Commits · 91944e8699dedf7df7fc03429a2f97fc27e338a7 · Roger Ferrer / llvm-epi-0.8

Apr 15, 2006
- pslldrm, psrawrm, etc. encoding bug · 91944e86
  Evan Cheng authored Apr 15, 2006
```
llvm-svn: 27721
```
  91944e86
- hsubp{s|d} encoding bug · 1220b31a
  Evan Cheng authored Apr 15, 2006
```
llvm-svn: 27720
```
  1220b31a
- Silly bug · 6222cf2a
  Evan Cheng authored Apr 15, 2006
```
llvm-svn: 27719
```
  6222cf2a
- Do not use movs{h|l}dup for a shuffle with a single non-undef node. · 65bb720a
  Evan Cheng authored Apr 15, 2006
```
llvm-svn: 27718
```
  65bb720a
- significant cleanups to code that uses insert/extractelt heavily. This builds · 39fac448
  Chris Lattner authored Apr 15, 2006
```
maximal shuffles out of them where possible.

llvm-svn: 27717
```
  39fac448
- Added SSE (and other) entries to foldMemoryOperand(). · 0ba896c7
  Evan Cheng authored Apr 14, 2006
```
llvm-svn: 27716
```
  0ba896c7
- Some clean up · 00a5b3d9
  Evan Cheng authored Apr 14, 2006
```
llvm-svn: 27715
```
  00a5b3d9
- Allow undef in a shuffle mask · 559c8ba4
  Chris Lattner authored Apr 14, 2006
```
llvm-svn: 27714
```
  559c8ba4
- Move these ctors out of line · 0875d945
  Chris Lattner authored Apr 14, 2006
```
llvm-svn: 27713
```
  0875d945
Apr 14, 2006
- Last few SSE3 intrinsics. · 5d247f81
  Evan Cheng authored Apr 14, 2006
```
llvm-svn: 27711
```
  5d247f81
- Teach scalarrepl to promote unions of vectors and floats, producing · 3323ce16
  Chris Lattner authored Apr 14, 2006
```
insert/extractelement operations.  This implements
Transforms/ScalarRepl/vector_promote.ll

llvm-svn: 27710
```
  3323ce16
- Misc. SSE2 intrinsics: clflush, lfench, mfence · 3bd60539
  Evan Cheng authored Apr 14, 2006
```
llvm-svn: 27699
```
  3bd60539
- We were not adjusting the frame size to ensure proper alignment when alloca / · e349d01a
  Evan Cheng authored Apr 14, 2006
```
vla are present in the function. This causes a crash when a leaf function
allocates space on the stack used to store / load with 128-bit SSE
instructions.

llvm-svn: 27698
```
  e349d01a
- New entry · 8d76f392
  Evan Cheng authored Apr 14, 2006
```
llvm-svn: 27697
```
  8d76f392
- Don't print out the install command for Intrinsics.gen unless VERBOSE mode. · ef56d92d
  Reid Spencer authored Apr 14, 2006
```
llvm-svn: 27696
```
  ef56d92d
- Make this assertion better · 086e986e
  Chris Lattner authored Apr 14, 2006
```
llvm-svn: 27695
```
  086e986e
- Move the rest of the PPCTargetLowering::LowerOperation cases out into · 4211ca91
  Chris Lattner authored Apr 14, 2006
```
separate functions, for simplicity and code clarity.

llvm-svn: 27693
```
  4211ca91
- Pull the VECTOR_SHUFFLE and BUILD_VECTOR lowering code out into separate · 19e9055e
  Chris Lattner authored Apr 14, 2006
```
functions, which makes the code much cleaner :)

llvm-svn: 27692
```
  19e9055e
- Implement value #'ing for vector operations, implementing · 68c650ca
  Chris Lattner authored Apr 14, 2006
```
Regression/Transforms/GCSE/vectorops.ll

llvm-svn: 27691
```
  68c650ca
- pcmpeq* and pcmpgt* intrinsics. · eb0063a3
  Evan Cheng authored Apr 14, 2006
```
llvm-svn: 27685
```
  eb0063a3
- psll*, psrl*, and psra* intrinsics. · 16287444
  Evan Cheng authored Apr 14, 2006
```
llvm-svn: 27684
```
  16287444
- Remove the .cvsignore file so this directory can be pruned. · 64f6c11c
  Reid Spencer authored Apr 13, 2006
```
llvm-svn: 27683
```
  64f6c11c
Apr 13, 2006
- Remove .cvsignore so that this directory can be pruned. · 497ecf68
  Reid Spencer authored Apr 13, 2006
```
llvm-svn: 27682
```
  497ecf68
- Handle some kernel code than ends in [0 x sbyte]. I think this is safe · 4aa30016
  Andrew Lenharth authored Apr 13, 2006
```
llvm-svn: 27672
```
  4aa30016
- Expand some code with temporary variables to rid ourselves of the warning · 709eaacb
  Reid Spencer authored Apr 13, 2006
```
about "dereferencing type-punned pointer will break strict-aliasing rules"

llvm-svn: 27671
```
  709eaacb
- Doh. PANDrm, etc. are not commutable. · a8431971
  Evan Cheng authored Apr 13, 2006
```
llvm-svn: 27668
```
  a8431971
- Force non-darwin targets to use a static relo model. This fixes PR734, · 883fb053
  Chris Lattner authored Apr 13, 2006
```
tested by CodeGen/Generic/vector.ll

llvm-svn: 27657
```
  883fb053
- add a note, move an altivec todo to the altivec list. · 5879efe0
  Chris Lattner authored Apr 13, 2006
```
llvm-svn: 27654
```
  5879efe0
- linear -> constant time · 92cf71f6
  Andrew Lenharth authored Apr 13, 2006
```
llvm-svn: 27652
```
  92cf71f6
- Add the README files to the distribution. · 9857229a
  Reid Spencer authored Apr 13, 2006
```
llvm-svn: 27651
```
  9857229a
- psad, pmax, pmin intrinsics. · ed399674
  Evan Cheng authored Apr 13, 2006
```
llvm-svn: 27647
```
  ed399674
- Various SSE2 packed integer intrinsics: pmulhuw, pavgw, etc. · 58dad559
  Evan Cheng authored Apr 13, 2006
```
llvm-svn: 27645
```
  58dad559
- X86 SSE2 supports v8i16 multiplication · e4f97ccf
  Evan Cheng authored Apr 13, 2006
```
llvm-svn: 27644
```
  e4f97ccf
- Update · d2eb6624
  Evan Cheng authored Apr 13, 2006
```
llvm-svn: 27643
```
  d2eb6624
- padds{b|w}, paddus{b|w}, psubs{b|w}, psubus{b|w} intrinsics. · b3fe00bd
  Evan Cheng authored Apr 13, 2006
```
llvm-svn: 27639
```
  b3fe00bd
- Naming inconsistency. · 0aab735a
  Evan Cheng authored Apr 13, 2006
```
llvm-svn: 27638
```
  0aab735a
- SSE / SSE2 conversion intrinsics. · c88afc36
  Evan Cheng authored Apr 12, 2006
```
llvm-svn: 27637
```
  c88afc36
Apr 12, 2006
- All "integer" logical ops (pand, por, pxor) are now promoted to v2i64. · 92232307
  Evan Cheng authored Apr 12, 2006
```
Clean up and fix various logical ops issues.

llvm-svn: 27633
```
  92232307
- Promote vector AND, OR, and XOR · 119266ea
  Evan Cheng authored Apr 12, 2006
```
llvm-svn: 27632
```
  119266ea
- Make sure CVS versions of yacc and lex files get distributed. · 7c8fef2c
  Reid Spencer authored Apr 12, 2006
```
llvm-svn: 27630
```
  7c8fef2c