Commits · 36d4092173179b25b931a522361cf6139a7c635a · Lorenzo Albano / LLVM bpEVL · GitLab

Nov 29, 2018
- Revert r347871 "Fix: Add support for TFE/LWE in image intrinsic" · c6603861
  David Stuttard authored Nov 29, 2018
  c6603861
- Add support for TFE/LWE in image intrinsics · de02e4b1
  David Stuttard authored Nov 29, 2018
  de02e4b1
Aug 28, 2018
- [AMDGPU] Add support for a16 modifiear for gfx9 · 1f334d00
  Ryan Taylor authored Aug 28, 2018
  1f334d00
Aug 01, 2018
- [AMDGPU] Optimize _L image intrinsic to _LZ when lod is zero · 894c8fd0
  Ryan Taylor authored Aug 01, 2018
  894c8fd0
Jun 21, 2018
- AMDGPU: Remove redundant MIMG instruction variants · 15745ba5
  Nicolai Haehnle authored Jun 21, 2018
  15745ba5
- AMDGPU: Remove old-style image intrinsics · db6911a6
  Nicolai Haehnle authored Jun 21, 2018
  db6911a6
- AMDGPU: Select MIMG instructions manually in SITargetLowering · 7a9c03f4
  Nicolai Haehnle authored Jun 21, 2018
  7a9c03f4
- AMDGPU: Refactor MIMG instruction TableGen using generic tables · 0ab200b6
  Nicolai Haehnle authored Jun 21, 2018
  0ab200b6
- AMDGPU: Pass AMDGPUSampleVariant to MIMG_{Sampler,Gather}(_WQM) · 2367f035
  Nicolai Haehnle authored Jun 21, 2018
  2367f035
- AMDGPU: Turn D16 for MIMG instructions into a regular operand · f2674319
  Nicolai Haehnle authored Jun 21, 2018
  f2674319
Jun 15, 2018
- AMDGPU: Make v4i16/v4f16 legal · 02dc7e19
  Matt Arsenault authored Jun 15, 2018
  02dc7e19
Jun 04, 2018
- AMDGPU: Make various NamedOperands upper case · 59198ed0
  Nicolai Haehnle authored Jun 04, 2018
  59198ed0
- TableGen: Streamline the semantics of NAME · 01d261f1
  Nicolai Haehnle authored Jun 04, 2018
  01d261f1
May 22, 2018
- AMDGPU: Make v2i16/v2f16 legal on VI · 1349a04e
  Matt Arsenault authored May 22, 2018
  1349a04e
Apr 04, 2018
- [AMDGPU][MC] Added support of 3-element addresses for MIMG instructions · a0b8cd03
  Dmitry Preobrazhensky authored Apr 04, 2018
  a0b8cd03
- AMDGPU: Dimension-aware image intrinsics · 2f5a7382
  Nicolai Haehnle authored Apr 04, 2018
  2f5a7382
Mar 28, 2018
- [AMDGPU][MC] Added PCK variants of image load/store instructions · 2456ac69
  Dmitry Preobrazhensky authored Mar 28, 2018
  2456ac69
Mar 12, 2018
- [AMDGPU][MC] Corrected GATHER4 opcodes · da4a7c01
  Dmitry Preobrazhensky authored Mar 12, 2018
  da4a7c01
Jan 26, 2018
- [AMDGPU][MC] Added support of 64-bit image atomics · 0b4eb1ea
  Dmitry Preobrazhensky authored Jan 26, 2018
  0b4eb1ea
- [AMDGPU][MC] Enabled disassembler for image atomic operations · 6cb42e76
  Dmitry Preobrazhensky authored Jan 26, 2018
  6cb42e76
Jan 18, 2018
- AMDGPU/SI: Add d16 support for image intrinsics. · 4737e892
  Changpeng Fang authored Jan 18, 2018
  4737e892
Dec 29, 2017
- AMDGPU: Remove mayLoad/hasSideEffects from MIMG stores · d94b63d7
  Matt Arsenault authored Dec 29, 2017
  d94b63d7
Dec 13, 2017
- AMDGPU: Partially fix disassembly of MIMG instructions · cad7fa85
  Matt Arsenault authored Dec 13, 2017
  cad7fa85
Dec 08, 2017
- AMDGPU: image_getlod and image_getresinfo do not read memory · 856777d8
  Matt Arsenault authored Dec 08, 2017
  856777d8
Oct 03, 2017
- AMDGPU: Remove global isGCN predicates · 90c7593a
  Matt Arsenault authored Oct 03, 2017
  90c7593a
Jul 04, 2017
- [AMDGPU] Fix latency of MIMG instructions · b83f5c99
  Marek Olsak authored Jul 04, 2017
  b83f5c99
Apr 04, 2017
- AMDGPU: Remove legacy image intrinsics · 236da200
  Matt Arsenault authored Apr 04, 2017
  236da200
Dec 20, 2016
- AMDGPU/SI: Add a MachineMemOperand to MIMG instructions · 244891d1
  Tom Stellard authored Dec 20, 2016
  244891d1
Nov 15, 2016
- [AMDGPU] TableGen: change individual instruction flags to bit type from bits<1> · c01faa38
  Sam Kolton authored Nov 15, 2016
  c01faa38
Nov 14, 2016
- AMDGPU/SI: Support data types other than V4f32 in image intrinsics · 8236fe10
  Changpeng Fang authored Nov 14, 2016
  8236fe10
Oct 28, 2016
- AMDGPU: Rename glc operand type · 4b6a6cc8
  Matt Arsenault authored Oct 28, 2016
  4b6a6cc8
Oct 12, 2016
- AMDGPU/SI: Change mimg intrinsic signatures · fac248cb
  Tom Stellard authored Oct 12, 2016
  fac248cb
Sep 01, 2016
- AMDGPU/SI: MIMG TD Refactoring. · b28fe030
  Changpeng Fang authored Sep 01, 2016
  b28fe030