Lines Matching full:radeonsi
70 - [radeonsi] Graphical corruption in Valve Source-based games
75 - Abort due to "Unknown intrinsic: intrinsic store_deref" (firefox, radeonsi)
81 - radeonsi: error allocating sparse buffers.. bug?
82 - [BISECTED/radeonsi] Flickering triangles with wayfire cube
87 - radeonsi: driver crashes with kwin-effects-cube
93 - [Bisected][RadeonSI] Blender crashes on startup
110 - [RadeonSI] Teamfortress 2 crashing frequently with mesa 22.1.1
149 - radeonsi: Hang on RX 5500 XT on latest mesa git
176 - eglExportDMABUFImageQueryMESA crashes on amdgpu gallium/drivers/radeonsi
181 - APU downgraded to OpenGL 3.0 since UBO size change in radeonsi
218 - radeonsi-stoney: Failing jobs during a CI bug
231 - radeonsi: KHR_debug thread safety
820 - radeonsi: Add nir_lower_poly_line_smooth pass for polygon and line smoothing
825 - radeonsi: Enable nir_lower_point_smooth lowering pass for point smoothing
1064 - radeonsi/vcn: add vcn 4.0 encode support
1065 - radeonsi/vcn: add vcn 4.0 encode fw interface version
1066 - radeonsi/vcn: add encode context ib for vcn 4.0
1067 - radeonsi/vcn: add session init ib for vcn 4.0
1068 - radeonsi/vcn: enable vcn 4.0 encode for gfx11 asic
1069 - radeonsi/vcn: use calculated max hierarchy depth for hevc enc
1072 - radeonsi/vcn/enc: enable cabac entropy mode
1519 - ac: port radeonsi tess factor calculations to common code.
1520 - radeonsi: port tess ring calcs to the common helper.
1607 - ci/radeonsi: add RoR and Nheko traces
2120 - radeonsi: port amdgcn_glslc build to meson
2250 - radeonsi: Use nir_fold_16bit_image_load_store_conversions.
2312 - radeonsi: Stop using nir_legalize_16bit_sampler_srcs.
2607 - radeonsi: r600: d3d12: st: Use NIR lowering for tg4 offset arrays instead of GLSL lowering
2673 - radeonsi/gfx11: VRS changes
2674 - radeonsi: save the fs constant buffer to the util blitter context
2697 - radeonsi/gfx11: update codec support for gfx11
2698 - radeonsi/vcn: add decode software ring support for gfx11
2699 - radeonsi/vcn: add jpeg decode support for gfx11
3512 - radeonsi: Set vertex_id_zero_based
3530 - radeonsi: Remove format desc null checks
3804 - gallium: move drivers/radeon/* into drivers/radeonsi/
3810 - radeonsi: set more state bits for disabled binning
3811 - radeonsi: don't compile monolithic PS for dual source blending
3812 - radeonsi: add an SQTT workaround for chips with disabled RBs
3813 - radeonsi: set PA_SC_EDGERULE according to bottom_edge_rule state
3814 - radeonsi: initialize line stipple config registers
3815 - radeonsi: fix ALPHA_IS_ON_MSB determination
3816 - radeonsi: handle an out-of-bounds sampler/image index better
3817 - radeonsi: don't treat skipped GS copy shader compilation as successful
3818 - radeonsi: dump shaders after si_shader_binary_upload to fix printed LDS stat
3819 - radeonsi: ignore AMD_DEBUG=nodcc if the texture is imported
3820 - radeonsi: run nir_io_add_const_offset_to_base for TES/TCS as late optimizations
3821 - radeonsi: scalarize IO instructions
3822 - radeonsi: add pipe_stream_output_info into si_shader_context
3823 - radeonsi: use nir_shader_get_entrypoint in si_nir_scan_shader
3824 - radeonsi: apply key.ge.opt.kill_{outputs,pointsize,clipdistance} in NIR
3825 - radeonsi: gather pipe_stream_output_info from NIR intrinsics
3826 - radeonsi: move most "info" fields from si_shader_selector into si_shader_info
3827 - radeonsi: move si_shader_info::stage into si_shader_selector
3829 - radeonsi: assign param export indices before compilation
3844 - ac,radeonsi: rework how scratch_waves is used and move it to ac_gpu_info.c
3848 - radeonsi: remove an obsolete comment about LLVM 14
3849 - radeonsi: fail to create pipe_screen if LLVM doesn't support the GPU
3851 - radeonsi/ci: add polaris11 baseline
3853 - util: change util_format_snorm8_to_sint8 -> snorm_to_sint for radeonsi
3859 - ac,radeonsi: report SCALED formats as unsupported by samplers and color buffers
3860 - radeonsi: always print the lower-case GPU name for radeonsi-run-tests.py
3861 - radeonsi: cleanup local includes
3862 - radeonsi: align the tessellation ring address to 2MB
3863 - radeonsi: remove unused si_cs_shader_state::uses_scratch
3864 - radeonsi: simplify setting COMPUTE_PGM_RSRC2.SCRATCH_EN
3865 - radeonsi: code style fix
3866 - radeonsi: add si_can_use_compute_blit
3867 - radeonsi: don't report R64_*INT as a sampler format because it doesn't work
3868 - radeonsi: don't allow broken compute image copies into A8R8_UNORM
3869 - radeonsi: add color buffer support for 2_10_10_10 pipe formats
3870 - radeonsi: add color buffer support for R9G9B9E5_FLOAT on gfx10.3
3871 - radeonsi: implement compute_copy_image between 1D_ARRAY and other texture types
3872 - radeonsi: cosmetic changes in si_compute_copy_image
3873 - radeonsi: assert format support in si_compute_copy_image
3874 - radeonsi: don't expose formats that CB supports, but not shaders
3875 - radeonsi: rename si_test_blit -> si_test_image_copy_region
3876 - radeonsi/test_image_copy: remove whole image copies
3877 - radeonsi/test_image_copy: remove linear and power-of-two size testing
3878 - radeonsi/test_image_copy: remove special SDMA codepaths
3879 - radeonsi/test_image_copy: simplify texture size generation
3880 - radeonsi/test_image_copy: separate generating a random pipe_resource
3881 - radeonsi/test_image_copy: separate printing pipe_resource into a function
3882 - radeonsi/test_image_copy: test all interesting texture targets
3883 - radeonsi/test_image_copy: test copying a block-compressed format
3884 - radeonsi/test_image_copy: test mipmaps
3885 - radeonsi/test_image_copy: test all formats
3886 - radeonsi/ci: update sienna_cichlid results
3887 - radeonsi: handle compressed formats in si_compute_copy_image
3888 - radeonsi: improve the subsampled copy_image calculation
3889 - radeonsi: remove compressed and subsampled gfx copy from resource_copy_region
3890 - radeonsi: remove si_create_sampler_view_custom and related code
3891 - radeonsi: remove si_create_surface_custom
3914 - radeonsi/gfx11: increase the hw screen offset alignment
3915 - radeonsi/gfx11: always allow DCC stores
3916 - radeonsi/gfx11: expect packed threadID VGPRs
3917 - radeonsi/gfx11: enable NGG-only draw paths
3918 - radeonsi/gfx11: enable arbitrary DCC format reinterpretation
3919 - radeonsi/gfx11: use the new TCS WaveID SGPR to compute vs_rel_patch_id
3920 - radeonsi/gfx11: program inst_pref_size for graphics
3921 - radeonsi/gfx11: buffer descriptor changes
3922 - radeonsi/gfx11: add CB deltas
3923 - ac,radeonsi/gfx11: remove FMASK loads
3924 - ac,radeonsi/gfx11: set SWIZZLE_ENABLE correctly
3925 - radeonsi/gfx11: don't set non-existent SPI_SHADER_USER_DATA_VS_x
3926 - radeonsi/gfx11: don't set COMPR for exports, use 0x3 channel mask instead
3927 - radeonsi/gfx11: implement attributes through memory
3928 - radeonsi/gfx11: scattered register deltas
3929 - radeonsi/gfx11: don't set non-existent CP_COHER_START_DELAY
3930 - radeonsi/gfx11: emit SQ_NON_EVENT for tessellation at the end of IBs
3931 - radeonsi/gfx11: don't use FLUSH_AND_INV_DB_META
3932 - radeonsi/gfx11: TF_RING_SIZE changed to a per-SE size
3933 - radeonsi/gfx11: limit MSAA color buffers to the RGBA channel order
3934 - radeonsi/gfx11: update the initialization of SGPR0/1 registers for HS and GS
3935 - radeonsi/gfx11: change LDS allocation granularity for PS
3936 - radeonsi/gfx11: don't count the non-existent scratch_byte_offset SGPR
3937 - radeonsi/gfx11: resolve MSAA using u_blitter
3938 - radeonsi/gfx11: mark streamout as unimplemented for now
3939 - radeonsi/gfx11: limit CP DMA to max 32KB sizes
3940 - radeonsi: inline si_cp_dma_prefetch in si_draw_vbo for lower overhead
3941 - radeonsi/gfx11: add a workaround for CB perf counters
3943 - radeonsi: remove tautologies when setting CU_EN
3951 - radeonsi: report 10_10_10_2 scaled formats as unsupported on gfx6-9 as well
3952 - radeonsi: report more missing/broken texture formats as unsupported on gfx6-9
3954 - radeonsi: wait before s_barrier in TCS epilog to fix LLVM 15
3960 - radeonsi: add si_get_tcs_epilog_key
3962 - radeonsi: inline si_nir_build_llvm
3964 - radeonsi: cleanups getting rid of the ambigous "epilogue" word
3965 - radeonsi: get rid of the ambiguous "prologue" word
3966 - radeonsi: try to group stage-specific code in si_llvm_translate_nir
3969 - radeonsi: rename RADEON_FLAG_UNCACHED -> RADEON_FLAG_GL2_BYPASS
3972 - radeonsi: use PIPE_RESOURCE_FLAG_UNMAPPABLE and DRIVER_INTERNAL more
3975 - radeonsi: use the new flag AMDGPU_GEM_CREATE_DISCARDABLE
3976 - radeonsi: use AMDGPU_VM_PAGE_NOALLOC to disable MALL (infinity cache)
3977 - radeonsi: fix a crash when failing to create a context
3978 - radeonsi: print an error when failing to create a context
3979 - radeonsi/gfx11: set BIG_PAGE for the attribute ring
3981 - radeonsi: remove GFX9_MERGED_NUM_USER_SGPR definition
3983 - radeonsi/gfx11: fix alpha-to-coverage with stencil or samplemask export
3984 - radeonsi/gfx11: don't insert shader code for GS_PIPELINE_STATS_EMU
3985 - radeonsi/gfx11: fix the value of VGT_GS_OUT_PRIM_TYPE at the beginning of IBs
3986 - Revert "radeonsi/gfx11: limit MSAA color buffers to the RGBA channel order"
3987 - radeonsi/gfx11: fix VM faults due to the attribute ring
3988 - radeonsi/gfx11: optimize attribute stores
3989 - radeonsi: merge all preamble states into one
3990 - radeonsi: move CS preamble emission into the winsys
3992 - radeonsi: fix resource_copy_region with ETC formats (e.g. for Stoney)
4010 - radeonsi: increase the max compute LDS size to 64KB for gfx7+
4011 - radeonsi: report correct maximum compute grid sizes
4012 - radeonsi: clamp against MAX_TEXEL_BUFFER_ELEMENTS correctly
4013 - radeonsi: change max TBO/SSBO sizes again and rework max alloc size
4014 - radeonsi: compute PIPE_CAP_MAX_TEXEL_BUFFER_ELEMENTS_UINT correctly
4015 - radeonsi: set the max UBO size same as the max SSBO size
4017 - radeonsi: print LDS size in bytes
4018 - radeonsi: don't use info.gs.invocations if it's not GS
4019 - radeonsi: fix uninitialized wait_mem_scratch_tmz
4020 - radeonsi: remove unused dword from wait_mem_scratch
4021 - radeonsi: fix polygon stippling without color and Z outputs (v2)
4022 - radeonsi: restructure PS no-export fixups
4023 - radeonsi: determine DB_SHADER_CONTROL in si_shader_ps
4024 - radeonsi: set INTERPOLATE_COMP_Z to 0 to work around an EQAA bug
4025 - radeonsi: add BREAK_BATCH at the beginning of IBs
4026 - radeonsi: simplify how pipeline statistic offsets are computed
4027 - radeonsi: rework how VS_STATE_BITS are set for VS, TES, and GS
4028 - radeonsi: rename and regroup VS_STATE definitions
4029 - radeonsi: move GS_STATE bits to the end to make space at the beginning
4030 - radeonsi: rework how vs_state_bits is set and unpacked
4031 - radeonsi: fix an NGG streamout hang with monolithic shaders
4032 - radeonsi: fix a crash in gfx10_sh_query_get_result_resource
4033 - radeonsi: unconditionally enable the streamout overflow query with NGG
4034 - radeonsi: inline gfx10_emit_streamout_begin/end
4035 - radeonsi: fix NGG streamout hang by allocating GDS in the right place
4036 - radeonsi: remove streamout code from shaders if no streamout buffers are bound
4038 - radeonsi: allocate GDS only once per process
4039 - radeonsi: allocate only 1 GDS OA counter for gfx10 NGG streamout
4040 - ac,radeonsi: don't export null from PS if it has no effect on gfx10+
4045 - radeonsi/gfx11: compile monolithic PS if it writes memory
4046 - radeonsi/gfx11: allocate more space for pipeline statistics
4047 - radeonsi/gfx11: synchronize correctly before setting SPI_ATTRIBUTE_RING_*
4048 - radeonsi/gfx11: don't use memory for waiting for cache flushes
4049 - radeonsi/gfx11: don't allocate unused wait_mem_scratch
4050 - radeonsi/gfx11: drop the ES vertex count requirement
4051 - radeonsi/gfx11: add missing register shadowing code
4052 - radeonsi/gfx11: rework GDS streamout code to single-lane and enable streamout
4055 - radeonsi: fix random PS wave size
4074 - radeonsi: remove workarounds for radeon DRM < 2.45.0
4086 - radeonsi: unify VGT_TESS_DISTRIBUTION programming
4087 - radeonsi: move SI_MAX_VRAM_MAP_SIZE to si_debug_options.h
4088 - radeonsi: call pipe->blit instead of util_blitter_blit after MSAA resolving
4089 - radeonsi: make si_launch_grid_internal static
4090 - radeonsi: add common helper si_launch_grid_internal_images that is more robust
4091 - radeonsi: remove compute-based DCC decompression because it's broken
4092 - radeonsi: don't do image stores with RGBX, L, LA, I, and SRGB formats
4093 - radeonsi: follow shader_info.float_controls_execution_mode (mostly)
4094 - radeonsi: add need_fmask_expand parameter into si_decompress_subresource
4095 - radeonsi: check for 16-bit hw support instead of relying on options.fp16
4096 - radeonsi: move compute-related code from si_blit.c to si_compute_blit.c
4097 - radeonsi: fold async_copy into the preceding conditional in si_blit
4098 - radeonsi: cosmetic changes around do_hardware_msaa_resolve
4099 - radeonsi/gfx11: use correct VGT_TESS_DISTRIBUTION settings
4100 - radeonsi/gfx11: set SAMPLE_MASK_TRACKER_WATERMARK = 15 and clean up
4101 - radeonsi/gfx11: use better PRIM_GRP_SIZE_GFX11 setting
4102 - radeonsi/gfx11: skip code in si_update_shaders that has no effect
4103 - radeonsi/gfx11: rename si_calc_inst_pref_size -> si_get_shader_prefetch_size
4104 - radeonsi/gfx11: enable shader prefetch except for initial chip revisions
4105 - radeonsi/gfx11: use a better workaround for the export conflict bug
4106 - radeonsi: make various blit functions non-static
4107 - radeonsi: allow texture_map to upload only 1 sample for MSAA instead of all
4108 - radeonsi: add a randomized blit test
4109 - radeonsi: move the no-AA small prim precision cull constant into an SGPR
4110 - radeonsi: move small prim precision computation out of si_emit_cull_state
4111 - radeonsi: don't assume that TC_ACTION_ENA invalidates L1 cache on gfx9
4976 - radeonsi: make si_pc_emit_shaders non static
4977 - radeonsi: add helpers to deal with spm counters
4978 - radeonsi: add spm counters setup code
4979 - radeonsi/sqtt: setup spm bo
4980 - radeonsi/sqtt: enable spm counters on gfx10+
4981 - radeonsi: make sure profile_peak is used before using sqtt
4984 - radeonsi: remove commutative_blend_add option
4985 - radeonsi: drop assume_no_z_fights option
4988 - radeonsi: enable PIPE_CAP_ALLOW_DRAW_OUT_OF_ORDER
4989 - radeonsi: fix VS kill_outputs handling
4990 - radeonsi: don't always strip NIR debug infos
4991 - radeonsi: generate unique shader name in si_get_nir_shader
4993 - radeonsi/gfx11: add assert in legacy vs path
4994 - radeonsi/gfx11: register changes
4995 - radeonsi/gfx11: image descriptor changes
4996 - radeonsi/tests: allow empty line and comments in csv files
4997 - radeonsi/tests: add a heuristic to pick the baseline
4998 - radeonsi/tests: reformat with black
5000 - radeonsi: flush VGT streamout like PAL
5001 - radeonsi: implement pipeline stats workaround
5005 - radeonsi: deduplicate query offsets
5006 - radeonsi: store the pipeline stats index
5007 - radeonsi: implement GL_GEOMETRY_SHADER_PRIMITIVES_EMITTED_ARB in shaders
5008 - radeonsi/test: update sienna_cichlid results
5009 - radeonsi/query: use the qbo correct size
5010 - radeonsi/ngg: reuse the pipeline stats buffer when using atomics
5011 - radeonsi: fix gs_invocation query with NGG
5012 - radeonsi: don't use wave32 for GE on gfx10 if culling is used
5015 - radeonsi: remove unused params from si_shader_select_with_key
5016 - radeonsi: scale the number of shader compiler threads
5017 - radeonsi: store shader variants in an array
5019 - radeonsi: fix glTexBuffer max size handling
5020 - radeonsi: wait for PS idle in si_set_framebuffer_state
5021 - radeonsi/tests: add a --slow option
5022 - radeonsi/tests: use a smaller tests-per-group value
5023 - radeonsi: don't use sel->nir in si_check_blend_dst_sampler_noop
5027 - radeonsi: enable use_waterfall_for_divergent_tex_samplers
5032 - radeonsi: set size in si_texture_get_handle
5034 - radeonsi: prevent recursion in si_decompress_dcc
5035 - radeonsi: add helper to use si_screen::aux_context
5036 - radeonsi: use helpers to access si_screen::aux_context
5044 - radeonsi: use LLVMBuildLoad2 where possible
5048 - radeonsi: use LLVMBuildLoad2 in llvm PS
5049 - radeonsi: use LLVMBuildLoad2 for inter-stage outputs loads
5050 - radeonsi: check last_dirty_buf_counter and dirty_tex_counter
5053 - radeonsi: prevent u_blitter recursion in si_update_ps_colorbuf0_slot
5054 - radeonsi: use nir_opt_large_constants earlier
5056 - radeonsi: invalidate L2 when using dcc stores
5060 - radeonsi: lower nir_intrinsic_sparse_residency_code_and
5080 - radeonsi: implement load_lshs_vertex_stride abi
5081 - radeonsi: add tcs_vgpr_only_inputs parameter to si_get_nir_shader
5083 - radeonsi: replace llvm ls/hs interface lds ops with nir lowered ones
5087 - radeonsi: deserialize nir binary in si_check_blend_dst_sampler_noop
5088 - radeonsi: use si_shader as parameter in si_get_nir_shader
5089 - radeonsi: support multi stage shader state creation in nir shaderlib
5090 - radeonsi: add si_create_passthrough_tcs
5091 - radeonsi: replace llvm based fixed tcs with nir
5092 - radeonsi: implement nir_intrinsic_load_tcs_num_patches_amd
5098 - radeonsi: preload tess offchip ring for tcs
5099 - radeonsi: implement nir_intrinsic_load_ring_tess_offchip(_offset)_amd
5100 - radeonsi: implement nir_intrinsic_load_tess_rel_patch_id_amd for both tcs and tes
5103 - radeonsi: replace llvm tcs output with nir lower pass
5104 - radeonsi: enable PIPE_CAP_GLSL_TESS_LEVELS_AS_INPUTS
5105 - radeonsi: set uses_vmem_load_other for nir_intrinsic_load_buffer_amd
5106 - radeonsi: replace llvm tes input load with nir lowering
5108 - radeonsi: implement two esgs ring nir intrinsic
5110 - radeonsi: set lds for gs/es to handle nir shared memory load/store
5113 - radeonsi: replace llvm es output with nir lowering
5116 - radeonsi: replace llvm gs input handle with nir lowering
5121 - radeonsi: fix tcs_out_lds_offsets arg alignment
5343 - radeonsi/vcn: Add support of array_mode for gfx11
5344 - radeonsi/vcn: update av1 decoding to support vcn4
5345 - radeonsi/vcn: prepare for unified queue in vcn4
5346 - radeonsi/vcn: support unified queue in vcn4
5402 - ac,radeonsi: add has_sqtt_auto_flush_mode_bug
5630 - gallium/va/radeonsi: Using private as a parameter name conflicts with C++ keywords
5674 - radeonsi: Set display_remote for non-refernced frames
5675 - radeonsi/vcn : update enc->dpb ref_use for index 0
5676 - Revert "radeonsi: Set display_remote for non-refernced frames"
5955 - radeonsi: Move NULL check before dereference.
5971 - radeonsi: Add support for EGL_IMG_context_priority
6033 - radeonsi/gfx11: make flat_scratch changes for graphics
6034 - radeonsi/gfx11: make flat_scratch changes for compute
6035 - radeonsi/gfx11: interp changes for 32bit
6036 - radeonsi/gfx11: interp changes for 16bit
6037 - radeonsi/gfx11: instruction cache line size is 128 bytes
6038 - radeonsi/gfx11: program inst_pref_size for compute
6039 - radeonsi/gfx11: program db render control register
6040 - radeonsi/gfx11: export alpha through mrtz for alpha-to-coverage if mrtz is there
6041 - ac,radeonsi/gfx11: swizzle MRT0/1 for dual source blending
6043 - radeonsi: remove tabs from code
6044 - radeonsi: no need to call si_pm4_clear_state() in si_pm4_free_state()
6048 - radeonsi/gfx11: use PIXEL_PIPE_STATE_DUMP event instead of ZPASS_DONE