- bd0468e nir: add options to lower nir_op_pack_[64/32]_* via nir_lower_alu_to_scalar() by Daniel Schürmann · 4 years ago
- 3d9ffdc nir/lower_memcpy: Don't mask the store by Jason Ekstrand · 3 years, 11 months ago
- 4ca38a1 nir/lower_clip_cull: Store array size for FS inputs by Connor Abbott · 4 years ago
- d972a6a nir: get rid of OOB dereferences in nir_lower_io_arrays_to_elements by Andrii Simiklit · 4 years ago
- 6795668 nir: Rename replicated-result dot-product instructions by Ian Romanick · 4 years, 3 months ago
- b739bb7 compile/nir: Correct printing dest_type by Gert Wollny · 3 years, 11 months ago
- 4735c8a nir/loop_analyze: adjust force unrolling to only include interesting modes by Rhys Perry · 3 years, 11 months ago
- 8cf0024 nir: Use a switch in nir_lower_explicit_io_instr by Caio Marcelo de Oliveira Filho · 4 years ago
- 33ccf0e nir: drop unused alpha_ref_float by Erik Faye-Lund · 5 years ago
- 42ee423 nir: drop support for using load_alpha_ref_float by Erik Faye-Lund · 5 years ago
- 2335200 nir: consider load_color intrinsics as both inputs and sysval in gathering by Marek Olšák · 4 years ago
- fdbc45d nir: Only validate in passes that might have changed things. by Eric Anholt · 3 years, 11 months ago
- ef68f74 nir/lower_io: Assert non-zero power-of-two alignments by Jason Ekstrand · 4 years ago
- d867e7c nir: Add an option to not lower source mods for f64/u64/i64. by Eric Anholt · 4 years, 2 months ago
- c730fea nir: Add a call to get a struct describing SSA liveness per instruction. by Eric Anholt · 4 years, 2 months ago
- a206b58 nir: Add a block start/end ip to live instr index metadata. by Eric Anholt · 3 years, 11 months ago
- 2f5d184 nir: Replace nir_ssa_def->live_index with nir_instr->index. by Eric Anholt · 4 years ago
- b6cb184 nir: Introduce nir_metadata_instr_index for nir_index_instr() being current. by Eric Anholt · 4 years ago
- d9cb9ff nir: Emit set_vertex_and_primitive_count for inactive streams. by Timur Kristóf · 3 years, 11 months ago
- f6d5dd8 nir: Add nir_intrinsic_terminate and nir_intrinsic_terminate_if by Caio Marcelo de Oliveira Filho · 4 years, 4 months ago
- f91b2fe nir/opt_load_store_vectorize: add some tests for discard/demote behaviour by Rhys Perry · 3 years, 11 months ago
- f8e971f nir/opt_load_store_vectorize: don't vectorize stores across demote by Rhys Perry · 3 years, 11 months ago
- 9b01598 nir/lower_io_to_scalar: update io semantics on per-component inst by Alejandro Piñeiro · 4 years ago
- f503699 nir/opt_algebraic: optimize unpack_half_2x16_split_x(ushr, a, 16) by Daniel Schürmann · 4 years ago
- 21422b1 nir/opt_uniform_atomics: remove useless returns by Rhys Perry · 3 years, 11 months ago
- f4c754b nir: add a nir_get_ubo_size intrinsic by Iago Toral Quiroga · 4 years ago
- 6004ad9 nir/lower_io: add an option to lower interpolateAt functions by Iago Toral Quiroga · 4 years, 1 month ago
- 10b79bf nir: include texture query lod as one of the ops that requires a sampler by Alejandro Piñeiro · 4 years, 2 months ago
- a7114f3 nir/opt_uniform_atomics: don't optimize atomics twice by Rhys Perry · 4 years ago
- bc43650 nir/opt_uniform_atomics: optimize image atomics by Rhys Perry · 4 years ago
- f83bc5b nir: add pass to optimize uniform atomics by Rhys Perry · 4 years ago
- 37b6b09 nir: allow divergence information to be updated when inserting instruction by Rhys Perry · 4 years ago
- e1120f2 nir: move divergence analysis options to nir_shader_compiler_options by Rhys Perry · 4 years ago
- 1a912a5 nir: add last_invocation intrinsic by Rhys Perry · 4 years ago
- 8850a63 radv/aco,nir/lower_subgroups: don't lower elect by Rhys Perry · 4 years ago
- c31abab nir: update ubo locations in nir_lower_uniforms_to_ubo by Mike Blumenkrantz · 4 years, 1 month ago
- 18f9fc9 spirv: add and use a generator id enum by Rhys Perry · 4 years ago
- 181d5f5 nir: Allow more deref modes in phis by Jason Ekstrand · 4 years ago
- e7127b3 nir/algebraic: optimize iand/ior of (n)eq zero when umax/umin not available by Jose Maria Casanova Crespo · 4 years ago
- 5f2671b nir: return progress from nir_lower_io_to_scalar_early by Rhys Perry · 4 years ago
- f11f4a2 nir: Add ability to count primitives per stream. by Timur Kristóf · 4 years ago
- aac5adc nir: Count vertices per stream. by Timur Kristóf · 4 years ago
- 70b94ad nir: Add ability to overwrite incomplete GS primitives. by Timur Kristóf · 4 years, 2 months ago
- c977c36 nir: Add ability to count emitted GS vertices per primitive. by Timur Kristóf · 4 years, 3 months ago
- 2be9901 nir: Add ability to count emitted GS primitives. by Timur Kristóf · 4 years, 3 months ago
- 06a5edf nir/opt_deref: Fix the vector bitcast optimization by Jason Ekstrand · 4 years ago
- 826a102 st/mesa: Add NV12 lowering to PIPE_FORMAT_R8_G8B8_420_UNORM by Kristian H. Kristensen · 4 years ago
- 2fa7c79 spirv: Move nir_lower_libclc to src/compiler/spirv by Jason Ekstrand · 4 years ago
- ef453f5 spirv: Add a shared libclc loader by Jason Ekstrand · 4 years ago
- 0ef2f1d nir: Fix unaligned pointer access by Tony Wasserka · 4 years ago
- 6a9dc75 nir: Fix undefined behavior due to signed integer multiplication overflows by Tony Wasserka · 4 years ago
- 3f1b35a nir: add new helper passes that lower uniforms to literals by Marek Olšák · 4 years ago
- 1e7d82c nir/algebraic: always lower idiv to shifts if bitops are allowed by Marek Olšák · 4 years ago
- 4c54f05 nir/constant_folding: init nir_const_value to zero by Samuel Pitoiset · 4 years ago
- 60825a5 nir/constant_folding: Fold load_deref of nir_var_mem_constant by Jason Ekstrand · 4 years ago
- 481b753 nir: Validate constant initializers by Jason Ekstrand · 4 years ago
- 1ada835 nir/constant_folding: Use nir_shader_instruction_pass by Jason Ekstrand · 4 years ago
- 9df1ff3 nir/constant_folding: Use the builder by Jason Ekstrand · 4 years ago
- b2e1fc8 nir: Add a pass to lower vec3s to vec4s by Jason Ekstrand · 4 years ago
- 9190f82 nir/opt_deref: Add an optimization for bitcasts by Jason Ekstrand · 4 years ago
- 80e6ac3 nir/opt_deref: Add an instruction type switch by Jason Ekstrand · 4 years, 1 month ago
- 769ede2 nir: Add component mask re-interpret helpers by Jason Ekstrand · 4 years ago
- 57e7c5f nir/opt_load_store_vectorize: Use bit sizes when checking mask compatibility by Jason Ekstrand · 4 years ago
- f6667cb nir: Add a memcpy optimization pass by Jason Ekstrand · 4 years ago
- e363da3 nir: Handle memcpy in copy_prop_vars and combine_stores by Jason Ekstrand · 4 years ago
- 100a5ac nir/find_array_copies: Properly discard copies for casts by Jason Ekstrand · 4 years ago
- 98bb74b nir: Fix a misspelling by Jason Ekstrand · 4 years ago
- cb95065 nir: Add lowering from regular ALU conversions to the intrinsic by Jason Ekstrand · 4 years ago
- 383ecfb nir: Add a passes for nir_intrinsic_convert_alu_types by Jason Ekstrand · 4 years ago
- d5cb51e nir: Add builder helpers for OpenCL type conversions by Jason Ekstrand · 4 years ago
- 588bb66 nir: Add a conversion and rounding intrinsic by Jason Ekstrand · 4 years ago
- 0aa08ae nir: Split NIR_INTRINSIC_TYPE into separate src/dest indices by Jason Ekstrand · 4 years ago
- e3f4655 nir: Make nir_lower_ubo_vec4() handle non-vec4-aligned loads. by Eric Anholt · 4 years ago
- 618556a nir: Drop the high_offset argument to the load_store_vectorizer filter. by Eric Anholt · 4 years ago
- 5f757bb nir: Make the load_store_vectorizer provide align_mul + align_offset. by Eric Anholt · 4 years ago
- ffbfc1e nir/nir_lower_uniforms_to_ubo: Set better alignments on our new instructions. by Eric Anholt · 4 years ago
- c88c89f nir: Print the alignment information on casts. by Eric Anholt · 4 years ago
- 6c1c571 nir: Document a bit about how align_mul/offset work. by Eric Anholt · 4 years ago
- 25ebd7f Revert "nir/lower_goto_if: Add a route::outside set" by Jason Ekstrand · 4 years ago
- 57c9fc3 nir/lower_goto_ifs: Always include level dom_frontiers in prev_frontier by Jason Ekstrand · 4 years ago
- 7749983 nir/lower_goto_ifs: Add asserts for SSA forks by Jason Ekstrand · 4 years ago
- dc010cb nir/lower_goto_ifs: Use rzalloc by Jason Ekstrand · 4 years ago
- 719c680 nir/dominance: Use _mesa_set_clear instead ofhand-rolling it by Jason Ekstrand · 4 years ago
- b6a4172 nir/lower_goto_ifs: Don't destroy SSA form in the process by Jason Ekstrand · 4 years ago
- 6f134a6 nir/validate: Improve the validation of blocks by Jason Ekstrand · 4 years ago
- 7dbb1f7 nir/cf: Better handle intra-block splits by Jason Ekstrand · 4 years ago
- 5e2e882 nir: Disallow goto and goto_if in clone and [de]serialize by Jason Ekstrand · 4 years ago
- 9a48ed8 nir/copy_propagate: Copy-prop into jump conditions by Jason Ekstrand · 4 years ago
- 7f0cd6f nir/opt_if: Use early returns in opt_if_merge() by Connor Abbott · 5 years ago
- 656e428 nir/opt_if: Remove open-coded nir_ssa_def_rewrite_uses() by Connor Abbott · 5 years ago
- c6f871b nir/lower_returns: Use nir control flow insertion helpers by Connor Abbott · 5 years ago
- ac7537f nir/lower_io: Add support for push constants by Jason Ekstrand · 4 years, 3 months ago
- 7a2b4ce nir: Allow creating variables with nir_var_mem_push_const. by Jason Ekstrand · 4 years, 3 months ago
- 51e2b31 nir: Handle per-view io in nir_io_add_const_offset_to_base() by Connor Abbott · 4 years ago
- bc8a5c0 nir: Add per_view to IO semantics by Connor Abbott · 4 years ago
- 5a88db6 nir/lower_io_arrays: Fix xfb_offset bug by Connor Abbott · 4 years, 1 month ago
- df955ce nir: Count i/o slots correctly for per-view variables by Connor Abbott · 4 years, 1 month ago
- ffe946d nir: Add nir_lower_multiview pass by Connor Abbott · 4 years, 2 months ago
- a0e35c7 nir/lower_io: change nir_io_add_const_offset_to_base to use bitfield modes by Samuel Pitoiset · 4 years ago
- e8c5f8b nir/lower_clip: Add i/o semantics for load/store intrinsics. by Eric Anholt · 4 years ago