Alex MacLean fb33af08e4 [NVPTX] Remove nvvm.ldg.global.* intrinsics (#112834) ...

Remove these intrinsics which can be better represented by load
instructions with `!invariant.load` metadata:

- llvm.nvvm.ldg.global.i
- llvm.nvvm.ldg.global.f
- llvm.nvvm.ldg.global.p

2024-10-27 16:14:13 -07:00

..

MCTargetDesc

[NVPTX] mangle symbols in debug info to conform to PTX restrictions. (#113216)

2024-10-22 13:01:40 -07:00

TargetInfo

…

cl_common_defines.h

…

CMakeLists.txt

…

NVPTX.h

[NVPTX] add an optional early copy of byval arguments (#113384)

2024-10-24 12:00:56 -07:00

NVPTX.td

[CUDA] Add support for CUDA-12.6 and sm_100 (#112028)

2024-10-14 11:51:05 -07:00

NVPTXAliasAnalysis.cpp

[NVPTX][AA] Traverse use-def chain to find non-generic addrspace (#106477)

2024-08-29 21:26:58 -07:00

NVPTXAliasAnalysis.h

…

NVPTXAllocaHoisting.cpp

…

NVPTXAllocaHoisting.h

…

NVPTXAsmPrinter.cpp

[NVPTX] Avoid repeated map lookups (NFC) (#111627)

2024-10-09 06:44:58 -07:00

NVPTXAsmPrinter.h

[NVPTX] add support for encoding PTX registers for DWARF (#109495)

2024-09-26 12:32:43 -06:00

NVPTXAssignValidGlobalNames.cpp

[NVPTX] mangle symbols in debug info to conform to PTX restrictions. (#113216)

2024-10-22 13:01:40 -07:00

NVPTXAtomicLower.cpp

…

NVPTXAtomicLower.h

…

NVPTXCtorDtorLowering.cpp

…

NVPTXCtorDtorLowering.h

…

NVPTXFrameLowering.cpp

[llvm] Consistently respect naked fn attribute in TargetFrameLowering::hasFP() (#106014)

2024-10-18 09:35:42 +04:00

NVPTXFrameLowering.h

[llvm] Consistently respect naked fn attribute in TargetFrameLowering::hasFP() (#106014)

2024-10-18 09:35:42 +04:00

NVPTXGenericToNVVM.cpp

…

NVPTXImageOptimizer.cpp

…

NVPTXInstrFormats.td

…

NVPTXInstrInfo.cpp

[TII][RISCV] Add renamable bit to copyPhysReg (#91179)

2024-08-27 10:08:43 +08:00

NVPTXInstrInfo.h

[TII][RISCV] Add renamable bit to copyPhysReg (#91179)

2024-08-27 10:08:43 +08:00

NVPTXInstrInfo.td

[NVPTX] Add support for clamped funnel shift intrinsics (#113228)

2024-10-22 16:39:44 -07:00

NVPTXIntrinsics.td

Reland "[NVVM] Upgrade nvvm.ptr.* intrinics to addrspace cast" (#110262)

2024-09-28 14:13:17 -07:00

NVPTXISelDAGToDAG.cpp

[NVPTX] Remove nvvm.ldg.global.* intrinsics (#112834)

2024-10-27 16:14:13 -07:00

NVPTXISelDAGToDAG.h

[NVPTX] Load/Store/Fence syncscope support (#106101)

2024-09-23 10:18:00 -07:00

NVPTXISelLowering.cpp

[NVPTX] Remove nvvm.ldg.global.* intrinsics (#112834)

2024-10-27 16:14:13 -07:00

NVPTXISelLowering.h

[NVPTX] Remove nvvm.ldg.global.* intrinsics (#112834)

2024-10-27 16:14:13 -07:00

NVPTXLowerAggrCopies.cpp

…

NVPTXLowerAggrCopies.h

…

NVPTXLowerAlloca.cpp

[NVPTX] Check Before inserting AddrSpaceCastInst in NVPTXLoweringAlloca (#106127)

2024-09-12 17:58:40 -04:00

NVPTXLowerArgs.cpp

[NVPTX] add an optional early copy of byval arguments (#113384)

2024-10-24 12:00:56 -07:00

NVPTXLowerUnreachable.cpp

[NVPTX] Fix NVPTXLowerUnreachable::isLoweredToTrap logic (#109730)

2024-09-24 18:44:59 -07:00

NVPTXMachineFunctionInfo.h

…

NVPTXMCExpr.cpp

…

NVPTXMCExpr.h

…

NVPTXPassRegistry.def

[NVPTX] add an optional early copy of byval arguments (#113384)

2024-10-24 12:00:56 -07:00

NVPTXPeephole.cpp

…

NVPTXPrologEpilogPass.cpp

…

NVPTXProxyRegErasure.cpp

[llvm][NVPTX] Fix RAUW bug in NVPTXProxyRegErasure (#105871)

2024-08-24 13:19:44 -04:00

NVPTXRegisterInfo.cpp

[NVPTX] fix debug register encoding of special %Depot register (#111596)

2024-10-08 19:39:16 -06:00

NVPTXRegisterInfo.h

[NVPTX] add support for encoding PTX registers for DWARF (#109495)

2024-09-26 12:32:43 -06:00

NVPTXRegisterInfo.td

…

NVPTXReplaceImageHandles.cpp

[NVPTX] Load/Store/Fence syncscope support (#106101)

2024-09-23 10:18:00 -07:00

NVPTXSubtarget.cpp

[NVPTX] Load/Store/Fence syncscope support (#106101)

2024-09-23 10:18:00 -07:00

NVPTXSubtarget.h

[NVPTX] Only run LowerUnreachable when necessary (#109868)

2024-10-10 12:57:43 -07:00

NVPTXTargetMachine.cpp

[NVPTX] add an optional early copy of byval arguments (#113384)

2024-10-24 12:00:56 -07:00

NVPTXTargetMachine.h

…

NVPTXTargetObjectFile.h

…

NVPTXTargetTransformInfo.cpp

[NVPTX] Add support for clamped funnel shift intrinsics (#113228)

2024-10-22 16:39:44 -07:00

NVPTXTargetTransformInfo.h

[LLVM] Use {} instead of std::nullopt to initialize empty ArrayRef (#109133)

2024-09-19 16:16:38 +01:00

NVPTXUtilities.cpp

[NVPTX] mangle symbols in debug info to conform to PTX restrictions. (#113216)

2024-10-22 13:01:40 -07:00

NVPTXUtilities.h

[NVPTX] mangle symbols in debug info to conform to PTX restrictions. (#113216)

2024-10-22 13:01:40 -07:00

NVVMIntrRange.cpp

…

NVVMReflect.cpp

…