summaryrefslogtreecommitdiff
path: root/src/compiler
AgeCommit message (Expand)AuthorFilesLines
2020-10-13scons: fix SPIR-V -> NIR buildRhys Perry1-0/+7
2020-10-13nir/opt_uniform_atomics: don't optimize atomics twiceRhys Perry1-0/+76
2020-10-13nir/opt_uniform_atomics: optimize image atomicsRhys Perry1-3/+12
2020-10-13nir: add pass to optimize uniform atomicsRhys Perry4-0/+261
2020-10-13nir: allow divergence information to be updated when inserting instructionRhys Perry3-26/+59
2020-10-13nir: move divergence analysis options to nir_shader_compiler_optionsRhys Perry2-18/+22
2020-10-13nir: add last_invocation intrinsicRhys Perry2-1/+3
2020-10-13radv/aco,nir/lower_subgroups: don't lower electRhys Perry2-0/+4
2020-10-13nir: update ubo locations in nir_lower_uniforms_to_uboMike Blumenkrantz1-1/+6
2020-10-13glsl: fix up location setting for variables pointing to a UBO's baseMike Blumenkrantz1-42/+45
2020-10-12android: fix SPIR-V -> NIR buildRhys Perry3-2/+8
2020-10-12spirv: replace discard with demote for incorrect HLSL->SPIR-V translationsRhys Perry3-3/+30
2020-10-12spirv: add and use a generator id enumRhys Perry6-2/+278
2020-10-11nir: Allow more deref modes in phisJason Ekstrand1-6/+7
2020-10-11android: util: Move xxd.py to utilMauro Rossi1-1/+1
2020-10-10nir/algebraic: optimize iand/ior of (n)eq zero when umax/umin not availableJose Maria Casanova Crespo2-4/+12
2020-10-09disk_cache: build option for disabled-by-defaultJohn Bates1-0/+21
2020-10-09nir: return progress from nir_lower_io_to_scalar_earlyRhys Perry2-7/+9
2020-10-09nir: Add ability to count primitives per stream.Timur Kristóf2-6/+19
2020-10-09nir: Count vertices per stream.Timur Kristóf2-14/+30
2020-10-09nir: Add ability to overwrite incomplete GS primitives.Timur Kristóf2-1/+79
2020-10-09nir: Add ability to count emitted GS vertices per primitive.Timur Kristóf3-6/+53
2020-10-09nir: Add ability to count emitted GS primitives.Timur Kristóf4-25/+74
2020-10-08nir/opt_deref: Fix the vector bitcast optimizationJason Ekstrand1-0/+4
2020-10-08st/mesa: Add NV12 lowering to PIPE_FORMAT_R8_G8B8_420_UNORMKristian H. Kristensen2-0/+22
2020-10-07spirv: Move nir_lower_libclc to src/compiler/spirvJason Ekstrand4-0/+118
2020-10-07spirv: Add a shared libclc loaderJason Ekstrand4-1/+371
2020-10-07util: Move xxd.py to utilJesse Natalie3-112/+2
2020-10-07glsl/xxd.py: fix importsDylan Baker1-2/+1
2020-10-07nir: Fix unaligned pointer accessTony Wasserka1-6/+9
2020-10-07nir: Fix undefined behavior due to signed integer multiplication overflowsTony Wasserka1-2/+7
2020-10-07nir: add new helper passes that lower uniforms to literalsMarek Olšák5-0/+214
2020-10-07nir/algebraic: always lower idiv to shifts if bitops are allowedMarek Olšák2-5/+2
2020-10-06nir/constant_folding: init nir_const_value to zeroSamuel Pitoiset1-0/+1
2020-10-06nir/constant_folding: Fold load_deref of nir_var_mem_constantJason Ekstrand1-0/+80
2020-10-06nir: Validate constant initializersJason Ekstrand1-0/+31
2020-10-06nir/constant_folding: Use nir_shader_instruction_passJason Ekstrand1-53/+16
2020-10-06nir/constant_folding: Use the builderJason Ekstrand1-51/+41
2020-10-06spirv: Make the clc_shader constJason Ekstrand1-1/+1
2020-10-03glsl: Initialize ast_node member field location.path in constructor.Vinson Lee1-0/+1
2020-10-02nir: Add a pass to lower vec3s to vec4sJason Ekstrand8-0/+248
2020-10-02nir/opt_deref: Add an optimization for bitcastsJason Ekstrand1-0/+159
2020-10-02nir/opt_deref: Add an instruction type switchJason Ekstrand1-13/+19
2020-10-02nir: Add component mask re-interpret helpersJason Ekstrand3-39/+78
2020-10-02nir/opt_load_store_vectorize: Use bit sizes when checking mask compatibilityJason Ekstrand1-2/+2
2020-10-02nir: Add a memcpy optimization passJason Ekstrand4-0/+227
2020-10-02nir: Handle memcpy in copy_prop_vars and combine_storesJason Ekstrand2-2/+5
2020-10-02nir/find_array_copies: Properly discard copies for castsJason Ekstrand1-11/+15
2020-10-01nir: Fix a misspellingJason Ekstrand2-2/+2
2020-10-02glsl: don't duplicate state vars as uniforms in the NIR linkerTimothy Arceri1-0/+6