Commits · e8c9d1eaf7a5cdfe9fb93f87b25a42a758485a59 · Roger Ferrer / llvm-epi · GitLab

Nov 24, 2014
- [mips][microMIPS] Implement LBU16, LHU16, LW16, SB16, SH16 and SW16 instructions · e8c9d1ea
  Jozef Kolek authored Nov 24, 2014
  e8c9d1ea
- [mips][microMIPS] Implement disassembler support for 16-bit instructions · ea22c4cf
  Jozef Kolek authored Nov 24, 2014
  ea22c4cf
- [X86] Improved target specific combine on VSELECT dag nodes. · 23e2cfa8
  Andrea Di Biagio authored Nov 24, 2014
  23e2cfa8
- InstCombine: Don't assume DataLayout is always available · b2a6e745
  David Majnemer authored Nov 24, 2014
  b2a6e745
Nov 23, 2014
- [X86] Fixes bug in build_vector v4x32 lowering · 9ef90647
  Michael Kuperstein authored Nov 23, 2014
  9ef90647
- Masked Vector Load and Store Intrinsics. · 9e5089a9
  Elena Demikhovsky authored Nov 23, 2014
  9e5089a9
- R600: Fix extloads of i1 on R600/Evergreen · 2a495975
  Matt Arsenault authored Nov 23, 2014
  2a495975
- R600/SI: Add additional tests for i1 loads · b7ebdffe
  Matt Arsenault authored Nov 23, 2014
  b7ebdffe
- R600/SI: Fix broken check lines and modernize prefixes · 79db0a70
  Matt Arsenault authored Nov 23, 2014
  79db0a70
- R600/SI: Fix missing -verify-machineinstrs on a test · 8499ea6a
  Matt Arsenault authored Nov 23, 2014
  8499ea6a
Nov 22, 2014
- InstCombine: Propagate exact for (sdiv X, Pow2) -> (udiv X, Pow2) · fb380557
  David Majnemer authored Nov 22, 2014
  fb380557
- InstCombine: Propagate exact for (sdiv X, Y) -> (udiv X, Y) · ec6e481b
  David Majnemer authored Nov 22, 2014
  ec6e481b
- InstCombine: Propagate exact for (sdiv -X, C) -> (sdiv X, -C) · fa4699e6
  David Majnemer authored Nov 22, 2014
  fa4699e6
- InstCombine: Propagate exact in (udiv (lshr X,C1),C2) -> (udiv x,C1<<C2) · a3aeb156
  David Majnemer authored Nov 22, 2014
  a3aeb156
- InstCombine: Propagate NSW/NUW for X*(1<<Y) -> X<<Y · 546f8106
  David Majnemer authored Nov 22, 2014
  546f8106
- InstCombine: Propagate NSW for -X * -Y -> X * Y · 8279a750
  David Majnemer authored Nov 22, 2014
  8279a750
- InstSimplify: Simplify (sub 0, X) -> X if it's NUW · 4efa9ff8
  David Majnemer authored Nov 22, 2014
  4efa9ff8
- [x86] Add some tests for a common unpack pattern of vector shuffle that · 8c44d86a
  Chandler Carruth authored Nov 22, 2014
  8c44d86a
- InstCombine: Preserve nsw when folding X*(2^C) -> X << C · 80c8f627
  David Majnemer authored Nov 22, 2014
  80c8f627
- InstCombine: Preserve nsw/nuw for ((X << C2)*C1) -> (X * (C1 << C2)) · fd4a6d2b
  David Majnemer authored Nov 22, 2014
  fd4a6d2b
- InstCombine: Preserve nsw for (mul %V, -1) -> (sub 0, %V) · 027bc809
  David Majnemer authored Nov 22, 2014
  027bc809
- [InstCombine] Re-commit of r218721 (Optimize icmp-select-icmp sequence) · ec6217c9
  Gerolf Hoflehner authored Nov 21, 2014
  ec6217c9
Nov 21, 2014
- Fix transformation of add with pc argument to adr for non-immediate · 02b13a8d
  Joerg Sonnenberger authored Nov 21, 2014
  02b13a8d
- [asan] remove old experimental code · 60ef25bd
  Kostya Serebryany authored Nov 21, 2014
  60ef25bd
- R600/SI: Add a failing test case for offset order in ds_read2 instructions · f1206edf
  Tom Stellard authored Nov 21, 2014
  f1206edf
- R600/SI: Emit s_mov_b32 m0, -1 before every DS instruction · a99ada52
  Tom Stellard authored Nov 21, 2014
  a99ada52
- R600/SI: Add SIFoldOperands pass · 6596ba79
  Tom Stellard authored Nov 21, 2014
  6596ba79
- [mips][microMIPS] This patch implements functionality in MIPS delay slot · 3b8ddb66
  Jozef Kolek authored Nov 21, 2014
  3b8ddb66
- R600/SI: Use hex notation for constant in test · 3ae58878
  Tom Stellard authored Nov 21, 2014
  3ae58878
- [Hexagon] Adding sxth instruction. · 310991c6
  Colin LeMahieu authored Nov 21, 2014
  310991c6
- [Hexagon] Adding sxtb instruction. Renaming some identically named classes... · 91ffec90
  Colin LeMahieu authored Nov 21, 2014
  91ffec90
- Debug Info: revert r222195, r222210 and r222239. · f0a582ba
  Manman Ren authored Nov 21, 2014
  f0a582ba
- Add a feature flag for slow 32-byte unaligned memory accesses [x86]. · 501890e9
  Sanjay Patel authored Nov 21, 2014
  501890e9
- [x86] Restructure the checking patterns for v16 and v32 avx2 vector · ce5a26b0
  Chandler Carruth authored Nov 21, 2014
  ce5a26b0
- [x86] Make the previous logic significantly less conservative and get · 6c4d1ea8
  Chandler Carruth authored Nov 21, 2014
  6c4d1ea8
- [DAG] Teach how to turn a build_vector into a shuffle if some of the operands are zero. · 0225b5bf
  Andrea Di Biagio authored Nov 21, 2014
  0225b5bf
- [x86] Teach the x86 vector shuffle lowering to detect mergable 128-bit · d2b19bc8
  Chandler Carruth authored Nov 21, 2014
  d2b19bc8
- [x86] Remove more windows line endings that slipped into this file... · 77e1a0ad
  Chandler Carruth authored Nov 21, 2014
  77e1a0ad
- [x86] Add a bunch of test cases to 256-bit shuffles that exercise · 61c7b625
  Chandler Carruth authored Nov 21, 2014
  61c7b625
- [X86] For Silvermont CPU use 16-bit division instead of 64-bit for small positive numbers · fd1731d8
  Alexey Volkov authored Nov 21, 2014
  fd1731d8