nouveau_fence_emit() can fail before and after initializing the
dma-fence and hence before and after initializing the dma-fence' kref.
In order to avoid nouveau_fence_emit() potentially failing before
dma-fence initialization pass the channel to nouveau_fence_new() already
and perform the required check before even allocating the fence.
While at it, restore the original behavior of nouveau_fence_new() and
add nouveau_fence_create() for separate (pre-)allocation instead. Always
splitting up allocation end emit wasn't a good idea in the first place.
Hence, limit it to the places where we actually need to pre-allocate.
Fixes: 7f2a0b50b2 ("drm/nouveau: fence: separate fence alloc and emit")
Signed-off-by: Danilo Krummrich <dakr@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230829223847.4406-1-dakr@redhat.com
Currently, NO_PREFETCH is passed implicitly through
drm_nouveau_gem_pushbuf_push::length and drm_nouveau_exec_push::va_len.
Since this is a direct representation of how the HW is programmed it
isn't really future proof for a uAPI. Hence, fix this up for the new
uAPI and split up the va_len field of struct drm_nouveau_exec_push,
such that we keep 32bit for va_len and 32bit for flags.
For drm_nouveau_gem_pushbuf_push::length at least provide
NOUVEAU_GEM_PUSHBUF_NO_PREFETCH to indicate the bit shift.
While at it, fix up nv50_dma_push() as well, such that the caller
doesn't need to encode the NO_PREFETCH flag into the length parameter.
Signed-off-by: Danilo Krummrich <dakr@redhat.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230823181746.3446-1-dakr@redhat.com
If a sched job depends on a dma-fence from a job from the same GPU
scheduler instance, but a different scheduler entity, the GPU scheduler
does only wait for the particular job to be scheduled, rather than for
the job to fully complete. This is due to the GPU scheduler assuming
that there is a scheduler instance per ring. However, the current
implementation, in order to avoid arbitrary amounts of kthreads, has a
single scheduler instance while scheduler entities represent rings.
As a workaround, set the DRM_SCHED_FENCE_DONT_PIPELINE for all
out-fences in order to force the scheduler to wait for full job
completion for dependent jobs from different entities and same scheduler
instance.
There is some work in progress [1] to address the issues of firmware
schedulers; once it is in-tree the scheduler topology in Nouveau should
be re-worked accordingly.
[1] https://lore.kernel.org/dri-devel/20230801205103.627779-1-matthew.brost@intel.com/
Signed-off-by: Danilo Krummrich <dakr@redhat.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collaboralcom>
Link: https://patchwork.freedesktop.org/patch/msgid/20230811010632.2473-1-dakr@redhat.com
This panel has been implemented in commit eae7488814 ("drm/panel-simple:
Add Innolux G156HCE-L01 panel entry") with a higher clock than the typical
one mentioned on the documentation to avoid flickering on the unit
tested. Testing on a different unit shows that the panel actually works
with the intended 70.93 MHz clock and even lower frequencies so the
flickering is likely caused either by a defective unit or by other
different components such as the bridge.
Fixes: eae7488814 ("drm/panel-simple: Add Innolux G156HCE-L01 panel entry")
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Tested-by: Marek Vasut <marex@denx.de> # MX8MM with LT9211
Reviewed-by: Marek Vasut <marex@denx.de>
[narmstrong: fixed commit id in commit msg]
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230814134024.397739-1-luca.ceresoli@bootlin.com
amdgpu:
- SDMA 6.1.0 support
- SMU 13.x fixes
- PSP 13.x fixes
- HDP 6.1 support
- SMUIO 14.0 support
- IH 6.1 support
- Coding style cleanups
- Misc display fixes
- Initial Freesync panel replay support
- RAS fixes
- SDMA 5.2 MGCG updates
- SR-IOV fixes
- DCN3+ gamma fix
- Revert zpos properly until IGT regression is fixed
- NBIO 7.9 fixes
- Use TTM to manage the doorbell BAR
- Async flip fix
- DPIA tracing support
- DCN 3.x TMDS HDMI fixes
- FRU fixes
amdkfd:
- Coding style cleanups
- SVM fixes
- Trap handler fixes
- Convert older APUs to use dGPU path like newer APUs
- Drop IOMMUv2 path as it is no longer used
radeon:
- Coding style cleanups
drm buddy:
- Fix debugging output
UAPI:
- A new memory pool was added to amdgpu_drm.h since we converted doorbell BAR management to use TTM,
but userspace is blocked from allocating from it at this point, so kind of not really anything new
here per se
Signed-off-by: Dave Airlie <airlied@redhat.com>
# -----BEGIN PGP SIGNATURE-----
#
# iHUEABYKAB0WIQQgO5Idg2tXNTSZAr293/aFa7yZ2AUCZNahZwAKCRC93/aFa7yZ
# 2KNjAP0UV2vJZjrze7OQI/YoI+40UlGjS81nKGlMIN3eR8nzvAD/c9McLJViL82R
# idEAK7tsr/MaCKoPAlED7CkUZiHNlQw=
# =4w7I
# -----END PGP SIGNATURE-----
# gpg: Signature made Sat 12 Aug 2023 07:00:23 AEST
# gpg: using EDDSA key 203B921D836B5735349902BDBDDFF6856BBC99D8
# gpg: Can't check signature: No public key
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230811211554.7804-1-alexander.deucher@amd.com
Mediatek DRM Next for Linux 6.6
1. Small mtk-dpi cleanups
2. DisplayPort: support eDP and aux-bus
3. Fix uninitialized symbol
4. Do not check for 0 return after calling platform_get_irq()
5. Convert to platform remove callback returning void
6. Fix coverity issues
7. Fix potential memory leak if vmap() fail
8. Fix void-pointer-to-enum-cast warning
9. Rid W=1 warnings from GPU
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230813152726.14802-1-chunkuang.hu@kernel.org
Add additional pixel formats for which blending is disabled when
DRM_MODE_BLEND_PIXEL_NONE is set.
Refactor the fourcc selection into a separate function to handle the
increased number of formats.
Signed-off-by: Damian Hobson-Garcia <dhobsong@igel.co.jp>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
The (large) rcar_du_modeset_init() function can fail for many reasons,
two of two involving probe deferral. Use dev_err_probe() in those code
paths to record the cause of the probe deferral, in order to help
debugging probe issues.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
1. Fix build warning message in mtk_disp_ovl_adaptor.c
>> drivers/gpu/drm/mediatek/mtk_disp_ovl_adaptor.c:415:10:
warning: cast to smaller integer type 'enum mtk_ovl_adaptor_comp_type'
from 'const void *' [-Wvoid-pointer-to-enum-cast]
type = (enum mtk_ovl_adaptor_comp_type)of_id->data;
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1 warning generated.
2. Also fix the same warning message in mtk_drm_drv.c
>> drivers/gpu/drm/mediatek/mtk_drm_drv.c:832:15:
warning: cast to smaller integer type 'enum mtk_ddp_comp_type'
from 'const void *' [-Wvoid-pointer-to-enum-cast]
comp_type = (enum mtk_ddp_comp_type)of_id->data;
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1 warning generated.
Signed-off-by: Jason-JH.Lin <jason-jh.lin@mediatek.com>
Fixes: 453c336463 ("drm/mediatek: Add ovl_adaptor support for MT8195")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202305042054.ZtWME9OU-lkp@intel.com/
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20230621075421.1982-1-jason-jh.lin@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
We need the domains in amdgpu_drm.h for the kernel driver to manage
the pool, but we don't want userspace using it until the code
is ready. So reject for now.
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Use the dGPU path instead. There were a lot of platform
issues with IOMMU in general on these chips due to windows
not enabling IOMMU at the time. The dGPU path has been
used for a long time with newer APUs and works fine. This
also paves the way to simplify the driver significantly.
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Tested-by: Mike Lothian <mike@fireburn.co.uk>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Use the dGPU path instead. There were a lot of platform
issues with IOMMU in general on these chips due to windows
not enabling IOMMU at the time. The dGPU path has been
used for a long time with newer APUs and works fine. This
also paves the way to simplify the driver significantly.
v2: use the dGPU queue manager functions
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Tested-by: Mike Lothian <mike@fireburn.co.uk>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is (mostly) ignored
and this typically results in resource leaks. To improve here there is a
quest to make the remove callback return void. In the first step of this
quest all drivers are converted to .remove_new() which already returns
void.
Trivially convert the mediatek drm drivers from always returning zero in
the remove callback to the void returning variant.
Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jyri Sarha <jyri.sarha@iki.fi>
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20230801110239.831099-8-u.kleine-koenig@pengutronix.de/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
drm_exec_prepare_obj() and drm_exec_prepare_array() both reserve
dma-fence slots and hence a dma_resv_list without ever freeing it.
Make sure to call drm_gem_private_object_fini() for each GEM object
passed to drm_exec_prepare_obj()/drm_exec_prepare_array() throughout the
test to fix this up.
While at it, remove some trailing empty lines.
Fixes: 9710631cc8 ("drm: add drm_exec selftests v4")
Signed-off-by: Danilo Krummrich <dakr@redhat.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230809225034.8803-1-dakr@redhat.com
Signed-off-by: Christian König <christian.koenig@amd.com>
