Commits · 056f02256935d943e08c8f475a1e87e32291ad9b · Roger Ferrer / llvm-epi-0.8 · GitLab

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

Apr 07, 2006
- For PR723: · 056f0225
  Reid Spencer authored Apr 07, 2006
  056f0225
- For PR723: · b65ade8b
  Reid Spencer authored Apr 07, 2006
  b65ade8b
- For PR723: · 080eb0b3
  Reid Spencer authored Apr 07, 2006
  080eb0b3
- For PR723: · 754cfecd
  Reid Spencer authored Apr 07, 2006
  754cfecd
- Ignore stuff built by "make dist". · 1a63f48c
  Reid Spencer authored Apr 07, 2006
  1a63f48c
- A saner workaround. I hope. · bb66ec26
  Evan Cheng authored Apr 07, 2006
  bb66ec26
- Temporary workaround for a Mac OSX specific issue. · 1e86e679
  Evan Cheng authored Apr 07, 2006
  1e86e679
- Added more shuffle tests · 085c61d8
  Evan Cheng authored Apr 07, 2006
  085c61d8
- We have an assertion that checks that we do not encode null values into the · 92b1a570
  Chris Lattner authored Apr 07, 2006
  92b1a570
- Constant fold extractelement(zero, x) -> zero · e4f9d7b2
  Chris Lattner authored Apr 07, 2006
  e4f9d7b2
- Fix inlining of insert/extract element constantexprs · 8ec0205d
  Chris Lattner authored Apr 07, 2006
  8ec0205d
- Add code to RemapOperand() to handle Instruction::ExtractElement and · 5c349b8b
  Evan Cheng authored Apr 07, 2006
  5c349b8b
- A MOVPS2SSmr, i.e. _mm_store_ss, encoding bug. · d8e1a01b
  Evan Cheng authored Apr 06, 2006
  d8e1a01b
- Remove int_x86_sse_loadh_ps and int_x86_sse_loadl_ps. These are now lowered · bc658994
  Evan Cheng authored Apr 06, 2006
  bc658994
- - movlp{s|d} and movhp{s|d} support. · c995b45f
  Evan Cheng authored Apr 06, 2006
  c995b45f
- New entries. · acf8b3c8
  Evan Cheng authored Apr 06, 2006
  acf8b3c8
- 1. If both vector operands of a vector_shuffle are undef, turn it into an undef. · 613996c5
  Evan Cheng authored Apr 06, 2006
  613996c5
- This may be overconservative, but it lets the new cfe compile · 1596a1b2
  Andrew Lenharth authored Apr 06, 2006
  1596a1b2
- Add an item · e61cfad8
  Chris Lattner authored Apr 06, 2006
  e61cfad8
- Make sure to return the result in the right type. · 466841dd
  Chris Lattner authored Apr 06, 2006
  466841dd
- Move this to lib/Analysis. · 477169e3
  Andrew Lenharth authored Apr 06, 2006
  477169e3
- Match vpku[hw]um(x,x). · a4bbfaed
  Chris Lattner authored Apr 06, 2006
  a4bbfaed
- Add testcases for vpku[hw]um(x,x) · b8fc5376
  Chris Lattner authored Apr 06, 2006
  b8fc5376
- Add the variable llvmgccmajvers to the site.exp file. This will contain · c5aca114
  Reid Spencer authored Apr 06, 2006
  c5aca114
- Avoid building the runtime libraries if llvm-gcc version 4 is detected. · aed84e46
  Reid Spencer authored Apr 06, 2006
  aed84e46
- Add support for matching vmrg(x,x) patterns · f38e0332
  Chris Lattner authored Apr 06, 2006
  f38e0332
- Add vmrg(x,x) tests · 8688b6be
  Chris Lattner authored Apr 06, 2006
  8688b6be
- For support of new GCC v4, obtain the full and major versions of the · 6c99b419
  Reid Spencer authored Apr 06, 2006
  6c99b419
Apr 06, 2006
- fix some linking problems with the new gcc · cee782d5
  Andrew Lenharth authored Apr 06, 2006
  cee782d5
- Add tests for vmrg[hl]* · ab7e5301
  Chris Lattner authored Apr 06, 2006
  ab7e5301
- These intrinsics are now lowered by the FE. · 26749fe6
  Chris Lattner authored Apr 06, 2006
  26749fe6
- Pattern match vmrg* instructions, which are now lowered by the CFE into shuffles. · d1dcb520
  Chris Lattner authored Apr 06, 2006
  d1dcb520
- Infer element types for shuffle masks · 726df0bb
  Chris Lattner authored Apr 06, 2006
  726df0bb
- rename a method, to avoid confusion with llvm intrinsics. · 09575a9b
  Chris Lattner authored Apr 06, 2006
  09575a9b
- test vperm promotion · 13f01c8a
  Chris Lattner authored Apr 06, 2006
  13f01c8a
- remove two done items · a4c727f1
  Chris Lattner authored Apr 06, 2006
  a4c727f1
- Lower vperm(x,y, mask) -> shuffle(x,y,mask) if mask is constant. This allows · e79d249c
  Chris Lattner authored Apr 06, 2006
  e79d249c
- this is no longer an intrinsic. · 6329bbf2
  Chris Lattner authored Apr 06, 2006
  6329bbf2
- Support pattern matching vsldoi(x,y) and vsldoi(x,x), which allows the f.e. to · 1d338191
  Chris Lattner authored Apr 06, 2006
  1d338191
- new testcase · 0373789c
  Chris Lattner authored Apr 06, 2006
  0373789c