Gitiles
Code Review Sign In
gerrit-public.fairphone.software / toolchain / llvm-project / 81fe1fbf4abd00b24e078a80cb15a2d2c1343f11 / llvm / lib / Analysis / TargetTransformInfo.cpp
  1. 2b166c5 [TTI] getOperandInfo - a broadcast shuffle means the result is OK_UniformValue by Simon Pilgrim · 7 years ago
  2. 077a42c [TTI] Make TargetTransformInfo::getOperandInfo static. NFCI. by Simon Pilgrim · 7 years ago
  3. 26e1c88 [TTI] Flip vector types in getShuffleCost SK_ExtractSubvector call by Simon Pilgrim · 7 years ago
  4. d0c7160 [CostModel] Add SK_ExtractSubvector handling to getInstructionThroughput (PR39368) by Simon Pilgrim · 7 years ago
  5. 34da6dd [LV] Support vectorization of interleave-groups that require an epilog under by Dorit Nuzman · 7 years ago
  6. 38bbf81 recommit 344472 after fixing build failure on ARM and PPC. by Dorit Nuzman · 7 years ago
  7. 5118c68 revert 344472 due to failures. by Dorit Nuzman · 7 years ago
  8. 8174368 [IAI,LV] Add support for vectorizing predicated strided accesses using masked by Dorit Nuzman · 7 years ago
  9. 29d80f0 [LoopVectorizer] Use TTI.getOperandInfo() by Jonas Paulsson · 7 years ago
  10. f78650a Remove trailing space by Fangrui Song · 7 years ago
  11. 667a5b5 [TargetTransformInfo] Add pow2 analysis for scalar constants by Simon Pilgrim · 7 years ago
  12. 2ca3360 [IR] move shuffle mask queries from TTI to ShuffleVectorInst by Sanjay Patel · 7 years ago
  13. 1193bbf Fix namespaces. No functionality change. by Benjamin Kramer · 7 years ago
  14. c0d53ab [CostModel] Cleanup isSingleSourceVectorMask to match other shuffle matchers. NFCI. by Simon Pilgrim · 7 years ago
  15. 32702cc [CostModel] Recognise REVERSE shuffle mask if the elements come from the second src by Simon Pilgrim · 7 years ago
  16. 54a138a [CostModel] Recognise BROADCAST shuffle mask if the elements come from the second src by Simon Pilgrim · 7 years ago
  17. e39fa6c [CostModel] Replace ShuffleKind::SK_Alternate with ShuffleKind::SK_Select (PR33744) by Simon Pilgrim · 7 years ago
  18. 51ef8dab Fix signed/unsigned warning. NFCI. by Simon Pilgrim · 7 years ago
  19. 0783921 [CostModel] Treat Identity shuffle masks as zero cost by Simon Pilgrim · 7 years ago
  20. 4162d77 [TTI] Add uniform/non-uniform constant Pow2 detection to TargetTransformInfo::getInstructionThroughput by Simon Pilgrim · 7 years ago
  21. 5f8f34e4 Remove \brief commands from doxygen comments. by Adrian Prantl · 7 years ago
  22. b4096eb [TTI, AArch64] Add transpose shuffle kind by Matthew Simpson · 7 years ago
  23. dfed941 [LV] Introduce TTI::getMinimumVF by Krzysztof Parzyszek · 7 years ago
  24. 8ad9a97 Plumb useAA through TargetTransformInfo to remove Transforms->CodeGen header dependency by David Blaikie · 7 years ago
  25. 5d93fdf [LV] Add TTI::shouldMaximizeVectorBandwidth to allow enabling it per target by Krzysztof Parzyszek · 8 years ago
  26. 0b377e0 [LSR] Allow giving priority to post-incrementing addressing modes by Krzysztof Parzyszek · 8 years ago
  27. d7c702b [LoopStrengthReduce, x86] don't add cost for a cmp that will be macro-fused (PR35681) by Sanjay Patel · 8 years ago
  28. 1f59ae3 Re-commit : [PowerPC] Add handling for ColdCC calling convention and a pass to mark by Zaara Syeda · 8 years ago
  29. c9dc7b4 Revert [PowerPC] This reverts commit rL322721 by Zaara Syeda · 8 years ago
  30. 8e951fd [PowerPC] Add handling for ColdCC calling convention and a pass to mark by Zaara Syeda · 8 years ago
  31. 29697c1 Revert r321377, it causes regression to https://reviews.llvm.org/P8055. by Guozhi Wei · 8 years ago
  32. 3325034 [SimplifyCFG] Don't do if-conversion if there is a long dependence chain by Guozhi Wei · 8 years ago
  33. 5fb624a [Memcpy Loop Lowering] Remove the fixed int8 lowering. by Sean Fertile · 8 years ago
  34. 0de1a4b [PartiallyInlineLibCalls][x86] add TTI hook to allow sqrt inlining to depend on arg rather than result by Sanjay Patel · 8 years ago
  35. b2c3eb8 [CodeGen][ExpandMemcmp] Allow memcmp to expand to vector loads (2). by Clement Courbet · 8 years ago
  36. cb8f632 [NVPTX] allow address space inference for volatile loads/stores. by Artem Belevich · 8 years ago
  37. 3344a21 Revert r314923: "Recommit : Use the basic cost if a GEP is not used as addressing mode" by Daniel Jasper · 8 years ago
  38. d40e03c Recommit : Use the basic cost if a GEP is not used as addressing mode by Jun Bum Lim · 8 years ago
  39. e76aa3b Revert "Use the basic cost if a GEP is not used as addressing mode" by Alex Shlyapnikov · 8 years ago
  40. 0e16a59 Use the basic cost if a GEP is not used as addressing mode by Jun Bum Lim · 8 years ago
  41. 2807c0a [CodeGenPrepare][NFC] Rename TargetTransformInfo::expandMemCmp -> TargetTransformInfo::enableMemCmpExpansion. by Clement Courbet · 8 years ago
  42. 6fd4391 [DivRempairs] add a pass to optimize div/rem pairs (PR31028) by Sanjay Patel · 8 years ago
  43. 62d6414 [TargetTransformInfo] Add a new public interface getInstructionCost by Guozhi Wei · 8 years ago
  44. 6dd29fc [SLP] Support for horizontal min/max reduction. by Alexey Bataev · 8 years ago
  45. d7eb619 Model cache size and associativity in TargetTransformInfo by Tobias Grosser · 8 years ago
  46. 6228aed [LSR / TTI / SystemZ] Eliminate TargetTransformInfo::isFoldableMemAccess() by Jonas Paulsson · 8 years ago
  47. 3e9b3eb [Cost] Rename getReductionCost() to getArithmeticReductionCost(), NFC. by Alexey Bataev · 8 years ago
  48. cef53dc [TTI] fixing a bug in the isLegalMaskedScatter API by Mohammed Agabaria · 8 years ago
  49. 024e319 [SystemZ, LoopStrengthReduce] by Jonas Paulsson · 8 years ago
  50. abdef9e [TTI] Refine the cost of EXT in getUserCost() by Haicheng Wu · 8 years ago
  51. 9cd1cdf Extend memcpy expansion in Transform/Utils to handle wider operand types. by Sean Fertile · 8 years ago
  52. 70ed78e [TargetTransformInfo, API] Add a list of operands to TTI::getUserCost by Evgeny Astigeevich · 8 years ago
  53. 66d9bdb [LoopUnroll] Pass SCEV to getUnrollingPreferences hook. NFCI. by Geoff Berry · 8 years ago
  54. 0f9c84c DivergencyAnalysis patch for review by Alexander Timofeev · 8 years ago
  55. b2a212c [Atomics][LoopIdiom] Recognize unordered atomic memcpy by Anna Thomas · 8 years ago
  56. f2b3b46 Fix PR23384 (part 2 of 3) NFC by Evgeny Stupachenko · 8 years ago
  57. 3a7578c [PPC] Inline expansion of memcmp by Zaara Syeda · 8 years ago
  58. 8624b7e [LoopVectorizer] Let target prefer scalar addressing computations. by Jonas Paulsson · 8 years ago
  59. e29686e [SLP] Enable 64-bit wide vectorization on AArch64 by Adam Nemet · 8 years ago
  60. 836b0f4 Add a late IR expansion pass for the experimental reduction intrinsics. by Amara Emerson · 8 years ago
  61. cf9daa3 Introduce experimental generic intrinsics for horizontal vector reductions. by Amara Emerson · 8 years ago
  62. 919f9e8 [InlineCost] Improve the cost heuristic for Switch by Jun Bum Lim · 8 years ago
  63. da74ed4 [LoopVectorizer, TTI] New method supportsEfficientVectorElementLoadStore() by Jonas Paulsson · 8 years ago
  64. fccc7d6 [SystemZ] TargetTransformInfo cost functions implemented. by Jonas Paulsson · 8 years ago
  65. dee5565 [CodeGenPrep] move aarch64-type-promotion to CGP by Jun Bum Lim · 8 years ago
  66. a48ea23 [TargetTransformInfo] getIntrinsicInstrCost() scalarization estimation improved by Jonas Paulsson · 9 years ago
  67. 42b6478 NVPTX: Refactor NVPTXInferAddressSpaces to check TTI by Matt Arsenault · 9 years ago
  68. 8e2f948 [TargetTransformInfo] Refactor and improve getScalarizationOverhead() by Jonas Paulsson · 9 years ago
  69. 2c96c43 [X86] updating TTI costs for arithmetic instructions on X86\SLM arch. by Mohammed Agabaria · 9 years ago
  70. 23599ba Currently isLikelyComplexAddressComputation tries to figure out if the given stride seems to be 'complex' and need some extra cost for address computation handling. by Mohammed Agabaria · 9 years ago
  71. dab4eae [PM] Change the static object whose address is used to uniquely identify by Chandler Carruth · 9 years ago
  72. 4df1cc0 [ARM] Don't convert switches to lookup tables of pointers with ROPI/RWPI by Oliver Stannard · 9 years ago
  73. 1c38681 Add new target hooks for LoadStoreVectorizer by Volkan Keles · 9 years ago
  74. 7a79422 [LoopStrenghtReduce] Refactoring and addition of a new target cost function. by Jonas Paulsson · 9 years ago
  75. 36e0d01 Consistently use FunctionAnalysisManager by Sean Silva · 9 years ago
  76. e9c32c7 Revert "[X86] Support the "ms-hotpatch" attribute." by Charles Davis · 9 years ago
  77. 0822aa1 [X86] Support the "ms-hotpatch" attribute. by Charles Davis · 9 years ago
  78. 6f937b1 LoadStoreVectorizer: Remove TargetBaseAlign. Keep alignment for stack adjustments. by Alina Sbirlea · 9 years ago
  79. 38c2cd0 This implements a more optimal algorithm for selecting a base constant in by Sjoerd Meijer · 9 years ago
  80. 327955e Add TLI.allowsMisalignedMemoryAccesses to LoadStoreVectorizer by Alina Sbirlea · 9 years ago
  81. 15f3e82 [TTI] Expose TTI::getGEPCost and use it in SLSR and NaryReassociate. by Jingyue Wu · 9 years ago
  82. 164a2aa6 [PM] Remove support for omitting the AnalysisManager argument to new by Chandler Carruth · 9 years ago
  83. 8dad57c TTI: Add hook for memory width to vectorize by Matt Arsenault · 9 years ago
  84. 82de7d3 Apply clang-tidy's misc-move-constructor-init throughout LLVM. by Benjamin Kramer · 9 years ago
  85. e5dfb08 [TTI] Add hook for vector extract with extension by Matthew Simpson · 9 years ago
  86. 8650a4d [TTI] Add getInliningThresholdMultiplier. by Justin Lebar · 9 years ago
  87. 5cb666a [ARM] Adding IEEE-754 SIMD detection to loop vectorizer by Renato Golin · 9 years ago
  88. 0f26b0a [CodeGen] Teach LLVM how to lower @llvm.{min,max}num to {MIN,MAX}NAN by David Majnemer · 9 years ago
  89. 709e304 [LoopDataPrefetch] Add TTI to limit the number of iterations to prefetch ahead by Adam Nemet · 10 years ago
  90. 6d8beec [LoopDataPrefetch/Aarch64] Allow selective prefetching of large-strided accesses by Adam Nemet · 10 years ago
  91. b4faf13 [PM] Implement the final conclusion as to how the analysis IDs should by Chandler Carruth · 10 years ago
  92. df0cd72 [PM] Appease mingw32's auto-import DLL build with minimal tweaks, with fix for clang. by NAKAMURA Takumi · 10 years ago
  93. ca04a1f Revert r262185, "[PM] Appease mingw32's auto-import DLL build with minimal tweaks." by NAKAMURA Takumi · 10 years ago
  94. de40e74 [PM] Appease mingw32's auto-import DLL build with minimal tweaks. by NAKAMURA Takumi · 10 years ago
  95. 3a63435 [PM] Introduce CRTP mixin base classes to help define passes and by Chandler Carruth · 10 years ago
  96. dadfbb5 [TTI] Add getPrefetchDistance from PPCLoopDataPrefetch, NFC by Adam Nemet · 10 years ago
  97. af76110 [TTI] Add getCacheLineSize by Adam Nemet · 10 years ago
  98. 5494698 Implemented cost model for masked gather and scatter operations by Elena Demikhovsky · 10 years ago
  99. 53da2fe Revert r243347 "Add TargetTransformInfo::isZExtFree." by Sanjoy Das · 10 years ago
  100. 0928585 Scalarizer for masked.gather and masked.scatter intrinsics. by Elena Demikhovsky · 10 years ago