summaryrefslogtreecommitdiffstats
path: root/src/amd
Commit message (Collapse)AuthorAgeFilesLines
* ac/nir: force unnormalized coordinates for RECTMarek Olšák2019-09-231-1/+3
| | | | | | This fixes VAAPI. Reviewed-by: Connor Abbott <[email protected]>
* ac/nir: port Z compare value clamping from radeonsiMarek Olšák2019-09-231-9/+25
| | | | | | This fixes some dEQP tests. Reviewed-by: Connor Abbott <[email protected]>
* ac: stop using PCI IDs for chip identificationMarek Olšák2019-09-231-15/+58
| | | | | | PCI IDs for amdgpu will be removed from Mesa. Acked-by: Pierre-Eric Pelloux-Prayer <[email protected]>
* ac/addrlib: fix chip identification for Vega10, Arcturus, Raven2, RenoirMarek Olšák2019-09-231-10/+5
| | | | | Cc: 19.2 <[email protected]> Acked-by: Pierre-Eric Pelloux-Prayer <[email protected]>
* aco: only emit waitcnt on loop continues if we there was some load or exportDaniel Schürmann2019-09-231-1/+1
| | | | Reviewed-by: Rhys Perry <[email protected]>
* amd: Build aco only if radv is enabledBas Nieuwenhuizen2019-09-211-1/+1
| | | | | | | | ACO depends on C++14, but radeonsi/radv with LLVM 8,9 do not. Let us only require it for RADV, since that is the only user. Fixes: a70a9987181 "radv/aco: Setup alternate path in RADV to support the experimental ACO compiler" Reviewed-by: Marek Olšák <[email protected]>
* radv: remove dead shared variablesDaniel Schürmann2019-09-191-1/+1
| | | | | | | LLVM does this anyway, but for ACO we need to do it in NIR. Reviewed-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/aco: enable VK_EXT_shader_demote_to_helper_invocationDaniel Schürmann2019-09-193-0/+8
| | | | | | | For now, this extension will only be enabled for ACO. Reviewed-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: enable clustered reductionsDaniel Schürmann2019-09-191-0/+1
| | | | | | | These work with both, LLVM and ACO. Reviewed-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/aco: Setup alternate path in RADV to support the experimental ACO compilerDaniel Schürmann2019-09-1911-103/+205
| | | | | | | | | | LLVM remains default and ACO can be enabled with RADV_PERFTEST=aco. Co-authored-by: Daniel Schürmann <[email protected]> Co-authored-by: Rhys Perry <[email protected]> Reviewed-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* aco: Initial commit of independent AMD compilerDaniel Schürmann2019-09-1931-0/+25572
| | | | | | | | | | | | | | | | | | | | | | ACO (short for AMD Compiler) is a new compiler backend with the goal to replace LLVM for Radeon hardware for the RADV driver. ACO currently supports only VS, PS and CS on VI and Vega. There are some optimizations missing because of unmerged NIR changes which may decrease performance. Full commit history can be found at https://github.com/daniel-schuermann/mesa/commits/backend Co-authored-by: Daniel Schürmann <[email protected]> Co-authored-by: Rhys Perry <[email protected]> Co-authored-by: Bas Nieuwenhuizen <[email protected]> Co-authored-by: Connor Abbott <[email protected]> Co-authored-by: Michael Schellenberger Costa <[email protected]> Co-authored-by: Timur Kristóf <[email protected]> Acked-by: Samuel Pitoiset <[email protected]> Acked-by: Bas Nieuwenhuizen <[email protected]>
* radv: Add DFSM support.Bas Nieuwenhuizen2019-09-181-5/+17
| | | | | | | | | | | | Apparently we already enabled it without having support ... Not sure if we also need to set disable_start_of_prim when the PS has memory writes, but this mirrors radeonsi. Doubles fillrate in my dual_quad_bench from ~16 pixels/cycles to ~32 pixels/cycle on a Raven. Reviewed-by: Samuel Pitoiset <[email protected]>
* radv: Disable dfsm by default even on Raven.Bas Nieuwenhuizen2019-09-182-3/+4
| | | | | | When actually implementing it, Talos on low is still 3% slower. Reviewed-by: Samuel Pitoiset <[email protected]>
* radv: Only break batch on framebuffer change with dfsm.Bas Nieuwenhuizen2019-09-181-1/+1
| | | | Reviewed-by: Samuel Pitoiset <[email protected]>
* radv: never kill a NGG GS shaderRhys Perry2019-09-181-1/+3
| | | | | | | | Seems to fix a hang with excessive vertex emissions when NGG is used for GS. Signed-off-by: Rhys Perry <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: fix VK_KHR_pipeline_executable_properties with NGG GSSamuel Pitoiset2019-09-181-4/+13
| | | | | | | | | | | No GS copy shader if a pipeline enables NGG GS. This fixes dEQP-VK.pipeline.executable_properties.graphics.*geometry_stage*. Fixes: 86864eedd2d ("radv: Implement radv_GetPipelineExecutablePropertiesKHR.") Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* ac: move ac_get_num_physical_vgprs into radeon_infoMarek Olšák2019-09-182-10/+2
| | | | | Reviewed-by: Timothy Arceri <[email protected]> Reviewed-by: Samuel Pitoiset <[email protected]>
* ac: move ac_get_num_physical_sgprs into radeon_infoMarek Olšák2019-09-184-15/+15
| | | | | Reviewed-by: Timothy Arceri <[email protected]> Reviewed-by: Samuel Pitoiset <[email protected]>
* ac: move ac_get_max_wave64_per_simd into radeon_infoMarek Olšák2019-09-183-17/+5
| | | | | Reviewed-by: Timothy Arceri <[email protected]> Reviewed-by: Samuel Pitoiset <[email protected]>
* ac: move num_sdp_interfaces into radeon_infoMarek Olšák2019-09-183-15/+16
| | | | | Reviewed-by: Timothy Arceri <[email protected]> Reviewed-by: Samuel Pitoiset <[email protected]>
* ac: move PBB MAX_ALLOC_COUNT into radeon_infoMarek Olšák2019-09-183-31/+34
| | | | | Reviewed-by: Timothy Arceri <[email protected]> Reviewed-by: Samuel Pitoiset <[email protected]>
* radv: fix loading 64-bit GS inputsSamuel Pitoiset2019-09-181-0/+35
| | | | | | | | | | | We have to load 2 32-bit integer and to cast correctly. This fixes crashes with gs-double-interpolator.vk_shader_test. Cc: 19.2 <[email protected]> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111734 Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: fix writing depth/stencil clear values to imageSamuel Pitoiset2019-09-181-3/+4
| | | | | | | | | Use the fastest way only if both aspects are used. Oops. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111728 Fixes: 218ce34962c ("radv: add mipmap support for the clear depth/stencil values") Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* ac: Remove DEBUG workaroundMichel Dänzer2019-09-171-6/+0
| | | | | | As of version 7, LLVM uses LLVM_DEBUG instead of just DEBUG. Reviewed-by: Timothy Arceri <[email protected]>
* radv: always emit a position export in gs copy shadersRhys Perry2019-09-161-1/+1
| | | | | | Signed-off-by: Rhys Perry <[email protected]> Fixes: f8d0337299f ('radv: add multiple streams support for the GS copy shader') Reviewed-by: Samuel Pitoiset <[email protected]>
* radv: keep GS threads with excessive emissions which could write to memoryRhys Perry2019-09-163-4/+16
| | | | | Signed-off-by: Rhys Perry <[email protected]> Reviewed-by: Samuel Pitoiset <[email protected]>
* radv/gfx10: disable unsupported transform feedback features for NGGSamuel Pitoiset2019-09-161-3/+3
| | | | | | | Mostly multiple streams and queries which have to be fixed/implemented. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: implement NGG streamoutSamuel Pitoiset2019-09-161-7/+514
| | | | | | | | It's still disabled by default because transform feedback randomly hangs and it seems like it's related to GDS (cf. RadeonSI). Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: make sure to wait for idle before clearing GDSSamuel Pitoiset2019-09-161-0/+8
| | | | | | | | | | Otherwise the next streamout operation will overwrite GDS. This can be improved by tracking if there is a streamout operation in flight. Currently the driver unconditionally flushes but that doesn't matter much as NGG streamout is disabled by default. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: make GDS idle when leaving the IBSamuel Pitoiset2019-09-161-0/+7
| | | | | | | | NGG streamout uses GDS and we have to make sure that another process isn't going to overwrite GDS while our shaders are busy. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: enable NGG_WAVE_ID_EN for NGG streamoutSamuel Pitoiset2019-09-161-0/+2
| | | | | | | | Otherwise the wave IDs are probably 0 and it hangs. NGG_WAVE_ID_EN generates wave IDs for GDS OA. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: gather GS output for VS as NGGSamuel Pitoiset2019-09-161-0/+2
| | | | | | | For streamout we have to the number of streamout outputs. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: compute the correct buffer size for NGG streamoutSamuel Pitoiset2019-09-161-1/+10
| | | | | | | It's used to determined the max emit per buffer. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: fix unnecessary LDS overallocation for NGG GSSamuel Pitoiset2019-09-161-7/+1
| | | | | | | Ported from RadeonSI. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: adjust the LDS size for VS/TES NGG streamoutSamuel Pitoiset2019-09-161-8/+9
| | | | | | | It should account for the number of streamout outputs. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: unconditionally declare scratch space for NGG streamout without GSSamuel Pitoiset2019-09-161-0/+22
| | | | | | | Streamout outputs are stored in the ESGS ring. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: adjust the GS NGG scratch size for streamoutSamuel Pitoiset2019-09-162-3/+19
| | | | | | | It needs more space for multiple streams. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: allocate GDS/OA buffer objects for NGG streamoutSamuel Pitoiset2019-09-163-4/+70
| | | | | | | This allocates two BOs for GFX10 NGG streamout. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: implement NGG streamout begin/end functionsSamuel Pitoiset2019-09-161-6/+105
| | | | | Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/gfx10: add an option to switch from legacy to NGG streamoutSamuel Pitoiset2019-09-167-24/+43
| | | | | | | | | | This internal option is turned off by default because NGG streamout still hangs. It seems like it's related to GDS as RadeonSI. That option will be turned on once all issues are resolved. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv/winsys: add support for GS and OA domainsSamuel Pitoiset2019-09-162-1/+7
| | | | | | | For NGG streamout which uses GDS. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* vulkan: add vk_x11_strict_image_count optionLionel Landwerlin2019-09-151-0/+1
| | | | | | | | | | | | | | | | | | This option strictly allocate the minImageCount given by the application at swapchain creation. This works around application that do not deal with the fact that the implementation allocates more images than the minimum specified. v2: Add values in default drirc (Bas) v3: specify engine name/version (Lionel) Signed-off-by: Lionel Landwerlin <[email protected]> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111522 Reviewed-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Eric Engestrom <[email protected]> Cc: 19.2 <[email protected]>
* driconfig: add a new engine name/version parameterLionel Landwerlin2019-09-151-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Vulkan applications can register with the following structure : typedef struct VkApplicationInfo { VkStructureType sType; const void* pNext; const char* pApplicationName; uint32_t applicationVersion; const char* pEngineName; uint32_t engineVersion; uint32_t apiVersion; } VkApplicationInfo; This enables the Vulkan implementations to apply workarounds based off matching this description. Here we add a new parameter for matching the driconfig options with the following : <device driver="anv"> <application engine_name_match="MyOwnEngine.*" engine_versions="10:12,40:42"> <option name="blaaah" value="true" /> </application> </device> v2: switch engine name match to use regexps v3: Verify that the regexec returns REG_NOMATCH for match failure (Eric) v4: Add missing bit that went to the following commit (Eric) Signed-off-by: Lionel Landwerlin <[email protected]> Reviewed-by: Eric Engestrom <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]> Cc: 19.2 <[email protected]>
* radv: store engine nameLionel Landwerlin2019-09-152-0/+16
| | | | | | | | | | | We'll use this later for a new driconfig matching parameter. v2: Avoid leak in device creation error case (Bas) Signed-off-by: Lionel Landwerlin <[email protected]> Reviewed-by: Eric Engestrom <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]> Cc: 19.2 <[email protected]>
* radv: fix allocating number of user sgprs if streamout is usedSamuel Pitoiset2019-09-131-1/+1
| | | | | | | | | | streamout_buffers is assigned after that function, so the previous fix was completely wrong. This probably fix something when streamout buffers and push constants are used/inlined in the same shader. Fixes: 378e2d24143 ("radv: fix computing number of user SGPRs for streamout buffers") Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* ac: replace HAVE_LLVM with LLVM_VERSION_MAJOR for atomic-optimizationsMarek Olšák2019-09-111-1/+1
| | | | trivial
* radv/gfx10: declare a LDS symbol for the NGG emit spaceSamuel Pitoiset2019-09-103-32/+19
| | | | | | | | | | | | | This fixes some interactions when NGG GS is enabled. It fixes: - dEQP-VK.clipping.user_defined.clip_cull_distance_dynamic_index.*geom* - dEQP-VK.tessellation.geometry_interaction.passthrough.* For some reasons, using the computed ESGS ring size randomly hangs with CTS. For now, just use the maximum LDS size for ESGS. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: calculate GFX9 GS and GFX10 NGG states before compiling shader variantsSamuel Pitoiset2019-09-101-35/+48
| | | | | Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: store the ESGS ring size as part of gfx10_ngg_infoSamuel Pitoiset2019-09-102-1/+3
| | | | | Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: store GFX10 NGG state as part of the shader infoSamuel Pitoiset2019-09-102-44/+46
| | | | | Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-20 19:28:01 +0000