Commits · bd159d32c43732a65a943333d04f46e5d30fb9c5 · Roger Ferrer / llvm-epi · GitLab

Nov 16, 2017
- Fix thinko in last commit. · 6348188e
  Eric Christopher authored Nov 16, 2017
  6348188e
- Add NDEBUG checks around LLVM_DUMP_METHOD functions for Wunused-function warnings. · 3148a1be
  Eric Christopher authored Nov 16, 2017
  3148a1be
- [globalisel][tablegen] Generate rule coverage and use it to identify untested rules · f76f3154
  Daniel Sanders authored Nov 16, 2017
  f76f3154
- Add backend name to Target to enable runtime info to be fed back into TableGen · 725584e2
  Daniel Sanders authored Nov 15, 2017
  725584e2
Nov 15, 2017
- AMDGPU: Replace i64 add/sub lowering · 301162c4
  Matt Arsenault authored Nov 15, 2017
  301162c4
- AMDGPU: Add separate definitions for DS insts without m0 use · 10c472dd
  Matt Arsenault authored Nov 15, 2017
  10c472dd
- AMDGPU: Don't use MUBUF vaddr if address may overflow · 45b98189
  Matt Arsenault authored Nov 15, 2017
  45b98189
- AMDGPU: Handle or in multi-use shl ptr combine · c8903125
  Matt Arsenault authored Nov 14, 2017
  c8903125
Nov 14, 2017
- AMDGPU: Error on stack size overflow · 9ba465a9
  Matt Arsenault authored Nov 14, 2017
  9ba465a9
- AMDGPU: Fix producing saveexec when the copy is spilled · 57c37b2d
  Matt Arsenault authored Nov 14, 2017
  57c37b2d
- AMDGPU: Fix not converting d16 load/stores to offset · 4b7938c6
  Matt Arsenault authored Nov 13, 2017
  4b7938c6
Nov 13, 2017
- AMDGPU: Implement computeKnownBitsForTargetNode for mbcnt · 4eea3f3d
  Matt Arsenault authored Nov 13, 2017
  4eea3f3d
- AMDGPU: Drop duplicate setOperationAction · b17f3204
  Jan Vesely authored Nov 13, 2017
  b17f3204
- AMDGPU: Preserve nuw in shl add ptr combine · e5e0c742
  Matt Arsenault authored Nov 13, 2017
  e5e0c742
- AMDGPU: Fix multi-use shl/add combine · fbe95335
  Matt Arsenault authored Nov 13, 2017
  fbe95335
- AMDGPU: Select d16 loads into low component of register · e1cd482f
  Matt Arsenault authored Nov 13, 2017
  e1cd482f
Nov 10, 2017
- AMDGPU/NFC: Split Processors.td into GCNProcessors.td and R600Processors.td · 27b0a033
  Konstantin Zhuravlyov authored Nov 10, 2017
  27b0a033
- [AMDGPU] Fix pointer info for lowering load/store for r600 for amdgiz environment · 35845f06
  Yaxun Liu authored Nov 10, 2017
  35845f06
- [AMDGPU] Fix pointer info for pseudo source for r600 · 920cc2f8
  Yaxun Liu authored Nov 10, 2017
  920cc2f8
Nov 09, 2017
- Fix "default label in switch which covers all enumeration values" warning · bee1964d
  Vitaly Buka authored Nov 09, 2017
  bee1964d
- AMDGPU: Merge BUFFER_STORE_DWORD_OFFEN/OFFSET into x2, x4 · 58410f37
  Marek Olsak authored Nov 09, 2017
  58410f37
- AMDGPU: Lower buffer store and atomic intrinsics manually · 5cec6419
  Marek Olsak authored Nov 09, 2017
  5cec6419
- AMDGPU: Merge BUFFER_LOAD_DWORD_OFFSET into x2, x4 · 4c421a2d
  Marek Olsak authored Nov 09, 2017
  4c421a2d
- AMDGPU: Merge BUFFER_LOAD_DWORD_OFFEN into x2, x4 · 6a0548ac
  Marek Olsak authored Nov 09, 2017
  6a0548ac
- AMDGPU: Merge S_BUFFER_LOAD_DWORD_IMM into x2, x4 · b953cc36
  Marek Olsak authored Nov 09, 2017
  b953cc36
- AMDGPU: Fold immediate offset into BUFFER_LOAD_DWORD lowered from SMEM · ffadcb74
  Marek Olsak authored Nov 09, 2017
  ffadcb74
Nov 08, 2017
- Target/TargetInstrInfo.h -> CodeGen/TargetInstrInfo.h to match layering · 3f833edc
  David Blaikie authored Nov 08, 2017
  3f833edc
- AMDGPU: Set correct sched model on v_mad_u64_u32 · 4709ab91
  Matt Arsenault authored Nov 08, 2017
  4709ab91
Nov 07, 2017
- AMDGPU: Remove redundant combine · 6119f800
  Matt Arsenault authored Nov 07, 2017
  6119f800
Nov 06, 2017
- AMDGPU: Select v_mad_u64_u32 and v_mad_i64_i32 · 4f6318fe
  Matt Arsenault authored Nov 06, 2017
  4f6318fe
- [IR] redefine 'UnsafeAlgebra' / 'reassoc' fast-math-flags and add 'trans' fast-math-flag · 629c4115
  Sanjay Patel authored Nov 06, 2017
  629c4115
- [AMDGPU] Change alloca addr space of r600 to 5 for amdgiz environment · cc56a8b1
  Yaxun Liu authored Nov 06, 2017
  cc56a8b1
- [AMDGPU] Fix assertion due to assuming pointer in default addr space is 32 bit · 1ac16619
  Yaxun Liu authored Nov 06, 2017
  1ac16619
Nov 04, 2017
- [AMDGPU] Remove hardcoded address space value from AMDGPULibFunc · 0d9673cf
  Yaxun Liu authored Nov 04, 2017
  0d9673cf
Nov 03, 2017
- Move TargetFrameLowering.h to CodeGen where it's implemented · 1be62f03
  David Blaikie authored Nov 03, 2017
  1be62f03
Nov 02, 2017
- AMDGPU: Fix warning discovered by r317266 [-Wunused-private-field] · 275a4f76
  Konstantin Zhuravlyov authored Nov 02, 2017
  275a4f76
- AMDGPU: Remove outdated fixme (it was already fixed) · b695cd41
  Konstantin Zhuravlyov authored Nov 02, 2017
  b695cd41
Nov 01, 2017
- AMDGPU: Fix set but not used warnings related to AMDGPUAS · 435151ad
  Konstantin Zhuravlyov authored Nov 01, 2017
  435151ad
- Fix warnings discovered by rL317076. [-Wunused-private-field] · 1657f2ad
  NAKAMURA Takumi authored Nov 01, 2017
  1657f2ad
- [AMDGPU] Clean up symbols in the global namespace. · f9ab3ddb
  Benjamin Kramer authored Oct 31, 2017
  f9ab3ddb