- ee1c69f glsl: Don't increase the iteration count when there are no terminators by Ian Romanick · 5 years ago
- 6e2ff10 glsl/nir: Fix copying 64-bit values in uniform storage by Caio Marcelo de Oliveira Filho · 5 years ago
- 390ff8a glsl/nir: Fix copying vector constant values by Caio Marcelo de Oliveira Filho · 5 years ago
- a8b0b6e nir: introduce lowering of bitfield_insert to bfm and a new opcode bitfield_select. by Daniel Schürmann · 5 years ago
- 1403c3a nir/algebraic: Use unsigned comparison when lowering bitfield insert/extract by Daniel Schürmann · 5 years ago
- 4eeb49e nir/algebraic: Remove unnecessary iand of [iu]bfe and bfm sources by Daniel Schürmann · 6 years ago
- 165b7f3 nir: define behavior of nir_op_bfm and nir_op_u/ibfe according to SM5 spec. by Daniel Schürmann · 6 years ago
- a74f256 nir/algebraic: add optimization pattern for ('ult', a, ('and', b, a)) and friends. by Daniel Schürmann · 6 years ago
- 5643445 nir/lower_tex: Add an assert() in nir_lower_txs_lod() by Boris Brezillon · 5 years ago
- 1213109 spirv: Restrict use of descriptor intrinsics to Vulkan by Caio Marcelo de Oliveira Filho · 5 years ago
- 81e51b4 nir: Make nir_constant a vector rather than a matrix by Jason Ekstrand · 5 years ago
- b019fe8 glsl/nir: Fix handling of 64-bit values in uniform storage by Jason Ekstrand · 5 years ago
- a54e397 spirv: Only copy needed components for OpSpecConstantOp by Jason Ekstrand · 5 years ago
- 96bb9c9 spirv: Use a single path for OpSpecConstantOp of OpVectorShuffle by Jason Ekstrand · 5 years ago
- 280e544 spirv: Use vtn_constan_uint() for array lengths and gather components by Jason Ekstrand · 5 years ago
- aa11c2e spirv: Add a vtn_constant_int helper by Jason Ekstrand · 5 years ago
- 93f4aa9 glsl/types: Add a real is_integer helper by Jason Ekstrand · 5 years ago
- f0920e2 glsl/types: Rename is_integer to is_integer_32 by Jason Ekstrand · 5 years ago
- 21a7e6d glsl/types: Ignore bit sizes in contains_integer() by Jason Ekstrand · 5 years ago
- 0d1fb38 glsl/types: Handle all bit sizes in glsl_type_is_integer by Jason Ekstrand · 5 years ago
- feb0cdc glsl/nir_opt_access: Update uniforms correctly when only vars change by Caio Marcelo de Oliveira Filho · 5 years ago
- d7ea433 glsl/nir: Fix getting the sampler dim when arrays are involved by Caio Marcelo de Oliveira Filho · 5 years ago
- 77be5b2 nir: Use reorderable access flag by Connor Abbott · 5 years ago
- a1c7379 nir: Add a helper to determine if an intrinsic can be reordered by Connor Abbott · 5 years ago
- f2d0e48 glsl/nir: Add optimization pass for access flags by Connor Abbott · 5 years ago
- c813c57 nir: Add reorderable memory access enum by Connor Abbott · 5 years ago
- 75063fb nir/copy_prop_vars: Ignore volatile accesses by Connor Abbott · 5 years ago
- 364996d glsl/nir: Propagate access qualifiers by Connor Abbott · 5 years ago
- 6f20643 nir: Allow qualifiers on copy_deref and image instructions by Connor Abbott · 5 years ago
- 47e7c69 nir: add a vectorization pass by Connor Abbott · 9 years ago
- 296c5fd nir/lower_tex: Add a way to lower TXS(non-0-LOD) instructions by Boris Brezillon · 5 years ago
- 0e489fd nir/lower_tex: Update ->sampler_dim value before calling get_texture_size() by Boris Brezillon · 5 years ago
- 352b1d9 nir/lower_tex: Actually report when projector lowering happened by Boris Brezillon · 5 years ago
- 3c10a27 glsl: Fix out of bounds read in shader_cache_read_program_metadata by Kenneth Graunke · 5 years ago
- 34d4b3e glsl: Set default precision on record members by Neil Roberts · 5 years ago
- 2354257 glsl/linker: Make precision matching optional in intrastage_match by Neil Roberts · 5 years ago
- 19b27a8 glsl/linker: Don’t check precision for shader interface by Neil Roberts · 5 years ago
- 230d1e8 compiler/types: Making comparing record precision optional by Neil Roberts · 5 years ago
- 2a25012 nir: detect more dynamically uniform expressions by Iago Toral Quiroga · 5 years ago
- 37b92b0 nir: Don't manually index intrinsic index enum by Connor Abbott · 5 years ago
- 7a858f2 spirv/nir: add support for AMD_shader_ballot and Groups capability by Daniel Schürmann · 6 years ago
- ea51275 nir: add intrinsics for AMD_shader_ballot by Daniel Schürmann · 6 years ago
- 1b89ebe nir/spirv: add support for the SubgroupBallotKHR SPIR-V capability by Daniel Schürmann · 7 years ago
- de56eba nir/spirv: add support for the SubgroupVoteKHR SPIR-V capability by Daniel Schürmann · 7 years ago
- 2cb5907 glsl: Check order and uniqueness of interlock functions by Caio Marcelo de Oliveira Filho · 5 years ago
- b7c9fc7 glsl: Make interlock builtins follow same compiler rules as barriers by Caio Marcelo de Oliveira Filho · 5 years ago
- fb21690 nir/opt_algebraic: Fix rules for imadsh_mix16 by Eduardo Lima Mitev · 5 years ago
- 440fe0e nir: fix s/&&/||/ typo by Eric Engestrom · 5 years ago
- 3addd7c nir_algebraic: Add basic optimizations for umul_low and imadsh_mix16 by Eduardo Lima Mitev · 5 years ago
- c27b375 nir/opcodes: Add new 'umul_low' and 'imadsh_mix16' opcodes by Eduardo Lima Mitev · 5 years ago
- 7a18ce0 glsl/loop_analysis: Don't search for NULL variables in the hash table by Jason Ekstrand · 5 years ago
- d96878a nir/propagate_invariant: Don't add NULL vars to the hash table by Jason Ekstrand · 5 years ago
- c7d1b52 nir: Combine lower_fmod16/32 back into a single lower_fmod. by Kenneth Graunke · 5 years ago
- edd45af nir: Drop lower_fmod64 option. by Kenneth Graunke · 5 years ago
- fe2fc30 nir: Don't replace the nir_shader when NIR_TEST_SERIALIZE=1 by Jason Ekstrand · 5 years ago
- 9eba6d9 nir: Don't replace the nir_shader when NIR_TEST_CLONE=1 by Jason Ekstrand · 5 years ago
- d2d3cc6 nir/algebraic: Simplify max(abs(a), 0.0) -> abs(a) by Alyssa Rosenzweig · 5 years ago
- d482a8f spirv: Update the OpenCL.std.h header by Caio Marcelo de Oliveira Filho · 5 years ago
- 5176805 spirv: Implement SPV_EXT_fragment_shader_interlock by Jason Ekstrand · 5 years ago
- b5aa76b spirv: Update the headers from latest Khronos master by Jason Ekstrand · 5 years ago
- 61de825 spirv: Like Uniform, do nothing for UniformId by Caio Marcelo de Oliveira Filho · 5 years ago
- b4eff83 spirv: Implement SpvOpCopyLogical by Caio Marcelo de Oliveira Filho · 5 years ago
- 81586e9 spirv: Generalize OpSelect by Caio Marcelo de Oliveira Filho · 5 years ago
- 1763029 spirv: Move OpSelect handling to a function by Caio Marcelo de Oliveira Filho · 5 years ago
- ea0e898 nir/vars_to_ssa: Handle UNDEF_NODE in more places by Caio Marcelo de Oliveira Filho · 5 years ago
- 1f8546b spirv: Implement OpPtrEqual, OpPtrNotEqual and OpPtrDiff by Caio Marcelo de Oliveira Filho · 5 years ago
- ca164ab nir: Add functions to subtract and compare addresses by Caio Marcelo de Oliveira Filho · 5 years ago
- 09cc338 nir: Add nir_ball_iequal() helper by Caio Marcelo de Oliveira Filho · 5 years ago
- 91672be nir: copy intrinsic type when lowering load input/uniform and store output by Jonathan Marek · 5 years ago
- 7559060 nir: Return nir_type_invalid for non-numeric base types by Caio Marcelo de Oliveira Filho · 5 years ago
- f387c2b nir: remove bool lowering from lower_int_to_float by Jonathan Marek · 5 years ago
- f6579ee nir: fix lower_{int,bool}_to_float for new mov opcode by Jonathan Marek · 5 years ago
- f889180 nir: add lower_bitshift option by Jonathan Marek · 5 years ago
- 887c2a6 nir: fix gather_ssa_types by Jonathan Marek · 5 years ago
- c127505 nir: add type information to load uniform/input and store output intrinsics by Jonathan Marek · 5 years ago
- 6016df2 nir: improvements to native_integers removal by Jonathan Marek · 5 years ago
- 3bd0733 nir/instr_set: Use _mesa_set_search_or_add() by Connor Abbott · 5 years ago
- 3ee2e84 nir: Rematerialize compare instructions by Ian Romanick · 5 years ago
- 336eab0 nir: Add a shallow clone function for nir_alu_instr by Ian Romanick · 5 years ago
- e24a784 nir: Actually propagate progress in nir_opt_move_load_ubo. by Bas Nieuwenhuizen · 5 years ago
- f1cb334 nir/split_vars: Properly bail in the presence of complex derefs by Jason Ekstrand · 5 years ago
- cc59503 nir/vars_to_ssa: Properly ignore variables with complex derefs by Jason Ekstrand · 6 years ago
- 911ea2c nir/vars_to_ssa: Use a non-null UNDEF_NODE pointer by Jason Ekstrand · 5 years ago
- e841946 nir/deref: Add a has_complex_use helper by Jason Ekstrand · 5 years ago
- 8948048 nir/dead_cf: Call instructions aren't dead by Jason Ekstrand · 5 years ago
- 5441d56 vtn: create cast with type stride. by Dave Airlie · 5 years ago
- 03ce12c nir: Accept nir_var_mem_global in derefs used by phis by Caio Marcelo de Oliveira Filho · 5 years ago
- e584fd8 nir: silence three compiler warnings seen with MinGW by Brian Paul · 5 years ago
- e45bf01 spirv: Change spirv_to_nir() to return a nir_shader by Caio Marcelo de Oliveira Filho · 5 years ago
- 8bdf5a0 nir: Allow derefs to be used as phi sources by Caio Marcelo de Oliveira Filho · 5 years ago
- 366811b nir/lower_non_uniform: safely iterate over blocks by Lionel Landwerlin · 5 years ago
- c31b442 st/nir: Re-vectorize shader IO by Kenneth Graunke · 5 years ago
- f2dc0f2 nir: Drop imov/fmov in favor of one mov instruction by Jason Ekstrand · 5 years ago
- 22421ca nir/builder: Merge nir_[if]mov_alu into one nir_mov_alu helper by Jason Ekstrand · 5 years ago
- cd73b61 nir/lower_to_source_mods: Stop turning add, sat, and neg into mov by Jason Ekstrand · 5 years ago
- 2a39788 nir/source_mods: Add a helpers for setting source modifiers by Jason Ekstrand · 5 years ago
- ddd08e1 nir/builder: Remove the use_fmov parameter from nir_swizzle by Jason Ekstrand · 5 years ago
- a482cf6 glsl: simplify resource list building code by Timothy Arceri · 5 years ago
- 9b9f703 spirv: Drop GOOGLE suffix from names incorporated to SPIR-V by Caio Marcelo de Oliveira Filho · 5 years ago
- 02d140c spirv: Pick the right bitsize when doing SpvUConvert by Caio Marcelo de Oliveira Filho · 5 years ago