AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
122,729 Commits 1 Branch 0 Tags
a44e63398b045f0a5f56e4d719d25a8501ab53cd
Commit Graph

122729 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Jason Ekstrand a44e63398b anv: Properly handle all sizes of specialization constants 
Closes: #2812
cc: mesa-stable@lists.freedesktop.org

Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4675>
 2020-04-24 09:23:59 +00:00
Jason Ekstrand 64e4297629 spirv: Allow constants and NULLs in SpvOpConvertUToPtr 
We were accidentally asserting that the value had to be a vtn_ssa_value
which isn't true if it, for instance, comes from a spec constant.

Fixes: fb282a68bc "spirv: Implement OpConvertPtrToU and OpConvertUToPtr"
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Acked-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4675>
 2020-04-24 09:23:59 +00:00
Eduardo Lima Mitev 4dc7b76276 anv/radv: Resolving 'GetInstanceProcAddr' should not require a valid instance 
Since vk_icdGetInstanceProcAddr() is wired through
vkGetInstanceProcAddr() in both drivers, we lost the ability for
'GetInstanceProcAddr' to resolve itself prior to having a valid
instance.

An upcoming spec change will fix that and allow
vkGetInstanceProcAddr() to resolve itself passing NULL as
instance. See https://gitlab.khronos.org/vulkan/vulkan/issues/2057
for details.

This patch implements the change in both radv and anvil.

CTS changes have already landed:
https://gitlab.khronos.org/Tracker/vk-gl-cts/issues/2278

vulkan-loader changes have also landed:
https://gitlab.khronos.org/Tracker/vk-gl-cts/issues/2278

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4273>
 2020-04-24 09:09:14 +00:00
Rhys Perry 665250e830 aco: fix v_or(s_lshl) and v_add(s_lshl) optimizations 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Fixes: d1621834f3
    ('aco: combine VALU and SALU into various VOP3 instructions')

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/2822
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4717>
 2020-04-24 08:55:19 +00:00
Timothy Arceri 58b8fbb824 glsl: remove some duplicate code from the nir uniform linker 
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4721>
 2020-04-24 08:31:01 +00:00
Timothy Arceri ffbec55072 glsl: some nir uniform linker fixes 
This fixes find_and_update_named_uniform_storage() for subroutines
and also updates num_shader_uniform_components for non opaque
uniforms.

The following patch will ensure this type of bug won't happen again.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4721>
 2020-04-24 08:31:01 +00:00
Lionel Landwerlin 9df1d92bbd drm-shim: stub syncobj wait ioctl 
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4726>
 2020-04-24 10:19:43 +03:00
Lionel Landwerlin 53f151f422 drm-shim: provide a valid fake syncobj handle at creation 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4726>
 2020-04-24 10:19:43 +03:00
Quentin Glidic 00f5ea9fdc meson: Use dependency.partial_dependency() 
It avoids calling pkg-config which was searched for in a wrong way, thus
breaking setup where unprefixed pkg-config was banned (e.g. on Exherbo).

Signed-off-by: Quentin Glidic <sardemff7+git@sardemff7.net>
Fixes: 53f9131205
       ("meson: fix getting cflags from pkg-config")

Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4701>
 2020-04-24 06:58:32 +00:00
Christian Gmeiner 7aaa0e5908 etnaviv: add anisotropic filter support 
I have not seen any usage of TEXTURE_FILTER_ANISOTROPIC in the cmd streams
from the binary blob. Maybe it gets used on some model/rev combinations.

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2888>
 2020-04-24 06:46:00 +00:00
Christian Gmeiner 1d4c191572 etnaviv: update headers from rnndb 
Update to etna_viv commit b40ec2a.

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2888>
 2020-04-24 06:46:00 +00:00
Christian Gmeiner 7d77295515 etnaviv: anisotropic filtering is supported starting with HALTI0 
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2888>
 2020-04-24 06:46:00 +00:00
Caio Marcelo de Oliveira Filho 7ee9f851e2 spirv: Update the headers from latest Khronos master 
This corresponds to 2ad0492fb00919d99500f1da74abf5ad3c870e4e ("Discuss
generator magic number reservations.")  in
https://github.com/KhronosGroup/SPIRV-Headers.

Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4682>
 2020-04-24 05:56:05 +00:00
Caio Marcelo de Oliveira Filho 5620c3efd8 spirv: Handle instruction aliases in vtn_gather_types 
Same solution as done in spirv_info generation.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4682>
 2020-04-24 05:56:05 +00:00
Tomeu Vizoso 8cba1a13fa gitlab-ci: Test Virgl with traces 
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4659>
 2020-04-24 05:37:06 +00:00
Tomeu Vizoso 5a5316ee1b gitlab-ci: Test OpenGL ES 3.1 on virgl 
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4659>
 2020-04-24 05:37:06 +00:00
Tomeu Vizoso 9b7c20b315 gitlab-ci: Allow test jobs to add options to the dEQP invocation 
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4659>
 2020-04-24 05:37:06 +00:00
Tomeu Vizoso 34ed5fff5b gitlab-ci: Update virglrenderer in the x86_test-gl image 
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4659>
 2020-04-24 05:37:06 +00:00
Alyssa Rosenzweig a3d2936a8e panfrost: The texture descriptor has a pointer to a trampoline 
Not to the texture itself, and can have a stride right after for linear
textures.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Signed-off-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/4680>
 2020-04-24 06:55:05 +02:00
Alyssa Rosenzweig 36d49b1fb1 panfrost: Identify texture layout field 
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Signed-off-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/4680>
 2020-04-24 06:55:02 +02:00
Alyssa Rosenzweig ad4024968e pan/decode: Remove is_zs weirdness 
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Signed-off-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/4680>
 2020-04-24 06:54:36 +02:00
Tomeu Vizoso e41894ba15 panfrost: Emit texture descriptor on bifrost 
Signed-off-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/4680>
 2020-04-24 06:53:42 +02:00
Tomeu Vizoso d3eb23adb5 panfrost: Emit sampler descriptor on bifrost 
Signed-off-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/4680>
 2020-04-24 06:53:39 +02:00
Alyssa Rosenzweig 497977bbe6 panfrost: decode textures and samplers on bifrost 
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Signed-off-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/4680>
 2020-04-24 06:53:34 +02:00
Alyssa Rosenzweig 0167391a1a panfrost: Add tentative bifrost_texture_descriptor 
It looks very similar to the Midgard texture descriptor, just with a
bunch of fields moved around and the whole descriptor flattened (so
basically just memory access optimizations, from what I can tell).

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Signed-off-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/4680>
 2020-04-24 06:53:30 +02:00
Alyssa Rosenzweig 81a31911dd panfrost: Set clear_color_[12] in the extra fb desc 
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Signed-off-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/4680>
 2020-04-24 06:53:26 +02:00
Tomeu Vizoso 0a0b670d63 panfrost: Clean up a bit the tiler structs for Bifrost 
And set a fixed hierarchy mask for now that seems to generally work.

Signed-off-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/4680>
 2020-04-24 06:53:21 +02:00
Eric Anholt 0d6019302e vc4: Use NIR shader's num_outputs for generating our new output. 
Simplifies the code (we don't have struct or matrix varyings that would
have previously made this code break), and makes sure we keep
s->num_outputs accurate.

Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4670>
 2020-04-23 18:52:46 +00:00
Eric Anholt 5593d80a2c freedreno/ir3: Fix sizing of the inputs/outputs array. 
If you have a struct, the var's base driver location is not the last
driver location that will be accessed in that var.  We have a shader
struct member with this number for us, already.  Fixes overflows in:

dEQP-GLES31.functional.program_interface_query.program_output.type.interface_blocks.out.named_block_explicit_location.struct.mat3x2

Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4670>
 2020-04-23 18:52:46 +00:00
Eric Anholt ac937bf878 freedreno/ir3: Fix driver_location of the added vertex_flags varying. 
It was ignoring the sizes of the output variables and assuming
single-slot, and failing to update num_outputs.

Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4670>
 2020-04-23 18:52:46 +00:00
Eric Anholt e82ce1852a gallium: Fix setup of pstipple frag coord var. 
If the last input was a struct or matrix, we would have overlapped driver
locations for our new position var.

Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4670>
 2020-04-23 18:52:46 +00:00
Eric Anholt 035fd4fb9f nir/lower_clip: Fix picking of unused driver locations. 
This fixes things when the last input/output is a struct or matrix.

Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4670>
 2020-04-23 18:52:46 +00:00
Eric Anholt 91668ae839 nir/lower_two_sided_color: Fix picking of new driver location. 
We have shader->num_inputs for "last used input + 1" already, which
respects struct/matrix varyings.

Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4670>
 2020-04-23 18:52:46 +00:00
Gert Wollny 49ce749d0e nir: Add umad24 and umul24 opcodes 
So far only the singed versions are defined.

v2: Make umad24 and umul24 non-driver specific (Eric Anholt)

v3: Take care of nir_builder and automatic lowering of the
    opcodes if they are not supported by the backend.

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4610>
 2020-04-23 18:23:04 +00:00
Gert Wollny 42aa348dad nir: Add r600 specific intrinsics for tesselation shader IO 
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4610>
 2020-04-23 18:23:04 +00:00
Eric Anholt e9add0c501 drm-shim: Let the driver choose to overwrite the first render node. 
When I was writing drm-shim, I was focused on the v3d kmsro case -- use my
intel device as the kmsro display device and add on a simulator-based v3d
device that we could render with.  But for the noop backends we use for
shader-db, it's a lot more useful to just overwrite the first render node
in the system so that you don't have to pass a -d <how many render nodes I
already have in my system> argument.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4664>
 2020-04-23 17:54:54 +00:00
Eric Anholt 5a8718f01b freedreno: Make the slice pitch be bytes, not pixels. 
Back in a2xx, HW pitches were in pixels, so storing that was reasonable.
Ever since then, the HW wants pitches in bytes, and we have only one
instance of using pitch in pixels in the code (a3xx sysmem path).

Flip things around so that only a2xx has to worry about the cpp for
looking at pitches.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4558>
 2020-04-23 16:37:50 +00:00
Eric Anholt bd76a24fd1 freedreno: Introduce a "cpp_shift" value for cpp divs/muls. 
This only converts part of the driver to use it, leaving the rest to the
following commit (which inspired this one).

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4558>
 2020-04-23 16:37:50 +00:00
Samuel Pitoiset 6a6e71524d radv: adjust the supported subgroup stages 
VK_SHADER_STAGE_ALL now includes all ray-tracing related stages.
Noticed while comparing vulkaninfo with some other drivers.

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/4679>
 2020-04-23 16:16:09 +00:00
Lionel Landwerlin efdb7fa9a8 anv: force whole EU array to be powered for perf queries 
Because of functional requirements for Gen11, when perf is enabled we
only power half the EU array.

This change forces it to enable everything.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Mark Janes <mark.a.janes@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4021>
 2020-04-23 15:55:59 +00:00
Lionel Landwerlin a7998371ed intel/perf: specify sseu configuration when supported 
Because of functional requirements for Gen11, when perf is enabled we
only power half the EU array.

This change forces it to enable everything.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Mark Janes <mark.a.janes@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4021>
 2020-04-23 15:55:59 +00:00
Lionel Landwerlin 8f152ed101 intel/perf: store default sseu configuration 
This is the powergating configuration of the EU array. The default is
everything powered.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Mark Janes <mark.a.janes@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4021>
 2020-04-23 15:55:59 +00:00
Lionel Landwerlin ea8cb79742 include/drm-uapi: bump headers 
From drm-next at the following commit :

  commit 1aa63ddf726ea049279989b93b69b57ce6efd75b
  Merge: 774f1eeb18b0 14d0066b8477
  Author: Dave Airlie <airlied@redhat.com>
  Date:   Wed Apr 22 10:40:34 2020 +1000

      Merge tag 'drm-misc-next-2020-04-14' of git://anongit.freedesktop.org/drm/drm-misc into drm-next

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Mark Janes <mark.a.janes@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4021>
 2020-04-23 15:55:59 +00:00
Samuel Pitoiset ff3f775476 radv: simplify checking for Navi1x chips 
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/4702>
 2020-04-23 15:54:32 +02:00
Rhys Perry 0d9fe0405f aco: improve code for 32-bit isign 
No shader-db changes on Navi.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4667>
 2020-04-23 12:39:33 +00:00
Rhys Perry d1621834f3 aco: combine VALU and SALU into various VOP3 instructions 
shader-db (Navi):
Totals from 2916 (2.28% of 127638) affected shaders:
SGPRs: 184427 -> 184283 (-0.08%); split: -0.10%, +0.02%
VGPRs: 143520 -> 143640 (+0.08%); split: -0.00%, +0.09%
CodeSize: 14913548 -> 14913288 (-0.00%); split: -0.00%, +0.00%
MaxWaves: 26034 -> 26012 (-0.08%)
Instrs: 2935435 -> 2930960 (-0.15%); split: -0.15%, +0.00%

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4667>
 2020-04-23 12:39:33 +00:00
Rhys Perry 607fb4153d aco: move call to store_output_to_temps in store_ls_or_es_output earlier 
Skips get_intrinsic_io_basic_offset()

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4667>
 2020-04-23 12:39:33 +00:00
Rhys Perry b497b774a5 aco: remove copy in load_input_from_temps() 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4667>
 2020-04-23 12:39:33 +00:00
Rhys Perry 2dc550202e aco: copy-propagate p_create_vector copies of vectors 
Instead of copying the operands of the other p_create_vector and labelling
the definition with label_vec, copy the operands and label it with
label_temp so that it can be copy-propagated.

This was found while removing a redundant copy in load_input_from_temps()
which removed duplicate p_create_vector instructions.

shader-db (Navi):
Totals from 139 (0.11% of 127638) affected shaders:
VGPRs: 8472 -> 7948 (-6.19%)
CodeSize: 514592 -> 512368 (-0.43%)
MaxWaves: 1089 -> 1195 (+9.73%)
Instrs: 100214 -> 99658 (-0.55%)
Cycles: 400856 -> 398632 (-0.55%)
VMEM: 15545 -> 15338 (-1.33%)
Copies: 5140 -> 4584 (-10.82%)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4667>
 2020-04-23 12:39:33 +00:00
Rhys Perry e4383b5c7f aco: decrease the uses of other copy operations after splitting/removing 
For copies like v[7:8] = v[8:9], what currently happens is:
- do_copy() will skip the second dword
- the uses of the second dword will be reduced to 0
- the copy operation will be removed from the map
and v8 will never be set to v9.

So just decrease the uses of other operations after splitting or removing
the current operation, so: "v8 = v9" will be split off, it's uses reduced
and then the new copy will be done in the next iteration.

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/4686>
 2020-04-23 11:39:23 +00:00