AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
163,834 Commits 1 Branch 0 Tags
77ecf9149c7fdadbb24b471785c4d5b4e285f2df
Commit Graph

163834 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Sviatoslav Peleshko 77ecf9149c anv: Defer flushing PIPE_CONTROL bits forbidden in CCS while in GPGPU mode 
Fixes: 313aeee8 ("anv: Use pending pipe control mechanism in flush_pipeline_select()
")

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7816
Signed-off-by: Sviatoslav Peleshko <sviatoslav.peleshko@globallogic.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20124>
 2022-12-03 00:10:32 +00:00
Jason Ekstrand cd5c66e165 dxil: Use nir_const_value_for_uint in dxil_nir_lower_int_samplers 
This change should avoid any accidental rounding issues because of
border colors getting stored in a float in dxil_wrap_sampler_state.  It
also switches us to using the correct helpers for nir_const_value so we
can avoid any weird uninitialized data failures that can be caused by
filling out the fields in the struct directly.

Fixes: b9c61379ab ("microsoft/compiler: translate nir to dxil")
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19689>
 2022-12-02 23:12:30 +00:00
Jason Ekstrand f3f1c28f8e r600/nir: Fix u64vec2 immediate lowering 
There were a couple of issues here:

 1. We should be using nir_const_value_for_uint instead of setting the
    union fields directly to ensure the rest of the union is zeroed.

 2. It was always filling out the first two components of val even if
    the incoming constant had 2 64-bit components.

Fixes: 165fb5117b ("r600/sfn: add lowering passes to get 64 bit ops lowered to 32 bit vec2")
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19689>
 2022-12-02 23:12:30 +00:00
Jason Ekstrand 49d86200e5 st/mesa: Use nir_const_value_for_bool() in ATIFS 
Fixes: 0a179bb6e2 ("st/mesa: Generate NIR for ATI_fragment_shader instead of TGSI.")
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19689>
 2022-12-02 23:12:30 +00:00
Jason Ekstrand e6de164e03 nir: Use nir_const_value_for_int in nir_lower_subgroups 
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7670
Fixes: e4e79de2a4 ("nir/subgroups: Support > 1 ballot components")
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19689>
 2022-12-02 23:12:30 +00:00
Konstantin Seurer ef168a57b9 llvmpipe: Use get_first_non_void_channel more often 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18634>
 2022-12-02 22:06:11 +00:00
Konstantin Seurer c94e3687d1 virgl: Use get_first_non_void_channel more often 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18634>
 2022-12-02 22:06:11 +00:00
Konstantin Seurer 9d4172f548 radeonsi: Use get_first_non_void_channel more often 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18634>
 2022-12-02 22:06:11 +00:00
Konstantin Seurer cc8b398c96 r600: Use get_first_non_void_channel more often 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18634>
 2022-12-02 22:06:11 +00:00
Konstantin Seurer 9a452a97d9 r300: Use get_first_non_void_channel more often 
Reviewed-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18634>
 2022-12-02 22:06:11 +00:00
Konstantin Seurer ad8de42ce5 radv: Use get_first_non_void_channel more often 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18634>
 2022-12-02 22:06:11 +00:00
Yiwei Zhang c2fdd6a2c2 venus: add VN_PERF_NO_CMD_BATCHING 
Make it easier to debug object lifetime issues in Venus.

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19830>
 2022-12-02 21:42:25 +00:00
Yiwei Zhang 739b20beda venus: fix draw cmd batch accounting 
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19830>
 2022-12-02 21:42:25 +00:00
Yiwei Zhang ea7c2d3d94 venus: add VN_PERF_NO_MEMORY_SUBALLOC 
Make it easier for memory alignment debugging.

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19830>
 2022-12-02 21:42:25 +00:00
Yiwei Zhang bda80458f9 venus: avoid no_async_buffer_create related caches with the perf option 
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19830>
 2022-12-02 21:42:25 +00:00
Konstantin Seurer 6397304519 radv: Only create bvh pipelines when using rt 
Saves some time when creating non-rt devices.

Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20110>
 2022-12-02 21:14:00 +00:00
Konstantin Seurer f89708c2c5 vulkan: Use vk_image_sanitize_extent 
We set the image type before the extend which allows us to use
vk_image_sanitize_extent.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20112>
 2022-12-02 20:46:33 +00:00
Jesse Natalie 8d8188ec59 microsoft/clc: Add a test for nested function-temp arrays 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20111>
 2022-12-02 09:53:09 -08:00
Jesse Natalie 88b2c2a5ca microsoft/compiler: Handle nested arrays correctly for emitting global consts 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20111>
 2022-12-02 09:53:09 -08:00
Jesse Natalie d4c70e483d compiler: Handle nested arrays correctly for computing CL size/alignment 
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20111>
 2022-12-02 09:52:44 -08:00
Jesse Natalie 4a976d371b gallium/u_debug_flush: Fix incompatible function signature warnings 
Looks like an MSVC update started complaining that pipe_error is
no longer a compatible return type for function pointers that are
supposed to be int.

Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20102>
 2022-12-02 17:15:30 +00:00
Konstantin Seurer 7fe515f6d4 radv/rra: Get rid of annoying memory aliasing warning 
Such cursed behavior is almost non existent in practise. When capturing
a Doom Eternal, this warning spams the output for no reason.

The warning is also unnecessary since we copy acceleration structures
right after building them now.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20047>
 2022-12-02 16:48:07 +00:00
Konstantin Seurer e2b7e478a5 radv/rra: Fix setting some offsets 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20047>
 2022-12-02 16:48:07 +00:00
Konstantin Seurer 79dcacfc04 radv/rra: Refactor rra_fill_accel_struct_header_common 
No need to re-do the offset calculation for every field.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20047>
 2022-12-02 16:48:07 +00:00
Konstantin Seurer bb6b45e26e radv/rra: Set the metadata size correctly 
Fixes: 5749806 ("radv: Add Radeon Raytracing Analyzer trace dumping utilities")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20047>
 2022-12-02 16:48:07 +00:00
Konstantin Seurer 0e3325dfb6 radv/rra: Remove an obsolete comment 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20047>
 2022-12-02 16:48:07 +00:00
Konstantin Seurer 94ec359ae5 radv/rra: Defer destroying accel struct data 
This allows us to dump acceleration structures that were destroyed
before present.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20047>
 2022-12-02 16:48:07 +00:00
Konstantin Seurer ae9c65a552 radv/rra: Copy accel structs directly after build 
This is the second step of decoupling acceleration structure dumping
from lifetimes. It also simplifies the logic a bit.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20047>
 2022-12-02 16:48:07 +00:00
Konstantin Seurer 08a85076e5 radv/rra: Introduce radv_rra_accel_struct_data 
This will be useful for dumping acceleration structures that were
destroyed before submit.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20047>
 2022-12-02 16:48:07 +00:00
Konstantin Seurer ff3ba5c74d radv: Add hash_table_foreach to .clang-format 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20047>
 2022-12-02 16:48:07 +00:00
Pedro J. Estébanez b2c4338862 microsoft/spirv_to_dxil: Properly handle load- and is_helper_invocation 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19908>
 2022-12-02 16:29:13 +00:00
Karol Herbst e22491c832 clc: fetch clang resource dir at runtime 
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19617>
 2022-12-02 15:38:44 +00:00
Karol Herbst cd2609b12c clc: generate sources only with with_microsoft_clc 
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19617>
 2022-12-02 15:38:44 +00:00
Danylo Piliaiev 5d025f4003 nir/nir_opt_offsets: Prevent offsets going above max 
In try_fold_load_store when trying to extract const addition from
non-const offset source, we should take into account that there is
already a constant base offset, which should count towards the limit.

The issue was found in "Monster Hunter: World" running on Turnip.

Fixes: cac6f633b2
("nir/opt_offsets: Use nir_ssa_scalar to chase offset additions.")

Well, the issue was present before this commit but it made a lot
of changes in surrounding code.

Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20099>
 2022-12-02 15:04:52 +00:00
David Heidelberg f7e76eee28 ci/amd: re-enable previously OOM tests 
Since we have ZRAM now, we can enable previously failing tests on OOM.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19535>
 2022-12-02 13:51:15 +00:00
David Heidelberg 3e80c5f1ec ci/kernel: enable ZRAM on all archs 
Let's enable ZRAM with 2G. Should help prevent peak OOM scenarios.

For more info see: https://www.kernel.org/doc/html/latest/admin-guide/blockdev/zram.html

Reviewed-by: Eric Engestrom <eric@igalia.com>
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19535>
 2022-12-02 13:51:15 +00:00
Bas Nieuwenhuizen 89663828ea aco: Don't use v_lshrrev_b64 for moves on GFX11. 
Looking at VOPD things, shifts are not very likely to get dual issued
but plain moves are. Looking at RDNA2 v_lshrrev_b64 are half the perf
of v_mov_b32 (but you need twice as many moves), so on GFX11 this likely
reaches the threshold where moves are faster.

Totals from 68400 (50.70% of 134906) affected shaders:

CodeSize: 275489516 -> 275459536 (-0.01%); split: -0.01%, +0.00%
Instrs: 51775474 -> 51991286 (+0.42%)
Latency: 589884847 -> 589066439 (-0.14%); split: -0.15%, +0.01%
InvThroughput: 127154986 -> 126037619 (-0.88%); split: -0.88%, +0.00%
Copies: 3756157 -> 3976193 (+5.86%)
Branches: 1259604 -> 1260072 (+0.04%)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19633>
 2022-12-02 13:25:57 +00:00
Bas Nieuwenhuizen 91fe2a2361 aco: Use more detailed wave64 timing for GFX10+. 
Also nabbed some dual issue stuff for GFX11 from LLVM.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19633>
 2022-12-02 13:25:57 +00:00
Eric Engestrom 09caa570b1 docs: update calendar and link releases notes for 22.3.0 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20129>
 2022-12-02 12:17:41 +00:00
Eric Engestrom 9d665ae439 docs: add release notes for 22.3.0 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20129>
 2022-12-02 12:17:41 +00:00
David Heidelberg 5797af2d8d Revert "ci: disable Collabora's LAVA lab for maintance" 
This reverts commit 3964a77454.

Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20128>
 2022-12-02 12:25:16 +01:00
Qiang Yu 2fb1097bac ac/nir/ngg: merge multi stream gs shader queries 
Before this commit each stream will emit a query block, now
we merge them to a single block.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20074>
 2022-12-02 09:38:07 +00:00
Lionel Landwerlin b7b91ae51e anv: enable VK_KHR_ray_tracing_maintenance1 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20011>
 2022-12-02 09:28:23 +00:00
Lionel Landwerlin d844fa4def anv: implement new queries for VK_KHR_ray_tracing_maintenance1 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20011>
 2022-12-02 09:28:23 +00:00
Lionel Landwerlin 4d05be49c2 anv: implement vkCmdTraceRaysIndirect2KHR 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20011>
 2022-12-02 09:28:23 +00:00
Lionel Landwerlin 675c5bd4cc anv: refactor ray tracing dispatch 
Preparing for vkCmdTraceRaysIndirect2KHR

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20011>
 2022-12-02 09:28:23 +00:00
Lionel Landwerlin df38426072 intel/rt/nir: add support for RayCullMaskKHR 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20011>
 2022-12-02 09:28:23 +00:00
Lionel Landwerlin 6202a2c6b4 intel/rt/nir: enable the trampoline shader to load the indirect ray shader bsr 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20011>
 2022-12-02 09:28:23 +00:00
Lionel Landwerlin af3f7948d1 anv: correctly predicate ray tracing 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 7479fe6ae0 ("anv: Implement vkCmdTraceRays and vkCmdTraceRaysIndirect")
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20011>
 2022-12-02 09:28:23 +00:00
Lionel Landwerlin 7d7c32de4c anv/genxml: make gen_rt more like other genxml files 
The main goal is to be able to generate genX_bits.h for those
structures so we can get generated field offsets.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20011>
 2022-12-02 09:28:23 +00:00