Boris Brezillon [Mon, 20 Jan 2020 15:05:31 +0000 (16:05 +0100)]
panfrost/midgard: Add f2f64 support
So we can convert floats into doubles.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3478>
Boris Brezillon [Mon, 20 Jan 2020 15:03:52 +0000 (16:03 +0100)]
panfrost/midgard: Factorize f2f and u2u handling
Those size conversion operations work the same way apart from f2f
using an fmov op code and u2u using an imov. Let's handle them in the
same case block to avoid code duplication.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3478>
Boris Brezillon [Mon, 20 Jan 2020 14:55:21 +0000 (15:55 +0100)]
panfrost/midgard: Make sure promote_fmov() only promotes 32-bit imovs
mir_constant_float() assumes we're dealing with 32-bit integers/floats,
which is only the case if reg_mode is equal to midgard_reg_mode_32.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3478>
Boris Brezillon [Mon, 20 Jan 2020 14:44:48 +0000 (15:44 +0100)]
panfrost/midgard: Rework mir_adjust_constants() to make it type/size agnostic
Right now, constant combining is not supported in 16 bit mode, and 64
bit mode is simply ignored. Let's rework the function to make it
type/bit-size agnostic.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3478>
Boris Brezillon [Mon, 20 Jan 2020 14:00:57 +0000 (15:00 +0100)]
panfrost/midgard: Use a union to manipulate embedded constants
Each instruction bundle can contain up to 16 constant bytes. The meaning
of those byte is instruction dependent: it depends on the instruction
native type (int, uint or float) and the instruction reg_mode (8, 16, 32
or 64 bit). Those different layouts can be exposed as a union to
facilitate constants manipulation.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3478>
Lionel Landwerlin [Wed, 22 Jan 2020 13:49:25 +0000 (15:49 +0200)]
anv: ensure prog params are initialized with 0s
As a result of
9baa33cef01f our backend compiler leaves params pretty
much untouched. So in order to avoid storing uninitialized values in
the shader cache blobs, just 0 out this array.
I've considered not even allocating this array which works on gen8+
but the vec4 backend still makes a copy of this array and so it
crashes on memcpy on HSW.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 9baa33cef01f ("anv: Rework push constant handling")
Reported-by: Tapani Pälli <tapani.palli@intel.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3516>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3516>
Alyssa Rosenzweig [Mon, 20 Jan 2020 21:01:53 +0000 (16:01 -0500)]
panfrost: Fix crash in compute variant allocation
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Fixes: d8a3501f1b2 ("panfrost: Dynamically allocate shader variants")
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3515>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3515>
Guido Günther [Wed, 22 Jan 2020 10:43:11 +0000 (11:43 +0100)]
etnaviv: drm: Don't miscalculate timeout
The current code overflows (s *
1000000000) for s >= 5 but that is
e.g. used in etna_bo_cpu_prep.
Signed-off-by: Guido Günther <agx@sigxcpu.org>
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3509>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3509>
Alexander van der Grinten [Sun, 15 Dec 2019 10:25:37 +0000 (11:25 +0100)]
egl: Fix _eglPointerIsDereferencable w/o mincore()
On platforms without mincore(), _eglPointerIsDereferencable()
currently just checks whether p != NULL. This is not sufficient:
In the Wayland platform code (i.e., in get_wl_surface_proxy()),
_eglPointerIsDereferencable() is called on the version field
of `struct wl_egl_window` which is 3 on current versions of
Wayland. This causes a segfault when trying to dereference p.
Fix this behavior by assuming that the first page of the
process is never dereferencable.
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3103>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3103>
Tapani Pälli [Tue, 21 Jan 2020 11:01:51 +0000 (13:01 +0200)]
egl/android: fix buffer_count for applications setting max count
Problem with previous solution was that it did not take account that
some applications may set a max count for buffers. Therefore we need to
query both min and max and clamp our setting based on that.
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2373
Fixes: be08e6a4496 ("egl/android: Restrict minimum triple buffering for android color_buffers")
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3480>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3480>
Timur Kristóf [Tue, 21 Jan 2020 14:34:23 +0000 (15:34 +0100)]
aco: Fix signedness compare warning.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3483>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3483>
Timur Kristóf [Tue, 21 Jan 2020 12:49:00 +0000 (13:49 +0100)]
aco: Fix maybe-uninitialized warnings.
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/3483>
Timur Kristóf [Tue, 21 Jan 2020 12:43:13 +0000 (13:43 +0100)]
aco: Fix -Wstringop-overflow warnings in aco_span.
GCC does not understand how aco_span works.
This patch fixes it by casting the aco_span's this pointer
to uintptr_t rather than to a char pointer, effectively
telling GCC not to try to figure it out.
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/3483>
Timur Kristóf [Tue, 21 Jan 2020 15:36:54 +0000 (16:36 +0100)]
radeon: Fix multiple definition error with radeon_debug
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3488>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3488>
Timur Kristóf [Tue, 21 Jan 2020 15:12:16 +0000 (16:12 +0100)]
gallium: Fix a couple of multiple definition warnings.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3488>
Timur Kristóf [Tue, 21 Jan 2020 15:08:21 +0000 (16:08 +0100)]
r600: Move get_pic_param to radeon_vce.c
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3488>
Timur Kristóf [Tue, 21 Jan 2020 15:04:33 +0000 (16:04 +0100)]
radeon: Move si_get_pic_param to radeon_vce.c
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3488>
Timur Kristóf [Tue, 21 Jan 2020 16:58:31 +0000 (17:58 +0100)]
intel/compiler: Fix array bounds warning on GCC 10.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Eric Anholt [Wed, 22 Jan 2020 00:56:34 +0000 (16:56 -0800)]
turnip: Add support for non-zero (still constant) UBO buffer indices.
This was actually all ready to go at this point, and just needed to
increment by the value.
Fixes dEQP-VK.binding_model.shader_access.primary_cmd_buf.uniform_buffer.*
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3504>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3504>
Jonathan Marek [Sun, 15 Dec 2019 23:56:47 +0000 (18:56 -0500)]
turnip: fix array/matrix varyings
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Eric Anholt <eric@anholt.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3109>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3109>
Jonathan Marek [Sun, 15 Dec 2019 23:55:39 +0000 (18:55 -0500)]
turnip: remove tu_sort_variables_by_location
nir_assign_io_var_locations already does sorting.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3109>
Jonathan Marek [Sun, 15 Dec 2019 23:54:26 +0000 (18:54 -0500)]
freedreno/ir3: allow inputs with the same location
turnip can have multiple inputs with the same location, and different
location_frac.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3109>
Matt Turner [Tue, 21 Jan 2020 23:23:39 +0000 (15:23 -0800)]
gitlab-ci: Skip ext_timer_query/time-elapsed
This test's result is unpredictable, so it may occasionally pass when we
expect it to fail, thus causing the CI pipeline to fail.
Reviewed-by: Eric Anholt <eric@anholt.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3498>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3498>
Matt Turner [Wed, 6 Nov 2019 18:14:45 +0000 (10:14 -0800)]
intel/compiler: Test compaction on Gen <= 12
With the previous commits we can now enable the unit test on Gen <= 12.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Tue, 12 Nov 2019 00:11:34 +0000 (16:11 -0800)]
intel/compiler: Validate fuzzed instructions
... before giving them to the instruction compactor.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Fri, 3 Jan 2020 20:59:32 +0000 (12:59 -0800)]
intel/compiler: Add unit tests for new EU validation checks
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Thu, 2 Jan 2020 22:44:16 +0000 (14:44 -0800)]
intel/compiler: Validate some instruction word encodings
Specifically, execution size, register file, and register type. I did
not add validation for vertical stride and width because I don't believe
it's possible to have an otherwise valid instruction with an invalid
vertical stride or width, due to all of the other regioning
restrictions.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Mon, 11 Nov 2019 23:19:07 +0000 (15:19 -0800)]
intel/compiler: Factor out brw_validate_instruction()
In order to fuzz test instructions, we first need to do some sanity
checking first. Factoring out this function allows us an easy way to
validate a single instruction.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Wed, 6 Nov 2019 18:05:48 +0000 (10:05 -0800)]
intel/compiler: Handle invalid compacted immediates
16-bit immediates need to be replicated through the 32-bit immediate
field, so we should never see one that isn't.
This does happen however in the fuzzer unit test, so returning false
allows the fuzzer to reject this case.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Thu, 2 Jan 2020 22:57:56 +0000 (14:57 -0800)]
intel/compiler: Handle invalid inputs to brw_reg_type_to_*()
Necessary to handle these cases when we test fuzzed instructions.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Tue, 7 Jan 2020 08:27:23 +0000 (00:27 -0800)]
intel/compiler: Split hw_type tables
Previously we were sharing tables between generations that were nearly
identical (i.e., Gen8 3-src adds HF support) and used a small bit of
code to handle the differences. This is kind of a mess if you want to
reject 64-bit types on platforms that don't support 64-bit types, so
split the tables, allowing each generation's table to list exactly what
it supports.
Acked-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Mon, 6 Jan 2020 21:17:49 +0000 (13:17 -0800)]
intel/compiler: Add a INVALID_{,HW_}REG_TYPE macros
Since the enum brw_reg_type is packed, comparisons with -1 don't work
directly, necessitating the cast. Add a macro to avoid this confusion.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Thu, 2 Jan 2020 22:54:34 +0000 (14:54 -0800)]
intel/compiler: Add NF some more places
Necessary to handle these cases when we test fuzzed instructions.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Wed, 6 Nov 2019 18:13:24 +0000 (10:13 -0800)]
intel/compiler: Limit compaction unit tests to specific gens
Two of the tests emit instructions with MRF destinations, and MRFs
aren't present on Gen7+. I think we were just lucky that this didn't
cause a problem earlier since we were running the tests on Gen7-9.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Tue, 21 Jan 2020 18:44:59 +0000 (10:44 -0800)]
intel/compiler: Don't disassemble align1 3-src operands on Gen < 10
Since the platforms don't support align1 3-src instructions, the
contents of these operands are not going to be meaningful. Just don't
print them to avoid hitting some assertions in brw_inst functions.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Mon, 13 Jan 2020 19:17:27 +0000 (11:17 -0800)]
intel/compiler: Split has_64bit_types into float/int
Gen7 has 64-bit floats but not 64-bit ints.
Acked-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Thu, 31 Oct 2019 17:27:48 +0000 (10:27 -0700)]
intel/compiler: Extract GEN_* macros into separate file
Will be used by the instruction compaction unit test.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Matt Turner [Thu, 16 Jan 2020 22:22:14 +0000 (14:22 -0800)]
intel/compiler: Use ARRAY_SIZE()
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2635>
Caio Marcelo de Oliveira Filho [Tue, 14 Jan 2020 20:03:22 +0000 (12:03 -0800)]
intel/fs: Don't emit control barrier if only one thread is used
When there's only one hardware thread (i.e. the dispatch width greater
or equal to the workgroup size), there's no need to use a barrier to
ensure all the invocations reach the same point in the shader, because
they are already running lock-step.
Results for SKL running Iris for shader-db tests with compute shaders
total sends in shared programs: 18361 -> 18339 (-0.12%)
sends in affected programs: 904 -> 882 (-2.43%)
helped: 9
HURT: 0
helped stats (abs) min: 1 max: 5 x̄: 2.44 x̃: 2
helped stats (rel) min: 0.84% max: 21.43% x̄: 7.82% x̃: 2.67%
95% mean confidence interval for sends value: -3.31 -1.58
95% mean confidence interval for sends %-change: -14.67% -0.97%
Sends are helped.
Shaders from Aztec Ruins, Car Chase, Manhattan and DeusEx are helped.
Results for ICL and TGL are similar to SKL.
Results for BDW are similar to SKL except for DeusEx shader that has a
workgroup size 16 but in BDW picks the SIMD8.
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3226>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3226>
Caio Marcelo de Oliveira Filho [Tue, 31 Dec 2019 09:01:27 +0000 (01:01 -0800)]
intel/fs: Don't emit fence for shared memory if only one thread is used
When there's only one hardware thread (i.e. the dispatch width greater
or equal to the workgroup size), there's no need to synchronize shared
memory access (SLM) since all the requests from a single thread are
already synchronized. In such case, we just add a scheduling fence.
To be able to identify that case for all platforms, move the handling
of platforms prior to Gen11 (which don't have a separate SLM fence)
after the optimization.
Results for SKL running Iris for shader-db tests with compute shaders
total sends in shared programs: 18395 -> 18361 (-0.18%)
sends in affected programs: 938 -> 904 (-3.62%)
helped: 9
HURT: 0
helped stats (abs) min: 1 max: 5 x̄: 3.78 x̃: 4
helped stats (rel) min: 1.56% max: 26.32% x̄: 10.33% x̃: 2.60%
95% mean confidence interval for sends value: -4.85 -2.71
95% mean confidence interval for sends %-change: -19.12% -1.54%
Sends are helped.
Shaders from Aztec Ruins, Car Chase, Manhattan and DeusEx are helped.
Results for ICL and TGL are similar to SKL.
Results for BDW are similar to SKL except for DeusEx shader that has a
workgroup size 16 but in BDW picks the SIMD8.
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3226>
Caio Marcelo de Oliveira Filho [Tue, 14 Jan 2020 20:22:47 +0000 (12:22 -0800)]
intel/fs: Add workgroup_size() helper
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3226>
Caio Marcelo de Oliveira Filho [Thu, 2 Jan 2020 23:27:58 +0000 (15:27 -0800)]
intel/fs: Add FS_OPCODE_SCHEDULING_FENCE
Like a SHADER_OPCODE_MEMORY_FENCE but doesn't doesn't generate any
assembly code.
Will be used when the compiler shouldn't reorder certain instructions
but there's no need to generate code for the HW to do it -- as the
ordering will be guaranteed by other means.
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3226>
Dongwon Kim [Wed, 15 Jan 2020 04:01:41 +0000 (20:01 -0800)]
gallium: check all planes' pipe formats in case of multi-samplers
Current code only checks whether first plane's format is supported
in case YUV format sampling is done by sampling each plane separately.
It would be safer to check other planes' as well.
Signed-off-by: Dongwon Kim <dongwon.kim@intel.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2863>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2863>
Kenneth Graunke [Fri, 17 Jan 2020 00:35:00 +0000 (16:35 -0800)]
anv: Drop some workarounds that are no longer necessary
These workarounds are no longer required by 10th Gen hardware.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3495>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3495>
Kenneth Graunke [Fri, 17 Jan 2020 00:34:10 +0000 (16:34 -0800)]
iris: Drop some workarounds which are no longer necessary
These workarounds are no longer required by 10th Gen hardware.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3495>
Eric Anholt [Thu, 16 Jan 2020 00:58:21 +0000 (16:58 -0800)]
turnip: Disable UBWC on images used as storage images.
The closed GL driver doesn't use UBWC on any storage images. It does tile
mostly (skipping tiling on writeonly images, it seems), but for freedreno
we've been enabling tiling in all cases and it's fine. We do need to
disable UBWC, as tests fail otherwise and just plugging in the equivalent
UBWC regs like we were setting up a texture isn't enough.
Fixes dEQP-VK.image.atomic_operations.*
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3433>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3433>
Eric Anholt [Mon, 9 Dec 2019 21:31:35 +0000 (13:31 -0800)]
turnip: Add limited support for storage images.
So far this doesn't handle the texture state-based storage image access
loads, and doesn't support descriptor arrays (same as SSBOs). The texture
side is more tricky, since we have another remapping table to work around.
This is enough to get some of dEQP-VK.image.atomic_operations.* working.
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3433>
Eric Anholt [Thu, 19 Dec 2019 00:30:37 +0000 (16:30 -0800)]
turnip: Refactor the intrinsic lowering.
Too many things in one function, split them out based on the intrinsic.
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3433>
Eric Anholt [Fri, 20 Dec 2019 00:47:08 +0000 (16:47 -0800)]
turnip: Fix some whitespace around binary operators.
Conforms to mesa style and the rest of turnip.
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3433>
Eric Anholt [Mon, 6 Jan 2020 23:13:04 +0000 (15:13 -0800)]
radeonsi: Drop PIPE_CAP_TGSI_ANY_REG_AS_ADDRESS.
Now that we don't expose TGSI, we can stop exposing the flag.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3493>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3493>
Eric Anholt [Mon, 6 Jan 2020 23:01:36 +0000 (15:01 -0800)]
r300: Remove a bunch of default handling of pipe caps.
u_screen will return 0 for all of these, which means that this is one
less driver to see in git grep when I'm checking who exposes a cap.
The exception is the texel/gather offsets and stream output
components, which will not be exposed since we don't expose the
corresponding GLSL version.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3493>
Eric Anholt [Mon, 6 Jan 2020 22:56:46 +0000 (14:56 -0800)]
r600: Remove a bunch of default handling of pipe caps.
u_screen will return 0 for all of these, which means that this is one
less driver to see in git grep when I'm checking who exposes a cap.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3493>
Eric Anholt [Mon, 6 Jan 2020 22:52:57 +0000 (14:52 -0800)]
radeonsi: Remove a bunch of default handling of pipe caps.
u_screen will return 0 for all of these, which means that this is one
less driver to see in git grep when I'm checking who exposes a cap.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3493>
Lionel Landwerlin [Tue, 21 Jan 2020 16:19:18 +0000 (18:19 +0200)]
anv: don't report error with other vendor DRM devices
Enumeration should just skip unsupported DRM devices.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 34c8621c3b37 ("anv: Allow enumerating multiple physical devices")
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2386
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3481>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3481>
Eric Anholt [Fri, 20 Dec 2019 22:02:55 +0000 (14:02 -0800)]
freedreno: Stop scattered remapping of SSBOs/images to IBOs.
Just make it be all SSBOs then all storage images. The remapping table
was there to make it so that the big gap present from gallium's atomic
lowering would get cleaned up, but that's no longer case. The table has
made it very hard to support Vulkan storage images, so it's time for it to
go.
This does mean that an SSBO/IBO that is only loaded (or size-queried) will
now occupy a slot in the table where it wouldn't before. This seems like
a minor cost compared to being able to drop this much logic.
With the remapping table gone, SSBO array handling for turnip just falls
out.
Fixes many array cases of
dEQP-VK.binding_model.shader_access.primary_cmd_buf.storage_buffer.*
Reviewed-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Jonathan Marek <jonathan@marek.ca> (turnip)
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3240>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3240>
Eric Anholt [Mon, 30 Dec 2019 20:01:25 +0000 (12:01 -0800)]
compiler: Add a note about how num_ssbos works in the program info.
These numbers are always confusing, and it's particularly so for this
field where it has a different meaning in different info structs.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3240>
Eric Anholt [Fri, 20 Dec 2019 21:52:06 +0000 (13:52 -0800)]
nir: Drop the ssbo_offset to atomic lowering.
The arguments passed in were:
- prog->info.num_ssbos
- prog->nir->info.num_ssbos
- arbitrary values for standalone compilers
The num_ssbos should match between the prog's info and prog->nir's info
until this lowering happens.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3240>
Eric Anholt [Fri, 20 Dec 2019 21:30:04 +0000 (13:30 -0800)]
gallium: Pack the atomic counters just above the SSBOs.
We carve out half the SSBO space for atomics, and we were just binding
them way up there. freedreno was then using a remapping table to map the
sparse buffer index back down, since space in the descriptor array is a
shared resource that may limit parallelism. That remapping table
generated inside of the ir3 compiler is getting thoroughly in the way of
implementing vulkan descriptor sets.
We will be able to get rid of the freedreno's remapping table, and
hopefully save shared resources on other hardware, by packing the atomics
tightly above the SSBOs (like i965 does). We already rebind the shader
buffers on program change if either the old or new program has SSBOs or
ABOs, so this doesn't necessarily increase the program state change cost
(the only cost increase I can come up with is if you're using the same
atomic counter without rebinding it across changes of programs with
varying SSBO counts, meaning it would now bounce around index space).
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3240>
Eric Anholt [Fri, 20 Dec 2019 17:02:07 +0000 (09:02 -0800)]
mesa: Make atomic lowering put atomics above SSBOs.
Gallium arbitrarily (it seems) put atomics below SSBOs, resulting in a
bunch of extra index management, and surprising shader code when you would
see your SSBOs up at index 16. It makes a lot more sense to see atomics
converted to SSBOs appear as magic high numbers.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3240>
Eric Anholt [Fri, 20 Dec 2019 22:26:44 +0000 (14:26 -0800)]
turnip: Refactor linkage state setup.
As I touch this for descriptor set reworks, I don't want to have to update
it twice.
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3240>
Timur Kristóf [Tue, 21 Jan 2020 16:11:22 +0000 (17:11 +0100)]
nouveau/nvc0: add extern keyword to nvc0_miptree_vtbl.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Tapani Pälli [Tue, 21 Jan 2020 14:48:11 +0000 (16:48 +0200)]
anv: initialize clear_color_is_zero_one
Fixes following valgrind warning:
==12508== Conditional jump or move depends on uninitialised value(s)
==12508== at 0x2CCD8B79: cmd_buffer_begin_subpass (genX_cmd_buffer.c:4599)
==12508== by 0x2CCDA72B: gen9_CmdBeginRenderPass (genX_cmd_buffer.c:5275)
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3487>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3487>
Boris Brezillon [Mon, 20 Jan 2020 21:00:48 +0000 (22:00 +0100)]
panfrost/midgard: Print the actual source register for store operations
Store operation use r26/r27 but have a word->reg set to 0 or 1 (base is
r26). Let's take this base offset into account in
print_load_store_instr().
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3482>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3482>
Alyssa Rosenzweig [Thu, 16 Jan 2020 15:43:03 +0000 (10:43 -0500)]
panfrost: Add pandecode entries for ASTC/ETC formats
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3414>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3414>
Icecream95 [Sat, 11 Jan 2020 06:19:45 +0000 (19:19 +1300)]
panfrost: Add ASTC texture formats
Acked-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3414>
Icecream95 [Sat, 11 Jan 2020 07:00:38 +0000 (20:00 +1300)]
panfrost: Add ETC1/ETC2 texture formats
Acked-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3414>
Alyssa Rosenzweig [Wed, 15 Jan 2020 18:15:01 +0000 (13:15 -0500)]
panfrost: Rework linear<--->tiled conversions
There's a lot going on here (it's a ton of commits squashed together
since otherwise this would be impossible to review...)
1. We have a fast path for linear->tiled for whole (aligned) tiles, but we
have to use a slow path for unaligned accesses. We can get a pretty
major win for partial updates by using this slow path simply on the
borders of the update region, and then hit the fast path for the
tile-aligned interior. This does require some shuffling.
2. Mark the LUTs constant, which allows the compiler to inline them,
which pairs well with loop unrolling (eliminating the memory accesses
and just becoming some immediates.. which are not as immediate on
aarch64 as I'd like..)
3. Add fast path for bpp1/2/8/16. These use the same algorithm and we
have native types for them, so may as well get the fast path.
4. Drop generic path for bpp != 1/2/8/16, since these formats are
generally awful and there's no way to tile them efficienctly and
honestly there's not a good reason too either. Lima doesn't support any
of these formats; Panfrost can make the opinionated choice to make them
linear.
5. Specialize the unaligned routines. They don't have to be fully
generic, they just can't assume alignment. So now they should be nearly
as fast as the aligned versions (which get some extra tricks to be even
faster but the difference might be neglible on some workloads).
6. Specialize also for the size of the tile, to allow 4x4 tiling as well
as 16x16 tiling. This allows compressed textures to be efficiently tiled
with the same routines (so we add support for tiling ASTC/ETC textures
while we're at it)
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Tested-by: Vasily Khoruzhick <anarsoul@gmail.com> #lima on Mali400
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3414>
Alyssa Rosenzweig [Tue, 14 Jan 2020 17:52:02 +0000 (12:52 -0500)]
panfrost,lima: De-Galliumize tiling routines
There's an implicit dependence on Gallium here that will add more
complexity than needed when testing/optimizing out of driver as well as
potentially Vulkanizing. We don't need a full pipe_box, just the x/y/w/h
properties directly.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Tested-by: Vasily Khoruzhick <anarsoul@gmail.com> #lima on Mali400
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3414>
Alyssa Rosenzweig [Tue, 14 Jan 2020 17:27:47 +0000 (12:27 -0500)]
panfrost: Compile tiling routines with -O3
These are major hot spots for panfrost and lima; better let the compiler
do its thing even on debug builds.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Tested-by: Vasily Khoruzhick <anarsoul@gmail.com> #lima on Mali400
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3414>
Bas Nieuwenhuizen [Tue, 21 Jan 2020 10:49:55 +0000 (11:49 +0100)]
radv: Remove syncobj_handle variable in header.
I strongly suspect it was supposed to be a typedef. However, used
nowhere, we should remove it.
Fixes: eaa56eab6da "radv: initial support for shared semaphores (v2)"
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2385
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3479>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3479>
Neil Armstrong [Tue, 15 Oct 2019 13:22:07 +0000 (15:22 +0200)]
gitlab-ci/lava: add pipeline information in the lava job name
In order to have more informations in the LAVA jobs list, add the
current pipeline URL and commit ref name in the LAVA job name.
Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2337>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2337>
Jan Zielinski [Mon, 20 Jan 2020 12:57:36 +0000 (13:57 +0100)]
gallium/gallivm: enable linking lp_bld_printf function with C++ code
To enable linking functions declared in lp_bld_printf.h file with C++,
we need to add appropriate macros to the header.
Reviewed-by: Krzysztof Raszkowski <krzysztof.raszkowski@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3470>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3470>
Danylo Piliaiev [Wed, 15 Jan 2020 16:56:42 +0000 (18:56 +0200)]
iris: Fix value of out-of-bounds accesses for vertex attributes
Having VERTEX_BUFFER_STATE.BufferSize greater than the size of
a bound vertex buffer allows shader to read uninitialized vertex
attributes from BO, instead of allowing hardware to return zeroes
on out-of-bounds access.
OpenGL spec "6.4 Effects of Accessing Outside Buffer Bounds" says:
"Robust buffer access can be enabled by creating a context with robust access
enabled through the window system binding APIs. When enabled, any command
unable to generate a GL error as described above, such as buffer object accesses
from the active program, will not read or modify memory outside of the data
store of the buffer object and will not result in GL interruption or termination.
Out-of-bounds reads may return values from within the buffer object or zero
values."
Fixes three webgl tests:
conformance/rendering/out-of-bounds-array-buffers.html
conformance2/rendering/out-of-bounds-index-buffers-after-copying.html
conformance2/rendering/element-index-uint.html
See #1996
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3427>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3427>
Vasily Khoruzhick [Thu, 19 Dec 2019 06:15:11 +0000 (22:15 -0800)]
ci: Re-enable CI for lima on mali450
Amend fails and skips lists basing on lists from Andreas Baierl,
shard mali400 job across two devices since it takes close to 10min
and rename jobs to lima-mali400-test and lima-mali450-test.
Also don't set MESA_GLES_VERSION_OVERRIDE=3.0 for lima since we don't support
GLES 3.0 and lower DEQP_PARALLEL to 3 for jobs on H3.
Keep mali400 jobs disabled atm since they take too much time to complete
and we also get some unexplicable failures in dEQP-GLES2.functional.default_vertex_attrib.*
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3163>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3163>
Vasily Khoruzhick [Fri, 17 Jan 2020 03:37:32 +0000 (19:37 -0800)]
ci: lava: pass CI_NODE_INDEX and CI_NODE_TOTAL to lava jobs
deqp-runner.sh uses it to determine whether we split job across multiple
devices and if we do what's the node index.
With this change we now can set 'parallel: N' in job description if we want
to split the job.
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3163>
Hyunjun Ko [Fri, 17 Jan 2020 07:23:03 +0000 (07:23 +0000)]
turnip: fix invalid VK_ERROR_OUT_OF_POOL_MEMORY
When VK_DESCRIPTOR_TYPE_SAMPLER is provided, it doesn't need to be
counted as a buffer count. Otherwise it leads to mismatch of allocated
buffer size, hitting VK_ERROR_OUT_OF_POOL_MEMORY finally.
Fixes: c39afe68f0390d45130c1317b3b7e65f55542c36
Also fixes amber tests:
./tests/cases/address_modes_float.amber
./tests/cases/address_modes_int.amber
./tests/cases/magfilter_linear.amber
./tests/cases/magfilter_nearest.amber
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Jan Vesely [Sun, 19 Jan 2020 02:27:01 +0000 (21:27 -0500)]
clover: Initialize Asm Parsers
Fixes piglits that use ADMGCN inline assembly:
program@execute@calls
program@execute@amdgcn-mubuf-negative-vaddr
CC: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Jason Ekstrand [Sat, 18 Jan 2020 05:52:50 +0000 (23:52 -0600)]
anv: Allow enumerating multiple physical devices
Instead of having a single physical device in anv_instance, have a
linked list of them. What we have now works today because we our GPUs
are build into the CPU and so you're guaranteed to only ever have one of
them. One day, that will change and we want ANV to be ready.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3461>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3461>
Jason Ekstrand [Sat, 18 Jan 2020 05:48:12 +0000 (23:48 -0600)]
anv: Re-arrange physical_device_init
This commit simply moves fetching the device info and checking if ANV
supports the device a bit higher up. This way we fail earlier and it'll
make error checking easier in the next commit.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3461>
Jason Ekstrand [Sat, 18 Jan 2020 05:45:31 +0000 (23:45 -0600)]
anv: Drop separate chipset_id fields
This already exists in gen_device_info. There's no reason to keep
duplicate copies.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3461>
Jason Ekstrand [Sat, 18 Jan 2020 05:17:48 +0000 (23:17 -0600)]
anv: Move the physical device dispatch table to anv_instance
We don't actually have genX versions of any physical device level
commands so we don't need the trampoline versions and we don't need to
have a separate table per physical device.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3461>
Jason Ekstrand [Sat, 18 Jan 2020 05:05:13 +0000 (23:05 -0600)]
anv: Drop the instance pointer from anv_device
There are very few times when we actually want to fetch the instance
from the anv_device. We can put up with a bit of pain there in exchange
for strongly discouraging people from doing this in general.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3461>
Jason Ekstrand [Sat, 18 Jan 2020 05:03:41 +0000 (23:03 -0600)]
anv: Stop allocating WSI event fences off the instance
Fixes: 16eb390834d "anv: add VK_EXT_display_control to anv driver [v5]"
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3461>
Jason Ekstrand [Sat, 18 Jan 2020 04:57:35 +0000 (22:57 -0600)]
anv: Take a device in anv_perf_warn
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3461>
Jason Ekstrand [Sat, 18 Jan 2020 04:43:06 +0000 (22:43 -0600)]
anv: Take an anv_device in vk_errorf
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3461>
Jason Ekstrand [Sat, 18 Jan 2020 04:23:30 +0000 (22:23 -0600)]
anv: Add an anv_physical_device field to anv_device
Having to always pull the physical device from the instance has been
annoying for almost as long as the driver has existed. It also won't
work in a world where we ever have more than one physical device. This
commit adds a new field called "physical" to anv_device and switches
every location where we use device->instance->physicalDevice to use the
new field instead.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3461>
Marek Olšák [Thu, 9 Jan 2020 01:21:04 +0000 (20:21 -0500)]
radeonsi/gfx10: enable GS fast launch for triangles and strips with NGG culling
Only non-indexed triangle lists and strips are supported. This increases
performance if there is something to cull.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Thu, 9 Jan 2020 21:09:47 +0000 (16:09 -0500)]
radeonsi/gfx10: rewrite late alloc computation
- Use conservative late alloc when the number of CUs <= 6.
- Move the late alloc GS register to the GS shader state, so that it can be
tuned for NGG culling.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Fri, 10 Jan 2020 00:12:36 +0000 (19:12 -0500)]
ac: add helper ac_build_triangle_strip_indices_to_triangle
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Mon, 30 Dec 2019 19:23:16 +0000 (14:23 -0500)]
radeonsi/gfx10: implement NGG culling for 4x wave32 subgroups
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Tue, 24 Dec 2019 18:50:06 +0000 (13:50 -0500)]
radeonsi/gfx10: move GE_PC_ALLOC setting to shader states
The value is not changed. I just use a different way to compute it.
The value will vary with NGG culling.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Sat, 4 Jan 2020 02:16:22 +0000 (21:16 -0500)]
radeonsi/gfx10: don't initialize VGPRs not used by NGG passthrough
v2: TES doesn't use the GS PrimitiveID
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Fri, 3 Jan 2020 22:07:38 +0000 (17:07 -0500)]
radeonsi/gfx10: merge main and pos/param export IF blocks into one if possible
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Fri, 3 Jan 2020 21:59:20 +0000 (16:59 -0500)]
radeonsi/gfx10: export primitives at the beginning of VS/TES
This decreases VGPR usage and will allow us to merge some IF blocks
in shaders.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Fri, 3 Jan 2020 21:20:40 +0000 (16:20 -0500)]
radeonsi/gfx10: move s_sendmsg gs_alloc_req to the beginning of shaders
This will allow us to merge some IF blocks in shaders.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Fri, 6 Dec 2019 01:46:30 +0000 (20:46 -0500)]
radeonsi/gfx10: correct VS PrimitiveID implementation for NGG
We didn't use the correct LDS pointer, though it probably doesn't matter,
because I think that nothing else is using LDS here.
This commit makes it consistent with all other esgs_ring use.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Tue, 24 Dec 2019 00:42:46 +0000 (19:42 -0500)]
radeonsi/gfx10: update comments and remove invalid TODOs
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Thu, 2 Jan 2020 23:41:26 +0000 (18:41 -0500)]
ac: add ac_build_readlane without optimization barrier
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Mon, 30 Dec 2019 19:08:45 +0000 (14:08 -0500)]
ac: add prefix bitcount functions
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Sat, 18 Jan 2020 00:55:13 +0000 (19:55 -0500)]
radeonsi: turn an assertion into return in si_nir_store_output_tcs
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>