1. 092acb0 Fix a typo: _MM_FLUSH_ZERO_OFF has the wrong value. rdar://10716672 by Bob Wilson · 13 years ago
  2. 34a1da4 Add AVX2 permute intrinsics. Also add parentheses on some macro arguments in other intrinsic headers. by Craig Topper · 13 years ago
  3. 32bae37 Fix vector macros to correctly check argument types. <rdar://problem/10261670> by Bob Wilson · 13 years ago
  4. 7c06f6b Tweak *mmintrin.h so that they don't make any bad assumptions about alignment (which probably has little effect in practice, but better to get it right). Make the load in _mm_loadh_pi and _mm_loadl_pi a single LLVM IR instruction to make optimizing easier for CodeGen. by Eli Friedman · 13 years ago
  5. 8e3ec9c Add 'may_alias' attribute. Noticed by Eli. by Bill Wendling · 13 years ago
  6. eed92a1 Represent the unaligned loads natively. These are converted into a call to the by Bill Wendling · 13 years ago
  7. 0c96430 LLVM doesn't always optimize away the four loads from this: by Bill Wendling · 13 years ago
  8. bae20d0 Remove comment that snuck in there. by Bill Wendling · 13 years ago
  9. 60cb5e9 Just use a native "load" instead of translating the builtin later. Clang can by Bill Wendling · 13 years ago
  10. 7acb953 Undo part of my previous commit to mm_malloc.h, going back to the use of by Chandler Carruth · 14 years ago
  11. 6999339 the mmx intrinsic for pshufw should map to the IR intrinsic, not by Chris Lattner · 14 years ago
  12. 551f37c fix incorrect MM_HINT_ definitions, PR8011 by Chris Lattner · 14 years ago
  13. d74e398 Fix some typos I made when adding alternate intrinsic names. by Chandler Carruth · 14 years ago
  14. fa38c81 Add alternate names for x86 SIMD intrinsics. These aren't as common, but show by Chandler Carruth · 14 years ago
  15. 79e5ab7 Headers: Fix quoting of macro arguments in a couple more places. by Daniel Dunbar · 14 years ago
  16. c2b9b36 Make macro safe in the face of users who pass "addr+64". Addition of void* is by Nick Lewycky · 14 years ago
  17. f805a6c add type casts to _mm_shuffle_ps so the front-end applies by Chris Lattner · 14 years ago
  18. 1bddbcb fix PR6658: inline isn't a keyword in C89 mode, use __inline__ instead. by Chris Lattner · 14 years ago
  19. 020f1ed Migrate typedefs to the top level of xmmintrin.h and remove the same by Eric Christopher · 15 years ago
  20. ef5ebf6 add two aliases for SSE functions we already have, fixing PR6316. by Chris Lattner · 15 years ago
  21. 21b91a3 fix PR6187: various load functions in xmmintrin.h were missing 'const'. by Chris Lattner · 15 years ago
  22. 9afb227 fix a bug in the _MM_TRANSPOSE4_PS definition, patch by Brian Sumner from PR6138 by Chris Lattner · 15 years ago
  23. 2c48345 _mm_xor_ps does a xor not a nxor. The other 'xor' builtins look fine, by Chris Lattner · 15 years ago
  24. c7d95dc Following gcc, hide the hack to include the SSE2 intrinsics from by Eli Friedman · 15 years ago
  25. 3eef3e1 xmmintrin needs to include emmintrin, Darwin system headers seem to depend on it by Daniel Dunbar · 15 years ago
  26. e0ae8bd Now that LLVM CodeGen can handle the generic variations a bit better, by Eli Friedman · 15 years ago
  27. 098136a Replace more calls to builtins with generic code. by Eli Friedman · 15 years ago
  28. 17d2e3a Fix some casts to work without -flax-vector-conversions. by Eli Friedman · 15 years ago
  29. 80c8004 Misc fixes to MMX/SSE intrinsics: a few small bug fixes, and getting rid by Eli Friedman · 15 years ago
  30. db7351a Add aliases for a couple of SSE intrinsics. Patch by Ed Schouten. by Eli Friedman · 15 years ago
  31. 79dcf5f Add 'cmp' SSE builtins and get rid of a bunch of other builtins. by Anders Carlsson · 15 years ago
  32. 7add547 fix misspelt attribute. by Chris Lattner · 15 years ago
  33. d99e31b Spell __always_inline__ correctly. by Daniel Dunbar · 16 years ago
  34. a837a07 Fix more xmmintrin.h typos by Anders Carlsson · 16 years ago
  35. a2f12ae Add the nodebug attribute to intrinsics by Anders Carlsson · 16 years ago
  36. dae4413 80col. by Mike Stump · 16 years ago
  37. 4cc4427 Fix some mistakes I made when I tried to decipher the Intel documentation of the MXCSR register by Anders Carlsson · 16 years ago
  38. bbd1fa2 Fix more bugs I discovered by Anders Carlsson · 16 years ago
  39. 62005c1 Fix some errors I noticed by Anders Carlsson · 16 years ago
  40. 12868cc Add a MS specific intrinsic. by Anders Carlsson · 16 years ago
  41. 4fd3e63 OK, all tests pass. Let's start using the SSE2 header by Anders Carlsson · 16 years ago
  42. 398082e Fix test failures noticed by Fariborz. by Anders Carlsson · 16 years ago
  43. 2a29dcc Flip the switch and start using the new xmmintrin.h by Anders Carlsson · 16 years ago[Renamed from lib/Headers/xmmintrin.devel.h]
  44. 8e28d24 Comment out _mm_insert_pi16 for now by Anders Carlsson · 16 years ago
  45. a6431dc Implement the last intrinsics, _mm_insert_pi16 is the last remaining one now. by Anders Carlsson · 16 years ago
  46. b595509 Add macros by Anders Carlsson · 16 years ago
  47. 50099cb Add misc intrinsics by Anders Carlsson · 16 years ago
  48. c1f9afd Add control register intrinsics by Anders Carlsson · 16 years ago
  49. 62af71c Add integer intrinsics by Anders Carlsson · 16 years ago
  50. dedad4e Add cacheability intrinsics by Anders Carlsson · 16 years ago
  51. 09b9305 Add store intrinsics by Anders Carlsson · 16 years ago
  52. a6ba001 Add set intrinsics by Anders Carlsson · 16 years ago
  53. 9770086 Add load intrinsics by Anders Carlsson · 16 years ago
  54. 1b76b80 _mm_cvtsi64_ss is 64-bit only, so wrap it in #ifdef __x86_64__ by Anders Carlsson · 16 years ago
  55. 4fcc313 Add a couple of conversion intrinsics by Anders Carlsson · 16 years ago
  56. f62c681 Add comparison intrinsics. by Anders Carlsson · 16 years ago
  57. 566d8da Check in an in-progress version of xmmintrin.h that I had lying around. by Anders Carlsson · 16 years ago