There are several problems in the current logic:
- it doesn't account for the WSI stride, and just assumes things are
tightly packed and the resource extent is aligned on a tile
- the format re-interpretation done in panfrost_mtk_detile_compute()
where the view format and the image format have a different block size
is not supposed to be supported, and that's something we'd like
to enforce now
- we write to textures that may have no descriptors bound to. This works
thanks to the robustness behind image stores/loads, but we'd probably
rather discard a bunch of instructions when Y/UV planes are copied
separately
- the linear to tiled coordinates conversion can be simplified by using
a combination of local_invocation_id+workgroup_id instead of the
global_invocation_id
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Tested-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Eric R. Smith <eric.smith@collabora.com>
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35317>
Those two types are completely different, and they don't use the same
key attributes. Let's provide two different helpers, and place the key
bits that are AFBC-specific into an struct inside a union. We also add
the modifier to the key to prevent AFBC/MTK_TILED collisions.
Not only this make things clearer, but we also no longer compile
MTK detiling shaders when AFBC shaders are requested (or the other way
around).
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35317>
Now that we use a precomp shader for indirect dispatch,
we can safely remove this.
Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32839>
Now that we have an implementation there, we can just call it.
This pass "dEQP-GLES31.functional.compute.indirect_dispatch.*".
Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32839>
This switch pan_perge to a macro to keep address space
information when working with OpenCL C.
Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32839>
Load the format enum and then just go through a series of :
if format == R16G16B16A16_UNORM
color = lower_r32g32_uint_tor_r16g16b16a16_unorm(color)
else if format == R16G16B16A16_SNORM
...
For Gfx12.5, there is no in-shader conversion.
For Gfx12/11, the in-shader conversion covers the following formats :
- ISL_FORMAT_R10G10B10A2_UNORM
- ISL_FORMAT_R10G10B10A2_UINT
- ISL_FORMAT_R11G11B10_FLOAT
For Gfx9, the following formats :
- ISL_FORMAT_R16G16B16A16_UNORM
- ISL_FORMAT_R16G16B16A16_SNORM
- ISL_FORMAT_R10G10B10A2_UNORM
- ISL_FORMAT_R10G10B10A2_UINT
- ISL_FORMAT_R8G8B8A8_UNORM
- ISL_FORMAT_R8G8B8A8_SNORM
- ISL_FORMAT_R16G16_UNORM
- ISL_FORMAT_R16G16_SNORM
- ISL_FORMAT_R11G11B10_FLOAT
- ISL_FORMAT_R8G8_UNORM
- ISL_FORMAT_R8G8_SNORM
- ISL_FORMAT_R16_UNORM
- ISL_FORMAT_R16_SNORM
- ISL_FORMAT_R8_UNORM
- ISL_FORMAT_R8_SNORM
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22524>
At draw time, if the number of BOs is bigger than 2048, the current
job submission is forced.
The 2048 limit has been validated to be big enough to not be reached
in most of the scenarios. Only a couple of CTS tests get over this
threshold.
So the new V3D_JOB_MAX_BO_HANDLE_COUNT is defines as 2048 and
V3D_JOB_MAX_BO_REFERENCED_SIZE is defined as 768MB.
This forced submission is useful to handle scenarios where the client
application is not calling glFlush() or where SwapBuffers() is a NOP
because of not having a window surface. In this case, the CLE can
grow indefinitely until the system runs out of memory resources.
This approach is followed by different drivers forcing the flush
of CL when it reaches a defined size because of HW limitations.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12227
Cc: mesa-stable
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Maíra Canal <mcanal@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35042>
Mark when at least one job for the current active FBO has already been
submitted since the last framebuffer state update.
With this we can apply TLB load invalidation only to the first
job that is submitted to the same FBO. Not applying TLB
loads invalidation on follow-up jobs targeting the same framebuffer
state.
With this we avoid doing incorrect invalidations when we force
a job submission for a reason not related with a new framebuffer bind.
Cc: mesa-stable
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35042>
We were allocating these, but never freeing the actual CSOs here.
Let's wire things up so we delete the data when we destroy the
hash-table. Because we don't have access to the context in that
callback, we can't call the pipe-level function to delete a CSO,
but luckily we don't actually need the context for the
driver-logic. So let's add an internal helper for that.
Fixes: ae3fb3089f ("panfrost: Add infrastructure for internal AFBC compute shaders")
Fixes: f39194cdd3 ("panfrost: support MTK 16L32S detiling")
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35336>
We were inserting twice here, once without holding the lock before
compiling the shaders, and once after while holding it. Let's remove
the insert without the lock.
Fixes: ae3fb3089f ("panfrost: Add infrastructure for internal AFBC compute shaders")
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35336>
This change:
1. Rename pan_image_wsi_layout to pan_image_layout_constraints.
1. Document that pan_image_layout_constraints::offset_B will be used to
pass planar plane offset for native images as well. Update
array_stride_B to exclude such offset accordingly.
2. Document that whether explicit layout is used further depends on if
the passed wsi_row_pitch_B is non-zero or not. Updated the checks to
base on the new boolean use_explicit_layout.
3. Update and document the intended slice offset_B behavior so that
native and imported images are aligned.
No behavior change from this commit since the client (panvk) hasn't set
the pan_image_layout_constraints->offset_B for planar plane yet.
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35299>
This change splits pan_image_layout_get_wsi_layout into separate queries
to prepare for making pan_image_wsi_layout an input layout constraint.
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35299>
Added a helper function MFAttachPipeResourceAsSampleExtension that converts pipe_resource into D3D12 resource, attached as IMFMediaBuffer to IMFSample.
Conditionally applied MFAttachPipeResourceAsSampleExtension to:
MFSampleExtension_FramePsnrYuv
MFSampleExtension_VideoEncodeQPMap
MFSampleExtension_VideoEncodeBitsUsedMap
Reviewed-by: Yubo Xie <yuboxie@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35264>
This code sets PANFROST_JD_REQ_CYCLE_COUNT when using LD_GCLK to enable
shader_clock
Fixes: c1ce2dcc ("pan/bi: Enable ARB_shader_clock extension support")
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Signed-off-by: Ashley Smith <ashley.smith@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35363>
Make AFBC work for NV12 and similar YUV formats that we already
supported.
Signed-off-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35015>
The way things are done at the moment, the pan_layout logic has no way
of knowing which plane is targeted, which either forces the user of
pan_image_layout_init() to patch the layout after the calculation based
on the YUV format.
These changes unblock unittests on YUV formats.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Tested-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35015>
So far we assumed tile rows would be tightly packed, so changing the
behavior without introducing a new set of AFBC modifiers is likely
to break some cases where the exported got the stride wrong. That's
still something we would like to enforce in the long run, so let's
start by adding a "PAN_MESA_DEBUG=strict_afbc" flag to exercise the
code supposed to force this behavior.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Tested-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35015>
Mali GPUs don't support DRM_FORMAT_MOD_MTK_16L_32S_TILE natively, so
it doesn't make sense to teach the common layer about this format,
especially since it would require adding more hacks to make sense of
the tiled -> linear view of such resources.
Given the only user of this format is the gallium driver, move the
MTK_TILED definitions there and start differentiating native and
emulated modifiers.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Tested-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35015>
Add one GLES 2 test to run in the deqp-runner suite of the
android-angle-lavapipe job to verify the version and renderer used.
Signed-off-by: Valentine Burley <valentine.burley@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35337>
This reverts commit ca087e2027.
It's discovered that the newest Rogue proprietary Vulkan driver does not
prevent us from running Zink (aside from an assert of geometryShader,
which is already changed to an if clause), and which fails previously is
because of some bugs in old driver blobs.
As the faulty blob versions are rejected now, revert this commit to
accept Rogue drivers.
Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34183>
Imagination closed blobs earlier than 24.1@6554834 has some issues about
NULL pCommandBuffers in VkSubmitInfo, which is present in VkSubmitInfo's
submitted by Zink.
Reject these drivers to prevent crashes, unless zink is enforced and the
user knows what they're doing.
Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34183>
The proprietary driver for Imagination Rogue-architecture GPUs does not
come with geometryShader support.
Change the assert for it to another if condition.
Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34183>
Make svga piglit job extend from vmware automatic (non-manual) rules pool
again. There was an issue with local nginx caching of FDO artifacts, which
led to jobs sporadically timing out on their http-download sections; the
issue has been resolved now.
Signed-off-by: Martin Krastev <martin.krastev@broadcom.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35341>
