summaryrefslogtreecommitdiff
path: root/llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp
AgeCommit message (Expand)Author
2020-02-11AMDGPU: Don't expand more special div cases in IRMatt Arsenault
2020-02-11AMDGPU: Directly use rcp intrinsic in idiv expansionsMatt Arsenault
2020-02-11AMDGPU: Don't create potentially dead rcp declarationsMatt Arsenault
2020-02-07AMDGPU: Enhancement on FDIV lowering in AMDGPUCodeGenPrepareChangpeng Fang
2020-01-23AMDGPU: Implement FDIV optimizations in AMDGPUCodeGenPrepareChangpeng Fang
2020-01-23AMDGPU: Check for other uses when looking through casted selectMatt Arsenault
2020-01-22AMDGPU: Fix typoMatt Arsenault
2020-01-22AMDGPU: Look through casted selects to constant fold bin opsMatt Arsenault
2020-01-22AMDGPU: Do binop of select of constant fold in AMDGPUCodeGenPrepareMatt Arsenault
2020-01-20[AMDGPU] Fix -Wunused-variable after e5823bf806ca9fa6f87583065b3898a2edabce57Fangrui Song
2020-01-20AMDGPU: Don't create weird sized integersMatt Arsenault
2019-11-19AMDGPU: Refactor treatment of denormal modeMatt Arsenault
2019-11-13Sink all InitializePasses.h includesReid Kleckner
2019-08-24AMDGPU: Preserve value name when inserting mul24 intrinsicMatt Arsenault
2019-08-24AMDGPU: Introduce a flag to disable mul24 intrinsic formationMatt Arsenault
2019-07-15AMDGPU: Add 24-bit mul intrinsicsMatt Arsenault
2019-02-01[opaque pointer types] Pass value type to LoadInst creation.James Y Knight
2019-01-19Update the file headers across all of the LLVM projects in the monorepoChandler Carruth
2018-10-08[IRBuilder] Fixup CreateIntrinsic to allow specifying Types to Mangle.Neil Henning
2018-08-31AMDGPU: Remove remnants of old address space mappingMatt Arsenault
2018-08-30[NFC] Rename the DivergenceAnalysis to LegacyDivergenceAnalysisNicolai Haehnle
2018-07-25[AMDGPU] Use AssumptionCacheTracker in the divrem32 expansionStanislav Mekhanoshin
2018-07-11AMDGPU: Refactor Subtarget classesTom Stellard
2018-06-28[AMDGPU] Early expansion of 32 bit udiv/uremStanislav Mekhanoshin
2018-06-28AMDGPU: Fix AMDGPUCodeGenPrepare using uninitialized AMDGPUAS structMatt Arsenault
2018-06-07AMDGPU: Try a lot harder to emit scalar loadsMatt Arsenault
2018-06-06[AMDGPU] Improve reciprocal handlingStanislav Mekhanoshin
2018-06-05AMDGPU: Preserve metadata when widening loadsMatt Arsenault
2018-05-01Remove \brief commands from doxygen comments.Adrian Prantl
2018-02-09Reapply "AMDGPU: Add 32-bit constant address space"Matt Arsenault
2018-02-07Revert "AMDGPU: Add 32-bit constant address space"Rafael Espindola
2018-02-07AMDGPU: Add 32-bit constant address spaceMarek Olsak
2017-11-06[IR] redefine 'UnsafeAlgebra' / 'reassoc' fast-math-flags and add 'trans' fas...Sanjay Patel
2017-07-26AMDGPU : Widen extending scalar loads to 32-bits.Wei Ding
2017-07-06[AMDGPU] Always use rcp + mul with fast mathStanislav Mekhanoshin
2017-06-06Sort the remaining #include lines in include/... and lib/....Chandler Carruth
2017-05-18[LegacyPassManager] Remove TargetMachine constructorsFrancis Visoiu Mistrih
2017-03-17AMDGPU: Cleanup control flow intrinsicsMatt Arsenault
2017-02-27AMDGPU: Support v2i16/v2f16 packed operationsMatt Arsenault
2017-02-01AMDGPU: Improve nsw/nuw/exact when promoting uniform i16 opsMatt Arsenault
2017-01-20[AMDGPU] Fix some Clang-tidy modernize and Include What You Use warnings; oth...Eugene Zelenko
2016-12-22AMDGPU: Allow rcp and rsq usage with f16Matt Arsenault
2016-12-06AMDGPU: Fix crash on i16 constant expressionMatt Arsenault
2016-10-07[AMDGPU] AMDGPUCodeGenPrepare: remove extra ';'Konstantin Zhuravlyov
2016-10-07[AMDGPU] Promote uniform (i1, i16] operations to i32Konstantin Zhuravlyov
2016-10-06[AMDGPU] Promote uniform i16 bitreverse intrinsic to i32Konstantin Zhuravlyov
2016-10-03[AMDGPU] Sign extend AShr when promoting (instead of zero extending)Konstantin Zhuravlyov
2016-10-01Use StringRef in Pass/PassManager APIs (NFC)Mehdi Amini
2016-09-28[AMDGPU] Promote uniform i16 ops to i32 ops for targets that have 16 bit inst...Konstantin Zhuravlyov
2016-07-26AMDGPU: Use rcp for fdiv 1, x with fpmath metadataMatt Arsenault
generated by cgit v1.2.3 (git 2.25.1) at 2026-10-26 05:07:10 +0000