Gitiles
Code Review
Sign In
gerrit-public.fairphone.software
/
toolchain
/
llvm-project
/
81fe1fbf4abd00b24e078a80cb15a2d2c1343f11
/
llvm
/
lib
/
CodeGen
/
AtomicExpandPass.cpp
66d9a75
[RISCV] Implement codegen for cmpxchg on RV32IA
by Alex Bradbury
· 7 years ago
79518b0
[AtomicExpandPass]: Add a hook for custom cmpxchg expansion in IR
by Alex Bradbury
· 7 years ago
21aea51
[RISCV] Codegen for i8, i16, and i32 atomicrmw with RV32A
by Alex Bradbury
· 7 years ago
3291f9a
[AtomicExpandPass] Widen partword atomicrmw or/xor/and before tryExpandAtomicRMW
by Alex Bradbury
· 7 years ago
f78650a
Remove trailing space
by Fangrui Song
· 7 years ago
cb0bab8
[CodeGen] Fix inconsistent declaration parameter name
by Fangrui Song
· 7 years ago
d34e60c
Rename DEBUG macro to LLVM_DEBUG.
by Nicola Zaghen
· 7 years ago
2fa1436
[IR][CodeGen] Remove dependency on EVT from IR/Function.cpp. Move EVT to CodeGen layer.
by Craig Topper
· 7 years ago
36a0f22
Fix layering by moving ValueTypes.h from CodeGen to IR
by David Blaikie
· 8 years ago
b3bde2e
Fix a bunch more layering of CodeGen headers that are in Target
by David Blaikie
· 8 years ago
f193332
[CodeGen] Fix some Clang-tidy modernize-use-default-member-init and Include What You Use warnings; other minor fixes (NFC).
by Eugene Zelenko
· 8 years ago
674d2c2
[Instruction] add moveAfter() convenience function; NFCI
by Sanjay Patel
· 8 years ago
bb80d3e
Enhance synchscope representation
by Konstantin Zhuravlyov
· 8 years ago
1527baa
CodeGen: Rename DEBUG_TYPE to match passnames
by Matthias Braun
· 8 years ago
8b61764
[LegacyPassManager] Remove TargetMachine constructors
by Francis Visoiu Mistrih
· 8 years ago
04de70d
[Atomic] Remove IsStore/IsLoad in the interface, and pass the instruction instead. NFC.
by Tim Shen
· 8 years ago
b518054
Rename AttributeSet to AttributeList
by Reid Kleckner
· 9 years ago
148a646
Support expanding partial-word cmpxchg to full-word cmpxchg in AtomicExpandPass.
by James Y Knight
· 9 years ago
d3f4c05
Move instances of std::function.
by Benjamin Kramer
· 9 years ago
be11bdc
Rename getLargestLegalIntTypeSize to getLargestLegalIntTypeSizeInBits(). NFC.
by Jun Bum Lim
· 9 years ago
bbb0aee6
NFC: unify clang / LLVM atomic ordering
by JF Bastien
· 9 years ago
fb9871b
Revert "NFC: unify clang / LLVM atomic ordering"
by JF Bastien
· 9 years ago
6ef3aa2
NFC: unify clang / LLVM atomic ordering
by JF Bastien
· 9 years ago
19f6cce
Add __atomic_* lowering to AtomicExpandPass.
by James Y Knight
· 9 years ago
d41b54b
This reverts commit r266002, r266011 and r266016.
by Rafael Espindola
· 9 years ago
b3ac75f
AtomicExpandPass: mark assert variable as used
by JF Bastien
· 9 years ago
00db547
Fix compile with GCC after r266002 (Add __atomic_* lowering to AtomicExpandPass)
by James Y Knight
· 9 years ago
b91d38c
Add __atomic_* lowering to AtomicExpandPass.
by James Y Knight
· 9 years ago
800f87a
NFC: make AtomicOrdering an enum class
by JF Bastien
· 9 years ago
01f2ca5
NFC: skip FenceInst up-front in AtomicExpandPass.
by James Y Knight
· 10 years ago
f44fc52
Tweak some atomics functions in preparation for larger changes; NFC.
by James Y Knight
· 10 years ago
d32f8e6
ARM: sink atomic release barrier as far as possible into cmpxchg.
by Tim Northover
· 10 years ago
1960cfd
[IR] Extend cmpxchg to allow pointer type operands
by Philip Reames
· 10 years ago
367fdd9
Restrict scope of variables [NFC]
by Philip Reames
· 10 years ago
2331901
Speculative fix for windows build
by Philip Reames
· 10 years ago
61a24ab
[IR] Add support for floating pointer atomic loads and stores
by Philip Reames
· 10 years ago
f520eff
AArch64: use ldxp/stxp pair to implement 128-bit atomic loads.
by Tim Northover
· 10 years ago
8f11e1a
CodeGen: Start removing implicit conversions to/from list iterators, NFC
by Duncan P. N. Exon Smith
· 10 years ago
07a844d
[AArch64] Emit clrex in the expanded cmpxchg fail block.
by Ahmed Bougacha
· 10 years ago
49b531a
[CodeGen] Fix AtomicExpand invalidation issue caused by r247429.
by Ahmed Bougacha
· 10 years ago
5246867
[CodeGen] Refactor TLI/AtomicExpand interface to make LLSC explicit.
by Ahmed Bougacha
· 10 years ago
9d67713
[CodeGen] Rename AtomicRMWExpansionKind to AtomicExpansionKind.
by Ahmed Bougacha
· 10 years ago
bd753c9
Fix an alignment error in `llvm::expandAtomicRMWToCmpXchg` without breaking the build where X86 isn't enabled.
by Richard Diamond
· 10 years ago
a02ac60
Revert "Divide the primitive size in bits by eight so the initial load's alignment is in bytes as expected. Tested with the included unit test."
by Renato Golin
· 10 years ago
559c1d7
Divide the primitive size in bits by eight so the initial load's alignment is in
by Richard Diamond
· 10 years ago
7ef9456
Write access test.
by Richard Diamond
· 10 years ago
e8aad29
Refactor AtomicExpand::expandAtomicRMWToCmpXchg into a standalone function.
by JF Bastien
· 10 years ago
f00654e
Revert r240137 (Fixed/added namespace ending comments using clang-tidy. NFC)
by Alexander Kornienko
· 10 years ago
70bc5f1
Fixed/added namespace ending comments using clang-tidy. NFC
by Alexander Kornienko
· 10 years ago
572e03a
Fix "the the" in comments.
by Eric Christopher
· 10 years ago
f14889e
Mutate TargetLowering::shouldExpandAtomicRMWInIR to specifically dictate how AtomicRMWInsts are expanded.
by JF Bastien
· 11 years ago
4e048eeb
Migrate AtomicExpandPass and DwarfEHPrepare to using a Function-ized getSubtargetImpl.
by Eric Christopher
· 11 years ago
b11a1b7
Cache the lookup of TargetLowering in the atomic expand pass.
by Eric Christopher
· 11 years ago
810739d
Lower idempotent RMWs to fence+load
by Robin Morisset
· 11 years ago
6dbbbc2
[X86] Make wide loads be managed by AtomicExpand
by Robin Morisset
· 11 years ago
dedef33
Add AtomicExpandPass::bracketInstWithFences, and use it whenever getInsertFencesForAtomic would trigger in SelectionDAGBuilder
by Robin Morisset
· 11 years ago
25c8e31
[X86] Use the generic AtomicExpandPass instead of X86AtomicExpandPass
by Robin Morisset
· 11 years ago
ed3d48f
Refactor AtomicExpandPass and add a generic isAtomic() method to Instruction
by Robin Morisset
· 11 years ago
a47cb41
Use target-dependent emitLeading/TrailingFence instead of the target-independent insertLeading/TrailingFence (in AtomicExpandPass)
by Robin Morisset
· 11 years ago
59c23cd
Rename AtomicExpandLoadLinked into AtomicExpand
by Robin Morisset
· 11 years ago
[Renamed (91%) from llvm/lib/CodeGen/AtomicExpandLoadLinkedPass.cpp]
d18cda6
Fix typos in comments
by Robin Morisset
· 11 years ago
d913448
Remove the TargetMachine forwards for TargetSubtargetInfo based
by Eric Christopher
· 11 years ago
55beb64
CodeGen: it turns out that NAND is not the same thing as BIC. At all.
by Tim Northover
· 11 years ago
c40e5ed
Add a new subtarget hook for whether or not we'd like to enable
by Eric Christopher
· 11 years ago
20b9f73
Atomics: make use of the "cmpxchg weak" instruction.
by Tim Northover
· 11 years ago
d039abd
Atomics: switch direction of cmpxchg comparison
by Tim Northover
· 11 years ago
420a216
IR: add "cmpxchg weak" variant to support permitted failure.
by Tim Northover
· 11 years ago
d623c52
Create macro INITIALIZE_TM_PASS.
by Jiangning Liu
· 11 years ago
66f09ad
[C++11] Use 'nullptr'.
by Craig Topper
· 11 years ago
b4ddc08
ARM & AArch64: make use of common cmpxchg idioms after expansion
by Tim Northover
· 11 years ago
353eda4
[C++] Use 'nullptr'.
by Craig Topper
· 11 years ago
1b9dde0
[Modules] Remove potential ODR violations by sinking the DEBUG_TYPE
by Chandler Carruth
· 11 years ago
037f26f2
Atomics: promote ARM's IR-based atomics pass to CodeGen.
by Tim Northover
· 11 years ago
[Renamed (63%) from llvm/lib/Target/ARM/ARMAtomicExpandPass.cpp]
70450c5
ARM: skip cmpxchg failure barrier if ordering is monotonic.
by Tim Northover
· 11 years ago
c882eb0
ARM: expand atomic ldrex/strex loops in IR
by Tim Northover
· 11 years ago