Lines Matching full:radeonsi

32 - GL_INTEL_blackhole_render on radeonsi
38 - GL_NV_shader_atomic_int64 on radeonsi
82 - \[radeonsi\] After 549ae5f84375dfadb86cfd465f0103acfae3249f commit Firefox Nightly Asan begins cr…
97 - Factorio v1.0 - Linux native - 64 bit - OpenGL/radeonsi - completely broken rendering
129 - \[Regression][Bisected][20.2][radeonsi\] American Truck Simulator continually allocates memory un…
130 - \[radeonsi\] bottom mips of height=1 2D texture is uninitialised after upload
174 - radeonsi: radeonsi crashes in Chrome on chromeos
849 - radeonsi: Work around Wasteland 2 bug.
852 - radeonsi: Put retile map in separate buffers.
853 - radeonsi: Move display dcc dirty tracking to framebuffer emission.
855 - radv,radeonsi: Disable compression on interop depth images
868 - radeonsi: Fix imports with displayable DCC.
1220 - CI: Don't run pixmark-piano twice on radeonsi
1430 - radeonsi: support for external buffers (ext_external_objects)
1503 - pipe-loader: Use real galliumvl if radeonsi is being linked.
1930 - radeonsi: set compiler flag lower_uniforms_to_ubo
2545 - radeonsi: enable support for GL_NV_shader_atomic_int64
2546 - radeonsi/gfx10: fix stream index for multi-stream overflow query
2723 - radeonsi: Only call nir_lower_var_copies at the end of the opt loop
3307 - radeonsi: enable ETC2 hw acceleration on Raven2
3309 - Revert "radeonsi: honor a user-specified pitch on gfx10.3"
3310 - radeonsi: use correct wave size in gfx10_ngg_calculate_subgroup_info
3311 - radeonsi: use the same units for esgs_ring_size and ngg_emit_size
3312 - radeonsi: increase minimum NGG vertex count requirement per workgroup on gfx 10.3
3313 - radeonsi: fix applying the NGG minimum vertex count requirement
3314 - radeonsi: don't count unusable vertices to the NGG LDS size
3315 - radeonsi: add a common function for getting the size of gs_ngg_scratch
3316 - radeonsi: remove the NGG hack decreasing LDS usage to deal with overflows
3317 - radeonsi: various fixes for gfx10.3
3318 - radeonsi: disable NGG culling on gfx10.3 because of hangs
3319 - radeonsi: fix compute-based culling with VERTEX_COUNTER_GDS_MODE == 1
3335 - radeonsi: fix tess levels coming as scalar arrays from SPIR-V
3345 - radeonsi: don't crash if input_usage_mask is 0 for a VS input
3346 - radeonsi: get color interpolation info from shader_info
3347 - radeonsi: clean up code for loading VS inputs
3349 - radeonsi: lower IO intrinsics - complete rewrite of input/output scanning
3350 - radeonsi: remove in/out/uniform variables from NIR after lowering IO
3351 - radeonsi: don't lower indirect IO in GLSL
3352 - radeonsi: don't execute LDS stores for TCS outputs that are never read
3353 - radeonsi: simplify handling color interp modes in si_emit_spi_map
3354 - radeonsi: change PIPE_SHADER to MESA_SHADER (si_shader_selector::type)
3355 - radeonsi: change PIPE_SHADER to MESA_SHADER (si_shader_context::type)
3356 - radeonsi: change PIPE_SHADER to MESA_SHADER (debug flags)
3357 - radeonsi: change PIPE_SHADER to MESA_SHADER (si_compile_llvm)
3358 - radeonsi: change PIPE_SHADER to MESA_SHADER (si_get_shader_part)
3359 - radeonsi: remove unused si_shader_context::type
3360 - radeonsi: change PIPE_SHADER to MESA_SHADER (si_shader_dump_disassembly)
3361 - radeonsi: precompute si_*_descriptors_idx in si_shader_selector
3362 - radeonsi: change PIPE_SHADER to MESA_SHADER (si_dump_descriptors)
3363 - radeonsi: remove si_shader_selector::type
3364 - compiler: add INTERP_MODE_COLOR for radeonsi
3365 - radeonsi: replace TGSI_INTERPOLATE with INTERP_MODE
3366 - radeonsi: replace TGSI_SEMANTIC with VARYING_SLOT and FRAG_RESULT
3367 - radeonsi: optimize out the loop in si_get_ps_input_cntl
3373 - radeonsi: remove redundant no-signed-zero-fp-math LLVM attribute
3374 - radeonsi: move nir_shader_compiler_options into si_screen
3375 - Revert "ac: generate FMA for inexact instructions for radeonsi"
3385 - radeonsi: stop using TGSI_PROPERTY_NEXT_SHADER
3386 - radeonsi: stop using TGSI_PROPERTY_VS_WINDOW_SPACE_POSITION / VS_BLIT_SGPRS_AMD
3387 - radeonsi: stop using TGSI_PROPERTY_TCS_VERTICES_OUT
3388 - radeonsi: stop using TGSI_PROPERTY_TES_POINT_MODE / TES_PRIM_MODE
3389 - radeonsi: stop using TGSI_PROPERTY_TES_SPACING
3390 - radeonsi: stop using TGSI_PROPERTY_TES_VERTEX_ORDER_CW
3391 - radeonsi: stop using TGSI_PROPERTY_GS\_\*
3392 - radeonsi: stop using TGSI_PROPERTY_CS\_\*
3393 - radeonsi: stop using TGSI_PROPERTY_FS_EARLY_DEPTH_STENCIL
3394 - radeonsi: stop using TGSI_PROPERTY_FS_POST_DEPTH_COVERAGE
3395 - radeonsi: stop using TGSI_PROPERTY_FS_COORD_PIXEL_CENTER
3396 - radeonsi: stop using TGSI_PROPERTY_FS_DEPTH_LAYOUT
3397 - radeonsi: stop using TGSI_PROPERTY_CS_LOCAL_SIZE
3398 - radeonsi: stop using TGSI_PROPERTY_FS_COLOR0_WRITES_ALL_CBUFS
3399 - radeonsi: remove \info::samplers_declared, image_buffers, msaa_images_declared
3400 - radeonsi: remove redundant si_shader_info::shader_buffers_declared
3401 - radeonsi: remove redundant si_shader_info::images_declared
3402 - radeonsi: remove redundant si_shader_info::const_buffers_declared
3403 - radeonsi: remove redundant si_shader_info:\*(clip|cull)\* fields
3404 - radeonsi: remove unused si_shader_info::uses_(vertexid|basevertex)
3405 - radeonsi: merge uses_persp_opcode_interp_sample/uses_linear_opcode_interp_sample
3406 - radeonsi: remove redundant si_shader_info::uses_kill
3407 - radeonsi: reduce type sizes in si_shader_selector
3408 - radeonsi: rename num_memory_instructions -\> num_memory_stores
3409 - radeonsi: remove redundant si_shader_info::writes_memory
3410 - radeonsi: remove redundant GS variables in si_shader_selector
3411 - radeonsi: remove redundant si_shader_selector::max_gs_stream
3412 - radeonsi: remove redundant si_shader_info::uses_derivatives
3413 - radeonsi: use shader_info::cs::local_size_variable to clean up some code
3414 - radeonsi: deduplicate setting key.mono.u.vs_export_prim_id
3415 - radeonsi: kill point size VS output if it's not used by the rasterizer
3416 - radeonsi: set outputs_written_before_ps for geometry shaders too
3417 - radeonsi: eliminate unused shader outputs for separate NGG geometry shaders
3418 - radeonsi: remove swizzle == ~0 dead code in si_llvm_load_input_gs
3419 - ac,radeonsi: lower 64-bit IO to 32 bits and remove all dead code
3420 - radeonsi: inline trivial PS functions
3432 - nir,radeonsi: move ffma fusing to late optimizations for better codegen
3433 - radeonsi: clean up ffma handling
3434 - Revert "radeonsi: set BIG_PAGE fields on gfx10.3"
3435 - Revert "radeonsi: move L2_CACHE_CONTROL registers into si_emit_framebuffer_state"
3436 - radeonsi: don't lower pack for better 16-bit vectorization
3437 - radeonsi: set flags for FP16 in shaders
3438 - radeonsi: implement 16-bit FS color outputs
3439 - radeonsi: vectorize IO for better ALU vectorization
3440 - radeonsi: don't scalarize 16-bit vec2 ALU opcodes
3441 - radeonsi: add 16-bit ALU vectorization
3446 - radeonsi: set TRUNC_COORD=0 for Total War: WARHAMMER to fix it
3447 - radeonsi: move debug options from si_disk_cache_create to si_get_ir_cache_key
3448 - radeonsi: remove KILL_PS_INF_INTERP/CLAMP_DIV_BY_ZERO, use screen::options
3451 - radeonsi: set KEEP_TOGETHER_ENABLE if needed
3452 - radeonsi: move binning parameters into si_screen
3453 - radeonsi: break a binning batch on a new PS if bins can use multiple state sets
3454 - radeonsi: add a tweak for PS wave CU utilization for gfx10.3
3457 - radeonsi: fuse or lower ffma optimally on all chips
3459 - radeonsi: use optimal order of operations when setting up a compute dispatch
3460 - radeonsi: call si_upload_graphics_shader_descriptors before the big conditional
3461 - radeonsi: move a displaced comment in si_draw_vbo
3462 - radeonsi: don't call emit_cache_flush after uploading bindless descriptors
3463 - radeonsi: reorganize the code around the gfx9 scissor bug
3464 - radeonsi: move si_upload_vertex_buffer_descriptors into si_state_draw.c
3465 - radeonsi: add unlikely statements into si_draw_vbo
3466 - radeonsi: lift the conditional for skipping si_upload_vertex_buffer_descriptors
3467 - radeonsi: always inline draw-related functions that have only one use
3474 - radeonsi: fix indirect dispatches with variable block sizes
3475 - radeonsi: call nir_shader_gather_info after lowering and optimizing NIR
3476 - radeonsi: use info.system_values_read
3477 - radeonsi: get information about FS color outputs from shader_info directly
3478 - radeonsi: get input/output usage flags from shader_info directly
3479 - radeonsi: run NIR optimizations that glsl_to_nir runs but other places might not
3480 - radeonsi: assume that constant load_local_group_size has been optimized out
3481 - radeonsi: remove redundant variables from struct si_compute
3482 - radeonsi: remove redundant info.uses_fbfetch
3489 - radeonsi: Fix dead lock with aux_context_lock in si_screen_clear_buffer.
3490 - radeonsi: simplify NGG culling enablement and add radeonsi_shader_culling option
3491 - radeonsi: kill disabled clip distances and planes at per-channel granularity
3492 - radeonsi: move si_set_active_descriptors_for_shader into si_update_common_shader_state
3493 - radeonsi: use staging buffer uploads for most VRAM buffers
3494 - radeonsi: call nir_lower_bool_to_int32 last because it breaks nir_opt_if
3495 - radeonsi: restructure si_pipe_set_constant_buffer
3503 - radeonsi: remove dead variable postponed_kill
3504 - radeonsi: implement GL_INTEL_blackhole_render
3524 - radeonsi: update the DMA perf test
3525 - radeonsi: disable SDMA on gfx6-7 and gfx10.3 to decrease CPU overhead
3526 - Revert "radeonsi/gfx10: disable vertex grouping"
3527 - radeonsi: don't disable NGG culling on gfx10.3
3528 - radeonsi: enable NGG culling by default on gfx10.3 dGPUs
3529 - radeonsi: optimize out LDS bank conflicts in the NGG culling shader
3530 - radeonsi: remove indirection when loading position at the end for NGG culling
3531 - radeonsi: write VS/TES system values into LDS after culling
3532 - radeonsi: pack LDS better for NGG culling
3533 - radeonsi: tweak LATE_ALLOC_GS numbers for faster NGG culling
3534 - radeonsi: enable NGG on Navi14 PRO cards
3535 - radeonsi: enable NGG culling by default on Navi1x PRO cards
3546 - Revert "radeonsi: use staging buffer uploads for most VRAM buffers"
3554 - radeonsi: implement inlinable uniforms
3558 - radeonsi: add num_draws parameter into si_need_gfx_cs_space
3559 - radeonsi don't get count from pipe_draw_info in si_num_prims_for_vertices
3560 - radeonsi: don't check info->count == 0
3561 - radeonsi: implement multi_draw but supporting only 1 draw
3562 - radeonsi: add support for multi draws
3563 - radeonsi: set NOT_EOP for back-to-back draws on gfx10+
3564 - radeonsi: implement multi_draw for compute-based primitive culling
3568 - radeonsi: fix min_direct_count value
3569 - radeonsi: do VGT_FLUSH when switching NGG -\> legacy on Sienna Cichlid
3570 - radeonsi: only do VGT_FLUSH for fast launch if previous draw was normal launch
3571 - radeonsi: determine correctly if switching from normal launch to fast launch
3572 - radeonsi: add options.inline_uniforms to the shader cache key
3575 - radeonsi: fix NGG streamout regression
3576 - radeonsi: fix scan_instruction for bindless inc_wrap/dec_wrap atomics
3582 - radeonsi: fix a memory leak in si_create_dcc_retile_cs
3583 - radeonsi: fix a nasty bug in si_pm4.c
3584 - radeonsi: disable WGP mode on gfx10.3 to prevent hangs
3653 - android: gallium/radeonsi: cleanup si_driinfo.h gen rules
3692 - ci: Use ignore_scheduled_pipelines anchor in .radeonsi-rules
3912 - radeonsi: enable PIPE_CAP_NO_CLIP_ON_COPY_TEX
3914 - radeonsi,driconf: add clamp_div_by_zero option
3915 - radeonsi: use radeonsi_clamp_div_by_zero for SPECviewperf13, Road Redemption
3929 - radeonsi: fix quant_mode selection for large negative values
3930 - radeonsi: fix guardband handling for large values
3933 - radeonsi: move GL vendor workaround to drirc
3934 - radeonsi: reduce PIPE_SHADER_CAP_MAX_CONST_BUFFER_SIZE value
3935 - radeonsi: change vendor name to AMD
3936 - radeonsi: force linear for textures with height=1 (gfx6-8)
3937 - radeonsi/tmz: use secure job if framebuffer has dcc
3938 - radeonsi/tmz: use secure job if using an encrypted z/s buffer
3939 - radeonsi/tmz: add safety assert when tmz is enabled
3940 - radeonsi/tmz: allocate depth/stencil buffers as encrypted
3941 - radeonsi: introduce SI_RESOURCE_FLAG_INTERNAL / RADEON_FLAG_DRIVER_INTERNAL
3944 - radeonsi/tmz: allow secure job if the app made a tmz allocation
3946 - radeonsi/tmz: fail si_texture_transfer_map if tex is encrypted
3947 - radeonsi/tmz: add tmz variant of sctx::wait_mem_scratch
3948 - radeonsi/tmz: add tmz variant for sctx::tess_rings
3949 - radeonsi: disable primitive discard if tmz is in use
3950 - radeonsi/tmz: add a tmz variant for sctx::eop_bug_scratch
3951 - radeonsi/tmz: add workaround for mpv/vaapi subtitles
3958 - radeonsi: honor PIPE_BIND_PROTECTED
3960 - radeonsi: enable PIPE_CAP_DEVICE_PROTECTED_CONTENT
3965 - radeonsi: fix RADEON_FLUSH flags conflicts
3970 - radeonsi/gfx10: flush gfx cs on ngg -\> legacy transition
3979 - radeonsi: fix syncobj wait timeout
3980 - radeonsi: fix user fence space when MCBP is enabled
3981 - radeonsi: fix max syncobj wait timeout
3982 - radeonsi: fix user fence GPU address
4050 - radeonsi: don't use nir_opt_access
4326 - ci: adjust RadeonSI rules
4335 - radeonsi: call nir_io_add_const_offset_to_base only once per shader
4359 - radeonsi: remove dead code in TCS/TES/GS since const_index is always 0
4360 - ac,radv,radeonsi: remove unused parameters in the shader ABI IO
4374 - ac,radv,radeonsi: stop multiplying driver_location by 4
4555 - ci: Run deqp-gles2 on RadeonSI
4556 - ci: Run deqp-gles3 and deqp-gles31 on RadeonSI