Commits · e35244cf636b684e0725738cab900ee8f024d044 · Roger Ferrer / llvm-epi · GitLab

Nov 08, 2015
- [WinEH] Update PHIs of CATCHRET successors · e35244cf
  David Majnemer authored Nov 08, 2015
  e35244cf
Nov 07, 2015
- [WinEH] Update exception pointer registers · f748c893
  Joseph Tremoulet authored Nov 07, 2015
  f748c893
- [AArch64][FastISel] Don't even try to select vector icmps. · cf49b523
  Ahmed Bougacha authored Nov 06, 2015
  cf49b523
- [X86] Fold (trunc (i32 (zextload i16))) into vbroadcast. · b49eb3ab
  Ahmed Bougacha authored Nov 06, 2015
  b49eb3ab
- [X86] SRL non-LSB extracts when folding to truncating broadcasts. · 05a0514b
  Ahmed Bougacha authored Nov 06, 2015
  05a0514b
- [X86] Don't fold non-LSB extracts into truncating broadcasts. · 68614a36
  Ahmed Bougacha authored Nov 06, 2015
  68614a36
Nov 06, 2015
- DAGCombiner: Check shouldReduceLoadWidth before combining (and (load), x) -> extload · 05691a67
  Tom Stellard authored Nov 06, 2015
  05691a67
- [WebAssembly] Use more explicit types in testcases. · 3cb66c85
  Dan Gohman authored Nov 06, 2015
  3cb66c85
- [WebAssembly] Add more explicit pushes to the tests. · 8d456e42
  Dan Gohman authored Nov 06, 2015
  8d456e42
- [ShrinkWrapping] Teach shrink-wrapping how to analyze RegMask. · 9a8efc08
  Quentin Colombet authored Nov 06, 2015
  9a8efc08
- Improved the operands commute transformation for X86-FMA3 instructions. · 4731bea3
  Andrew Kaylor authored Nov 06, 2015
  4731bea3
- [WebAssembly] Make expression-stack pushing explicit · 4b96d8d1
  Dan Gohman authored Nov 06, 2015
  4b96d8d1
- AMDGPU: Create emergency stack slots during frame lowering · 0c90e950
  Matt Arsenault authored Nov 06, 2015
  0c90e950
- AMDGPU: Add pass to detect used kernel features · 3931948b
  Matt Arsenault authored Nov 06, 2015
  3931948b
- AMDGPU: Hack for VS_32 register pressure · 623e6fd4
  Matt Arsenault authored Nov 06, 2015
  623e6fd4
- [WinEH] Mark funclet entries and exits as clobbering all registers · b8fd162f
  Reid Kleckner authored Nov 06, 2015
  b8fd162f
- [AArch64]Enable the narrow ld promotion only on profitable microarchitectures · 22fe15ee
  Jun Bum Lim authored Nov 06, 2015
  22fe15ee
- Bring r252305 back with a test fix. · 889d7bb4
  Rafael Espindola authored Nov 06, 2015
  889d7bb4
- [mips] Define patterns for the atomic_{load,store}_{8,16,32,64} nodes. · b04672ca
  Vasileios Kalintiris authored Nov 06, 2015
  b04672ca
- AMDGPU/SI: Emit HSA kernels with symbol type STT_AMDGPU_HSA_KERNEL · 1e1b05db
  Tom Stellard authored Nov 06, 2015
  1e1b05db
- Revert r252249 (and r252255, r252258), "[WinEH] Clone funclets with multiple parents" · 9947cace
  NAKAMURA Takumi authored Nov 06, 2015
  9947cace
- Temporarily disable flaky checks in wineh-multi-parent-cloning. · f05a87df
  Andrew Kaylor authored Nov 06, 2015
  f05a87df
- [WinEH] Clone funclets with multiple parents · 29cd5765
  Andrew Kaylor authored Nov 06, 2015
  29cd5765
Nov 05, 2015
- DI: Reverse direction of subprogram -> function edge. · d4bff303
  Peter Collingbourne authored Nov 05, 2015
  d4bff303
- [WinEH] Fix funclet prologues with stack realignment · 6ddae310
  Reid Kleckner authored Nov 05, 2015
  6ddae310
- [WebAssembly] Update wasm builtin functions to match spec changes. · d7ffb919
  Dan Gohman authored Nov 05, 2015
  d7ffb919
- Add cfi instr for CFA calculation when movpc is expanded to call and pop · 99fba3c1
  Petar Jovanovic authored Nov 05, 2015
  99fba3c1
- [WebAssembly] Rename ior operator to or to match the spec · 8a76b04a
  Derek Schuff authored Nov 05, 2015
  8a76b04a
- revert rev. 252153 due to build failure on ubuntu · f99c054e
  Asaf Badouh authored Nov 05, 2015
  f99c054e
- [X86][AVX512] add comi with Sae · 7fdabf0a
  Asaf Badouh authored Nov 05, 2015
  7fdabf0a
- AMDGPU: Fix assert when legalizing atomic operands · a40450cb
  Matt Arsenault authored Nov 05, 2015
  a40450cb
- [WinEH] Fix establisher param reg in CLR funclets · 6afccf61
  Joseph Tremoulet authored Nov 05, 2015
  6afccf61
- AMDGPU: Add missing v2f64 fadd tests · b0f87ed6
  Matt Arsenault authored Nov 05, 2015
  b0f87ed6
Nov 04, 2015
- [x86] Teach the shrink-wrapping hooks to do the proper thing with Win64. · 421723cd
  Quentin Colombet authored Nov 04, 2015
  421723cd
- [X86][SSE] Add general memory folding for (V)INSERTPS instruction · 7e6606f4
  Simon Pilgrim authored Nov 04, 2015
  7e6606f4
- Created new X86 FMA3 opcodes (FMA*_Int) that are used now for lowering of scalar FMA intrinsics. · e41a8c41
  Andrew Kaylor authored Nov 04, 2015
  e41a8c41
- [ARM] Combine CMOV into BFI where possible · e7d679cf
  James Molloy authored Nov 04, 2015
  e7d679cf
- [X86] DAGCombine should not introduce FILD in soft-float mode · b34de722
  Michael Kuperstein authored Nov 04, 2015
  b34de722
- [StatepointLowering] Remove distinction between call and invoke safepoints · 35fe6920
  Igor Laevsky authored Nov 04, 2015
  35fe6920
Nov 03, 2015
- Address nit · cd9488d5
  Derek Schuff authored Nov 03, 2015
  cd9488d5