Commits · 75f12cc3fe53f3d991e9d1500b6df5d7f7d064a0 · Roger Ferrer / llvm-epi-0.8 · GitLab

This is an archived project. Repository and other project resources are read-only.

Dec 20, 2011
- Mark ARM eh_sjlj_dispatchsetup as clobbering all registers. Radar 10567930. · 75f12cc3
  Bob Wilson authored Dec 20, 2011
  75f12cc3
- ARM assembly shifts by zero should be plain 'mov' instructions. · e2ca9e5b
  Jim Grosbach authored Dec 20, 2011
  e2ca9e5b
- Now that PR11464 is fixed, reapply the patch to fix PR11464, · 9eb3f004
  Chris Lattner authored Dec 20, 2011
  9eb3f004
- fix PR11464 by preventing the linker from mapping two different struct types... · 5e3bd972
  Chris Lattner authored Dec 20, 2011
  5e3bd972
- Move tests to FileCheck. · 3bfaefe9
  Evan Cheng authored Dec 19, 2011
  3bfaefe9
- ARM assembly parsing and encoding support for LDRD(label). · 8648c101
  Jim Grosbach authored Dec 19, 2011
  8648c101
Dec 19, 2011
- Add a test case for r146900. · 37c45db1
  Akira Hatanaka authored Dec 19, 2011
  37c45db1
- Add patterns for matching immediates whose lower 16-bit is cleared. These · db47e0c4
  Akira Hatanaka authored Dec 19, 2011
  db47e0c4
- ARM NEON two-operand aliases for VPADD. · 64f4de29
  Jim Grosbach authored Dec 19, 2011
  64f4de29
- Remove definitions of double word shift plus 32 instructions. Assembler or · 2a232d81
  Akira Hatanaka authored Dec 19, 2011
  2a232d81
- Remove the restriction on the first operand of the add node in SelectAddr. · 3c9f3363
  Akira Hatanaka authored Dec 19, 2011
  3c9f3363
- ARM NEON implied destination aliases for VMAX/VMIN. · 9ae4fc03
  Jim Grosbach authored Dec 19, 2011
  9ae4fc03
- ARM NEON relax parse time diagnostics for alignment specifiers. · cef98cdd
  Jim Grosbach authored Dec 19, 2011
  cef98cdd
Dec 18, 2011
- Allow inlining of functions with returns_twice calls, if they have the · d6cb7649
  Joerg Sonnenberger authored Dec 18, 2011
  d6cb7649
Dec 17, 2011
- Revert 146728 as it's causing failures on some of the external bots as well as · 5e5bee4c
  Chad Rosier authored Dec 17, 2011
  5e5bee4c
- Revert r146822 at Pete Cooper's request as it broke clang self hosting. · 8b3deabd
  Kevin Enderby authored Dec 17, 2011
  8b3deabd
- SimplifyCFG now predicts some conditional branches to true or false depending... · eadf124d
  Pete Cooper authored Dec 17, 2011
  eadf124d
- Deleting the json-bench-test until I understand why it is flaky. · 3c2848ea
  Manuel Klimek authored Dec 17, 2011
  3c2848ea
- Fix a CPSR liveness tracking bug introduced when I converted IT block to bundle. · 903231bc
  Evan Cheng authored Dec 17, 2011
  903231bc
- Add back the MC bits of 126425. Original patch by Nathan Jeffords. I added the · d3df3d35
  Rafael Espindola authored Dec 17, 2011
  d3df3d35
- Make sure that the lower bits on the VSELECT condition are properly set. · da07b3ad
  Lang Hames authored Dec 17, 2011
  da07b3ad
- The powers that be have decided that LLVM IR should now support 16-bit · 518cda42
  Dan Gohman authored Dec 17, 2011
  518cda42
- When recursing for the original size of a type, stop if we are at a · 27886c6c
  Eric Christopher authored Dec 16, 2011
  27886c6c
- Fix off-by-one error in bucket sort. · 9790187b
  Jakob Stoklund Olesen authored Dec 16, 2011
  9790187b
Dec 16, 2011
- Hexagon: Fix a nasty order-of-initialization bug. · 9ca2e729
  Benjamin Kramer authored Dec 16, 2011
  9ca2e729
- Adds a JSON parser and a benchmark (json-bench) to catch performance regressions. · 2c899a18
  Manuel Klimek authored Dec 16, 2011
  2c899a18
- By popular demand, link up types by name if they are isomorphic and one is an · 3fdf98c6
  Chris Lattner authored Dec 16, 2011
  3fdf98c6
- Don't try to match 'unpackl/h v, v' for 32xi8 and 16xi16 when only AVX1 is... · a4d411cb
  Craig Topper authored Dec 16, 2011
  a4d411cb
- [asan] add a test for instrumenting globals · 561dade5
  Kostya Serebryany authored Dec 16, 2011
  561dade5
- Make sure we correctly note the existence of an i8 immediate for vblendvps and... · 64944090
  Eli Friedman authored Dec 15, 2011
  64944090
Dec 15, 2011
- ARM NEON VCLE is an alias for VCGE w/ the source operands reversed. · a47294e2
  Jim Grosbach authored Dec 15, 2011
  a47294e2
- ARM NEON VTBL/VTBX assembly parsing and encoding. · 4a5c8873
  Jim Grosbach authored Dec 15, 2011
  4a5c8873
- Add missing zmovl AVX patterns which were causing crashes. · 41dbf59e
  Chad Rosier authored Dec 15, 2011
  41dbf59e
- Fix assert in LowerBUILD_VECTOR for v16i16 type on AVX. · 75ed9dcb
  Chad Rosier authored Dec 15, 2011
  75ed9dcb
- Set specific target cpu for testcase. · 918f976e
  Lang Hames authored Dec 15, 2011
  918f976e
- Added test case for r146671. · 2d6d3a2f
  Lang Hames authored Dec 15, 2011
  2d6d3a2f
- Add a test case to make sure that the nop really does follow the bl on ppc64 elf · 750366f0
  Hal Finkel authored Dec 15, 2011
  750366f0
- Fix test. · ef7b2f25
  Eli Friedman authored Dec 15, 2011
  ef7b2f25
- Make constant folding for GEPs a bit more aggressive. · a45ab503
  Eli Friedman authored Dec 15, 2011
  a45ab503
- Don't try to form FGETSIGN after legalization; it is possible in some cases,... · 2ec82496
  Eli Friedman authored Dec 15, 2011
  2ec82496