Alyssa Rosenzweig [Mon, 1 Jun 2020 22:26:03 +0000 (18:26 -0400)]
panfrost: Prefer sysval for gl_PointCoord on Bifrost
It's like gl_FragCoord. Still not implemented. This unfortunately makes
point sprites a lot more complicated.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5290>
Alyssa Rosenzweig [Mon, 1 Jun 2020 21:36:35 +0000 (17:36 -0400)]
pan/bi: Disassemble gl_PointCoord reads.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5290>
Alyssa Rosenzweig [Tue, 2 Jun 2020 00:34:34 +0000 (20:34 -0400)]
panfrost: Explicitly convert to 32-bit for logic-ops
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reported-by: Icecream95 <ixn@keemail.me>
Fixes: 19b4e586f62 ("panfrost: Switch to pan_lower_framebuffer")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5289>
Alyssa Rosenzweig [Mon, 1 Jun 2020 22:32:41 +0000 (18:32 -0400)]
panfrost: Readd MIDGARD_SHADERLESS quirk to t760
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reported-by: Icecream95 <ixn@keemail.me>
Fixes: e53d27de61b ("panfrost: Add quirks for blend shader types")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5289>
Francisco Jerez [Fri, 29 May 2020 23:57:01 +0000 (16:57 -0700)]
iris: Extend iris_context dirty state flags to 128 bits.
We're nearly out of dirty bits, and some patches pending review on
GitLab no longer apply due to that. Make room for them by splitting
off shader stage-specific bits into a separate stage_dirty mask.
An alternative would be to split compute-related bits into a separate
mask, but that would prevent the '<< stage' indexing done in various
parts of the driver from working.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5279>
Francisco Jerez [Fri, 29 May 2020 23:54:35 +0000 (16:54 -0700)]
iris: Simplify iris_batch_prepare_noop().
This makes iris_batch_prepare_noop() return a boolean instead of
passing through the relevant set of dirty flags. It will make it
easier to change the representation of dirty flags.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5279>
Rob Clark [Wed, 3 Jun 2020 18:34:09 +0000 (11:34 -0700)]
nir/lower_tex: fixes for fp16 yuv lowering
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3079
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5318>
Rob Clark [Wed, 3 Jun 2020 19:12:54 +0000 (12:12 -0700)]
nir/builder: add bitsize conversion helpers
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5318>
Rob Clark [Wed, 3 Jun 2020 19:09:59 +0000 (12:09 -0700)]
nir: extract out convert_to_bitsize() helper
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5318>
Rob Clark [Wed, 3 Jun 2020 19:07:52 +0000 (12:07 -0700)]
nir: get_base_type() should return enum type
Needed by the next patch, for c++ code which is more strict about
conversions between integers and enums.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5318>
Alyssa Rosenzweig [Tue, 2 Jun 2020 18:12:29 +0000 (14:12 -0400)]
panfrost: Handle writes_memory correctly
We need to pass it thru to EARLY_Z and WRITES_GLOBAL instead of ignoring
and assuming respectively. Nontrivial performance fix.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5300>
Alyssa Rosenzweig [Tue, 2 Jun 2020 18:05:34 +0000 (14:05 -0400)]
panfrost: Document MALI_WRITES_GLOBAL bit
We've been setting this unconditionally -- oops!
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5300>
Alyssa Rosenzweig [Tue, 2 Jun 2020 18:03:58 +0000 (14:03 -0400)]
panfrost: Update MALI_EARLY_Z description
Via the ES3.1 early-z testing force, I've confirmed this bit is e-z.
I've also confirmed e-z must be disabled for global writes, as expected.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5300>
Marcin Ślusarz [Wed, 3 Jun 2020 15:00:38 +0000 (17:00 +0200)]
iris: remove unused iris_bo->swizzle_mode
Signed-off-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5313>
Samuel Pitoiset [Tue, 26 May 2020 14:21:44 +0000 (16:21 +0200)]
aco: sign-extend input/identity for 16-bit subgroup ops on GFX6-GFX7
16-bit subgroup ops are implemented with 32-bit instructions
on GFX6-GFX7.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5227>
Samuel Pitoiset [Tue, 26 May 2020 07:38:27 +0000 (09:38 +0200)]
aco: fix subdword copies on GFX6-GFX7
SDWA is only GFX8+. Use v_mov_b32 since the upper 16 bits don't matter.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5227>
Samuel Pitoiset [Tue, 26 May 2020 14:06:14 +0000 (16:06 +0200)]
aco: implement 16-bit nir_intrinsic_quad_* on GFX6-GFX7
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5227>
Samuel Pitoiset [Mon, 25 May 2020 17:59:57 +0000 (19:59 +0200)]
aco: implement 16-bit reduce operations on GFX6-GFX7
No fp16 on GFX6-GFX7.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5227>
Alyssa Rosenzweig [Tue, 2 Jun 2020 23:30:56 +0000 (19:30 -0400)]
pan/bi: Handle vectorized load_const
In preparation for 16-bit vectors.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5307>
Alyssa Rosenzweig [Tue, 2 Jun 2020 23:29:25 +0000 (19:29 -0400)]
pan/bi: Passthrough second argument of F32_TO_F16
At the NIR level this is a second vector source of the first (only)
argument; at the BIR level this is a pair of scalars.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5307>
Alyssa Rosenzweig [Tue, 2 Jun 2020 23:28:55 +0000 (19:28 -0400)]
pan/bi: Pack second argument of F32_TO_F16
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5307>
Alyssa Rosenzweig [Tue, 2 Jun 2020 23:28:03 +0000 (19:28 -0400)]
pan/bi: Fix SEL.16 swizzle
2 scalar arguments, not 1 vector.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5307>
Alyssa Rosenzweig [Tue, 2 Jun 2020 23:27:47 +0000 (19:27 -0400)]
pan/bi: Handle SEL with vec3 16-bit
Otherwise we end up with a missing argument.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5307>
Alyssa Rosenzweig [Mon, 25 May 2020 18:00:17 +0000 (14:00 -0400)]
panfrost: Passthrough NATIVE loads/stores
Now that we handle load_output directly, this works for e.g. RGB565 on
Midgard.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5285>
Alyssa Rosenzweig [Mon, 1 Jun 2020 18:14:33 +0000 (14:14 -0400)]
pan/mdg: Handle regular nir_intrinsic_load_output
Instead of the vendored version. Only for blend shaders at the moment,
frag shaders fb_fetch has a lot more going on.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5285>
Alyssa Rosenzweig [Mon, 25 May 2020 18:46:53 +0000 (14:46 -0400)]
pan/mdg: Allow f2u8 and friends thru
Now that we can handle destination sizes directly, this keeps us from
needing to chew through so many conversions.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5285>
Alyssa Rosenzweig [Mon, 25 May 2020 18:46:40 +0000 (14:46 -0400)]
pan/mdg: Handle f2u8
This is similar to f2u16.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5285>
Alyssa Rosenzweig [Mon, 25 May 2020 18:19:24 +0000 (14:19 -0400)]
pan/mdg: Fold roundmode into applicable instructions
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5285>
Alyssa Rosenzweig [Mon, 25 May 2020 18:19:11 +0000 (14:19 -0400)]
pan/mdg: Implement *_rtz conversions with roundmode
Use rte as the canonical type.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5285>
Alyssa Rosenzweig [Mon, 25 May 2020 18:11:04 +0000 (14:11 -0400)]
pan/mdg: Lower roundmodes
So now we can use the IR field semantically.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5285>
Alyssa Rosenzweig [Mon, 1 Jun 2020 18:06:44 +0000 (14:06 -0400)]
pan/mdg: Add opcode roundmode property
When the output is rounded in a specified direction.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5285>
Alyssa Rosenzweig [Mon, 25 May 2020 18:05:34 +0000 (14:05 -0400)]
pan/mdg: Add roundmode enum
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5285>
Alyssa Rosenzweig [Mon, 25 May 2020 17:19:43 +0000 (13:19 -0400)]
pan/mdg: Distinguish blend shaders in internal shader-db
Since these shaders are purely internal, the optimization criteria are a
bit different, so it's worth calling attention to this when dumping.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5285>
Icecream95 [Tue, 2 Jun 2020 02:14:12 +0000 (14:14 +1200)]
panfrost: Only use AFBC YTR with RGB and RGBA
The "lossless colorspace transform" is lossy for R and RG formats.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5293>
Icecream95 [Tue, 2 Jun 2020 02:13:03 +0000 (14:13 +1200)]
panfrost: Decode AFBC flag bits
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5293>
Timothy Arceri [Tue, 5 May 2020 04:24:46 +0000 (14:24 +1000)]
glsl: when NIR linker enable use it to resize uniform arrays
Here we turn on uniform array resizing in the NIR linker and disable
the GLSL IR resizing pass when the NIR linker is enabled.
This will potentially make uniform arrays smaller due to NIR
optimising away more uniform uses.
Shader-db results (SKL):
total instructions in shared programs:
14947192 ->
14944093 (-0.02%)
instructions in affected programs: 138088 -> 134989 (-2.24%)
helped: 822
HURT: 4
total cycles in shared programs:
324868402 ->
324794597 (-0.02%)
cycles in affected programs:
3904170 ->
3830365 (-1.89%)
helped: 2333
HURT: 1485
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4910>
Timothy Arceri [Tue, 5 May 2020 04:22:04 +0000 (14:22 +1000)]
glsl: gather uniform dereference info before main linking loop
We want to gather information for all stages here before the main
linking loop. In the following patch we will use to information
to reduce the size of uniform arrays where possible.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4910>
Timothy Arceri [Tue, 5 May 2020 04:18:23 +0000 (14:18 +1000)]
glsl: add update_array_sizes() helper to the NIR uniform linker
This will be used to reduce the size of uniform arrays and replace
the current glsl ir pass. Doing this in NIR allows us to better
optimise the size of uniform arrays.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4910>
Timothy Arceri [Tue, 5 May 2020 04:13:51 +0000 (14:13 +1000)]
glsl: add struct to gather more info about uniform array access
This will be used in the following patches to allow the linker
to resize uniform arrays based on array dereferences.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4910>
Timothy Arceri [Wed, 29 Apr 2020 04:15:12 +0000 (14:15 +1000)]
util: add BITSET_LAST_BIT() helper
This is the reverse of BITSET_FFS()
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4910>
Timothy Arceri [Wed, 6 May 2020 04:01:41 +0000 (14:01 +1000)]
i965: call brw_nir_lower_uniforms() after uniform linking is complete
i965 currently uses the NIR uniform linker for spirv support. Until
now the only reason there has been no issue with calling the
lowering pass before the linker is because no garbage collection
is done between the calls.
An upcoming change to the linker will add an optimisation to resize
unform arrays where possible. Because lowering causes the array
defs to no longer be used the new optimisation ends up resizing the
arrays to 0. To fix this we move the lowering call after the
linking calls.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4910>
Simon Ser [Thu, 28 May 2020 10:27:07 +0000 (12:27 +0200)]
gbm: document that gbm_bo_map exposes a linear view
Drivers (Gallium, i965) expose a linear view of the buffer via
gbm_bo_map.
Signed-off-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5238>
Danylo Piliaiev [Fri, 29 May 2020 13:20:45 +0000 (16:20 +0300)]
glsl: Don't replace lrp pattern with lrp if arguments are not floats
We don't have "lrp(int, int, int)" and validation of ir_triop_lrp
fails down the road.
Fixes: 8d37e991
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3059
Signed-off-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Tested-by: Witold Baryluk <witold.baryluk@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5257>
Boris Brezillon [Tue, 5 May 2020 08:31:02 +0000 (10:31 +0200)]
spirv: Use scoped barriers for SpvOpControlBarrier
If use_scoped_barrier is set to true, we don't have to split the control
and memory barriers.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4900>
Boris Brezillon [Tue, 5 May 2020 08:18:29 +0000 (10:18 +0200)]
intel/compiler: Extract control barriers from scoped barriers
Add a lowering pass extracting all control barriers embedded in scoped
barriers into proper control barriers so we can get rid of the logic
inserting control barriers when an SpvOpControlBarrier with WorkGroup
scope is parsed in spirv_to_nir().
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4900>
Boris Brezillon [Tue, 5 May 2020 07:13:20 +0000 (09:13 +0200)]
nir: Replace the scoped_memory barrier by a scoped_barrier
SPIRV OpControlBarrier can have both a memory and a control barrier
which some hardware can handle with a single instruction. Let's
turn the scoped_memory_barrier into a scoped barrier which can embed
both barrier types. Note that control-only or memory-only barriers can
be supported through this new intrinsic by passing NIR_SCOPE_NONE to the
unused barrier type.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Suggested-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4900>
Boris Brezillon [Thu, 23 Apr 2020 12:16:43 +0000 (14:16 +0200)]
spirv: Split the vtn_emit_scoped_memory_barrier() logic
We are about to add support for scoped control+memory barriers. Let's
move the convert from SPIRV to NIR enums logic in helpers so we can
easily re-use them.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4900>
Samuel Pitoiset [Fri, 29 May 2020 18:26:00 +0000 (20:26 +0200)]
radv: enable zero VRAM for all VKD3D (DX12->VK) games
To fix rendering issues with Metro Exodus, RE2 and 3 and probably
more titles. It seems the default behaviour of DX12 anyways.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3064
Cc: <mesa-stable@lists.freedesktop.org>
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/5262>
Samuel Pitoiset [Fri, 29 May 2020 18:02:49 +0000 (20:02 +0200)]
radv: enable zero VRAM for Doom Eternal
That fixes some rendering issues. Probably some unitialized data
from the game.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3064
Cc: <mesa-stable@lists.freedesktop.org>
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/5262>
Timothy Arceri [Wed, 3 Jun 2020 00:23:32 +0000 (10:23 +1000)]
gitlab-ci: bump piglit checkout commit
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4797>
Timothy Arceri [Mon, 4 May 2020 04:33:56 +0000 (14:33 +1000)]
glsl/spirv: remove dead uniforms in spirv nir linker
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4797>
Timothy Arceri [Wed, 29 Apr 2020 02:20:47 +0000 (12:20 +1000)]
glsl: remove dead uniforms in the nir linker
This is now possible as we do uniform linking via a nir based linker.
Shader-db results for IRIS (SKL):
total instructions in shared programs:
14947192 ->
14946397 (<.01%)
instructions in affected programs: 39498 -> 38703 (-2.01%)
helped: 230
HURT: 18
total cycles in shared programs:
324868402 ->
324847058 (<.01%)
cycles in affected programs: 706701 -> 685357 (-3.02%)
helped: 599
HURT: 449
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4797>
Timothy Arceri [Thu, 28 May 2020 01:08:42 +0000 (11:08 +1000)]
glsl: add can_remove_uniform() helper to the NIR linker
This helper reflects the rules we follow in the GLSL IR linker when
deciding if we can remove a dead uniform. This check is required to
avoid regressions when turning on NIR dead uniform clean up in the
following patch.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4797>
Timothy Arceri [Thu, 28 May 2020 00:59:28 +0000 (10:59 +1000)]
nir: add callback to nir_remove_dead_variables()
This allows us to do API specific checks before removing variable
without filling nir_remove_dead_variables() with API specific code.
In the following patches we will use this to support the removal
of dead uniforms in GLSL.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4797>
Timothy Arceri [Thu, 30 Apr 2020 03:36:02 +0000 (13:36 +1000)]
nir: add glsl_get_ifc_packing() helper
This will be used in the following patch.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4797>
Alyssa Rosenzweig [Tue, 2 Jun 2020 22:12:14 +0000 (18:12 -0400)]
pan/mdg: Don't double-replicate blend on T720
We already do this unconditionally in NIR.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5305>
Bas Nieuwenhuizen [Mon, 1 Jun 2020 23:44:52 +0000 (01:44 +0200)]
radv: Use common gfx10_format_table.h
Save some python code and build time, as well as some code duplication.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5291>
Bas Nieuwenhuizen [Mon, 1 Jun 2020 22:45:14 +0000 (00:45 +0200)]
radv: Include gfx10_format_table.h only from a single source file.
The radeonsi variant has everything in the header, so lets not
include it everywhere.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5291>
Bas Nieuwenhuizen [Mon, 1 Jun 2020 22:55:22 +0000 (00:55 +0200)]
radeonsi: Define gfx10_format in the common header.
So we don't have to have multiple definitions of the struct when
sharing with radv.
While at it put the table properly in a C file so we don't have to
deal with multiple definitions, and the struct definition isn't
in generated source.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5291>
Bas Nieuwenhuizen [Mon, 1 Jun 2020 21:49:22 +0000 (23:49 +0200)]
amd/common,radeonsi: Move gfx10_format_table to common.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5291>
Bas Nieuwenhuizen [Mon, 1 Jun 2020 22:34:28 +0000 (00:34 +0200)]
radeonsi: Explicitly map Z16_UNORM_S8_UINT to None for GFX10.
We should always use separate planes for textures with this format.
Fixes: 273ead81f1a "util/format: Add VK_FORMAT_D16_UNORM_S8_UINT."
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5291>
Daniel Stone [Tue, 2 Jun 2020 18:59:00 +0000 (19:59 +0100)]
Revert "CI: Disable Panfrost T720/T760"
Switches have been rewired, VLANs have been reconfigured, network
elements with non-functional remote management have been removed from
racks and thrown on desks in anger.
This reverts commit
ae6e1aee7d1bd49ae494b8a25ca33d092a3a145a.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5301>
Christian Gmeiner [Sat, 9 May 2020 19:49:52 +0000 (21:49 +0200)]
ci: bare-metal: make it possible to use a script for serial
Makes it possible to use e.g. a ser2net script to talk to the devices.
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5268>
Erik Faye-Lund [Thu, 28 May 2020 12:41:17 +0000 (14:41 +0200)]
zink: Use store_dest_raw instead of storing an uint
I cleaned up the other similar call-sites, but somehow missed this one.
There's nothing different with this, so let's also fix this.
Fixes: 16339646f03 ("zink/spirv: rename functions a bit")
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5250>
Oschowa [Wed, 27 May 2020 10:09:20 +0000 (12:09 +0200)]
radv: Explicitly cast TIMESTAMP_NOT_READY value to uin32_t where needed.
Fixes a clang warning.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5228>
Oschowa [Wed, 27 May 2020 10:00:19 +0000 (12:00 +0200)]
aco: Use correct reference type in for-range-loop.
Fixes a clang warning.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5228>
Oschowa [Fri, 22 May 2020 10:52:05 +0000 (12:52 +0200)]
aco: Don't std::move temporary object.
Fixes the following clang warning:
mesa/src/amd/compiler/aco_optimizer.cpp:2928:15: warning: moving a temporary object prevents copy elision [-Wpessimizing-move]
ctx.uses = std::move(dead_code_analysis(program));
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5228>
Oschowa [Fri, 22 May 2020 10:40:29 +0000 (12:40 +0200)]
aco: Don't declare 'Block' as class, but define as struct.
Fixes clang warnings.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5228>
Oschowa [Fri, 22 May 2020 10:37:27 +0000 (12:37 +0200)]
radv: Don't take absolute value of unsigned type.
Fixes clang warnings.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5228>
Timur Kristóf [Tue, 26 May 2020 23:31:35 +0000 (01:31 +0200)]
radv/aco: Always enable subgroup shuffle.
It is now supported by both backends on all hw.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5223>
Timur Kristóf [Tue, 26 May 2020 23:28:03 +0000 (01:28 +0200)]
aco: Implement subgroup shuffle on GFX6-7.
GFX6 and GFX7 don't have the ds_bpermute (or permute) instruction,
but we would like to support subgroup shuffle on these old GPUs.
So we introduce a new pseudio instruction which will be lowered
to an "unrolled loop" that emulates bpermute on GFX6 and GFX7
using readlane instructions, while also respecting the exec mask
thanks to v_cmpx.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5223>
Timur Kristóf [Tue, 26 May 2020 23:22:28 +0000 (01:22 +0200)]
aco/gfx10: Refactor of GFX10 wave64 bpermute.
The emulated GFX10 wave64 bpermute no longer needs a linear_vgpr,
so we don't consider it a reduction anymore. Additionally, the
code is slightly reorganized in preparation for the GFX6 emulated
bpermute.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5223>
Marek Olšák [Tue, 26 May 2020 09:39:08 +0000 (05:39 -0400)]
radeonsi: add a hack to disable TRUNC_COORD for shadow samplers
This fixes dEQP-GLES3.functional.shaders.texture_functions.textureprojlodoffset.sampler2dshadow_vertex.
This is probably a dEQP bug.
Fixes: d573d1d82524b8a2e5f56938069cabc0f0176a0e
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5209>
Marek Olšák [Thu, 21 May 2020 09:13:25 +0000 (05:13 -0400)]
radeonsi: pass at most 3 images and/or shader buffers via user SGPRs for compute
This should slightly decrease shader lifetime.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5209>
Marek Olšák [Thu, 21 May 2020 08:18:19 +0000 (04:18 -0400)]
radeonsi: remove const_buffers_declared hacks
This was a bug that was uncovered by
4553fc66a5f23607c2e872d8ac8755c747bd0bd2.
Piglit: spec@arb_uniform_buffer_object@maxblocks
Fixes: 4553fc66a5f23607c2e872d8ac8755c747bd0bd2
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5209>
Marek Olšák [Sun, 24 May 2020 12:34:30 +0000 (08:34 -0400)]
radeonsi: remove unused leftover code for INDIRECT_BUFFER inside IBs
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5209>
Marek Olšák [Sun, 24 May 2020 19:52:26 +0000 (15:52 -0400)]
nir: gather which images are MSAA
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5209>
Marek Olšák [Thu, 21 May 2020 09:13:01 +0000 (05:13 -0400)]
nir: gather which images are buffers
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5209>
Marek Olšák [Tue, 26 May 2020 08:36:33 +0000 (04:36 -0400)]
nir: don't count samplers and images in interface blocks
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5209>
Marek Olšák [Mon, 11 May 2020 07:01:50 +0000 (03:01 -0400)]
ac/nir: support v2f16 derivatives
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5003>
Marek Olšák [Mon, 11 May 2020 20:31:58 +0000 (16:31 -0400)]
ac/nir: set the second v_cvt_pkrtz argument to undef if it's unused
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5003>
Marek Olšák [Sun, 10 May 2020 03:00:44 +0000 (23:00 -0400)]
ac/nir: select v_cvt_pkrtz for all conversions from f32 to f16 for radeonsi
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5003>
Marek Olšák [Sun, 3 May 2020 23:30:31 +0000 (19:30 -0400)]
ac/nir: handle nir_op_[fiu]2[fiu]mp opcodes
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5003>
Marek Olšák [Mon, 4 May 2020 13:30:40 +0000 (09:30 -0400)]
ac/nir: support 16-bit data in image opcodes
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5003>
Marek Olšák [Mon, 4 May 2020 14:43:07 +0000 (10:43 -0400)]
ac/nir: support 16-bit data in buffer_load_format opcodes
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5003>
Marek Olšák [Mon, 4 May 2020 14:55:08 +0000 (10:55 -0400)]
ac/nir: remove type and num_channels args from ac_build_buffer_store_common
They were only used for type overloading where we can just use
the type of data.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5003>
Marek Olšák [Mon, 4 May 2020 11:01:04 +0000 (07:01 -0400)]
ac/nir: support vector types in the type suffix of overloaded intrinsics
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5003>
Marek Olšák [Mon, 4 May 2020 13:27:49 +0000 (09:27 -0400)]
ac/nir: use more types from ac_llvm_context
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5003>
Marek Olšák [Sun, 3 May 2020 23:27:27 +0000 (19:27 -0400)]
ac: rename has_double_rate_fp16 -> has_packed_math_16bit
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5003>
Marek Olšák [Sun, 10 May 2020 21:05:00 +0000 (17:05 -0400)]
gallium: add shader caps INT16 and FP16_DERIVATIVES
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5002>
Marek Olšák [Sun, 10 May 2020 22:00:55 +0000 (18:00 -0400)]
glsl: lower samplers with highp coordinates correctly
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5002>
Marek Olšák [Sun, 10 May 2020 00:57:25 +0000 (20:57 -0400)]
glsl: lower the precision of imageLoad
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5002>
Marek Olšák [Sun, 10 May 2020 01:38:34 +0000 (21:38 -0400)]
glsl: lower mediump partial derivatives
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5002>
Marek Olšák [Sat, 9 May 2020 02:42:43 +0000 (22:42 -0400)]
glsl: lower mediump integer types to int16 and uint16
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5002>
Marek Olšák [Sat, 9 May 2020 02:16:42 +0000 (22:16 -0400)]
glsl: handle int16 and uint16 types and add instructions for mediump
v2: add more changes to ir_validate.cpp
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5002>
Marek Olšák [Tue, 5 May 2020 01:52:09 +0000 (21:52 -0400)]
glsl: treat lowp as mediump when lowering builtins
This seems to have been missed.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5002>
Marek Olšák [Mon, 4 May 2020 00:10:57 +0000 (20:10 -0400)]
nir: add options::vectorize_vec2_16bit to limit vectorization to vec2 16
for hardware that is scalar but can do 2 16-bit operations on low and high
16 bits of registers at once.
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5002>
Marek Olšák [Mon, 11 May 2020 06:56:47 +0000 (02:56 -0400)]
nir: fix lower_wpos for 16-bit fddy
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5002>
Marek Olšák [Mon, 11 May 2020 03:04:23 +0000 (23:04 -0400)]
nir: lower int16 and uint16 in nir_lower_mediump_outputs
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5002>
Marek Olšák [Mon, 11 May 2020 04:42:51 +0000 (00:42 -0400)]
nir: add int16 and uint16 type helpers
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5002>
projects / mesa.git / log