summaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers/radeonsi/si_shader.c
Commit message (Expand)AuthorAgeFilesLines
* radeonsi: use faster integer division for instance divisorsMarek Olšák2018-10-161-28/+24
* ac: define all address spaces properlyMarek Olšák2018-10-061-3/+3
* radeonsi: avoid sending GS_EMIT in shaders without outputsJózef Kucia2018-10-021-3/+6
* ac: use iN_0/1 constantsMarek Olšák2018-09-101-2/+2
* radeonsi: fix tess/gs fetchs for new swizzle.Dave Airlie2018-08-311-5/+8
* gallivm: allow to pass two swizzles into fetches.Dave Airlie2018-08-301-3/+4
* radeonsi: add TGSI_SEMANTIC_CS_USER_DATA for reading up to 4 SGPRs with TGSIMarek Olšák2018-08-291-0/+11
* ac,radeonsi: use ac_build_gather_values moreMarek Olšák2018-08-211-5/+3
* ac,radeonsi: use ac_build_fmadMarek Olšák2018-08-211-12/+5
* radeonsi: use ac_build_imadMarek Olšák2018-08-211-37/+17
* ac: add ac_build_s_barrierMarek Olšák2018-08-211-3/+1
* radeonsi: print the shader stage name when printing LLVM IRMarek Olšák2018-08-211-1/+2
* radeonsi: use is_merged shader in si_prolog_get_rw_buffersMarek Olšák2018-08-211-18/+14
* radeonsi: don't use emit_data->args in build_interp_intrinsicMarek Olšák2018-08-141-19/+13
* radeonsi: remove fetch_args callbacks for ALU instructionsMarek Olšák2018-08-141-45/+28
* radeonsi: don't load block dimensions into SGPRs if they are not variableMarek Olšák2018-08-071-1/+2
* amd: remove support for LLVM 5.0Marek Olšák2018-08-031-24/+0
* ac,radeonsi: reduce optimizations for complex compute shaders on older APUs (v2)Marek Olšák2018-08-011-5/+24
* radeonsi: reduce LDS stalls by 40% for tessellationMarek Olšák2018-07-231-4/+4
* radeonsi: rename si_compiler -> ac_llvm_compilerDave Airlie2018-07-041-12/+12
* radeonsi: implement vertex color clamping for tess and GSMarek Olšák2018-06-281-28/+69
* radeonsi: move VS_STATE_SGPR before draw SGPRsMarek Olšák2018-06-281-7/+7
* radeonsi: don't use malloc in si_generate_gs_copy_shaderMarek Olšák2018-06-281-10/+2
* radeonsi: store compute local_size into tgsi_shader_infoMarek Olšák2018-06-281-1/+2
* radeonsi/gfx9: insert the barrier between merged shaders inside the if blockMarek Olšák2018-06-281-5/+13
* radeonsi: stop using lp_build_emit_llvm_unary/binaryMarek Olšák2018-06-251-9/+8
* radeonsi: stop using lp_build_allocMarek Olšák2018-06-251-4/+3
* radeonsi: use gallivm lessMarek Olšák2018-06-251-2/+2
* radeonsi: stop using lp_bld_intr.hMarek Olšák2018-06-251-16/+18
* radeonsi: remove last uses of lp_build_context::undefMarek Olšák2018-06-251-6/+6
* radeonsi: stop using lp_bld_arit.hMarek Olšák2018-06-251-37/+31
* radeonsi: stop using lp_build_gather_valuesMarek Olšák2018-06-251-23/+22
* radeonsi: clean up some #includesMarek Olšák2018-06-251-5/+1
* radeonsi: clean up passing the is_monolithic flag for compilationMarek Olšák2018-06-251-16/+14
* radeonsi: fix passing gl_ClipVertex for GS and tessMarek Olšák2018-05-251-1/+4
* radeonsi: fix color inputs/outputs for GS and tessMarek Olšák2018-05-251-11/+19
* radeonsi: skip ES output stores for undefined output componentsMarek Olšák2018-05-181-0/+3
* radeonsi: add struct si_compiler containing LLVMTargetMachineRefMarek Olšák2018-04-271-33/+33
* radeonsi: use r600_resource() typecast helperMarek Olšák2018-04-271-2/+1
* ac/radv/radeonsi: refactor max simd waves into common code.Dave Airlie2018-04-241-11/+1
* radeonsi/nir: fix crash in test involving the sample maskNicolai Hähnle2018-04-201-1/+2
* radeonsi: add support for VegaMMarek Olšák2018-04-181-0/+1
* radeonsi: add shader binary padding for UMRMarek Olšák2018-04-101-3/+15
* radeonsi: move definitions out of r600_pipe_common.hMarek Olšák2018-04-051-1/+1
* radeonsi: update copyrightsMarek Olšák2018-04-051-0/+1
* radeonsi/gfx9: fix bad LLVM params in monolithic LS+HSMarek Olšák2018-04-031-1/+5
* radeonsi: implement GL_KHR_blend_equation_advancedMarek Olšák2018-04-021-2/+2
* radeonsi: rename unpack_param -> si_unpack_paramMarek Olšák2018-04-021-27/+27
* radeonsi/nir: fix explicit component packing for geom/tess doublesTimothy Arceri2018-04-021-8/+11
* radeonsi/nir: gather buffers declared more accurately and use const fast pathTimothy Arceri2018-04-021-0/+7