AgeCommit message (Expand)AuthorFilesLines
2021-07-14intel/compiler: Update block IPs once in opt_cmod_propagation21.2-branchpointIan Romanick1-8/+14
2021-07-14intel/compiler: Update block IPs once in register_coalesceIan Romanick1-1/+3
2021-07-14intel/compiler: Update block IPs once in dead_code_eliminateIan Romanick1-1/+3
2021-07-14intel/compiler: Add cfg_t::adjust_block_ips() methodIan Romanick1-0/+21
2021-07-14intel/compiler: Add the ability to defer IP updates in backend_instruction::r...Ian Romanick4-4/+22
2021-07-14aco/live_var_analysis: change worklist to a single integerDaniel Schürmann1-10/+6
2021-07-14aco/insert_waitcnt: Remove many unnecessary wait_imm.combine()Daniel Schürmann1-37/+23
2021-07-14aco/isel: avoid unnecessary calls to nir_unsigned_upper_bound()Daniel Schürmann1-1/+1
2021-07-14broadcom/compiler: implement gl_PrimitiveID in FS without a GSIago Toral Quiroga7-7/+50
2021-07-14v3d: better scissor trackingIago Toral Quiroga5-4/+82
2021-07-14zink: respect line_rectangular stateErik Faye-Lund1-12/+12
2021-07-14zink: check for right featureErik Faye-Lund1-2/+2
2021-07-14vulkan: allocate host-visible memory for swapchain imagesErik Faye-Lund1-1/+1
2021-07-14vulkan: do not map zero-sized region of memoryErik Faye-Lund1-1/+1
2021-07-14anv/allocator: Use list->u64 in free_list_pushJason Ekstrand1-1/+1
2021-07-14llvmpipe: Handle NULL views in llvmpipe_cleanup_stage_samplingJoshua Ashton1-4/+5
2021-07-14freedreno+turnip: Add a6xx gen4 supportRob Clark2-0/+37
2021-07-14freedreno+turnip: Get device name from device-info tableRob Clark6-26/+43
2021-07-14freedreno+turnip: Add has_8bpp_ubwcRob Clark6-10/+24
2021-07-14freedreno+turnip: Add has_cp_reg_writeRob Clark6-6/+38
2021-07-14turnip: Get has_sample_locations from fd_dev_infoRob Clark3-1/+4
2021-07-14turnip: Get has_tex_filter_cubic from fd_dev_infoRob Clark3-2/+5
2021-07-14turnip: Get indirect_draw_wfm_quirk from fd_dev_infoRob Clark3-9/+16
2021-07-14turnip: Get storage_16bit from fd_dev_infoRob Clark5-5/+11
2021-07-14turnip: Drop unused vshs_workgroup paramRob Clark3-7/+4
2021-07-14freedreno/ir3: Get reg_size_vec4 from fd_dev_infoRob Clark3-7/+8
2021-07-14freedreno/ir3: Get tess_use_shared from fd_dev_infoRob Clark4-3/+8
2021-07-14turnip: Convert fd_dev_info to const pointerRob Clark10-32/+31
2021-07-14freedreno: Convert fd_dev_info to const pointerRob Clark11-39/+37
2021-07-14freedreno: Generate device-info tables at build timeRob Clark11-134/+338
2021-07-14freedreno: Rename *_dev_infoRob Clark9-10/+10
2021-07-14freedreno/common: unhardcode CCU color cache offsetJonathan Marek15-41/+58
2021-07-14freedreno/a6xx: larger gmem_page_align from tile align instead of gpu idJonathan Marek2-7/+1
2021-07-14tu: remove workaround for conditional rendering + hw binningJonathan Marek1-30/+0
2021-07-14iris: Use simple_mtx in the bufmgr.Kenneth Graunke1-32/+28
2021-07-14panfrost: Initialise the blend equation in create_blend_stateIcecream951-1/+1
2021-07-14crocus: use simple_mtx in the bufmgrDave Airlie1-32/+29
2021-07-13radv, aco, ac/nir: Tweak position export scheduling for NGG culling.Timur Kristóf5-3/+20
2021-07-13ac/nir: Reuse uniforms from top part of culling shaders.Timur Kristóf1-0/+163
2021-07-13ac/nir: Analyze culling shaders to remember which inputs are used when.Timur Kristóf2-0/+99
2021-07-13ac/nir: Reuse the repacked output positions of culling shaders.Timur Kristóf1-1/+66
2021-07-13radv: Run algebraic optimizations before NGG lowering.Timur Kristóf1-0/+3
2021-07-13radv: Support NGG culling with new perftest environment variable.Timur Kristóf8-5/+64
2021-07-13aco: Implement NGG culling related intrinsics.Timur Kristóf2-2/+98
2021-07-13radv: New shader args for NGG culling settings and viewport.Timur Kristóf7-5/+315
2021-07-13radv: Expose radv_get_viewport_xform in radv_private.hTimur Kristóf2-5/+8
2021-07-13ac/nir: Implement NGG deferred attribute culling in NIR.Timur Kristóf2-12/+557
2021-07-13ac/nir: Use a ballot that matches the wave size during NGG lowering.Timur Kristóf1-5/+10
2021-07-13ac/nir: Add a NIR port of ac_llvm_cull.Timur Kristóf3-0/+168
2021-07-13nir: Add AMD specific intrinsics for NGG shader based culling.Timur Kristóf2-1/+31