linux.git/drivers/gpu/drm/xe, branch v6.8-rc2

drm/xe: Remove PVC from xe_wa kunit tests

2024-01-24T10:13:55+00:00

Since the PCI IDs for PVC weren't added to the xe driver, the xe_wa
tests should not try to create a fake PVC device since they can't find
the right PCI ID. Fix bugs when running kunit:

	# xe_wa_gt: ASSERTION FAILED at drivers/gpu/drm/xe/tests/xe_wa_test.c:111
	Expected ret == 0, but
	    ret == -19 (0xffffffffffffffed)
	[FAILED] PVC (B0)
	# xe_wa_gt: ASSERTION FAILED at drivers/gpu/drm/xe/tests/xe_wa_test.c:111
	Expected ret == 0, but
	    ret == -19 (0xffffffffffffffed)
	[FAILED] PVC (B1)
	# xe_wa_gt: ASSERTION FAILED at drivers/gpu/drm/xe/tests/xe_wa_test.c:111
	Expected ret == 0, but
	    ret == -19 (0xffffffffffffffed)
	[FAILED] PVC (C0)

Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Signed-off-by: Lucas De Marchi 
Reviewed-by: Matt Roper 
Link: https://patchwork.freedesktop.org/patch/msgid/20240123031242.3548724-1-lucas.demarchi@intel.com
(cherry picked from commit ab5ae65fb25d06c38a6617a628b964828adb4786)
Signed-off-by: Thomas Hellström

drm/xe/vm: bugfix in xe_vm_create_ioctl

2024-01-24T10:13:41+00:00

Fix xe_vm_create_ioctl routine not freeing the vm-id allocated to it
when the function fails.

Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Signed-off-by: Moti Haimovski 
Reviewed-by: Matthew Brost 
Reviewed-by: Tomer Tayar 
Signed-off-by: Matt Roper 
Link: https://patchwork.freedesktop.org/patch/msgid/20240122102424.4008095-1-mhaimovski@habana.ai
(cherry picked from commit f6bf0424cadc27d7cf6a049d2db960e4b52fa513)
Signed-off-by: Thomas Hellström

drm/xe/xe2: Use XE_CACHE_WB pat index

2024-01-24T10:13:33+00:00

The pat table entry associated with XE_CACHE_WB is coherent whereas
XE_CACHE_NONE is non coherent. Migration expects the coherency
with cpu therefore use the coherent entry XE_CACHE_WB for
buffers not supporting compression. For read/write to flat ccs region
the issue is not related to coherency with cpu. The hardware expects
the pat index associated with GPUVA for indirect access to be
compression enabled hence use XE_CACHE_NONE_COMPRESSION.

v2
- Fix the argument to emit_pte, pass the bool directly. (Thomas)

v3
- Rebase
- Update commit message (Matt)

v4
- Add a Fixes: tag. (Thomas)

Cc: Matt Roper 
Cc: Thomas Hellström 
Fixes: 65ef8dbad1db ("drm/xe/xe2: Update emit_pte to use compression enabled PAT index")
Signed-off-by: Himal Prasad Ghimiray 
Reviewed-by: Thomas Hellström 
Signed-off-by: Thomas Hellström 
Link: https://patchwork.freedesktop.org/patch/msgid/20240119041826.1670496-1-himal.prasad.ghimiray@intel.com
(cherry picked from commit 6a02867560f77328ae5637b70b06704b140aafa6)
Signed-off-by: Thomas Hellström

drm/xe/display: Avoid calling readq()

2024-01-24T10:13:25+00:00

readq() is not available in 32bits and i915_gem_object_read_from_page()
is supposed to allow reading arbitrary sizes determined by the `size`
argument. Currently the only caller only passes a size == 8 so the
second problem is not that big. Migrate to calling
memcpy()/memcpy_fromio() to allow possible changes in the display side
and to fix the build on 32b architectures.

v2: Use memcpy/memcpy_fromio directly rather than using iosys-map with
    the same size == 8 bytes restriction (Matt Roper)

Fixes: 44e694958b95 ("drm/xe/display: Implement display support")
Signed-off-by: Lucas De Marchi 
Reviewed-by: Matt Roper 
Link: https://patchwork.freedesktop.org/patch/msgid/20240119001612.2991381-4-lucas.demarchi@intel.com
(cherry picked from commit 406663f777bee53e9ad93dc080c333d4655ab7de)
Signed-off-by: Thomas Hellström

drm/xe/mmio: Cast to u64 when printing

2024-01-24T10:13:15+00:00

resource_size_t uses %pa format in printk since the size varies
depending on build options. However to keep the io_size/physical_size
addition in the same call we can't pass the address without adding yet
another variable in these function. Simply cast it to u64 and keep using
%llx.

Fixes: 286089ce6929 ("drm/xe: Improve vram info debug printing")
Cc: Oak Zeng 
Cc: Michael J. Ruhl 
Cc: Matthew Brost 
Cc: Rodrigo Vivi 
Reviewed-by: Matt Roper 
Signed-off-by: Lucas De Marchi 
Link: https://patchwork.freedesktop.org/patch/msgid/20240119001612.2991381-3-lucas.demarchi@intel.com
(cherry picked from commit 6d8d038364d8ec573e9dc0872e17bee1e5f12490)
Signed-off-by: Thomas Hellström

drm/xe: Use _ULL for u64 division

2024-01-24T10:13:06+00:00

Use DIV_ROUND_UP_ULL() so it also works on 32bit build.

Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Reviewed-by: Matt Roper 
Signed-off-by: Lucas De Marchi 
Link: https://patchwork.freedesktop.org/patch/msgid/20240119001612.2991381-2-lucas.demarchi@intel.com
(cherry picked from commit 7b5bdb447b14930b9ef3e39bd301937889c60c96)
Signed-off-by: Thomas Hellström

drm/xe: Use a NULL pointer instead of 0.

2024-01-24T10:12:59+00:00

The last argument of xe_pcode_read() is a pointer. Use NULL instead of 0.

Fixes: 92d44a422d0d ("drm/xe/hwmon: Expose card reactive critical power")
Cc: Rodrigo Vivi 
Signed-off-by: Thomas Hellström 
Reviewed-by: Francois Dugast 
Link: https://patchwork.freedesktop.org/patch/msgid/20240117134048.165425-6-thomas.hellstrom@linux.intel.com
(cherry picked from commit 79f8eacbdf9dad7ead39b3319e31e12d4dc6529e)
Signed-off-by: Thomas Hellström

drm/xe/dmabuf: Make xe_dmabuf_ops static

2024-01-24T10:12:45+00:00

It is not referenced outside of the xe_dma_buf.c source file.

Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Cc: Rodrigo Vivi 
Cc: Matthew Brost 
Signed-off-by: Thomas Hellström 
Reviewed-by: Francois Dugast 
Link: https://patchwork.freedesktop.org/patch/msgid/20240117134048.165425-2-thomas.hellstrom@linux.intel.com
(cherry picked from commit e2dc52f849f8694bdabb75127164c9df622af459)
Signed-off-by: Thomas Hellström

drm/xe: display support should not depend on EXPERT

2024-01-15T20:40:32+00:00

Remove the DRM_XE_DISPLAY config dependency on EXPERT. I can only
presume the idea was only experts should be able to disable it, but the
effect is the opposite.

Reported-by: Eero Tamminen 
Reviewed-by: Francois Dugast 
Signed-off-by: Jani Nikula 
Link: https://patchwork.freedesktop.org/patch/msgid/20240111104716.3548744-1-jani.nikula@intel.com
(cherry picked from commit 1c7531f50eaa425eca8ff726287b8df3a4a51e55)
Signed-off-by: Thomas Hellström

drm/xe: Fix bounds checking in __xe_bo_placement_for_flags()

2024-01-15T14:37:03+00:00

Requesting all memory regions on PVC will fill bo->placements up to
XE_BO_MAX_PLACEMENTS. The subsequent call to try_add_stolen() will trip
over the bounds checking even though XE_PL_STOLEN is not expected to
be used in this case.

This is hit with igt@xe_exec_fault_mode@once-basic-prefetch:
    xe 0000:8c:00.0: [drm] Assertion `*c < (sizeof(bo->placements) / sizeof((bo->placements)[0]) + ((int)(sizeof(struct { int:(-!!(__builtin_types_compatible_p(typeof((bo->placements)), typeof(&(bo->placements)[0])))); }))))` failed!
    WARNING: CPU: 30 PID: 6161 at drivers/gpu/drm/xe/xe_bo.c:203 __xe_bo_placement_for_flags+0x218/0x240 [xe]

Is fixed here by moving the bounds checks closer to where we actually
write into the bo->placement array.

Fixes: 8c54ee8a8606 ("drm/xe: Ensure that we don't access the placements array out-of-bounds")
Link: https://patchwork.freedesktop.org/patch/msgid/20240111002111.10190-1-brian.welty@intel.com
Signed-off-by: Matthew Brost 
Signed-off-by: Brian Welty 
Reviewed-by: Matthew Brost 
(cherry picked from commit 52e3fa3e3ea3ee05e32c1a8d72bb3ae306a4da64)
Signed-off-by: Thomas Hellström