Commits · 895dba9714d5d44b8fa2e687d7f6a7965fff168d · Lorenzo Albano / LLVM bpEVL · GitLab

Apr 17, 2006
- assign stable opcodes to the various altivec ops. · 895dba97
  Chris Lattner authored Apr 17, 2006
  895dba97
- PPCPerfectShuffle.h is autogenerated, don't include it in the LOC counts. · 77a3ca63
  Chris Lattner authored Apr 17, 2006
  77a3ca63
- Check in a table, generated by llvm-PerfectShuffle, of optimal shuffles · 07a3d01a
  Chris Lattner authored Apr 17, 2006
  07a3d01a
- Rename BuildShuffleTable -> PerfectShuffle · ac66b4a1
  Chris Lattner authored Apr 17, 2006
  ac66b4a1
- rename the table · 156c85ea
  Chris Lattner authored Apr 17, 2006
  156c85ea
- Initial checking of a perfect shuffle generation program for 4-element · 2951955d
  Chris Lattner authored Apr 17, 2006
  2951955d
Apr 16, 2006
- Better way to splat v2f64 · c6f665f5
  Evan Cheng authored Apr 16, 2006
  c6f665f5
- movduprm, movshduprm bugs · 20712dee
  Evan Cheng authored Apr 16, 2006
  20712dee
- Encoding bugs · 3064f9aa
  Evan Cheng authored Apr 16, 2006
  3064f9aa
- Can't fold loads into alias vector SSE ops used for scalar operation. The load · 685ddd81
  Evan Cheng authored Apr 16, 2006
  685ddd81
- Implement a TODO: have the legalizer canonicalize a bunch of operations to · 06a21ba9
  Chris Lattner authored Apr 16, 2006
  06a21ba9
- Add support for promoting stores from one legal type to another, allowing us · 91226e57
  Chris Lattner authored Apr 16, 2006
  91226e57
- Make the BUILD_VECTOR lowering code much more aggressive w.r.t constant vectors. · fa5aa396
  Chris Lattner authored Apr 16, 2006
  fa5aa396
- Fix a bug in the 'shuffle(undef,x,mask) -> shuffle(x, undef,mask')' xform · 9095186d
  Chris Lattner authored Apr 16, 2006
  9095186d
- Canonicalize shuffle(undef,x,mask) -> shuffle(x, undef,mask'). · 34cebe78
  Chris Lattner authored Apr 16, 2006
  34cebe78
- Fix a crash when faced with a shuffle vector that has an undef in its mask. · 24acbe46
  Chris Lattner authored Apr 15, 2006
  24acbe46
- Add patterns for matching vnots with bit converted inputs. Most of these will · 873202fa
  Chris Lattner authored Apr 15, 2006
  873202fa
- Add a new vnot_conv predicate for matching vnot's where the allones vector is · 41df12ff
  Chris Lattner authored Apr 15, 2006
  41df12ff
- Make these predicates return true for bit_convert(buildvector)'s as well as · 7e7ad593
  Chris Lattner authored Apr 15, 2006
  7e7ad593
Apr 15, 2006
- More encoding bugs · 8f1d8013
  Evan Cheng authored Apr 15, 2006
  8f1d8013
- pslldrm, psrawrm, etc. encoding bug · 91944e86
  Evan Cheng authored Apr 15, 2006
  91944e86
- hsubp{s|d} encoding bug · 1220b31a
  Evan Cheng authored Apr 15, 2006
  1220b31a
- Silly bug · 6222cf2a
  Evan Cheng authored Apr 15, 2006
  6222cf2a
- Do not use movs{h|l}dup for a shuffle with a single non-undef node. · 65bb720a
  Evan Cheng authored Apr 15, 2006
  65bb720a
- significant cleanups to code that uses insert/extractelt heavily. This builds · 39fac448
  Chris Lattner authored Apr 15, 2006
  39fac448
- Added SSE (and other) entries to foldMemoryOperand(). · 0ba896c7
  Evan Cheng authored Apr 14, 2006
  0ba896c7
- Some clean up · 00a5b3d9
  Evan Cheng authored Apr 14, 2006
  00a5b3d9
- Allow undef in a shuffle mask · 559c8ba4
  Chris Lattner authored Apr 14, 2006
  559c8ba4
- Move these ctors out of line · 0875d945
  Chris Lattner authored Apr 14, 2006
  0875d945
- These instructions always return a packed vector. Improve the class... · 5876571a
  Chris Lattner authored Apr 14, 2006
  5876571a
Apr 14, 2006
- Last few SSE3 intrinsics. · 5d247f81
  Evan Cheng authored Apr 14, 2006
  5d247f81
- Teach scalarrepl to promote unions of vectors and floats, producing · 3323ce16
  Chris Lattner authored Apr 14, 2006
  3323ce16
- New testcase, checking to see we can turn this code: · 1c0ef438
  Chris Lattner authored Apr 14, 2006
  1c0ef438
- Adding back vector instructions to keep in mainline. · b138bbe8
  Tanya Lattner authored Apr 14, 2006
  b138bbe8
- Removing unimplemented vector instructions from language referrence. · 8333c7d7
  Tanya Lattner authored Apr 14, 2006
  8333c7d7
- Correct the Superclasses list for GlobalVariable and Function to indicate · 83ee73fd
  Reid Spencer authored Apr 14, 2006
  83ee73fd
- Allow an option, -with-externals, to be specified that provides the location · 9a20587c
  Reid Spencer authored Apr 14, 2006
  9a20587c
- Misc. SSE2 intrinsics: clflush, lfench, mfence · 3bd60539
  Evan Cheng authored Apr 14, 2006
  3bd60539
- We were not adjusting the frame size to ensure proper alignment when alloca / · e349d01a
  Evan Cheng authored Apr 14, 2006
  e349d01a
- New entry · 8d76f392
  Evan Cheng authored Apr 14, 2006
  8d76f392