summaryrefslogtreecommitdiff
AgeCommit message (Expand)AuthorFilesLines
2018-06-15i965/fs: Propagate conditional modifiers from not instructionsIan Romanick1-1/+61
2018-06-15i965/fs: Rearrange code to remove most of the gotosIan Romanick1-11/+3
2018-06-15i965/fs: Refactor propagation of conditional modifiers from compares to addsIan Romanick1-57/+80
2018-06-15i965/fs: Optimize OR with 0 into a MOVIan Romanick1-1/+2
2018-05-22intel/eu: Set EXECUTE_1 when setting the rounding mode in cr0Jason Ekstrand1-0/+2
2018-04-23i965/fs: Return mlen * 8 for size_read() for INTERPOLATE_AT_*Jason Ekstrand1-0/+2
2018-04-20i965/fs: retype offset_reg to UD at load_ssboJose Maria Casanova Crespo1-1/+1
2018-03-26i965/fs: Propagate conditional modifiers from compares to addsIan Romanick2-5/+400
2018-03-26i965/fs: Allow cmod propagation when src0 is a uniform or shader inputIan Romanick1-1/+2
2018-03-08i965/fs: Merge CMP and SEL into CSEL on Gen8+Ian Romanick2-0/+107
2018-03-08i965/fs: Add infrastructure for generating CSEL instructions.Kenneth Graunke8-1/+34
2018-03-07i965/fs: Add support for nir_intrinsic_shuffleJason Ekstrand7-0/+150
2018-03-07i965/fs: Support nir_intrinsic_vote_feqJason Ekstrand1-0/+6
2018-03-07i965/fs: Implement basic SPIR-V subgroup intrinsicsJason Ekstrand2-0/+26
2018-03-02Revert "i965/fs: Predicate byte scattered writes if needed"Francisco Jerez1-14/+1
2018-02-28i965/fs: Support 16-bit store_ssbo with VK_KHR_relaxed_block_layoutJose Maria Casanova Crespo1-7/+15
2018-02-28i965/fs: Support 16-bit do_read_vector with VK_KHR_relaxed_block_layoutJose Maria Casanova Crespo1-14/+37
2018-02-28i965/fs: shuffle_32bit_load_result_to_16bit_data now skips componentsJose Maria Casanova Crespo3-3/+6
2018-02-28isl/i965/fs: SSBO/UBO buffers need size padding if not multiple of 32-bitJose Maria Casanova Crespo3-2/+62
2018-01-25i965/fs: Reset the register file to VGRF in lower_integer_multiplicationJason Ekstrand1-5/+10
2018-01-11i965/fs: Use UW types when using V immediatesJason Ekstrand2-5/+5
2018-01-11Revert "Revert "i965/fs: Use align1 mode on ternary instructions on Gen10+""Matt Turner1-4/+8
2018-01-11i965/fs: Add/use functions to convert to 3src_align1 vstride/hstrideMatt Turner1-28/+41
2017-12-22Revert "i965/fs: Use align1 mode on ternary instructions on Gen10+"Anuj Phogat1-8/+4
2017-12-08i965/fs: Handle !supports_pull_constants and push UBOs properlyJason Ekstrand1-1/+1
2017-12-08i965/fs: Rewrite assign_constant_locationsJason Ekstrand1-133/+185
2017-12-06i965/fs: Use untyped_surface_read for 16-bit load_ssboJose Maria Casanova Crespo1-7/+20
2017-12-06i965/fs: Optimize 16-bit SSBO stores by packing two into a 32-bit regJose Maria Casanova Crespo1-15/+43
2017-12-06i965/fs: Enables 16-bit load_ubo with samplerJason Ekstrand1-7/+14
2017-12-06i965/fs: Helpers for un/shuffle 16-bit pairs in 32-bit componentsJose Maria Casanova Crespo2-0/+71
2017-12-06i965/fs: Use byte scattered read for 16-bit load_ssboJose Maria Casanova Crespo1-1/+13
2017-12-06i965/fs: Add byte scattered read message and fs supportJose Maria Casanova Crespo9-1/+94
2017-12-06i965/fs: Predicate byte scattered writes if neededAlejandro Piñeiro1-1/+14
2017-12-06i965/fs: Use byte_scattered_write on 16-bit store_ssboAlejandro Piñeiro1-20/+45
2017-12-06i965/fs: Add byte scattered write message and fs supportJose Maria Casanova Crespo9-0/+118
2017-12-06i965/fs: Add remove_extra_rounding_modes optimizationAlejandro Piñeiro3-0/+39
2017-12-06i965/fs: Enable rounding mode on f2f16 opsAlejandro Piñeiro1-0/+18
2017-12-06i965/fs: Define new shader opcode to set rounding modesAlejandro Piñeiro5-0/+62
2017-12-06i965/fs: Handle 32-bit to 16-bit conversionsAlejandro Piñeiro1-0/+25
2017-12-06i965/fs: Remove BRW_REGISTER_TYPE_HF assert at get_exec_typeAlejandro Piñeiro1-3/+0
2017-11-21i965/fs: Check ADD/MAD with immediates in satprop unit testMatt Turner1-1/+125
2017-11-21i965/fs: Handle negating immediates on MADs when propagating saturatesMatt Turner1-2/+8
2017-11-14i965/fs: Fix extract_i8/u8 to a 64-bit destinationMatt Turner1-2/+23
2017-11-14i965/fs: Split all 32->64-bit MOVs on CHV, BXT, GLKMatt Turner1-4/+4
2017-11-07i965/fs: Get rid of the early return in brw_compile_csJason Ekstrand1-13/+14
2017-11-07i965/fs/nir: Don't stomp 64-bit values to D in get_nir_srcJason Ekstrand1-13/+24
2017-11-07i965/fs/nir: Minor refactor of store_outputJason Ekstrand1-4/+3
2017-11-07i965/fs: Return a fs_reg from shuffle_64bit_data_for_32bit_writeJason Ekstrand2-29/+12
2017-11-07i965/fs/nir: Simplify 64-bit store_outputJason Ekstrand1-19/+6
2017-10-25i965/fs: Add brw_reg_type_from_bit_size utility methodAlejandro Piñeiro1-5/+64
generated by cgit v1.2.3 (git 2.39.1) at 2024-04-26 21:13:35 +0000