Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / 2fcdfb448debafcea8b22c02e2370ea8a5da8bad / lib / Target / X86 / X86ISelLowering.cpp
  1. 2fcdfb4 by David Greene · 15 years ago
  2. 74a579d by David Greene · 15 years ago
  3. 6b38126 by David Greene · 15 years ago
  4. f125a29 by David Greene · 15 years ago
  5. a5f2601 by David Greene · 15 years ago
  6. 3f4be4f Target/X86: Tweak allocating shadow area (aka home) on Win64. It must be enough for caller to allocate one. by NAKAMURA Takumi · 15 years ago
  7. fb840c9 lib/Target/X86/X86ISelLowering.cpp: Introduce a new variable "IsWin64". No functional changes. by NAKAMURA Takumi · 15 years ago
  8. 2763538 Target/X86: Fix whitespace. by NAKAMURA Takumi · 15 years ago
  9. 66c357d by David Greene · 15 years ago
  10. ffc5533 by David Greene · 15 years ago
  11. ccacdc1 by David Greene · 15 years ago
  12. c38a03e by David Greene · 15 years ago
  13. f297c93 Fix PR9127 by reversing the operands even if they have more then one use. by Rafael Espindola · 15 years ago
  14. 2bffee2 Patches to build EFI with Clang/LLVM. By Carl Norum. by Evan Cheng · 15 years ago
  15. e9a7ea6 Keep track of incoming argument's location while emitting LiveIns. by Devang Patel · 15 years ago
  16. 54d8eba by David Greene · 15 years ago
  17. cfe33c4 by David Greene · 15 years ago
  18. 9158509 by David Greene · 15 years ago
  19. 7754f85 Target/X86: Tweak win64's tailcall. by NAKAMURA Takumi · 15 years ago
  20. e5fffe9 Fix whitespace. by NAKAMURA Takumi · 15 years ago
  21. 9cd3da4 fix PR8981, a crash trying to form a conditional inc with a floating point compare. by Chris Lattner · 15 years ago
  22. 16c29b5 Rename TargetFrameInfo into TargetFrameLowering. Also, put couple of FIXMEs and fixes here and there. by Anton Korobeynikov · 15 years ago
  23. c9df025 Simplify a bunch of isVirtualRegister() and isPhysicalRegister() logic. by Jakob Stoklund Olesen · 15 years ago
  24. 55d4200 Recognize inline asm 'rev /bin/bash, ' as a bswap intrinsic call. by Evan Cheng · 15 years ago
  25. a5e1362 Revert r122955. It seems using movups to lower memcpy can cause massive regression (even on Nehalem) in edge cases. I also didn't see any real performance benefit. by Evan Cheng · 15 years ago
  26. 461f1fc Use movups to lower memcpy and memset even if it's not fast (like corei7). by Evan Cheng · 15 years ago
  27. 0521928 Re-implement r122936 with proper target hooks. Now getMaxStoresPerMemcpy by Evan Cheng · 15 years ago
  28. e915ff3 X86: Lower a select directly to a setcc_carry if possible. by Benjamin Kramer · 15 years ago
  29. 7d6fe13 Add some x86 specific dagcombines for conditional increments. by Benjamin Kramer · 15 years ago
  30. f1b4eaf rename MVT::Flag to MVT::Glue. "Flag" is a terrible name for by Chris Lattner · 15 years ago
  31. 672fb62 Implement feedback from Bruno on making pblendvb an x86-specific ISD node in addition to being an intrinsic, and convert by Nate Begeman · 15 years ago
  32. 23a0199 now that addc/adde are gone, "ADDC" in the X86 backend uses EFLAGS results, by Chris Lattner · 15 years ago
  33. d8ff7ec use for loop over types. by Chris Lattner · 15 years ago
  34. 5b85654 Change the X86 backend to stop using the evil ADDC/ADDE/SUBC/SUBE nodes (which by Chris Lattner · 15 years ago
  35. a0fd0d5 Prevents PerformShuffleCombine from creating a node with an illegal type after legalize types by Mon P Wang · 15 years ago
  36. c19d1c3 improve the setcc -> setcc_carry optimization to happen more by Chris Lattner · 15 years ago
  37. 481eebc simplify some code to just reuse a setcc if we can instead of by Chris Lattner · 15 years ago
  38. 2a786eb now that generic vector types aren't selected onto MMX operations, by Chris Lattner · 15 years ago
  39. e019ec1 reduce copy/paste programming with the power of for loops. by Chris Lattner · 15 years ago
  40. a34b3cf X86 supports i8/i16 overflow ops (except i8 multiplies), we should by Chris Lattner · 15 years ago
  41. b65c175 Add support for matching psign & plendvb to the x86 target by Nate Begeman · 15 years ago
  42. 2ea8ee7 Formalize the notion that AVX and SSE are non-overlapping extensions from the compiler's point of view. Per email discussion, we either want to always use VEX-prefixed instructions or never use them, and are taking "HasAVX" to mean "Always use VEX". Passing -mattr=-avx,+sse42 should serve to restore legacy SSE support when desirable. by Nate Begeman · 15 years ago
  43. d8c0536 Rewrite the darwin tlv support to use a chain and return to copying by Eric Christopher · 15 years ago
  44. 2040961 Stop confusing people, it's not really a chain, or a tumor. by Eric Christopher · 15 years ago
  45. 8bce7cc Remove extraneous copy from DAG conversion for darwin tls. This was by Eric Christopher · 15 years ago
  46. 9637d5b Teach X86ISelLowering that the second result of X86ISD::UMUL is a flags by Chris Lattner · 15 years ago
  47. b20e0b1 it turns out that when ".with.overflow" intrinsics were added to the X86 by Chris Lattner · 15 years ago
  48. 96908b1 generalize the previous check to handle -1 on either side of the by Chris Lattner · 15 years ago
  49. a2b5600 Improve an integer select optimization in two ways: by Chris Lattner · 15 years ago
  50. 1292c22 Add patterns for the x86 popcnt instruction. by Benjamin Kramer · 15 years ago
  51. 9838396 Simplify code. No functionality change. by Benjamin Kramer · 15 years ago
  52. 1bf891a Fix and re-enable tail call optimization of expanded libcalls. by Evan Cheng · 15 years ago
  53. 3816978 I don't think it makes any sense to assert that the target supports SSE3 here. by Duncan Sands · 15 years ago
  54. 28cd48f Speculatively disable x86 portion of r120501 to appease the x86_64 buildbot. by Evan Cheng · 15 years ago
  55. 3d2125c Enable sibling call optimization of libcalls which are expanded during by Evan Cheng · 15 years ago
  56. 41c902f Fix insertion point in pcmp expander. by Eric Christopher · 15 years ago
  57. 82be220 Fix some cleanups from my last patch. by Eric Christopher · 15 years ago
  58. 228232b Rewrite mwait and monitor support and custom lower arguments. by Eric Christopher · 15 years ago
  59. d652dbe Move lowering of TLS_addr32 and TLS_addr64 to X86MCInstLower. by Rafael Espindola · 15 years ago
  60. 5bf7c53 Lower TLS_addr32 and TLS_addr64. by Rafael Espindola · 15 years ago
  61. bf17cfa Renaming ISD::BIT_CONVERT to ISD::BITCAST to better reflect the LLVM IR concept. by Wesley Peck · 15 years ago
  62. d0c3817 Move hasFP() and few related hooks to TargetFrameInfo. by Anton Korobeynikov · 15 years ago
  63. 1e61e69 add targetoperand flags for jump tables, constant pool and block address by Chris Lattner · 15 years ago
  64. 142b531 move the pic base symbol stuff up to MachineFunction by Chris Lattner · 15 years ago
  65. 4fd0ea0 simplify getPICBaseSymbol a bit. by Chris Lattner · 15 years ago
  66. 948cf02 Recognise 32-bit ror-based bswap implementation used by uclibc by Peter Collingbourne · 15 years ago
  67. 9836118 Support ; as asm separator by Peter Collingbourne · 15 years ago
  68. 0e03456 Remove possibly useful info from comment, per Chris. by Dale Johannesen · 15 years ago
  69. cdfad36 Simplify uses of MVT and EVT. An MVT can be compared directly by Duncan Sands · 15 years ago
  70. 4590766 Factorize the duplicated logic for choosing the right argument by Duncan Sands · 15 years ago
  71. 44ab89e Inline asm multiple alternative constraints development phase 2 - improved basic logic, added initial platform support. by John Thompson · 15 years ago
  72. 6dad10e x86-Win32: Switch ftol2 calling convention from stdcall to C. by Michael J. Spencer · 15 years ago
  73. d155d7e An stdcall function calling a non-stdcall function by Dale Johannesen · 15 years ago
  74. 1e1ca0b Add parentheses to pacify gcc, which warns otherwise. by Duncan Sands · 15 years ago
  75. e9c253e X86: Add alloca probing to dynamic alloca on Windows. Fixes PR8424. by Michael J. Spencer · 15 years ago
  76. e93d99c Remove Synthesizable from the Type system; as MMX vector by Dale Johannesen · 15 years ago
  77. 94f7eeb X86: Add MS-CRT libcalls. by Michael J. Spencer · 15 years ago
  78. 87b8665 Fix Whitespace. by Michael J. Spencer · 15 years ago
  79. 52b4505 Combine these together - should probably have some text associated by Eric Christopher · 15 years ago
  80. 65b65d6 Mark variable 'NoImplicitFloatOps' used only in an assert as used. by Nick Lewycky · 15 years ago
  81. 320afb8 Initial va_arg support for x86-64. Patch by David Meyer! by Dan Gohman · 15 years ago
  82. 1a2cf3b Fixes bug 8297: i386 cmpxchg8b, missing MachineMemOperand by Andrew Trick · 15 years ago
  83. 335b806 X86: Call ulldiv and ftol2 on Windows instead of their libgcc eqivilents. by Michael J. Spencer · 15 years ago
  84. 92bf38c X86: MinGW should always use libgcc on Windows. by Michael J. Spencer · 15 years ago
  85. 1802a9f X86: Call _alldiv instead of __divdi3 on Windows (excluding cygwin). by Michael J. Spencer · 15 years ago
  86. ec38de2 Fix Whitespace. by Michael J. Spencer · 15 years ago
  87. ec37b00 Recommit 116056, now with the missing file... by Cameron Esfahani · 15 years ago
  88. 5900201 reverting 116056: win64_params.ll may need to be conditionalized? by Andrew Trick · 15 years ago
  89. d7b8d91 Small patch to restore home register stack space allocation for the Win64 case. Add test case. This code eventually needs to be tighter, since it's always allocating it, even in leaf routines. by Cameron Esfahani · 15 years ago
  90. 835580f Canonicalize X86ISD::MOVDDUP nodes to v2f64 to make sure all cases match. Also eliminate unneeded isel patterns. rdar://8520311 by Evan Cheng · 15 years ago
  91. e7beda1 va_args support for Win64. Patch by Cameron! by Anton Korobeynikov · 15 years ago
  92. 0488fb6 Massive rewrite of MMX: by Dale Johannesen · 15 years ago
  93. 8094578 improve indentation by Chris Lattner · 15 years ago
  94. 722d315 This code should never fire on non-darwin subtargets. by Eric Christopher · 15 years ago
  95. c451051 We can't return SSE/MMX vectors if SSE is disabled. by Dale Johannesen · 15 years ago
  96. eafca4e Attempt to fix llvm-gcc build. It was crashing when building gcov.o for an by Bob Wilson · 15 years ago
  97. f93b90c reimplement elf TLS support in terms of addressing modes, eliminating SegmentBaseAddress. by Chris Lattner · 15 years ago
  98. 492a43e convert the last 4 X86ISD nodes that should have memoperands to have them. by Chris Lattner · 15 years ago
  99. 2156b79 give X86ISD::FNSTCW16m a memoperand, since it touches memory. It only by Chris Lattner · 15 years ago
  100. 0729093 give FP_TO_INT16_IN_MEM and friends a memoperand. They are only by Chris Lattner · 15 years ago