Gitiles
Code Review
Sign In
review.blissroms.org
/
platform_external_llvm
/
ab494c93854b8593bc4c6a9212e4bd1ea826e8ae
ab494c9
[MC] Function stack size section.
by Sean Eveson
· 8 years ago
360f53a
Revert r319423: [MC] Function stack size section.
by Sean Eveson
· 8 years ago
11acbde
[ARM GlobalISel] Bail out for byval
by Diana Picus
· 8 years ago
7384652
[CodeGen] Print "%vreg0" as "%0" in both MIR and debug output
by Francis Visoiu Mistrih
· 8 years ago
996d6dd
[X86][AVX512] Tag binop/rounding/sae instructions scheduler classes
by Simon Pilgrim
· 8 years ago
5e536b7
[MC] Function stack size section.
by Sean Eveson
· 8 years ago
a90c32e
[DAGCombine] Refactor ReduceLoadWidth
by Sam Parker
· 8 years ago
86e196b7
Support generic lowering of vector bswap
by Serge Guelton
· 8 years ago
b785e26
[X86][AVX512] Tag RCP/RSQRT/GETEXP instructions scheduler classes
by Simon Pilgrim
· 8 years ago
6bb5dbc
[dsymutil] Exclude namespace from ifdef in CFBundle
by Jonas Devlieghere
· 8 years ago
bdc984c
[dsymutil] Upstream getBundleInfo implementation
by Jonas Devlieghere
· 8 years ago
50f1f94
Revert rL319407: [SROA] enable splitting for non-whole-alloca loads and stores
by Hiroshi Inoue
· 8 years ago
431a76f
[SystemZ] Bugfix in adjustSubwordCmp.
by Jonas Paulsson
· 8 years ago
39be023
[SROA] enable splitting for non-whole-alloca loads and stores
by Hiroshi Inoue
· 8 years ago
f6395de
[X86] Optimize avx2 vgatherqps for v2f32 with v2i64 index type.
by Craig Topper
· 8 years ago
d60f6e4
[X86] Make sure we don't remove sign extends of masks with AVX2 masked gathers.
by Craig Topper
· 8 years ago
fe376dc
[XRay][docs] Update documentation on new default for xray_naive_log=
by Dean Michael Berris
· 8 years ago
9d522dc
- Removed unused lamba (IsReturnBlock) causing build bots to fail for r319398
by Graham Yiu
· 8 years ago
106d893
With PGO information, we can do more aggressive outlining of cold regions in the inline candidate function. This contrasts with the scheme of keeping only the 'early return' portion of the inline candidate and outlining the rest of the function as a single function call.
by Graham Yiu
· 8 years ago
eb4cb7a
[libFuzzer] mention one more trophie in the Linux Kernel
by Kostya Serebryany
· 8 years ago
1de313a
AMDGPU: Allow negative MUBUF vaddr for gfx9
by Matt Arsenault
· 8 years ago
fb10287
Check alignment in getSectionContentsAsArray.
by Rafael Espindola
· 8 years ago
310bfcc
[Coverage] Use the most-recent completed region count (PR35437)
by Vedant Kumar
· 8 years ago
b45fa1a
LowerTypeTests: Deduplicate code. NFC.
by Peter Collingbourne
· 8 years ago
1edbc9f
LowerTypeTests: Remove unnecessary cast. NFC.
by Peter Collingbourne
· 8 years ago
0a78cd8
[X86] Remove some questionable looking code that seems to be looking through a VZEXT to create a larger VSEXT.
by Craig Topper
· 8 years ago
4ee291a
First step towards more human-friendly PPC assembler output:
by Joerg Sonnenberger
· 8 years ago
066e89a
[WebAssembly] Update test expectations for gcc torture tests
by Sam Clegg
· 8 years ago
7448686
[CodeView] Factor some code out of TypeTableBuilder.
by Zachary Turner
· 8 years ago
f1a319a
[SelectionDAG][X86] Teach promotion legalization for fp_to_sint/fp_to_uint to insert an assertsext/assertzext based on the original type
by Craig Topper
· 8 years ago
685c102
[WebAssembly] Fix fptoui lowering bounds
by Dan Gohman
· 8 years ago
9bb5163
Add libstd++-4.8 exceptions to ubsan_blacklist.txt
by Sam Clegg
· 8 years ago
433d310
[Hexagon] Remove HexagonISD::PACKHL
by Krzysztof Parzyszek
· 8 years ago
743fb60
[Hexagon] Create helpers extractVector and insertVector in lowering
by Krzysztof Parzyszek
· 8 years ago
5c698e3
[X86][AVX512] Tag RCP/RSQRT/GETEXP instructions scheduler classes (REVERSION)
by Simon Pilgrim
· 8 years ago
7fe0c00
Make TypeTableBuilder inherit from TypeCollection.
by Zachary Turner
· 8 years ago
4c20574
Fix line endings in llvm-pdbutil.cpp
by Zachary Turner
· 8 years ago
ae007cec
[X86][AVX512] Tag RCP/RSQRT/GETEXP instructions scheduler classes
by Simon Pilgrim
· 8 years ago
02c5c78
[X86][AVX512] Tag 3OP (shuffles, double-shifts and GFNI) instructions scheduler classes
by Simon Pilgrim
· 8 years ago
1782012
[ARM][DAG] Revert Disable post-legalization store merge for ARM
by Nirav Dave
· 8 years ago
3a42cac
[cmake] Replace -Wall with /W4 in clang-cl options now that -Wall aliases -Weverything
by Greg Bedwell
· 8 years ago
e081e35
Make check-lit tests respect LLVM_LIT_TOOLS_DIR
by Greg Bedwell
· 8 years ago
1a60893
[Power9] add more tests for D38287; NFC
by Zaara Syeda
· 8 years ago
0221a14
[InstCombine] add tests for select-of-constants; NFC
by Sanjay Patel
· 8 years ago
1c969b0
[X86][AVX512] Add itinerary argument to all AVX512_maskable_* wrappers. NFCI
by Simon Pilgrim
· 8 years ago
d3ee065
Add opt-viewer testing
by Adam Nemet
· 8 years ago
ff77244
Reverted r319315 because of unused functions (due to PPR not yet being
by Sander de Smalen
· 8 years ago
bc89401
[X86][AVX512] Tag VPERMILV instruction scheduler class
by Simon Pilgrim
· 8 years ago
a150a5a
[AArch64][SVE] Asm: Add SVE predicate register definitions and parsing support
by Sander de Smalen
· 8 years ago
c7436ce
[ARM GlobalISel] Fix selecting G_BRCOND
by Diana Picus
· 8 years ago
abf3849
[X86][AVX512] Setup unary (PABS/VPLZCNT/VPOPCNT/VPCONFLICT/VMOV*DUP) instruction scheduler classes
by Simon Pilgrim
· 8 years ago
e211444
[AMDGPU][MC][GFX9] Corrected mapping of GFX9 v_add/sub/subrev_u32
by Dmitry Preobrazhensky
· 8 years ago
7dfadbc
[X86][SSE] Merged sse2_unpack and sse2_unpack PUNPCK instruction templates. NFCI.
by Simon Pilgrim
· 8 years ago
7bef62d
[X86][SSE] Merged sse2_pack and sse2_pack_y PACKSS/PACKUS instruction templates. NFCI.
by Simon Pilgrim
· 8 years ago
dd9c0f5
[SCEV][NFC] Break from loop after we found first non-Phi in getAddRecExprPHILiterally
by Max Kazantsev
· 8 years ago
3aae870
[ARM] Add support for armv7e-m to the .arch directive
by Oliver Stannard
· 8 years ago
ef36801
[CGP] Enable complex addr mode
by Serguei Katkov
· 8 years ago
3bee26a
Comment fix in SelectionDAG.h
by Jonas Paulsson
· 8 years ago
acbaefa
[X86] Remove setOperationAction Promote for ISD::SINT_TO_FP MVT::v8i16/v16i8/v16i16.
by Craig Topper
· 8 years ago
bee1015
[SCEV][NFC] Remove condition that can never happen due to check few lines above
by Max Kazantsev
· 8 years ago
c2659f1
[CGP] Fix common type handling in optimizeMemoryInst
by Serguei Katkov
· 8 years ago
98e4604
[PowerPC] Relax the checking on AND/AND8 in isSignOrZeroExtended.
by Sean Fertile
· 8 years ago
c7ecbb2
AMDGPU: Use stricter regexes for add instructions
by Matt Arsenault
· 8 years ago
542c83c
[Modules] Add textual headers for recently added .def files
by Bruno Cardoso Lopes
· 8 years ago
ff838de
DAG: Add nuw when splitting loads and stores
by Matt Arsenault
· 8 years ago
c97df23
llvm-dwarfdump: honor the --show-children option when dumping a specific DIE.
by Adrian Prantl
· 8 years ago
9516b8f
AMDGPU: Select DS insts without m0 initialization
by Matt Arsenault
· 8 years ago
0a7ff1a
Rollback r319176.
by Don Hinton
· 8 years ago
3c8bef9
[X86] Promote fp_to_sint v16f32->v16i16/v16i8 to avoid scalarization.
by Craig Topper
· 8 years ago
f125f8f
Fix a warning.
by Zachary Turner
· 8 years ago
4418c09
Revert "Add opt-viewer testing"
by Adam Nemet
· 8 years ago
115c91a
[X86] Add test cases for fptosi v16f32->v16i8/v16i16 to show scalarization.
by Craig Topper
· 8 years ago
dc4f5ff
[NFC] Minor cleanups in CodeView TypeTableBuilder.
by Zachary Turner
· 8 years ago
2677b3d
[X86] Mark ISD::FP_TO_UINT v16i8/v16i16 as Promote under AVX512 instead of legal. Fix infinite loop in op legalization when promotion requires 2 steps.
by Craig Topper
· 8 years ago
00e3d22
[X86] Regenerate avx512-schedule test.
by Craig Topper
· 8 years ago
932b477
AMDGPU: Enable IPRA
by Matt Arsenault
· 8 years ago
c32b942
[X86] Tag CLFLUSHOPT with same scheduling behaviour as CLFLUSH
by Simon Pilgrim
· 8 years ago
8d8e7d4
[globalisel][tablegen] Fix PR35375 by sign-extending the table value to match getConstantVRegVal()
by Daniel Sanders
· 8 years ago
5bf4f12
[X86] Add CLFLUSHOPT schedule tests
by Simon Pilgrim
· 8 years ago
3eb9d89
[X86][SSE] Add SSE_SHUFP OpndItins
by Simon Pilgrim
· 8 years ago
56f4da8
[X86] Test clflushopt intrinsic on 32 and 64-bit targets
by Simon Pilgrim
· 8 years ago
d51777d
[X86][SSE] Add SSE_UNPCK/SSE_PUNPCK OpndItins
by Simon Pilgrim
· 8 years ago
1ee3007
[X86][SSE] Use SSE_PACK OpndItins in PACKSS/PACKUS instruction definitions
by Simon Pilgrim
· 8 years ago
999c141
Remove this test
by Adam Nemet
· 8 years ago
a5d8c5b
Fix VS2017 narrowing conversion warning. NFCI
by Simon Pilgrim
· 8 years ago
3f37b38
[X86] Remove unused variable.
by Craig Topper
· 8 years ago
1f98469
Demote this opt remark to DEBUG.
by Adam Nemet
· 8 years ago
45174c1
[X86] Remove code from combineUIntToFP that tried to favor UINT_TO_FP if legal when zero extending from vXi8/vX816.
by Craig Topper
· 8 years ago
0bec7f3
[X86] Remove custom lowering for uint_to_fp from vXi8/vXi16.
by Craig Topper
· 8 years ago
7f3e660
[globalisel][tablegen] Add support for importing G_ATOMIC_CMPXCHG, G_ATOMICRMW_* rules from SelectionDAG.
by Daniel Sanders
· 8 years ago
ee2e970
SROA: Don't create variable fragments that are outside of the variable.
by Adrian Prantl
· 8 years ago
d3fc21c
[cmake] Pass LLVM_USE_LINKER flag when building host tools, e.g.,
by Don Hinton
· 8 years ago
4ff6226
[SLP] Additional test for PR35354, NFC.
by Alexey Bataev
· 8 years ago
7affb6d
[Hexagon] Use stable sort for HexagonShuffler to remove non-deterministic ordering
by Mandeep Singh Grang
· 8 years ago
02dd787
[aarch64][globalisel] Add missing tests from r319216
by Daniel Sanders
· 8 years ago
336e099
[PowerPC] Allow tail calls of fastcc functions from C CallingConv functions.
by Sean Fertile
· 8 years ago
722a86f
[aarch64][globalisel] Define G_ATOMIC_CMPXCHG and G_ATOMICRMW_* and make them legal
by Daniel Sanders
· 8 years ago
c0c70f9
[SelectionDAG] Make sorting predicate stronger to remove non-deterministic ordering
by Mandeep Singh Grang
· 8 years ago
219ddf4
[X86][SSE] Add SSE_HADDSUB/SSE_PABS/SSE_PALIGN OpndItins
by Simon Pilgrim
· 8 years ago
5b7a591
[X86] In lowerVectorShuffleAsElementInsertion, if were able to find a scalar i8 or i16 and need to zero extend it, make sure we use a vXi32 type of the full vector width.
by Craig Topper
· 8 years ago
Next »