Log - 0099beb51fe3f73a90ae5c6fcb3354796dce94e6 - toolchain/llvm-project

aee9771 [ARM/AArch64] Update costs for interleaved accesses with wide types by Matthew Simpson · 9 years ago
2c96c43 [X86] updating TTI costs for arithmetic instructions on X86\SLM arch. by Mohammed Agabaria · 9 years ago
330e1b8 [AArch64] Consider all vector types for FeatureSlowMisaligned128Store by Evandro Menezes · 9 years ago
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
2c8de19 [AArch64] Guard Misaligned 128-bit store penalty by subtarget feature by Matthew Simpson · 9 years ago
651cff4 AArch64: Do not test for CPUs, use SubtargetFeatures by Matthias Braun · 9 years ago
47bd399 Add parentheses to silence buildbot warning by Matthew Simpson · 10 years ago
e5dfb08 [TTI] Add hook for vector extract with extension by Matthew Simpson · 10 years ago
1428d41 [LoopDataPrefetch] Centralize the tuning cl::opts under the pass by Adam Nemet · 10 years ago
709e304 [LoopDataPrefetch] Add TTI to limit the number of iterations to prefetch ahead by Adam Nemet · 10 years ago
6d8beec [LoopDataPrefetch/Aarch64] Allow selective prefetching of large-strided accesses by Adam Nemet · 10 years ago
53e758f [Aarch64] Add pass LoopDataPrefetch for Cyclone by Adam Nemet · 10 years ago
921ad01 [AArch64] Reduce vector insert/extract cost for Kryo by Matthew Simpson · 10 years ago
cd2be7f [AArch64] Add support for Qualcomm Kryo CPU. by Chad Rosier · 10 years ago
97564c3 [AArch64][ARM] Don't base interleaved op legality on type alloc size. by Ahmed Bougacha · 10 years ago
7c6692de [EarlyCSE] IsSimple vs IsVolatile naming clarification (NFC) by Philip Reames · 10 years ago
343af07 [Aarch64] Add cost for missing extensions. by Matthew Simpson · 10 years ago
4b27576 Remove templates from CostTableLookup functions. All instantiations had the same type. by Craig Topper · 10 years ago
ee0c859 Convert cost table lookup functions to return a pointer to the entry or nullptr instead of the index. by Craig Topper · 10 years ago
7bf52c9 Use MVT::SimpleValueType instead of MVT in template parameter. NFC by Craig Topper · 10 years ago
272d6a5 Call the version of ConvertCostTableLookup that takes a statically sized array rather than pointer and size. NFC by Craig Topper · 10 years ago
a3e27ed [CostModel][AArch64] Remove amortization factor for some of the vector select instructions by Silviu Baranga · 10 years ago
b322aa6 [CostModel][AArch64] Increase cost of vector insert element and add missing cast costs by Silviu Baranga · 10 years ago
93205eb [TTI] Make the cost APIs in TargetTransformInfo consistently use 'int' by Chandler Carruth · 10 years ago
7581d22 [ARM/AArch64] Fix cost model for interleaved accesses by Silviu Baranga · 10 years ago
a749f2a Remove getDataLayout() from TargetLowering by Mehdi Amini · 10 years ago
44ede33 Make TargetLowering::getPointerTy() taking DataLayout as an argument by Mehdi Amini · 10 years ago
7ec8ee3 [AArch64] Lower interleaved memory accesses to ldN/stN intrinsics. This patch also adds a function to calculate the cost of interleaved memory accesses. by Hao Liu · 10 years ago
d0ca8d7 [AArch64] Revert r239711 again. We need to discuss how to share code between AArch64 and ARM backend. by Hao Liu · 10 years ago
cb070e3 [AArch64] Match interleaved memory accesses into ldN/stN instructions. by Hao Liu · 10 years ago
65d37e6 This reverts commit r239529 and r239514. by Rafael Espindola · 10 years ago
4566d18 [AArch64] Match interleaved memory accesses into ldN/stN instructions. by Hao Liu · 10 years ago
062c744 [X86] Disable loop unrolling in loop vectorization pass when VF is 1. by Wei Mi · 10 years ago
aef6841 [AArch64] Enable partial & runtime unrolling on cortex-a57 by Kevin Qin · 11 years ago
7149aab Make some non-constant static variables non-static or fully const. by Benjamin Kramer · 11 years ago
ab5cb36 [multiversion] Remove the function parameter from the unrolling by Chandler Carruth · 11 years ago
93dcdc4 [PM] Switch the TargetMachine interface from accepting a pass manager by Chandler Carruth · 11 years ago
705b185 [PM] Change the core design of the TTI analysis to use a polymorphic by Chandler Carruth · 11 years ago
f9327d6 Commoning of target specific load/store intrinsics in Early CSE. by Chad Rosier · 11 years ago
72a799a [AArch64] Enable partial & runtime unrolling on cortex-a57. by Kevin Qin · 11 years ago
70d54ac [AArch64] Improve cost model to handle sdiv by a pow-of-two. by Chad Rosier · 11 years ago
7b0abb8 [AArch64] Revert r216141 for cyclone by Gerolf Hoflehner · 11 years ago
b653de1 Rename getMaximumUnrollFactor -> getMaxInterleaveFactor; also rename option names controlling this variable. by Sanjay Patel · 11 years ago
7f33ff7 Allow vectorization of division by uniform power of 2. by Karthik Bhat · 11 years ago
a88896b [LoopVectorize] Up the maximum unroll factor to 4 for AArch64 by James Molloy · 11 years ago
2b8933c Teach the SLP Vectorizer that keeping some values live over a callsite can have a cost. by James Molloy · 11 years ago
d913448 Remove the TargetMachine forwards for TargetSubtargetInfo based by Eric Christopher · 11 years ago
dbecc3b AArch64: improve handling & modelling of FP_TO_XINT nodes. by Tim Northover · 11 years ago
ef0d760 AArch64: improve vector [su]itofp handling. by Tim Northover · 11 years ago
3b0846e AArch64/ARM64: move ARM64 into AArch64's place by Tim Northover · 11 years ago[Renamed (86%) from llvm/lib/Target/ARM64/ARM64TargetTransformInfo.cpp]
89f1880 Fix typo. by Eric Christopher · 11 years ago
062a2ba [C++] Use 'nullptr'. Target edition. by Craig Topper · 12 years ago
a4ee178 [ARM64] Enable feature predicates for NEON / FP / CRYPTO. by Kevin Qin · 12 years ago
84e68b2 [Modules] Fix potential ODR violations by sinking the DEBUG_TYPE by Chandler Carruth · 12 years ago
e302883 AArch64/ARM64: add non-scalar lowering for more FCVT operations. by Tim Northover · 12 years ago
cf03068 [ARM64] Never hoist the shift value of a shift instruction. by Juergen Ributzka · 12 years ago
6e17aa4 [ARM64] Fix the cost model for cheap large constants. by Juergen Ributzka · 12 years ago
c5e0099 ARM64: add an explicit cast to silence a silly warning by Saleem Abdulrasool · 12 years ago
48c8c07d [ARM64] Fix immediate cost calculation for types larger than i64. by Juergen Ributzka · 12 years ago
c11e8b6 [Constant Hoisting][ARM64] Enable constant hoisting for ARM64. by Juergen Ributzka · 12 years ago
00ed996 ARM64: initial backend import by Tim Northover · 12 years ago