Alyssa Rosenzweig [Thu, 9 Jul 2020 21:35:24 +0000 (17:35 -0400)]
panfrost: Overhaul tilebuffer allocations
Based on the colour buffers in use, we need to select a tile size
allowing either 128-bits of storage per pixel or 512-bits. Based on the
size chosen, we scale the offsets into the tilebuffer. Likewise, we need
to calculate offsets based on bpp (with special cases) rather than
picking an average case.
Fixes regressions that otherwise would be caused by the next commit.
v2: Fix colour clears (Icecream95).
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5858>
Alyssa Rosenzweig [Tue, 14 Jul 2020 20:58:57 +0000 (16:58 -0400)]
panfrost: Call util_blitter_save_fragment_constant_buffer_slot
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5858>
Dave Airlie [Mon, 13 Jul 2020 05:44:24 +0000 (15:44 +1000)]
llvmpipe: fix position offset interpolation
pos offset only applies to the gl_FragPos input, when I refactored
I messed that up, only use pos_offset for the position inputs
and use 0.5 otherwise.
This fixes:
GTF-GL45.gtf30.GL3Tests.fragment_coord_conventions.fragment_coord_conventions_multisample
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5926>
Dave Airlie [Tue, 3 Dec 2019 06:01:37 +0000 (16:01 +1000)]
llvmpipe: fix stencil only formats.
Currently the test crashes with LLVM errors
Stored value type does not match pointer operand type!
store <8 x i32> %s_dst, <8 x i8>* %261
Change the stored type for 8-bit stencil formats.
Fixes:
GTF-GL45.gtf44.GL31Tests.texture_stencil8.texture_stencil8_gl44
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5926>
Thong Thai [Fri, 10 Jul 2020 18:03:26 +0000 (14:03 -0400)]
radeonsi: use PIPE_FORMAT_P010 for 10-bit VP9 decoding
Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leoliu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5848>
Rhys Perry [Thu, 16 Jul 2020 10:08:50 +0000 (11:08 +0100)]
aco: add add_interference() helper
This won't add interferences between spill ids of different types and will
exit early if there's already an interference.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5805>
Rhys Perry [Tue, 7 Jul 2020 12:11:07 +0000 (13:11 +0100)]
aco: use unordered_set for spill id interferences
Seems to be faster.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5805>
Rhys Perry [Tue, 7 Jul 2020 12:10:38 +0000 (13:10 +0100)]
aco: rewrite graph coloring in spiller
I don't think this is much of an optimization in the typical case, but for
very complex shaders this should work much better.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5805>
Rhys Perry [Tue, 7 Jul 2020 17:16:47 +0000 (18:16 +0100)]
aco: fix underestimated pressure in spiller when a phi has a killed def
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5805>
Alyssa Rosenzweig [Thu, 9 Jul 2020 17:42:25 +0000 (13:42 -0400)]
panfrost: Use Midgard-specific reloads
v2: Be more explicit about sampler types. Prefer the term "load" to
"resolve" to match VK convention. Generate shaders for MRT 8x. Blit
shader generation adds about 6ms to startup cost. We could cache thes.
shaders to disk if we needed to (or indeed, ship binaries).
v3: Fallback on u_blitter on Bifrost so Bifrost continues to work.
KHR_partial_update support is mostly no-oped on Bifrost now, but that's
okay for now - compositors are still functional.
v4: Specialize on multisample state as well to enable reloads of MSAA
textures. This requires 2x the shader variants, so I assume we're up to
12ms startup cost for generation. Annoying. Also fix interactions with
depth- or stencil-only clears of combined depth-stencil surfaces.
v5: Cache to the device (screen) instead of the context, reducing
duplicated work in apps that create many contexts (e.g. Chromium)
v6: Squash in KHR_partial_update cleanup to fix intermediate
regressions on a few tests.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5824>
Indrajit Kumar Das [Fri, 10 Jul 2020 06:15:22 +0000 (11:45 +0530)]
st/mesa: use fragment shader to copy stencil buffer
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5836>
Erik Faye-Lund [Fri, 10 Jul 2020 11:12:40 +0000 (13:12 +0200)]
mesa/main: use p_atomic_inc_return instead of locking
There's no good reason for using a mutex here, as we have a simpler
primitive; atomic integers. So let's use that instead, to simplify
things a bit.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5901>
Yevhenii Kharchenko [Wed, 1 Jul 2020 14:36:18 +0000 (17:36 +0300)]
st/mesa: fix corrupted texture levels, when adding more levels than expected
Some of existing texture levels can be corruted,
after calling 'glTexImage' with param 'level' higher than
max expected value 'floor(log2(max(width, height, depth)))'.
To fix we prevent overwriting image buffer pointer
in 'st_texture_object', if it was already allocated
for multiple mip-levels storage.
Fixes piglit test: 'arb_copy_image add-illegal-levels'
Signed-off-by: Yevhenii Kharchenko <yevhenii.kharchenko@globallogic.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5785>
Erik Faye-Lund [Fri, 10 Jul 2020 10:31:51 +0000 (12:31 +0200)]
gallium/util: do not use _MTX_INITIALIZER_NP on Windows
We already have another way of initializing these, so it's just a matter
of avoiding _MTX_INITIALIZER_NP here.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5902>
Samuel Pitoiset [Tue, 14 Jul 2020 09:16:45 +0000 (11:16 +0200)]
radv/winsys: pass the buffer list via the CS ioctl for less CPU overhead
The legacy path requires one more ioctl to create the buffer
list and this is more costly for the CPU.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5899>
Samuel Pitoiset [Tue, 14 Jul 2020 07:48:56 +0000 (09:48 +0200)]
radv/winsys: replace alloca() by malloc() everywhere
To remove the mix of alloca() and malloc().
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5899>
Neil Roberts [Tue, 7 Jul 2020 13:49:27 +0000 (15:49 +0200)]
v3d/compiler: Lower geometry output store base into offset src
When generating the VPM write instruction for geometry shader outputs,
emit_store_output_gs ends up adding the base and offset arguments
together with an ADD instruction. The addition was done at the VIR level
after scheduling so it always ends up right next to the corresponding
stvpm instruction. Most of the time the offset is constant but nothing
does any constant folding at the VIR level.
This patch makes it instead fold the addition into the offset at the NIR
level in v3d_nir_lower_io so that the NIR-level constant folding can get
rid of the addition most of the time.
v2: Use nir_iadd_imm to simplify the code. (Eric Anholt)
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5825>
Marek Olšák [Tue, 30 Jun 2020 20:57:23 +0000 (16:57 -0400)]
radeonsi: prevent a gfx10_ngg_calculate_subgroup_info failure for TES+NGG GS
arb_tessellation_shader-tes-gs-max-output -small -scan 1 50 -auto -fbo
doesn't pass, but at least all shaders are compiled successfully.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5700>
Dave Airlie [Thu, 16 Jul 2020 02:49:28 +0000 (12:49 +1000)]
ci: bump piglit checkout for dsa tests
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5896>
Dave Airlie [Tue, 14 Jul 2020 03:15:36 +0000 (13:15 +1000)]
mesa: change dsa texture error codes for GL 4.6
GL 4.6 changed error code for when the effective target of the
texture is illegal. Since it's not an illegal enum they modified
it to be an illegal operation.
However the CTS test for this is missing support for two cases,
I'm chasing that up, but I expect this will cause a CTS regression
for anyone who runs this test. I'm leaning on the side of being
compliant rather than passing the test until the test is fixed.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5896>
Alyssa Rosenzweig [Wed, 15 Jul 2020 17:10:02 +0000 (13:10 -0400)]
panfrost: Extract panfrost_batch_reserve_framebuffer
We need to trigger it explicitly for reloads without draws (for Z^S
reload which is an edge case).
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5929>
Alyssa Rosenzweig [Wed, 15 Jul 2020 21:35:58 +0000 (17:35 -0400)]
panfrost: Track surfaces drawn per-batch
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5929>
Alyssa Rosenzweig [Wed, 15 Jul 2020 16:02:32 +0000 (12:02 -0400)]
panfrost: Set zs_samples as necessary
Fixes MSAA Z/S.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5929>
Alyssa Rosenzweig [Wed, 15 Jul 2020 15:38:39 +0000 (11:38 -0400)]
panfrost: Handle per-sample shading
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5929>
Alyssa Rosenzweig [Mon, 13 Jul 2020 23:17:33 +0000 (19:17 -0400)]
panfrost: Add rectangle subtraction algorithm
For better supporting KHR_partial_update.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5929>
Alyssa Rosenzweig [Wed, 15 Jul 2020 15:57:35 +0000 (11:57 -0400)]
panfrost: Identify zs_samples field
For MSAA depth/stencil.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5929>
Alyssa Rosenzweig [Wed, 15 Jul 2020 15:39:08 +0000 (11:39 -0400)]
panfrost: Include sample count in payload estimates
Otherwise we might not reserve enough space.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5929>
Alyssa Rosenzweig [Wed, 15 Jul 2020 01:36:59 +0000 (21:36 -0400)]
panfrost: Add MALI_PER_SAMPLE bit
For gl_SampleID reads.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5929>
Alyssa Rosenzweig [Thu, 9 Jul 2020 17:41:15 +0000 (13:41 -0400)]
panfrost: Expose panfrost_get_blend_shader
It is needed to produce a blend shader for blits.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5929>
Alyssa Rosenzweig [Fri, 3 Jul 2020 17:17:35 +0000 (13:17 -0400)]
panfrost: Force Z/S writeback
This is unfortunately necessary for conformance at this stage.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5929>
Alyssa Rosenzweig [Wed, 15 Jul 2020 13:56:24 +0000 (09:56 -0400)]
pan/mdg: Implement gl_SampleID
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5929>
Alyssa Rosenzweig [Wed, 15 Jul 2020 01:39:06 +0000 (21:39 -0400)]
pan/mdg: Identify per-sample interpolation mode
So this is what .interp0 was this whole time.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5929>
Alyssa Rosenzweig [Fri, 10 Jul 2020 19:56:17 +0000 (15:56 -0400)]
pan/mdg: Bump compiler RT maximum
We don't actually support MRT 8x yet but we would like to soon, so bump
it in the compiler.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5929>
Roman Stratiienko [Wed, 15 Jul 2020 19:26:44 +0000 (22:26 +0300)]
Android: Fixes for Q and R
Fix Android-Q build:
- Use AOSP prebuilt bison by specifying $(BISON) variable
- Use AOSP prebuilt flex by specifying $(LEX) variable
Fix Android-R build:
- Add M4 environmet variable for Android R and higher (See [1])
[1] - https://cs.android.com/android/_/android/platform/build/+/
2bfffb9f48a78de12faf5da77424c0cecb70d6eb:Changes.md;dlc=
997661002af1282d938e88c3c723037e42e5d283
Signed-off-by: Roman Stratiienko <r.stratiienko@gmail.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Tested-by: Mauro Rossi <issor.oruam@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5894>
Dave Airlie [Tue, 14 Apr 2020 04:00:11 +0000 (14:00 +1000)]
llvmpipe/format: fix snorm conversion
This fixes:
GTF-GL45.gtf33.GL3Tests.vertex_type_2_10_10_10_rev.vertex_type_2_10_10_10_rev_conversion
and
piglit
spec/arb_texture_view/rendering-formats/clear gl_rgba8_snorm as gl_r32f
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5820>
Dave Airlie [Tue, 14 Apr 2020 05:28:05 +0000 (15:28 +1000)]
gallivm/sample: always square rho before fast log2
The fast log2 works better if rho is squared, i.e.
fast_log2(sqrt(2)) == 0.4
0.5 * fast_log2(2) == 0.5
so just square rho, and always divide by 2 afterwards.
Fixes:
GTF-GL45.gtf30.GL3Tests.sgis_texture_lod.sgis_texture_lod_basic_lod_selection
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5820>
Rhys Kidd [Wed, 15 Jul 2020 13:04:32 +0000 (23:04 +1000)]
nv50_2d: regenerate envytools-based rnndb headers
The headers hadn't been regenerated from envytools in a long time,
and there were a few minor divergences.
Based on envytools commit
c20929ed0f3be18b8419f7332ee22d905feb6589
Among other things, rnndb has changed naming to G80/etc, for now
I've not tackled switching that over and replaced the nvidia
codenames back to the chip ids that mesa uses with the following:
$ sed -i 's/G80_2D/NV50_2D/g' rnndb/graph/g80_2d.xml.h
$ sed -i 's/GF100_2D/NVC0_2D/g' rnndb/graph/g80_2d.xml.h
No other modifications of the headergen'd headers was done, which
was helped by the differing #define's being unutilised presently.
Signed-off-by: Rhys Kidd <rhyskidd@gmail.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5920>
Samuel Pitoiset [Sun, 12 Jul 2020 11:40:14 +0000 (13:40 +0200)]
radv: destroy the base object if VkCreateInstance() failed
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5868>
Samuel Pitoiset [Sun, 12 Jul 2020 11:34:39 +0000 (13:34 +0200)]
radv: destroy the base object if VkAllocateCommandBuffers() failed
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5868>
Samuel Pitoiset [Fri, 10 Jul 2020 15:29:21 +0000 (17:29 +0200)]
radv: destroy the base object if VkCreateFence() failed
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5868>
Samuel Pitoiset [Fri, 10 Jul 2020 15:25:00 +0000 (17:25 +0200)]
radv: destroy the base object if VkCreateSemaphore() failed
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5868>
Samuel Pitoiset [Fri, 10 Jul 2020 15:14:30 +0000 (17:14 +0200)]
radv: destroy the base object if VkCreateEvent() failed
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5868>
Samuel Pitoiset [Fri, 10 Jul 2020 15:11:10 +0000 (17:11 +0200)]
radv: destroy the base object if VkCreateBuffer() failed
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5868>
Samuel Pitoiset [Fri, 10 Jul 2020 14:18:37 +0000 (16:18 +0200)]
radv: destroy the base object if VkCreateImage() failed
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5868>
Samuel Pitoiset [Fri, 10 Jul 2020 14:04:26 +0000 (16:04 +0200)]
radv: destroy the base object if VkCreateRenderPass*() failed
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5868>
Samuel Pitoiset [Fri, 10 Jul 2020 13:52:38 +0000 (15:52 +0200)]
radv: destroy the base object if VkCreateQueryPool() failed
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5868>
Pierre-Eric Pelloux-Prayer [Tue, 7 Jul 2020 09:08:50 +0000 (11:08 +0200)]
mesa/st: release debug_output after destroying the context
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3230
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/2218
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5789>
Pierre-Eric Pelloux-Prayer [Tue, 7 Jul 2020 09:06:21 +0000 (11:06 +0200)]
mesa: add bool param to _mesa_free_context_data
The param controls whether _mesa_destroy_debug_output should be called or not.
No functional changes; this will be used by the next commit.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5789>
Pierre-Eric Pelloux-Prayer [Tue, 7 Jul 2020 09:00:58 +0000 (11:00 +0200)]
mesa: rename _mesa_free_errors_data
Use the _mesa_init_XXX / _mesa_destroy_XXX pattern to clearly associate
the 2 functions.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5789>
Michel Dänzer [Tue, 14 Jul 2020 09:12:19 +0000 (11:12 +0200)]
gitlab-ci: Fix "triggered by Marge for a merge request" rule
The commit below changed the rule such that it accidentally also applied
to the non-MR pipelines created by Marge, resulting in Marge triggering
twice as many jobs as necessary.
Fixes: 549b4a3dd4ab "gitlab-ci: Automatically run pipelines for Marge
Bot pre-merge only"
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5898>
Anuj Phogat [Thu, 9 Jul 2020 22:56:25 +0000 (15:56 -0700)]
intel/ehl: Add new PCI-IDs
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Anuj Phogat [Thu, 9 Jul 2020 22:15:13 +0000 (15:15 -0700)]
intel/ehl: Rename gen_device_info struct
Renaming makes it easier to relate a pciid with device configuration.
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Anuj Phogat [Thu, 9 Jul 2020 22:47:36 +0000 (15:47 -0700)]
intel/ehl: Use macro GEN11_LP_FEATURES in device info
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Anuj Phogat [Thu, 9 Jul 2020 22:00:21 +0000 (15:00 -0700)]
intel/ehl: Use GEN11_URB_MIN_MAX_ENTRIES in device info
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Icecream95 [Thu, 25 Jun 2020 10:56:14 +0000 (22:56 +1200)]
panfrost: Dual source blend support
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5620>
Icecream95 [Fri, 10 Jul 2020 11:28:21 +0000 (23:28 +1200)]
pan/mdg: Skip z/s combining for dual-source writes
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5620>
Icecream95 [Thu, 25 Jun 2020 10:21:50 +0000 (22:21 +1200)]
pan/mdg: Dual source blend input/writeout support
We write to r2, which is preseved through to the blend shader, from
where it is read. We won't worry about MRT to keep things simple.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5620>
Icecream95 [Mon, 6 Jul 2020 11:52:40 +0000 (23:52 +1200)]
pan/mdg: Add a nir pass to reorder store_output intrinsics
Real writeout stores, which break execution, need to be moved to after
dual-source stores, which are just standard register writes.
v2: Don't move stores forward, to avoid moving them to before where
their source is written.
v3: Only reorder past dual-source stores.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5620>
Icecream95 [Thu, 25 Jun 2020 10:50:15 +0000 (22:50 +1200)]
gallium: Dual source support in blend_factor_to_shader
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5620>
Icecream95 [Thu, 25 Jun 2020 10:48:32 +0000 (22:48 +1200)]
compiler: Add dual-source factors to blend_factor
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5620>
Rob Clark [Tue, 14 Jul 2020 17:36:04 +0000 (10:36 -0700)]
freedreno/ir3: DCE unused arrays
Letting unused arrays stick around confuses RA, which assigns vreg names
to the unused arrays, but then does not precolor them (because they are
unused). This leads to an assert in ra_select_reg_merged():
skqp: ../src/freedreno/ir3/ir3_ra.c:589: name_to_instr: Assertion '!name_is_array(ctx, name)' failed.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3262
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5907>
Rob Clark [Mon, 13 Jul 2020 21:49:01 +0000 (14:49 -0700)]
freedreno/a6xx: don't enable early-z/lrz if no z-test
But if shader explicitly asks for early-fragment-tests, obey it's
wishes.
Fixes a handful of skia (skqp) CTS fails (9.0_r12)
* gles_bug593049
* gles_circular_arcs_fill
* gles_circular_arcs_stroke_and_fill_square
* gles_clippedcubic2
* gles_complexclip2_path_bw
* gles_complexclip2_rrect_bw
* gles_complexclip3_complex
* gles_complexclip3_simple
* gles_crbug_691386
* gles_cubicclosepath
* gles_cubicpath
* gles_degeneratesegments
* gles_filltypespersp
* gles_innershapes_bw
* gles_inverse_paths
* gles_mixedtextblobs
* gles_onebadarc
* gles_quadclosepath
* gles_quadpath
* gles_rrect_clip_bw
* gles_scale-pixels
* gles_scaledstrokes
* gles_squarehair
* gles_strokes_zoomed
* gles_windowrectangles
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5907>
Rob Clark [Sat, 11 Jul 2020 21:22:08 +0000 (14:22 -0700)]
freedreno/ir3/ra: be better at failing
It doesn't happen much. But it's annoying when we hit an impossible
condition deep in RA 90% thru a long test run. Add some ra_assert()/
ra_unreachable() helper macros so we can bail cleanly and fail RA.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5907>
Rob Clark [Sat, 11 Jul 2020 20:27:36 +0000 (13:27 -0700)]
freedreno/a6xx: bail instead of crash for compile fails
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5907>
Rob Clark [Sat, 11 Jul 2020 20:21:36 +0000 (13:21 -0700)]
freedreno/ir3: make compile fails more visible
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5907>
Rob Clark [Sat, 11 Jul 2020 20:21:57 +0000 (13:21 -0700)]
freedreno/ir3: add missing VS driver params
Some of these only used by turnip so far, this is just for clarity.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5907>
Mike Blumenkrantz [Wed, 17 Jun 2020 21:40:04 +0000 (17:40 -0400)]
zink: enable tgsi texcoord pipe cap
this requires some modifications to the ntv slot remapping, as we definitely
don't want to reserve another 25% of the available slots for the (deprecated)
texcoord varyings
now we remap VARYING_SLOT_TEX[n] to the last available slots, which lets us avoid
needing to do permanent reservation, and we check to make sure that we haven't
seen the corresponding texcoord varying any time we emit a non-texcoord varying in
that slot
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5551>
Karol Herbst [Tue, 5 Jun 2018 12:12:55 +0000 (14:12 +0200)]
nouveau: expose HMM
v2: moved caps
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Pierre Moreau <dev@pmoreau.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5906>
Karol Herbst [Wed, 27 May 2020 14:29:06 +0000 (16:29 +0200)]
nvc0: support PIPE_CAP_RESOURCE_FROM_USER_MEMORY_COMPUTE_ONLY
v2: rework by adding a new buffer status
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5906>
Karol Herbst [Tue, 5 May 2020 13:09:50 +0000 (15:09 +0200)]
gallium: add PIPE_CAP_RESOURCE_FROM_USER_MEMORY_COMPUTE_ONLY
With the current UAPI we only support user pointers from the compute
engines, so we need a way to express that in gallium.
v2: fix typos
v3: add allows_user_pointers helper
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Pierre Moreau <dev@pmoreau.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5906>
Karol Herbst [Sun, 29 Jul 2018 20:06:18 +0000 (22:06 +0200)]
nouveau: enable HMM
v2: move declarations into libdrm
v3: fix typos
rework handling of how much memory to reserve
v4: remove unused parameter
unmap cutout on error and when the screen is destroyed
v5: move into screen_create
enable HMM only if CL gets enabled
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5906>
Karol Herbst [Tue, 30 Jun 2020 10:26:43 +0000 (12:26 +0200)]
ci: bump libdrm to 2.4.102
Since version 2.4.101 there are only xz archives hence the bz2 to xz
change.
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5906>
Karol Herbst [Tue, 14 Jul 2020 15:49:00 +0000 (17:49 +0200)]
ci: need to install wget in order to download libdrm
Fixes: dcd171f5e9b ("gitlab-ci: More stable URL for kernel and ramdisk artifacts, for LAVA")
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5906>
Jesse Natalie [Tue, 26 May 2020 19:21:33 +0000 (12:21 -0700)]
nir: Support load/store of temps as scratch in nir_lower_explicit_io
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5889>
Jesse Natalie [Tue, 26 May 2020 19:20:20 +0000 (12:20 -0700)]
nir: When nir_lower_vars_to_explicit_types is run on temps, update scratch_size
To allow interop with other scratch ops, append any remaining temp vars
to the end of any already-allocated scratch space.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5889>
Jesse Natalie [Tue, 14 Jul 2020 17:38:32 +0000 (10:38 -0700)]
nir_lower_io: Add addr_format_is_offset helper
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5889>
Jonathan Marek [Sat, 11 Jul 2020 19:55:12 +0000 (15:55 -0400)]
freedreno/regs: update a6xx PC regs
Update some registers in the 0x9800-0xa000 range.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5870>
Jonathan Marek [Sat, 11 Jul 2020 17:03:41 +0000 (13:03 -0400)]
freedreno/regs: update a6xx VPC regs
Update some registers in the 0x9000-0x95ff range.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5870>
Jonathan Marek [Sat, 11 Jul 2020 16:29:38 +0000 (12:29 -0400)]
freedreno/regs: update a6xx RB regs
Update some registers in the 0x8c00-0x8dff range.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5870>
Jonathan Marek [Thu, 9 Jul 2020 18:59:04 +0000 (14:59 -0400)]
freedreno/regs: update a6xx GRAS registers
Update some registers in the 0x8000-0x87ff range.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5870>
Jonathan Marek [Tue, 14 Jul 2020 14:58:56 +0000 (10:58 -0400)]
gitlab-ci: re-enable arm64_a630_vk
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5904>
Jonathan Marek [Thu, 18 Jun 2020 20:24:26 +0000 (16:24 -0400)]
turnip: remove use of tu_cs_entry for draw states
The tu_cs_entry struct doesn't match well what we want for SET_DRAW_STATE
and CP_INDIRECT_BUFFER (requires extra steps to get iova and size), so
start phasing it out.
Additionally, use newly added tu_cs_draw_state where it doesn't require any
effort (it requires a fixed size, but gets rid of the extra end_sub_stream)
Note this also changes the behavior of CmdBindDescriptorSets for compute to
emit directly in cmd->cs instead of doing through a CP_INDIRECT.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5558>
Jonathan Marek [Sat, 20 Jun 2020 00:43:50 +0000 (20:43 -0400)]
turnip: fix inconsistencies with tu6_load_state_size
The next patch assumes the correct size is returned in tu6_emit_load_state.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5558>
Jonathan Marek [Thu, 18 Jun 2020 19:41:31 +0000 (15:41 -0400)]
turnip: emit compute pipeline directly in CmdBindPipeline
There's no need to defer it, and can get rid DIRTY_COMPUTE_PIPELINE.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5558>
Jonathan Marek [Thu, 18 Jun 2020 17:07:48 +0000 (13:07 -0400)]
turnip: use DIRTY SDS bit to avoid making copies of pipeline load state ib
Some testing showed that the DIRTY bit has the desired behavior, so use it
to make things a bit simpler.
Note in CmdBindPipeline, having the TU_CMD_DIRTY_DESCRIPTOR_SETS behind a
if(pipeline->layout->dynamic_offset_count) was wrong.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5558>
Mike Blumenkrantz [Mon, 13 Jul 2020 20:35:10 +0000 (16:35 -0400)]
zink: try to handle multisampled null buffers
I don't have any tests for this that I've run into yet, so this is mostly
just guessing
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5686>
Mike Blumenkrantz [Fri, 29 May 2020 20:23:49 +0000 (16:23 -0400)]
zink: handle empty attachments
create an empty buffer and surface to reuse for the fb attachment here
this fixes most of the arb_framebuffer_object tests in piglit
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5686>
Icecream95 [Tue, 14 Jul 2020 03:06:09 +0000 (15:06 +1200)]
pan/mdg: Fix non-debug compiliation
Fixes error when the assert is optimized out:
../src/panfrost/midgard/midgard_compile.c: In function ‘output_load_rt_addr’:
../src/panfrost/midgard/midgard_compile.c:1644:1: error: control reaches end of non-void function [-Werror=return-type]
}
Closes #3270
Fixes: 7781d2c2ead ("pan/mdg: Support MRT in output load lowering")
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5895>
Michel Dänzer [Tue, 14 Jul 2020 15:08:10 +0000 (17:08 +0200)]
Revert https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4580
It broke the CI pipeline on master:
https://gitlab.freedesktop.org/mesa/mesa/-/jobs/
3604314
https://gitlab.freedesktop.org/mesa/mesa/-/jobs/
3604315
Revert for now, to allow other MRs to be merged.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5905>
Karol Herbst [Tue, 5 Jun 2018 12:12:55 +0000 (14:12 +0200)]
nouveau: expose HMM
v2: moved caps
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Pierre Moreau <dev@pmoreau.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4580>
Karol Herbst [Wed, 27 May 2020 14:29:06 +0000 (16:29 +0200)]
nvc0: support PIPE_CAP_RESOURCE_FROM_USER_MEMORY_COMPUTE_ONLY
v2: rework by adding a new buffer status
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4580>
Karol Herbst [Tue, 5 May 2020 13:09:50 +0000 (15:09 +0200)]
gallium: add PIPE_CAP_RESOURCE_FROM_USER_MEMORY_COMPUTE_ONLY
With the current UAPI we only support user pointers from the compute
engines, so we need a way to express that in gallium.
v2: fix typos
v3: add allows_user_pointers helper
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Pierre Moreau <dev@pmoreau.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4580>
Karol Herbst [Sun, 29 Jul 2018 20:06:18 +0000 (22:06 +0200)]
nouveau: enable HMM
v2: move declarations into libdrm
v3: fix typos
rework handling of how much memory to reserve
v4: remove unused parameter
unmap cutout on error and when the screen is destroyed
v5: move into screen_create
enable HMM only if CL gets enabled
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4580>
Karol Herbst [Tue, 30 Jun 2020 10:26:43 +0000 (12:26 +0200)]
ci: bump libdrm to 2.4.102
Since version 2.4.101 there are only xz archives hence the bz2 to xz
change.
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4580>
Mike Blumenkrantz [Sun, 14 Jun 2020 05:44:38 +0000 (01:44 -0400)]
zink: block resolve blits for depth/stencil buffers
"The format features of dstImage must contain VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT"
- vkCmdResolveImage spec
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5888>
Mike Blumenkrantz [Sun, 14 Jun 2020 04:30:55 +0000 (00:30 -0400)]
zink: block vkCmdBlitImage usage for multi sampled blits
this is prohibited by spec
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5888>
Mike Blumenkrantz [Sun, 14 Jun 2020 04:03:31 +0000 (00:03 -0400)]
zink: try copy_region hook for blits where we can't do a regular blit or resolve
in cases where the formats match, we can likely just pass this through for now
fixes a ton of spec@!opengl 1.1@depthstencil-default_fb-blit piglit tests
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5888>
Erik Faye-Lund [Fri, 10 Jul 2020 11:08:30 +0000 (13:08 +0200)]
mesa/main: use call_once instead of open-coding
We already have a utility for this, so let's use that instead.
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5879>
Erik Faye-Lund [Fri, 10 Jul 2020 11:05:32 +0000 (13:05 +0200)]
mesa/main: factor out one-time-init into a helper
This will make the next commit a bit cleaner.
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5879>
Karol Herbst [Mon, 13 Jul 2020 14:24:20 +0000 (16:24 +0200)]
gv100/ir: set ftz bit on floating point operations
Fixes Unigine Heavens ambient occlusion
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5881>
projects / mesa.git / log