buffers.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250403102440.266113-15-matthew.auld@intel.com
|
|
Some users apply PINNED and some don't when using pin_map(). The pin in
pin_map() should imply PINNED so just unconditionally apply it and clean
up all users.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250403102440.266113-14-matthew.auld@intel.com
|
|
With the idea of having more pinned objects using the blitter engine
where possible, during suspend/resume, mark the pinned objects which
can be done during the late phase once submission/migration has been
setup. Start out simple with lrc and page-tables from userspace.
v2:
- s/early_restore/late_restore; early restore was way too bold with too
many places being impacted at once.
v3:
- Split late vs early into separate lists, to align with newly added
apply-to-pinned infra.
v4:
- Rebase.
v5:
- Make sure we restore the late phase kernel_bo_present in igpu.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250403102440.266113-13-matthew.auld@intel.com
|
|
For kernel BOs we don't clear the CCS state on creation, therefore we
should be careful to ignore it when copying pages. In a future patch we
opt for using the copy path here for kernel BOs, so this now needs to be
considered.
v2:
- Drop bogus asserts (CI)
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250403102440.266113-12-matthew.auld@intel.com
|
|
Not all BOs need to be restored on resume / d3cold exit, add
XE_BO_FLAG_PINNED_NO_RESTORE which skips restoring of BOs rather just
allocates VRAM for the BO. This should slightly speedup resume / d3cold
exit flows.
Marking GuC ADS, GuC CT, GuC log, GuC PC, and SA as NORESTORE.
v2:
- s/WONTNEED/NORESTORE (Vivi)
- Rebase on newly added g2g and backup object flow
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250403102440.266113-11-matthew.auld@intel.com
|
|
Currently we move pinned objects, relying on the fact that the lpfn/fpfn
will force the placement to occupy the same pages when restoring.
However this then limits all such pinned objects to be contig
underneath. In addition it is likely a little fragile moving pinned
objects in the first place. Rather than moving such objects rather copy
the page contents to a secondary system memory object, that way the VRAM
pages never move and remain pinned. This also opens the door for
eventually having non-contig pinned objects that can also be
saved/restored using blitter.
v2:
- Make sure to drop the fence ref.
- Handle NULL bo->migrate.
v3:
- Ensure we add the copy fence to the BOs, otherwise backup_obj can
be freed before pipelined copy finishes.
v4:
- Rebase on newly added apply-to-pinned infra.
Link: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/1182
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250403102440.266113-10-matthew.auld@intel.com
|
|
Add Wa_16025250150 for the Xe2_HPG (graphics version: 20.01) platforms.
It is a permanent workaround, and applicable on all the steppings.
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Signed-off-by: Aradhya Bhatia <aradhya.bhatia@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250325134421.1489416-1-aradhya.bhatia@intel.com
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
|
|
The structure was missing a proper kerneldoc header and once
that was added a number of typos and errors became obvious.
Fix those.
Reported-by: Lucas De Marchi <lucas.demarchi@intel.com>
Closes: https://lore.kernel.org/intel-xe/x53tcs5bjldw6lcorjemuheklxcmepdvr2u7lvt3hpqrzqoc4h@nsu6hs25taqj/
Fixes: b2d4b03b03a7 ("drm/xe: Make the PT code handle placement per PTE rather than per vma / range")
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250402122924.25526-1-thomas.hellstrom@linux.intel.com
|
|
Define PMU events for GT frequency (actual and requested). The
instantaneous values for these frequencies will be displayed.
Following PMU events are being added:
xe_0000_00_02.0/gt-actual-frequency/ [Kernel PMU event]
xe_0000_00_02.0/gt-requested-frequency/ [Kernel PMU event]
Standard perf commands can be used to monitor GT frequency:
$ perf stat -e xe_0000_00_02.0/gt-requested-frequency,gt=0/ -I1000
1.001229762 1483 Mhz xe_0000_00_02.0/gt-requested-frequency,gt=0/
2.006175406 1483 Mhz xe_0000_00_02.0/gt-requested-frequency,gt=0/
v2: Use locks while storing/reading samples, keep track of multiple
clients (Lucas) and other general cleanup.
v3: Review comments (Lucas) and use event counts instead of mask for
active events.
v4: Add freq events to event_param_valid method (Riana)
v5: Use instantaneous values instead of aggregating (Lucas)
v6: Obtain fwake at init for freq events as well and use non fwake
variant method for reading requested freq to avoid lockdep issues (Lucas)
v7: Review comments (Rodrigo, Lucas)
Cc: Riana Tauro <riana.tauro@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://lore.kernel.org/r/20250331204827.2535393-1-vinay.belgaumkar@intel.com
|
|
The context of each engine starts with a 4k memory space for the
"Per-process HW status page" (PPHWSP). In xe_gt_lrc_size(), we have been
implicitly accounting for that page in the switch statement on the
engine class.
Since the PPHWSP is common to all engines, let's extract that into it's
own assignment. That makes the context structure more explicit in the
code and aligns better with the descriptions in Bspec.
Another advantage of keeping it separate is that now the sizes used in
the switch statement match the sizes we calculate engine-specific
context images, which have their own Bspec pages.
Bspec: 67296, 60159, 45554
Signed-off-by: Gustavo Sousa <gustavo.sousa@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://lore.kernel.org/r/20250328-explicit-pphwsp-size-in-xe_gt_lrc_size-v1-1-ceb9ce7c8bc1@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
|
|
Historically, the Vertex Fetcher unit has not been an L3 client. That
meant that, when a buffer containing vertex data was written to, it was
necessary to issue a PIPE_CONTROL::VF Cache Invalidate to invalidate any
VF L2 cachelines associated with that buffer, so the new value would be
properly read from memory.
Since Tigerlake and later, VERTEX_BUFFER_STATE and 3DSTATE_INDEX_BUFFER
have included an "L3 Bypass Enable" bit which userspace drivers can set
to request that the vertex fetcher unit snoop L3. However, unlike most
true L3 clients, the "VF Cache Invalidate" bit continues to only
invalidate the VF L2 cache - and not any associated L3 lines.
To handle that, PIPE_CONTROL has a new "L3 Read Only Cache Invalidation
Bit", which according to the docs, "controls the invalidation of the
Geometry streams cached in L3 cache at the top of the pipe." In other
words, the vertex and index buffer data that gets cached in L3 when
"L3 Bypass Disable" is set.
Mesa always sets L3 Bypass Disable so that the VF unit snoops L3, and
whenever it issues a VF Cache Invalidate, it also issues a L3 Read Only
Cache Invalidate so that both L2 and L3 vertex data is invalidated.
xe is issuing VF cache invalidates too (which handles cases like CPU
writes to a buffer between GPU batches). Because userspace may enable
L3 snooping, it needs to issue an L3 Read Only Cache Invalidate as well.
Fixes significant flickering in Firefox on Meteorlake, which was writing
to vertex buffers via the CPU between batches; the missing L3 Read Only
invalidates were causing the vertex fetcher to read stale data from L3.
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/4460
Fixes: 6ef3bb60557d ("drm/xe: enable lite restore")
Cc: stable@vger.kernel.org # v6.13+
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250330165923.56410-1-rodrigo.vivi@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
|
|
Commit b4b05e53b550 ("drm/xe/guc_pc: Retry and wait longer for GuC PC
start"), leads to the following Smatch static checker warning:
drivers/gpu/drm/xe/xe_guc_pc.c:1073 xe_guc_pc_start()
warn: missing error code here? '_dev_err()' failed. 'ret' = '0'
Fixes: b4b05e53b550 ("drm/xe/guc_pc: Retry and wait longer for GuC PC start")
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Closes: https://lore.kernel.org/intel-xe/1454a5f1-ee18-4df1-a6b2-a4a3dddcd1cb@stanley.mountain/
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250328181752.26677-1-rodrigo.vivi@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
|
|
This error message is only applicable for platforms using
GuC submission - to warn the user if the GuC they are using
or the platform they are running doesn't have this information
to provide to userspace about the platform. When forcing
execlist submission, which is something only used for debug,
the user is running at their own risk and should understand
the limitations of running without GuC.
v2 (John/Lucas): Don't print an info message with execlists
Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Jagmeet Randhawa <jagmeet.randhawa@intel.com>
Link: https://lore.kernel.org/r/20250328154236.9216-1-stuart.summers@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
|
|
The error capture list in the ADS is initially allocated using a
placeholder size. When the actual size is determinied later on, there
is a debug print about the new size. However, the wording is such that
some people see it as an unexpected thing and therefore a potential
problem. So re-word it to be a little less concerning.
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250325203211.3907890-1-John.C.Harrison@Intel.com
|
|
Building the xe driver for i386 results in a link time warning:
x86_64-linux-ld: drivers/gpu/drm/xe/xe_migrate.o: in function `xe_migrate_vram':
xe_migrate.c:(.text+0x1e15): undefined reference to `__udivdi3'
Avoid this by using DIV_U64_ROUND_UP() instead of DIV_ROUND_UP(). The driver
is unlikely to be used on 32=bit hardware, so the extra cost here is not
too important.
Fixes: 9c44fd5f6e8a ("drm/xe: Add migrate layer functions for SVM support")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20250324210612.2927194-1-arnd@kernel.org
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
|
|
The dump on a dead CT tries to emulate the devcoredump formatting (it
would use devcoredump code directly but that requires more re-work to
happen - work in progress). So update the print of the dead CT reason
code to match the format of the 'reason' string that was added to the
actual devcoredump a little while ago.
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250325203111.3907426-1-John.C.Harrison@Intel.com
|
|
According to pci core guidelines, pci_save_config is recommended when the
driver explicitly needs to set the pci power state. As of now xe kmd is
only doing pci_save_config while entering to s2idle/s3 state, which makes
pci core think that device driver has already applied required pci power
state. This leads to GPU remain in D0 state. To fix the issue setting
the pci power state to D3Cold.
Fixes:dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250327161914.432552-1-badal.nilawar@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
|
|
Sysfs_ops needs to be defined on all directories which
can have attr files with set/get method. Add sysfs_ops
to even those directories which is currently empty but
would have attr files with set/get method in future.
Leave .default with default sysfs_ops as it will never
have setter method.
V2(Himal/Rodrigo):
- use single sysfs_ops for all dir and attr with set/get
- add default ops as ./default does not need runtime pm at all
Fixes: 3f0e14651ab0 ("drm/xe: Runtime PM wake on every sysfs call")
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250327122647.886637-1-tejas.upadhyay@intel.com
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
|
|
Extend Wa_14022293748, Wa_22019794406 to Xe3_LPG
Signed-off-by: Julia Filipchuk <julia.filipchuk@intel.com>
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://lore.kernel.org/r/20250325224310.1455499-1-julia.filipchuk@intel.com
|
|
The intent of the error path in xe_migrate_clear is to wait on locally
generated fence and then return. The code is waiting on m->fence which
could be the local fence but this is only stable under the job mutex
leading to a possible UAF. Fix code to wait on local fence.
Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Cc: stable@vger.kernel.org
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250311182915.3606291-1-matthew.brost@intel.com
|
|
The RCU_MODE_FIXED_SLICE_CCS_MODE setting is not getting invoked
in the gt reset path after the ccs_mode setting by the user.
Add it to engine register update list (in hw_engine_setup_default_state())
which ensures it gets set in the gt reset and engine reset paths.
v2: Add register update to engine list to ensure it gets updated
after engine reset also.
Fixes: 0d97ecce16bd ("drm/xe: Enable Fixed CCS mode setting")
Cc: stable@vger.kernel.org
Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250327185604.18230-1-niranjana.vishwanathapura@intel.com
|
|
When the size of the range invalidated is larger than
rounddown_pow_of_two(ULONG_MAX),
The function macro roundup_pow_of_two(length) will hit an out-of-bounds
shift [1].
Use a full TLB invalidation for such cases.
v2:
- Use a define for the range size limit over which we use a full
TLB invalidation. (Lucas)
- Use a better calculation of the limit.
[1]:
[ 39.202421] ------------[ cut here ]------------
[ 39.202657] UBSAN: shift-out-of-bounds in ./include/linux/log2.h:57:13
[ 39.202673] shift exponent 64 is too large for 64-bit type 'long unsigned int'
[ 39.202688] CPU: 8 UID: 0 PID: 3129 Comm: xe_exec_system_ Tainted: G U 6.14.0+ #10
[ 39.202690] Tainted: [U]=USER
[ 39.202690] Hardware name: ASUS System Product Name/PRIME B560M-A AC, BIOS 2001 02/01/2023
[ 39.202691] Call Trace:
[ 39.202692] <TASK>
[ 39.202695] dump_stack_lvl+0x6e/0xa0
[ 39.202699] ubsan_epilogue+0x5/0x30
[ 39.202701] __ubsan_handle_shift_out_of_bounds.cold+0x61/0xe6
[ 39.202705] xe_gt_tlb_invalidation_range.cold+0x1d/0x3a [xe]
[ 39.202800] ? find_held_lock+0x2b/0x80
[ 39.202803] ? mark_held_locks+0x40/0x70
[ 39.202806] xe_svm_invalidate+0x459/0x700 [xe]
[ 39.202897] drm_gpusvm_notifier_invalidate+0x4d/0x70 [drm_gpusvm]
[ 39.202900] __mmu_notifier_release+0x1f5/0x270
[ 39.202905] exit_mmap+0x40e/0x450
[ 39.202912] __mmput+0x45/0x110
[ 39.202914] exit_mm+0xc5/0x130
[ 39.202916] do_exit+0x21c/0x500
[ 39.202918] ? lockdep_hardirqs_on_prepare+0xdb/0x190
[ 39.202920] do_group_exit+0x36/0xa0
[ 39.202922] get_signal+0x8f8/0x900
[ 39.202926] arch_do_signal_or_restart+0x35/0x100
[ 39.202930] syscall_exit_to_user_mode+0x1fc/0x290
[ 39.202932] do_syscall_64+0xa1/0x180
[ 39.202934] ? do_user_addr_fault+0x59f/0x8a0
[ 39.202937] ? lock_release+0xd2/0x2a0
[ 39.202939] ? do_user_addr_fault+0x5a9/0x8a0
[ 39.202942] ? trace_hardirqs_off+0x4b/0xc0
[ 39.202944] ? clear_bhb_loop+0x25/0x80
[ 39.202946] ? clear_bhb_loop+0x25/0x80
[ 39.202947] ? clear_bhb_loop+0x25/0x80
[ 39.202950] entry_SYSCALL_64_after_hwframe+0x76/0x7e
[ 39.202952] RIP: 0033:0x7fa945e543e1
[ 39.202961] Code: Unable to access opcode bytes at 0x7fa945e543b7.
[ 39.202962] RSP: 002b:00007ffca8fb4170 EFLAGS: 00000293
[ 39.202963] RAX: 000000000000003d RBX: 0000000000000000 RCX: 00007fa945e543e3
[ 39.202964] RDX: 0000000000000000 RSI: 00007ffca8fb41ac RDI: 00000000ffffffff
[ 39.202964] RBP: 00007ffca8fb4190 R08: 0000000000000000 R09: 00007fa945f600a0
[ 39.202965] R10: 0000000000000000 R11: 0000000000000293 R12: 0000000000000000
[ 39.202966] R13: 00007fa9460dd310 R14: 00007ffca8fb41ac R15: 0000000000000000
[ 39.202970] </TASK>
[ 39.202970] ---[ end trace ]---
Fixes: 332dd0116c82 ("drm/xe: Add range based TLB invalidations")
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: <stable@vger.kernel.org> # v6.8+
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> #v1
Link: https://lore.kernel.org/r/20250326151634.36916-1-thomas.hellstrom@linux.intel.com
|
|
Change the scope of the migrate subsystem to be dev managed instead of
drm managed.
The parent pci struct &device, that the xe struct &drm_device is a part
of, gets removed when a hot unplug is triggered, which causes the
underlying iommu group to get destroyed as well.
The migrate subsystem, which handles the lifetime of the page-table tree
(pt) BO, doesn't get a chance to keep the BO back during the hot unplug,
as all the references to DRM haven't been put back.
When all the references to DRM are indeed put back later, the migrate
subsystem tries to put back the pt BO. Since the underlying iommu group
has been already destroyed, a kernel NULL ptr dereference takes place
while attempting to keep back the pt BO.
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/3914
Suggested-by: Thomas Hellstrom <thomas.hellstrom@intel.com>
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Signed-off-by: Aradhya Bhatia <aradhya.bhatia@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250326151929.1495972-1-aradhya.bhatia@intel.com
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
|
|
With SVM, ranges forwarded to the PT code for binding can, mostly
due to races when migrating, point to both VRAM and system / foreign
device memory. Make the PT code able to handle that by checking,
for each PTE set up, whether it points to local VRAM or to system
memory.
v2:
- Fix system memory GPU atomic access.
v3:
- Avoid the UAPI change. It needs more thought.
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Link: https://lore.kernel.org/r/20250326080551.40201-6-thomas.hellstrom@linux.intel.com
|
|
The drm_pagemap functionality does not depend on the device having
recoverable pagefaults available. So allow xe_migrate_vram() also for
such devices. Even if this will have little use in practice, it's
beneficial for testin multi-device SVM, since a memory provider could
be a non-pagefault capable gpu.
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250326080551.40201-5-thomas.hellstrom@linux.intel.com
|
|
On device unbind, migrate exported bos, including pagemap bos to
system. This allows importers to take proper action without
disruption. In particular, SVM clients on remote devices may
continue as if nothing happened, and can chose a different
placement.
The evict_flags() placement is chosen in such a way that bos that
aren't exported are purged.
For pinned bos, we unmap DMA, but their pages are not freed yet
since we can't be 100% sure they are not accessed.
All pinned external bos (not just the VRAM ones) are put on the
pinned.external list with this patch. But this only affects the
xe_bo_pci_dev_remove_pinned() function since !VRAM bos are
ignored by the suspend / resume functionality. As a follow-up we
could look at removing the suspend / resume iteration over
pinned external bos since we currently don't allow pinning
external bos in VRAM, and other external bos don't need any
special treatment at suspend / resume.
v2:
- Address review comments. (Matthew Auld).
v3:
- Don't introduce an external_evicted list (Matthew Auld)
- Add a discussion around suspend / resume behaviour to the
commit message.
- Formatting fixes.
v4:
- Move dma-unmaps of pinned kernel bos to a dev managed
callback to give subsystems using these bos a chance to
clean them up. (Matthew Auld)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250326080551.40201-4-thomas.hellstrom@linux.intel.com
|
|
If drm_gpusvm_migrate_to_devmem() succeeds, if a cpu access happens to the
range the bo may be freed before xe_bo_unlock(), causing a UAF.
Since the reference is transferred, use xe_svm_devmem_release() to
release the reference on drm_gpusvm_migrate_to_devmem() failure,
and hold a local reference to protect the UAF.
Fixes: 2f118c949160 ("drm/xe: Add SVM VRAM migration")
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250326080551.40201-3-thomas.hellstrom@linux.intel.com
|
|
Don't rely on CONFIG_DRM_GPUSVM because other drivers may enable it
causing us to compile in SVM support unintentionally.
Also take the opportunity to leave more code out of compilation if
!CONFIG_DRM_XE_GPUSVM and !CONFIG_DRM_XE_DEVMEM_MIRROR
v3:
- Fixes for compilation errors on 32-bit. This changes the Kconfig
logic a bit.
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250326080551.40201-2-thomas.hellstrom@linux.intel.com
|
|
One additional BMG PCI ID has been added to the spec; make sure our
driver recognizes devices with this ID properly.
Bspec: 68090
Cc: stable@vger.kernel.org # v6.12+
Reviewed-by: Clint Taylor <Clinton.A.Taylor@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250325224709.4073080-2-matthew.d.roper@intel.com
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
|
|
Add fault injection for xe_oa_alloc_regs to allow it to fail while
executing xe_oa_add_config_ioctl().
This need to be added as it cannot be reached by injecting error through
IOCTL arguments.
Signed-off-by: Nakshtra Goyal <nakshtra.goyal@intel.com>
Reviewed-by: Francois Dugast <francois.dugast@intel.com>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Link: https://lore.kernel.org/r/20250227102339.2859726-1-nakshtra.goyal@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
|
|
Enable per-function engine activity stats when VF's are enabled
and disable when VF's are disabled
v2: fix commit message
remove reset stats from pf config (Michal)
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250311071759.2117211-4-riana.tauro@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
|
|
Add PMU support for per-function engine activity stats.
per-function engine activity is enabled when VF's are enabled.
If 2 VF's are enabled, then the applicable function values are
0 - PF engine activity
1 - VF1 engine activity
2 - VF2 engine activity
This can be read from perf tool as shown below
./perf stat -e xe_<bdf>/engine-active-ticks,gt=0,engine_class=0,
engine_instance=0,function=1/ -I 1000
v2: fix documentation (Umesh)
remove global for functions (Lucas, Michal)
v3: fix commit message
move function_id checks to same place (Michal)
v4: fix comment (Umesh)
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250311071759.2117211-3-riana.tauro@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
|
|
Add support for function level engine activity stats.
Engine activity stats are enabled when VF's are enabled
v2: remove unnecessary initialization
move offset to improve code readability (Umesh)
remove global for function engine activity (Lucas)
v3: fix commit message (Michal)
v4: remove enable function parameter
fix kernel-doc (Umesh)
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250311071759.2117211-2-riana.tauro@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
|
|
There are extra spaces in xe_vm_bind_ioctl_validate_bo(), remove those.
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250320211519.632432-1-dev@lankhorst.se
Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
|
|
Allow fault injection in a function used during initialization by
xe_hw_engine_class_sysfs_init() so that its error handling can be
tested.
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250314105050.636983-1-francois.dugast@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
|
|
WARNING: unmet direct dependencies detected for FB_IOMEM_HELPERS
Depends on [n]: HAS_IOMEM [=y] && FB_CORE [=n]
Selected by [m]:
- DRM_XE_DISPLAY [=y] && HAS_IOMEM [=y] && DRM [=m] && DRM_XE [=m] && DRM_XE [=m]=m [=m] && HAS_IOPORT [=y]
DRM_XE_DISPLAY requires FB_IOMEM_HELPERS, but the dependency FB_CORE is
missing, selecting FB_IOMEM_HELPERS if DRM_FBDEV_EMULATION is set as
other drm drivers.
Fixes: 44e694958b95 ("drm/xe/display: Implement display support")
Signed-off-by: Yue Haibing <yuehaibing@huawei.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250323114103.1960511-1-yuehaibing@huawei.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
|
|
Allow to test if driver behaves correctly when xe_pcode_probe_early()
fails. Note that this is not sufficient for testing survivability mode
as it's still required to read the hw to check for errors, which doesn't
happen on an injected failure.
To complete the early probe coverage, allow injection in the other
functions as well: xe_mmio_probe_early() and xe_device_probe_early().
Reviewed-by: Francois Dugast <francois.dugast@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250314-fix-survivability-v5-3-fdb3559ea965@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
|
|
Commit d40f275d96e8 ("drm/xe: Move survivability entirely to xe_pci")
tried to follow the logic: initialize everything needed and if
everything succeeds, set the flag that it's enabled. While it fixed some
corner cases of those calls failing, it was wrong for setting the flag
after the call to xe_heci_gsc_init(): that function does a different
initialization for survivability mode.
Fix that and add comments about this being done on purpose.
Suggested-by: Riana Tauro <riana.tauro@intel.com>
Fixes: d40f275d96e8 ("drm/xe: Move survivability entirely to xe_pci")
Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250314-fix-survivability-v5-2-fdb3559ea965@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
|
|
Commit d40f275d96e8 ("drm/xe: Move survivability entirely to xe_pci")
moved the survivability handling to be done entirely in the xe_pci
layer. However there are some issues with that approach:
1) Survivability mode needs at least the mmio initialized, otherwise it
can't really read a register to decide if it should enter that state
2) SR-IOV mode should be initialized, otherwise it's not possible to
check if it's VF
Besides, as pointed by Riana the check for
xe_survivability_mode_enable() was wrong in xe_pci_probe() since it's
not a bool return.
Fix that by moving the initialization to be entirely in the xe_device
layer, with the correct dependencies handled: only after mmio and sriov
initialization, and not triggering it on error from
wait_for_lmem_ready(). This restores the trigger behavior before that
commit. The xe_pci layer now only checks for "is it enabled?",
like it's doing in xe_pci_suspend()/xe_pci_remove(), etc.
Cc: Riana Tauro <riana.tauro@intel.com>
Fixes: d40f275d96e8 ("drm/xe: Move survivability entirely to xe_pci")
Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250314-fix-survivability-v5-1-fdb3559ea965@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
|
|
The different GTs on a device can be very different. Right now for all
platforms the same firmware is loaded in each GT, however future
platforms may benefit from loading a different file depending on the GT
type.
Based on previous patch by John Harrison <John.C.Harrison@Intel.com>.
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250307-xe-per-gt-fw-v1-2-459574d76400@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
|
|
The `entries` variable is used to loop through the array - it's supposed
to be const, but not static.
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250307-xe-per-gt-fw-v1-1-459574d76400@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
|
|
Introduce and use a helper to iterate over the various pinned bo lists.
There are a couple of slight functional changes:
1) GGTT maps are now performed with the bo locked.
2) If the per-bo callback fails, keep the bo on the original list.
v2:
- Skip unrelated change in xe_bo.c
Cc: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250321133709.75327-1-thomas.hellstrom@linux.intel.com
|
|
The WA requires KMD to disable DOP clock gating during a semaphore
wait and also ensure that idle delay for every CS is lower than the
idle wait time in the PWRCTX_MAXCNT register. Default values for these
registers already comply with this restriction.
v2: Store timestamp_base in gt info and other comments (Daniele)
v3: Skip WA check for VF
v4: Review comments (Matt Roper)
v5: Cleanup the clock functions and use reg_field_get (Matt Roper)
v6: Fix checkpatch issue
v7: Fix CI issue
Cc: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250320175123.3026754-1-vinay.belgaumkar@intel.com
|
|
The `struct ttm_resource->placement` contains TTM_PL_FLAG_* flags, but
it was incorrectly tested for XE_PL_* flags.
This caused xe_dma_buf_pin() to always fail when invoked for
the second time. Fix this by checking the `mem_type` field instead.
Fixes: 7764222d54b7 ("drm/xe: Disallow pinning dma-bufs in VRAM")
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: "Thomas Hellström" <thomas.hellstrom@linux.intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Nirmoy Das <nirmoy.das@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: intel-xe@lists.freedesktop.org
Cc: <stable@vger.kernel.org> # v6.8+
Signed-off-by: Tomasz Rusinowicz <tomasz.rusinowicz@intel.com>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250218100353.2137964-1-jacek.lawrynowicz@linux.intel.com
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
|
|
Backmerging to bring in the xe shrinker from drm-next.
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
|
|
Add hwmon support for fan1_input, fan2_input and fan3_input attributes,
which will expose fan speed of respective channels in RPM when supported
by hardware. With this in place we can monitor fan speed using lm-sensors
tool.
v2: Rely on platform checks instead of mailbox error (Aravind, Rodrigo)
v3: Introduce has_fan_control flag (Rodrigo)
Signed-off-by: Raag Jadav <raag.jadav@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250312085909.755073-1-raag.jadav@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
|