Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Implement grouped conv interface #80870

Closed
wants to merge 8,927 commits into from
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
8927 commits
Select commit Hold shift + click to select a range
abd0bf1
Test commit
dsandersllvm Jun 6, 2024
bf572a5
Revert "Test commit"
dsandersllvm Jun 6, 2024
28e57cd
[clang-format] Don't format comments in SkipMacroDefinitionBody (#94425)
owenca Jun 6, 2024
ea63530
[RuntimeDyld][ELF] Fix unwanted sign extension. (#94482)
al45tair Jun 6, 2024
9fc29d3
[LoongArch] Add a hook to sign extend i32 ConstantInt operands of phi…
heiher Jun 6, 2024
5ddc841
[RISCV] Support select/merge like ops for bf16 vectors when have Zvfb…
jacquesguan Jun 6, 2024
95e1431
[libc][math][c23] Implement fmaxf16 and fminf16 function (#94131)
HendrikHuebner Jun 6, 2024
9e25be5
[lldb] Fix inconsistencies in DWARFExpression errors (#94554)
JDevlieghere Jun 6, 2024
f2165ae
[libc][math][c23] Add {nextafter,nexttoward,nextup,nextdown}f16 C23 m…
overmighty Jun 6, 2024
b22873d
[NFC] Remove unused value (#94439)
jacquesguan Jun 6, 2024
6a4e4f6
[lldb/crashlog] Always load Application Specific Backtrace Thread ima…
medismailben Jun 6, 2024
725b792
[WebAssembly] Set IS_64 flag correctly on __indirect_function_table i…
sbc100 Jun 6, 2024
07c0c26
[serialization] no transitive decl change (#92083)
ChuanqiXu9 Jun 6, 2024
c719881
Revert "[RISCV] Support select/merge like ops for bf16 vectors when h…
joker-eph Jun 6, 2024
3052bcc
[flang] Add GETCWD runtime and lowering intrinsics implementation (#9…
JumpMasterJJ Jun 6, 2024
85fd90b
[clang] Implement a __is_bitwise_cloneable builtin type trait. (#86512)
hokein Jun 6, 2024
16c3e1a
[LoongArch] Adjust LA64 data layout by using n32:64 in layout string …
heiher Jun 6, 2024
8ffa33f
[MLIR][LLVM] Improve module translation comment (NFC) (#94577)
Dinistro Jun 6, 2024
b6c4da3
[clang] NFCI: Make ASTContext optional in the AST text dumper again (…
mizvekov Jun 6, 2024
ab331bb
[InstCombine] Add more tests for select equivalence fold (NFC)
nikic Jun 6, 2024
d836ae8
[InstCombine] Only requite not-undef in select equiv fold
nikic Jun 6, 2024
942e935
[ValueTracking] Make undef element check more precise
nikic Jun 6, 2024
fa7e78c
[LoopUnroll] Consider convergence control tokens when unrolling (#91715)
ssahasra Jun 6, 2024
11675cb
[SDPatternMatch] Do not use std::forward and rvalue references (NFC) …
nikic Jun 6, 2024
ae8398c
[InstCombine] Add transforms `(icmp spred (and X, Y), X)` if `X` or `…
goldsteinn Jun 4, 2024
2b0061c
[ARM] vabd.ll - regenerate test checks
RKSimon Jun 6, 2024
43a52d5
[ARM] vaba.ll - regenerate test checks
RKSimon Jun 6, 2024
39027b5
[MC][RISCV] relocations.s - add missing opcode to test check
RKSimon Jun 6, 2024
5ed1246
[DebugInfo][SelectionDAG] Fix position of salvaged 'dangling' DBG_VAL…
OCHyams Jun 6, 2024
94cc5f4
[clang][Interp] Handle lvalue APValues in visitAPValueInitializer()
tbaederr Jun 6, 2024
15b6e55
[llvm-reduce] Remove DIGlobalVariableExpressions from DICompileUnit's…
OCHyams Jun 6, 2024
0f88be8
[flang][OpenMP] Fix privatization when critical is present (#94441)
luporl Jun 6, 2024
16f7316
[clang-repl] Lay the foundation of pretty printing for C. (#89811)
vgvassilev Jun 6, 2024
376f0d5
[PowerPC] Add test to show alignment of toc-data symbol is changed. NFC.
Jun 6, 2024
65a7389
[lldb] Disable TestPtyServer API test when remote testing (#94587)
slydiman Jun 6, 2024
5791862
[ARM] Add neon_vabd.ll based off aarch64 tests
RKSimon Jun 6, 2024
80694e4
[DAG] visitSUB - update the ABS matching code to use SDPatternMatch a…
RKSimon Jun 6, 2024
40333fc
[flang][CodeGen][NFC] Reduce boilerplatre for ExternalNameConversion …
tblah Jun 6, 2024
c3bc314
[PowerPC] Adjust operand order of ADDItoc to be consistent with other…
Jun 6, 2024
75a1c58
[clang][Interp] Member Pointers (#91303)
tbaederr Jun 6, 2024
453205a
[gn build] Port a86c1e7175d4
llvmgnsyncbot Jun 6, 2024
1cb3b5c
[AMDGPU] Move INIT_EXEC lowering from SILowerControlFlow to SIWholeQu…
jayfoad Jun 6, 2024
73af086
[AMDGPU] Implement variadic functions by IR lowering (#93362)
JonChesterfield Jun 6, 2024
85c4dd6
Revert "[Analyzer][CFG] Correctly handle rebuilt default arg and defa…
bgra8 Jun 6, 2024
6ec798c
[clang][Interp] Handle one-past-the-end pointers in SubPtr
tbaederr Jun 6, 2024
8d48e5c
Revert "[serialization] no transitive decl change (#92083)"
ChuanqiXu9 Jun 6, 2024
13eb6de
Revert "Reapply "[Clang][CWG1815] Support lifetime extension of tempo…
bgra8 Jun 6, 2024
8a3e4b1
[AArch64][SME] Add calling convention for __arm_get_current_vg (#93963)
kmclaughlin-arm Jun 6, 2024
2640ff6
[gn build] Port 8516f54e6a98
llvmgnsyncbot Jun 6, 2024
6299b18
[GlobalIsel] Combine G_VSCALE (#94096)
tschuett Jun 6, 2024
bfc793a
[bazel] Port for 8516f54e6a984a79d1a988ec66a414b5cfbd7ec7
hokein Jun 6, 2024
1b84df2
[libc] Disable varargs tests on gpu, nvptx can't build them yet
JonChesterfield Jun 6, 2024
aa10468
[Transforms] Fix -Wunused-variable in ExpandVariadics.cpp (NFC)
DamonFool Jun 6, 2024
cec4763
[ARM] Don't block tail-predication from unrelated VPT blocks. (#94239)
davemgreen Jun 6, 2024
0b8566b
[clang-tidy] Remove redundant LINK_LIBS (#94588)
nikic Jun 6, 2024
7b0f3ad
[libc][math] Temporarily disable nexttowardf16 on aarch64 due to clan…
lntue Jun 6, 2024
cdc9c01
[DAG] expandABS - add missing FREEZE in abs(x) -> smax(x,sub(0,x)) ex…
RKSimon Jun 6, 2024
5fd7a9b
[flang][OpenMP] Make object identity more precise (#94495)
kparzysz Jun 6, 2024
ffebcd0
[clang][Sema] Add missing scope flags to Scope::dumpImpl (#94529)
kparzysz Jun 6, 2024
7a4c101
[ConstraintElim] Add set of tests where a loop iv is used in exit.
fhahn Jun 6, 2024
7a2a155
[LoongArch] Allow f16 codegen with expansion to libcalls (#94456)
xen0n Jun 6, 2024
8aebc05
[workflows] Add scan-build to ci-ubuntu-22.04 container (#94543)
tstellar Jun 6, 2024
e60c668
[X86][AMX] Checking AMXProgModel in X86LowerTileCopy (#94358)
phoebewang Jun 6, 2024
c33b869
[Libomptarget] Rework device initialization and image registration (#…
jhuber6 Jun 6, 2024
ba9c7b4
[X86] Fix pipe resources for HADD/SUB instructions
RKSimon Jun 6, 2024
1dce655
[X86] Fix pipe resources for FP HADD/SUB instructions
RKSimon Jun 6, 2024
093818b
[Clang][AMDGPU] Use `I` to decorate imm argument for `__builtin_amdgc…
shiltian Jun 6, 2024
4782ad7
[libc] Enable varargs tests for AMDGPU targets
jhuber6 Jun 6, 2024
2be0989
[NVPTX] Revamp NVVMIntrRange pass (#94422)
AlexMaclean Jun 6, 2024
8fe8ede
[bazel] Fix layering check violation for nextafter_test_template targ…
hokein Jun 6, 2024
48c9a27
[AArch64] Override isLSRCostLess, take number of instructions into ac…
huntergr-arm Jun 6, 2024
b4896c9
Fix -Wunused-variable in SemaAMDGPU.cpp in release build, NFC
hokein Jun 6, 2024
2450e72
[NVPTX] Remove unused private field in NVVMIntrRange.cpp (NFC)
DamonFool Jun 6, 2024
52edac7
[lldb] Fix ThreadPlanStepOverRange name in log message (#94611)
mariannems Jun 6, 2024
a6444dd
[OpenMP][NFC] Fix warning for OpenMP standalone build (#93463)
jprotze Jun 6, 2024
5ddd8e7
[RISCV] Fix duplicate test cases for G_UNMERGE_VALUES (#94622)
spaits Jun 6, 2024
0b27a2e
[mlir][tensor] Implement constant folder for tensor.pad (#92691)
sabauma Jun 6, 2024
27a3150
[BOLT][DWARF][NFC] Refactor GDB Index into a new file (#94405)
sayhaan Jun 6, 2024
158494e
[AArch64] Add support for Qualcomm Oryon processor (#91022)
wxz2020 Jun 6, 2024
79653ce
[SPIR-V] Add validation to the test case with get_image_array_size/ge…
VyacheslavLevytskyy Jun 6, 2024
25f2565
[ConstraintElim] Add induction tests with different start values.
fhahn Jun 6, 2024
0e92e4f
[clang][Interp][NFC] Return a valid SourceInfo for Function PCs
tbaederr Jun 6, 2024
5332122
DAG: Pass flags to FoldConstantArithmetic (#93663)
arsenm Jun 6, 2024
9881528
[WebAssembly] Fix element segments in wasm64 object files (#94617)
sbc100 Jun 6, 2024
4691b20
[MLIR] Fix generic assembly syntax for ArrayAttr containing hex float…
joker-eph Jun 6, 2024
b70d25d
[mlir] Add pack/unpack transpose foldings for linalg.generic ops, fix…
Max191 Jun 6, 2024
a9cf149
[gn build] Port 2a6efe6a49e0
llvmgnsyncbot Jun 6, 2024
659318e
[Bazel] Generate LLVM_HAS_XYZ_TARGET macros in llvm config (#94476)
apaszke Jun 6, 2024
a8ce798
[CodeGen] Use std::bitset for MachineFunctionProperties (#94627)
aengelke Jun 6, 2024
696a2f5
[X86] Skip AMX type lowering when AMX is not used (#92910)
aengelke Jun 6, 2024
ac3f92a
RegisterCoalescer: Remove unnecessary maybe_unused
arsenm Jun 1, 2024
3c1d2a0
[libc][math][c23] Fix implicit conversion in smoke tests for {fmax,fm…
overmighty Jun 6, 2024
19dd633
[libc][math][c23] Add {fmaximum,fminimum}{,_mag,_mag_num,_num} C23 ma…
overmighty Jun 6, 2024
7650125
[llvm][ScheduleDAG] Set a fixed size for Sched::Preference (#94523)
jroelofs Jun 6, 2024
b50875b
[NFC][libc++][test][AIX] fix SIMD test XFAIL for clang before 19 (#94…
daltenty Jun 6, 2024
a05f49a
[clang] Fix handling of adding a file with the same name as an existi…
jensmassberg Jun 6, 2024
99fcd1b
[clang][Interp] Always decay root array pointers to the first element
tbaederr Jun 6, 2024
8af229c
[AIX] use LIBPATH on AIX instead of LD_LIBRARY_PATH (#94602)
Jun 6, 2024
b874ae3
[AMDGPU] Update removeFnAttrFromReachable to accept array of Fn Attrs…
skc7 Jun 6, 2024
0031c27
[lldb] Refactor ReadRegisterValueAsScalar to return an llvm::Error (N…
JDevlieghere Jun 6, 2024
275a662
[lldb] Refactor ResolveLoadAddress to return an llvm::Expected (NFC) …
JDevlieghere Jun 6, 2024
879a468
Revert "[lldb][DebugNames] Only skip processing of DW_AT_declarations…
labath Jun 6, 2024
3a2fbf4
[lldb] Test case for the bug in #92328
labath Jun 6, 2024
00a43ed
[AMDGPU] Add a new target gfx1152 (#94534)
shiltian Jun 6, 2024
815412f
[bazel] Fix smoke/BUILD.bazel layering in nextafter_test_template (#9…
rupprecht Jun 6, 2024
13f5b56
DAG: Improve fminimum/fmaximum vector expansion logic (#93579)
arsenm Jun 6, 2024
ce63390
[clang-tidy] Fix crash in readability-container-size-empty (#94527)
PiotrZSL Jun 6, 2024
30693e5
fixup cuda-builtin-vars.cu broken in IntrRange change (#94639)
AlexMaclean Jun 6, 2024
8fb6945
[libc] at_quick_exit function implemented (#94317)
aaryanshukla Jun 6, 2024
ee48f68
[AMDGPU] Fix GFX1152 ELF arch
shiltian Jun 6, 2024
c3bfbfa
[HWASan] add optimization remarks for ignoreAccess (#94551)
fmayer Jun 6, 2024
7842384
[gtest] Enable zos for death test support (#94623)
abhina-sree Jun 6, 2024
fd8684c
[clang][Interp] Diagnose functions without body like undefined ones
tbaederr Jun 6, 2024
e070bda
[InstCombine] Add tests for folding multiuse `(icmp eq/ne (or X, Y), …
goldsteinn Sep 13, 2023
f655b95
[InstCombine] Folding multiuse `(icmp eq/ne (or X, Y), Y)` for 2 uses…
goldsteinn Sep 13, 2023
ae24158
[clang][RISCV] Update vcpop.v C interface to follow the nameing conve…
4vtomat Jun 6, 2024
5a832da
[MemProf] Remove context id set from nodes and recompute on demand (#…
teresajohnson Jun 6, 2024
fea5d46
[flang] Add reduction semantics to fir.do_loop (#93934)
khaki3 Jun 6, 2024
3d79450
[libc][FixedVector] Add more helper methods (#94278)
PiJoules Jun 6, 2024
993247b
[clang][NFC] fix name lookup for llvm::json::Value in SymbolGraphSeri…
yuxuanchen1997 Jun 6, 2024
19fd58a
[memprof] Use std::unique_ptr instead of std::optional (#94655)
kazutakahirata Jun 6, 2024
580a488
[AMDGPU] Promote immediate offset to atomics (#94043)
rampitec Jun 6, 2024
7e5a6ca
[ELF] Keep non-alloc orphan sections at the end
MaskRay Jun 6, 2024
5a31f5e
[bazel] Port for 649edb8eb25e82e3ac6fce4788f51759636229ec
hokein Jun 6, 2024
c7b1d07
[lldb/crashlog] Remove aarch64 requirement on crashlog tests (#94553)
medismailben Jun 6, 2024
215c9f1
[Offload] Fix missing `abs` function for test
jhuber6 Jun 6, 2024
91bf0b0
[LLVM] Do not require shell for some tests (#94595)
jayfoad Jun 6, 2024
dae0098
NFC: resolve TODO in LLVM dialect conversions (#91497)
christopherbate Jun 6, 2024
093ff68
[lldb] Include memory stats in statistics summary (#94671)
bulbazord Jun 6, 2024
fc0f7bf
[Offload] Use the kernel argument size directly in AMDGPU offloading …
jhuber6 Jun 6, 2024
91b0814
[RISCV][InsertVSETVLI] Check for undef register operand directly [nfc]
preames Jun 6, 2024
3df6c97
[ELF] Test non-alloc orphan that does not the RF_NOT_ADDR_SET rank flag
MaskRay Jun 6, 2024
d7fddb4
[ProfileData] Remove swapToHostOrder (#94665)
kazutakahirata Jun 6, 2024
898bd85
[clang] Fix flag typo in comment
kastiglione Jun 6, 2024
15701c1
[memprof] Use std::vector<Frame> instead of llvm::SmallVector<Frame> …
kazutakahirata Jun 6, 2024
66026f0
[lldb] Fix TestStatisticsAPI after 9293fc798152 (#94683)
bulbazord Jun 6, 2024
ff9f8a7
[clang-repl] Disable new test after #89811
vitalybuka Jun 6, 2024
37a2cc0
[libc] fixed target issue with exit_handler (#94678)
aaryanshukla Jun 6, 2024
f99ee69
[sanitizer] Make CHECKs in bitvector more precise (NFC) (#94630)
nikic Jun 6, 2024
f09cac8
[sanitizer_common] Change allocator base in test case for compatibili…
thurstond Jun 6, 2024
ec12488
[compiler-rt] Map internal_sigaction to __sys_sigaction on FreeBSD (#…
arichardson Jun 6, 2024
9905f64
[llvm/IR] Fix module build issue following e57308b063bb (NFC) (#94580)
medismailben Jun 6, 2024
a7c205d
[memprof] Add CallStackRadixTreeBuilder (#93784)
kazutakahirata Jun 6, 2024
625bd35
[clang-tidy]fix crashing when self include cycles for misc-header-inc…
HerrCai0907 Jun 6, 2024
300e13b
[libc++][hardening] Finish documenting hardening. (#92021)
var-const Jun 6, 2024
31f7172
[ELF] Improve -r section group tests
MaskRay Jun 7, 2024
ceb964d
[clang-format] Fix a bug in `AlignAfterOpenBracket: DontAlign` (#94561)
owenca Jun 7, 2024
720509b
[dfsan] Add test case for sscanf (#94700)
thurstond Jun 7, 2024
906734b
[ELF] Simplify code. NFC
MaskRay Jun 7, 2024
9ab25b2
[SPIRV] Fix -Wunused-but-set-variable. NFC
MaskRay Jun 7, 2024
f36f6a5
[InstCombine] Add tests for expanding `foldSelectValueEquivalence`; NFC
goldsteinn Apr 10, 2024
3865e74
[InstCombine] Improve coverage of `foldSelectValueEquivalence` for co…
goldsteinn Jun 5, 2024
b7c3fdb
[RISCV] Unify all the code that adds unaligned-scalar/vector-mem to F…
topperc Jun 7, 2024
f14147d
[clang-format][NFC] Remove an else after a return statement (#94548)
Khufu0 Jun 7, 2024
1b8d567
[clang-format]: Annotate colons found in inline assembly (#92617)
gedare Jun 7, 2024
12b9224
[Driver][test] Rename loongarch-{default-,}features.c
SixWeining Jun 7, 2024
59bed9c
[serialization] no transitive decl change (#92083)
ChuanqiXu9 Jun 6, 2024
c5b5dcd
[llvm][ScheduleDAG] Re-arrange SUnit's members to make it smaller (#9…
jroelofs Jun 7, 2024
b0281f1
Revert "[serialization] no transitive decl change (#92083)"
ChuanqiXu9 Jun 7, 2024
09ce478
[AMDGPU] Auto-generating lit test patterns (NFC) (#93837)
cdevadas Jun 7, 2024
634fbfb
[AMDGPU] Auto-generated some lit test patterns (NFC). (#94310)
cdevadas Jun 7, 2024
eb3090e
[NewPM][CodeGen] Port `regallocfast` to new pass manager (#94426)
paperchalice Jun 7, 2024
638074f
[NFC][LoongArch] Pre-commit tests for hoisting of PseudoLA*
heiher Jun 7, 2024
9928aa4
[test] Don't generate `regalloc-amdgpu.s` in #94426 (#94722)
paperchalice Jun 7, 2024
95f34a7
[clang-tidy] refactor misc-header-include-cycle (#94697)
HerrCai0907 Jun 7, 2024
e085ae5
[libc++] Fix naming of value template parameters (#76888)
philnik777 Jun 7, 2024
a65c853
Fix spurious non-strict availability warning (#94377)
Xazax-hun Jun 7, 2024
5694e29
[mlir][tensor] Fix FoldTensorCastProducerOp for multiple result opera…
pashu123 Jun 7, 2024
e8ac511
[MC] MCSection::dump: print section name
MaskRay Jun 7, 2024
da152a0
[clang][Interp] Improve APValue machinery
tbaederr Jun 7, 2024
47e0905
[X86] Assign AVX10_1 feature priority to align with gcc. (#94557)
FreddyLeaf Jun 7, 2024
e47cb50
[lldb] Split ValueObject::CreateChildAtIndex into two functions (#94455)
labath Jun 7, 2024
908d925
Revert "[X86] Assign AVX10_1 feature priority to align with gcc. (#94…
FreddyLeaf Jun 7, 2024
ed0d45e
[memprof] Use std::move in ContextEdge::ContextEdge (NFC) (#94687)
kazutakahirata Jun 7, 2024
4d849a4
[ORC] Switch ExecutionSession::ErrorReporter to use unique_function.
lhames Jun 7, 2024
137038f
[LoongArch] Set isReMaterializable on LU{12,32,52}I.D/ADDI.D and {X}O…
heiher Jun 7, 2024
b0d738c
[SCEV] Use insert_or_assign() (NFC)
nikic Jun 7, 2024
9516710
[LoongArch] Add a pass to rewrite rd to r0 for non-computational inst…
heiher Jun 7, 2024
6fc4e97
[clang][Interp][NFC] Add GetPtrFieldPop opcode
tbaederr Jun 7, 2024
7375a39
[analyzer][NFC] Factor out NoOwnershipChangeVisitor (#94357)
Szelethus Jun 7, 2024
ea0fcca
Reland "[RISCV] Support select/merge like ops for bf16 vectors when h…
jacquesguan Jun 6, 2024
5a0978c
[docs] Fix benchmarking tips (#94724)
maekawatoshiki Jun 7, 2024
711196a
[test] Don't generate extra file for regalloc-amdgpu.mir test.
hokein Jun 7, 2024
b1fafc4
[clang][Interp] Remove StoragKind limitation in Pointer assign operators
tbaederr Jun 7, 2024
c0635ee
[NFC][LoongArch] Update test for #94590
heiher Jun 7, 2024
e336acf
[MLIR] Translate DIStringType. (#94480)
abidh Jun 7, 2024
08bf183
[clangd] Fix crash with null check for Token at Loc (#94528)
usx95 Jun 7, 2024
7db1232
[flang][Transforms][NFC] Remove boilerplate from vscale range pass (#…
tblah Jun 7, 2024
7ce3900
[PowerPC] modify the frameaddress case, NFC
Jun 7, 2024
8d54dc1
[PowerPC] return correct frame address for frameaddress intrinsic
Jun 7, 2024
538584d
[ARM] Add NEON support for ISD::ABDS/ABDU nodes. (#94504)
RKSimon Jun 7, 2024
a74cf9d
[CMake][Release] Use the TXZ cpack generator for binaries (#90138)
tstellar Jun 7, 2024
e173fa7
[DebugInfo] Add DW_OP_LLVM_extract_bits (#93990)
john-brawn-arm Jun 7, 2024
b37c6bd
Add checks before hoisting out in loop pipelining (#90872)
fotiskoun Jun 7, 2024
85cbf2f
[clang][Interp][NFC] Properly assign block pointer Pointee
tbaederr Jun 7, 2024
f0fde2b
[clang][Interp] Fix refers_to_enclosing_variable_or_capture DREs
tbaederr Jun 7, 2024
66dad78
[SimplifyCFG] Remove bogus UTC line from test (NFC)
nikic Jun 7, 2024
405d7d5
[SimplifyCFG] Regenerate switch to lookup tests (NFC)
nikic Jun 7, 2024
afcd18f
[mlir][vector] Add n-d deinterleave lowering (#94237)
mub-at-arm Jun 7, 2024
3f99c0d
[ARM] r11 is reserved when using -mframe-chain=aapcs (#86951)
ostannard Jun 7, 2024
671bcef
[DAG] Always allow folding XOR patterns to ABS pre-legalization (#94601)
RKSimon Jun 7, 2024
9ecb812
fix(mlir/**.py): fix comparison to None (#94019)
e-kwsm Jun 7, 2024
0601711
[ARM] Add support for Cortex-R52+ (#94633)
jthackray Jun 7, 2024
2fd4477
[NFC][LoongArch] Update test for #94590
heiher Jun 7, 2024
14cd171
[clang][test] Skip interpreter value test on Arm 32 bit
DavidSpickett Jun 7, 2024
126837c
[gn build] Port e622996eddfb
llvmgnsyncbot Jun 7, 2024
7f3b593
[Flang] Handle the newly-added "Reserved" FramePointerKind for 1a5239…
hokein Jun 7, 2024
4e4eb43
[clang][SPIR-V] Add support for AMDGCN flavoured SPIRV (#89796)
AlexVlx Jun 7, 2024
9c42b20
[BOLT][NFC] Infailable fns return void (#92018)
urnathan Jun 7, 2024
35fbc3f
[CodeGen][SDAG] Remove CombinedNodes SmallPtrSet (#94609)
aengelke Jun 7, 2024
6db6f7e
[clang][Interp] Check ConstantExpr results for initialization
tbaederr Jun 7, 2024
c1a3bf7
[clang][Interp][NFC] Fix a const-correctness warning
tbaederr Jun 7, 2024
a647101
[clang][Interp] Limit lambda capture lazy visting to actual captures
tbaederr Jun 7, 2024
a9b37d7
[serialization] no transitive decl change (#92083)
ChuanqiXu9 Jun 6, 2024
a0dcaf2
[AMDGPU] Fix interaction between WQM and llvm.amdgcn.init.exec (#93680)
jayfoad Jun 7, 2024
01be0a3
[Frontend][OpenMP] Sort all the things in OMP.td, NFC (#94653)
kparzysz Jun 7, 2024
26ba412
[flang][OpenMP] Lower `target .. private(..)` to `omp.private` ops (#…
ergawy Jun 7, 2024
4407e67
[libc] Correctly pass the C++ standard to NVPTX internal builds
jhuber6 Jun 7, 2024
df12b11
[mlir][linalg] Support lowering unpack with outer_dims_perm (#94477)
ryan-holt-1 Jun 7, 2024
f0cdc72
[mlir] Add reshape propagation patterns for tensor.pad (#94489)
Max191 Jun 7, 2024
e4f8c4e
[mlir] Fix bugs in expand_shape patterns after semantics changes (#94…
Max191 Jun 7, 2024
b9d3565
[ARM] Clean up neon_vabd.ll, vaba.ll and vabd.ll tests a bit. NFC
davemgreen Jun 7, 2024
678428a
[arm64] Add tan intrinsic lowering (#94545)
farzonl Jun 7, 2024
50bec57
[AArch64] Add addp from shuffles tests. NFC
davemgreen Jun 7, 2024
86dd2c9
[Clang] Add timeout for GPU detection utilities (#94751)
jhuber6 Jun 7, 2024
1b239ca
[RISCV] Codegen support for XCVmem extension (#76916)
ChunyuLiao Jun 7, 2024
58c7def
[MachineOutliner] Sort by Benefit to Cost Ratio (#90264)
xuanzhang816 Jun 7, 2024
dc9c2df
[memprof] Clean up IndexedMemProfReader (NFC) (#94710)
kazutakahirata Jun 7, 2024
8d913d5
[flang] lower SIZE and SIZEOF for assumed-ranks (#94684)
jeanPerier Jun 7, 2024
637baa5
[memprof] Use CallStackRadixTreeBuilder in the V3 format (#94708)
kazutakahirata Jun 7, 2024
ad12734
[mlir][vector] Remove Emulated Sub-directory (#94742)
mub-at-arm Jun 7, 2024
7f5aeb1
[gn] port 33a6ce18373ff (check-clang obj2yaml dep)
nico Jun 7, 2024
aae32f6
[gn] port cb7690af09b95 (ntdll dep)
nico Jun 7, 2024
5b14f6d
[KnownBits] Remove `hasConflict()` assertions (#94568)
c8ef Jun 7, 2024
1508a3d
[libc++][test][AIX] Only XFAIL atomic tests for before clang 19 (#94646)
jakeegan Jun 7, 2024
53615ae
[AArch64] Add patterns for add(uzp1(x,y), uzp2(x, y)) -> addp.
davemgreen Jun 7, 2024
3a93ccc
[LV] Add test with dead load and vector pointer.
fhahn Jun 7, 2024
aaa160e
[libc++][regex] Correctly adjust match prefix for zero-length matches…
var-const Jun 7, 2024
8bb019d
Reapply PR/87550 (#94625)
oontvoo Jun 7, 2024
7e2707f
[libc++] Undeprecate shared_ptr atomic access APIs (#92920)
nico Jun 7, 2024
fd08cef
[Reassociate] shifttest.ll - generate test checks to replace custom g…
RKSimon Jun 7, 2024
3aa3f3a
[flang][runtime] add SHAPE runtime interface (#94702)
jeanPerier Jun 7, 2024
8a0529a
[flang][OpenMP] Add `--openmp-enable-delayed-privatization-staging` f…
ergawy Jun 7, 2024
e854b11
[OpenMP] Fix passing target id features to AMDGPU offloading (#94765)
jhuber6 Jun 7, 2024
624a743
Fixed grammatical error in "enum specifier" error msg #94443 (#94592)
kper Jun 7, 2024
f7d4ecb
[clang] always use resolved arguments for default argument deduction …
mizvekov Jun 7, 2024
d931adf
Check if LLD is built when checking if lto_supported (#92752)
jvoung Jun 7, 2024
ae23164
[mlir][vector][NFC] Make function name more meaningful in lit tests. …
hanhanW Jun 7, 2024
b928554
update
srcarroll Jun 7, 2024
df0747c
Merge branch 'main' into implement-grouped-conv-interface
srcarroll Jun 7, 2024
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions lldb/include/lldb/API/SBDebugger.h
Original file line number Diff line number Diff line change
Expand Up @@ -57,6 +57,8 @@ class LLDB_API SBDebugger {

static const char *GetBroadcasterClass();

static bool SupportsLanguage(lldb::LanguageType language);

lldb::SBBroadcaster GetBroadcaster();

/// Get progress data from a SBEvent whose type is eBroadcastBitProgress.
Expand Down
1 change: 1 addition & 0 deletions lldb/include/lldb/Symbol/TypeSystem.h
Original file line number Diff line number Diff line change
Expand Up @@ -209,6 +209,7 @@ class TypeSystem : public PluginInterface,
// TypeSystems can support more than one language
virtual bool SupportsLanguage(lldb::LanguageType language) = 0;

static bool SupportsLanguageStatic(lldb::LanguageType language);
// Type Completion

virtual bool GetCompleteType(lldb::opaque_compiler_type_t type) = 0;
Expand Down
4 changes: 4 additions & 0 deletions lldb/source/API/SBDebugger.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -1742,3 +1742,7 @@ bool SBDebugger::InterruptRequested() {
return m_opaque_sp->InterruptRequested();
return false;
}

bool SBDebugger::SupportsLanguage(lldb::LanguageType language) {
return TypeSystem::SupportsLanguageStatic(language);
}
11 changes: 11 additions & 0 deletions lldb/source/Symbol/TypeSystem.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -335,3 +335,14 @@ TypeSystemMap::GetTypeSystemForLanguage(lldb::LanguageType language,
}
return GetTypeSystemForLanguage(language);
}

bool TypeSystem::SupportsLanguageStatic(lldb::LanguageType language) {
if (language == eLanguageTypeUnknown)
return false;

LanguageSet languages =
PluginManager::GetAllTypeSystemSupportedLanguagesForTypes();
if (languages.Empty())
return false;
return languages[language];
}
52 changes: 50 additions & 2 deletions lldb/tools/lldb-dap/DAP.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -65,16 +65,64 @@ DAP::DAP()

DAP::~DAP() = default;

void DAP::PopulateExceptionBreakpoints() {
llvm::call_once(initExceptionBreakpoints, [this]() {
exception_breakpoints = {};
if (lldb::SBDebugger::SupportsLanguage(lldb::eLanguageTypeC_plus_plus)) {
exception_breakpoints->emplace_back("cpp_catch", "C++ Catch",
lldb::eLanguageTypeC_plus_plus);
exception_breakpoints->emplace_back("cpp_throw", "C++ Throw",
lldb::eLanguageTypeC_plus_plus);
}
if (lldb::SBDebugger::SupportsLanguage(lldb::eLanguageTypeObjC)) {
exception_breakpoints->emplace_back("objc_catch", "Objective-C Catch",
lldb::eLanguageTypeObjC);
exception_breakpoints->emplace_back("objc_throw", "Objective-C Throw",
lldb::eLanguageTypeObjC);
}
if (lldb::SBDebugger::SupportsLanguage(lldb::eLanguageTypeSwift)) {
exception_breakpoints->emplace_back("swift_catch", "Swift Catch",
lldb::eLanguageTypeSwift);
exception_breakpoints->emplace_back("swift_throw", "Swift Throw",
lldb::eLanguageTypeSwift);
}
});
}

ExceptionBreakpoint *DAP::GetExceptionBreakpoint(const std::string &filter) {
for (auto &bp : exception_breakpoints) {
// PopulateExceptionBreakpoints() is called after g_dap.debugger is created
// in a request-initialize.
//
// But this GetExceptionBreakpoint() method may be called before attaching, in
// which case, we may not have populated the filter yet.
//
// We also cannot call PopulateExceptionBreakpoints() in DAP::DAP() because
// we need SBDebugger::Initialize() to have been called before this.
//
// So just calling PopulateExceptionBreakoints(),which does lazy-populating
// seems easiest. Two other options include:
// + call g_dap.PopulateExceptionBreakpoints() in lldb-dap.cpp::main()
// right after the call to SBDebugger::Initialize()
// + Just call PopulateExceptionBreakpoints() to get a fresh list everytime
// we query (a bit overkill since it's not likely to change?)
PopulateExceptionBreakpoints();
assert(exception_breakpoints.has_value() &&
"exception_breakpoints must have been populated");

for (auto &bp : *exception_breakpoints) {
if (bp.filter == filter)
return &bp;
}
return nullptr;
}

ExceptionBreakpoint *DAP::GetExceptionBreakpoint(const lldb::break_id_t bp_id) {
for (auto &bp : exception_breakpoints) {
// See comment in the other GetExceptionBreakpoint().
PopulateExceptionBreakpoints();
assert(exception_breakpoints.has_value() &&
"exception_breakpoints must have been populated");

for (auto &bp : *exception_breakpoints) {
if (bp.bp.GetID() == bp_id)
return &bp;
}
Expand Down
5 changes: 4 additions & 1 deletion lldb/tools/lldb-dap/DAP.h
Original file line number Diff line number Diff line change
Expand Up @@ -156,7 +156,8 @@ struct DAP {
std::unique_ptr<std::ofstream> log;
llvm::StringMap<SourceBreakpointMap> source_breakpoints;
FunctionBreakpointMap function_breakpoints;
std::vector<ExceptionBreakpoint> exception_breakpoints;
std::optional<std::vector<ExceptionBreakpoint>> exception_breakpoints;
llvm::once_flag initExceptionBreakpoints;
std::vector<std::string> init_commands;
std::vector<std::string> pre_run_commands;
std::vector<std::string> post_run_commands;
Expand Down Expand Up @@ -228,6 +229,8 @@ struct DAP {

llvm::json::Value CreateTopLevelScopes();

void PopulateExceptionBreakpoints();

/// \return
/// Attempt to determine if an expression is a variable expression or
/// lldb command using a hueristic based on the first term of the
Expand Down
6 changes: 4 additions & 2 deletions lldb/tools/lldb-dap/lldb-dap.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,7 @@
#include <cstdio>
#include <cstdlib>
#include <cstring>
#include <optional>
#include <sys/stat.h>
#include <sys/types.h>
#if defined(_WIN32)
Expand Down Expand Up @@ -1586,6 +1587,7 @@ void request_initialize(const llvm::json::Object &request) {
bool source_init_file = GetBoolean(arguments, "sourceInitFile", true);

g_dap.debugger = lldb::SBDebugger::Create(source_init_file, log_cb, nullptr);
g_dap.PopulateExceptionBreakpoints();
auto cmd = g_dap.debugger.GetCommandInterpreter().AddMultiwordCommand(
"lldb-dap", "Commands for managing lldb-dap.");
if (GetBoolean(arguments, "supportsStartDebuggingRequest", false)) {
Expand Down Expand Up @@ -1621,7 +1623,7 @@ void request_initialize(const llvm::json::Object &request) {
body.try_emplace("supportsEvaluateForHovers", true);
// Available filters or options for the setExceptionBreakpoints request.
llvm::json::Array filters;
for (const auto &exc_bp : g_dap.exception_breakpoints) {
for (const auto &exc_bp : *g_dap.exception_breakpoints) {
filters.emplace_back(CreateExceptionBreakpointFilter(exc_bp));
}
body.try_emplace("exceptionBreakpointFilters", std::move(filters));
Expand Down Expand Up @@ -2476,7 +2478,7 @@ void request_setExceptionBreakpoints(const llvm::json::Object &request) {
// Keep a list of any exception breakpoint filter names that weren't set
// so we can clear any exception breakpoints if needed.
std::set<std::string> unset_filters;
for (const auto &bp : g_dap.exception_breakpoints)
for (const auto &bp : *g_dap.exception_breakpoints)
unset_filters.insert(bp.filter);

for (const auto &value : *filters) {
Expand Down
1 change: 1 addition & 0 deletions llvm/lib/Target/SPIRV/SPIRVEmitIntrinsics.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -1396,6 +1396,7 @@ bool SPIRVEmitIntrinsics::runOnFunction(Function &Func) {
continue;

insertAssignPtrTypeIntrs(I, B);
deduceOperandElementType(I);
insertAssignTypeIntrs(I, B);
insertPtrCastOrAssignTypeInstr(I, B);
insertSpirvDecorations(I, B);
Expand Down
37 changes: 37 additions & 0 deletions mlir/include/mlir/Dialect/Linalg/IR/LinalgInterfaces.h
Original file line number Diff line number Diff line change
Expand Up @@ -28,6 +28,8 @@ namespace mlir {
namespace linalg {
class IteratorTypeAttr;
class LinalgOp;
class ConvolutionOpInterface;
class GroupedConvolutionOpInterface;
class GenericOp;

namespace detail {
Expand Down Expand Up @@ -133,6 +135,38 @@ std::optional<Value> isaFillOpInterface(GenericOp genericOp);

namespace detail {

// Common implementations for ConvolutionOpInterface
namespace convolution_impl {
// Returns strides as a vector.
SmallVector<int64_t, 2> getStrides(ConvolutionOpInterface op);
// Returns dilations as a vector.
SmallVector<int64_t, 2> getDilations(ConvolutionOpInterface op);
// Region builder for basic convolution
void regionBuilder(ImplicitLocOpBuilder &b, Block &block,
ArrayRef<NamedAttribute> attrs);
// Region builder for basic quantized convolution
void quantizedRegionBuilder(ImplicitLocOpBuilder &b, Block &block,
ArrayRef<NamedAttribute> attrs);
void getEffects(
Operation *op,
SmallVectorImpl<SideEffects::EffectInstance<MemoryEffects::Effect>>
&effects);
ParseResult parse(OpAsmParser &parser, OperationState &result,
bool isQuantized = false);
void print(LinalgOp op, OpAsmPrinter &p);
} // namespace convolution_impl

// Common implementations for GroupedConvolutionOpInterface
namespace grouped_convolution_impl {
int64_t getSpatialRank(GroupedConvolutionOpInterface op);
ArrayAttr createCommonIndexingMaps(
MLIRContext *ctx, int64_t numSpatial,
const SmallVector<SmallVector<utils::GroupedConvDim>> &layouts,
const SmallVectorImpl<int64_t> &strides,
const SmallVectorImpl<int64_t> &dilations);
ArrayAttr getIteratorTypes(GroupedConvolutionOpInterface op);
} // namespace grouped_convolution_impl

/// Returns true if the block contains a contraction of the following form:
///
/// %0 = <elemwise>(permutation-of(cu(block-argument-0),
Expand Down Expand Up @@ -189,6 +223,9 @@ LogicalResult verifyContractionInterface(Operation *op);
/// Verify that `op` conforms to the ConvolutionOpInterface.
LogicalResult verifyConvolutionInterface(Operation *op);

/// Verify that `op` conforms to the GroupedConvolutionOpInterface.
LogicalResult verifyGroupedConvolutionInterface(Operation *op);

/// Verify that `op` conforms to the FillOpInterface.
LogicalResult verifyFillInterface(Operation *op);

Expand Down
95 changes: 95 additions & 0 deletions mlir/include/mlir/Dialect/Linalg/IR/LinalgInterfaces.td
Original file line number Diff line number Diff line change
Expand Up @@ -175,6 +175,101 @@ def LinalgConvolutionOpInterface : OpInterface<"ConvolutionOpInterface"> {
return $_op.getOperation()->getOperand(1);
}]
>,
InterfaceMethod<
/*desc=*/"Return the spatial rank.",
/*retTy=*/"int64_t",
/*methodName=*/"getSpatialRank",
/*args=*/(ins),
/*methodBody=*/"",
/*defaultImplementation=*/[{
// Most convolution's inputs have batch, channel and spatial dims
return cast<ShapedType>(image().getType()).getRank() - 2;
}]
>
];
}

def LinalgGroupedConvolutionOpInterface : OpInterface<"GroupedConvolutionOpInterface", [
LinalgConvolutionOpInterface]> {
let description = [{
A grouped convolution is defined in general terms:
1. It is a convolution as defined by `ConvolutionOpInterface`.
2. Operands have a the following distinct dimensions (excluding batch in input/output): group, channel, spatial
3. `input_rank == kernel_rank == output_rank` (including batch in input/output)
4. Reductions are along the input channel and spatial dimensions while group, output channel
and output spatial dimensions are parallel.
}];
let cppNamespace = "::mlir::linalg";
let verify = [{ return detail::verifyGroupedConvolutionInterface($_op); }];
let methods = [
InterfaceMethod<[{
Returns the groups position for the input.
}],
"SmallVector<SmallVector<::mlir::utils::GroupedConvDim>>", "getLayoutsEnums", (ins)
>,
InterfaceMethod<[{
Returns the groups position for the input.
}],
"int64_t", "getInputGroupsPosition", (ins)
>,
InterfaceMethod<[{
Returns the channel position for the input.
}],
"int64_t", "getInputChannelPosition", (ins)
>,
InterfaceMethod<[{
Returns the channel position for the output.
}],
"int64_t", "getOutputChannelPosition", (ins)
>,
InterfaceMethod<[{
Get number of groups.
}],
"int64_t", "getNumGroups", (ins),
/*methodBody=*/[{}],
/*defaultImplementation=*/[{
return cast<ShapedType>($_op.image().getType()).getShape()[$_op.getInputGroupsPosition() - 1];
}]>,
InterfaceMethod<[{
Get number of input channels.
}],
"int64_t", "getNumInputChannels", (ins),
/*methodBody=*/[{}],
/*defaultImplementation=*/[{
return cast<ShapedType>($_op.image().getType()).getShape()[$_op.getInputChannelPosition()];
}]>,
InterfaceMethod<[{
Get number of output channels.
}],
"int64_t", "getNumOutputChannels", (ins),
/*methodBody=*/[{}],
/*defaultImplementation=*/[{
return cast<ShapedType>($_op.getDpsInits()[0].getType()).getShape()[$_op.getOutputChannelPosition()];
}]>,
InterfaceMethod<[{
Returns indexing maps for any spatial dimension.
}],
"::mlir::ArrayAttr", "getIteratorTypes", (ins),
/*methodBody=*/[{}],
/*defaultImplementation=*/[{
return detail::grouped_convolution_impl::getIteratorTypes($_op);
}]>,
InterfaceMethod<[{
Returns strides.
}],
"::llvm::SmallVector<int64_t, 2>", "getStridesVector", (ins),
/*methodBody=*/[{}],
/*defaultImplementation=*/[{
return detail::convolution_impl::getStrides($_op);
}]>,
InterfaceMethod<[{
Returns dilations.
}],
"::llvm::SmallVector<int64_t, 2>", "getDilationsVector", (ins),
/*methodBody=*/[{}],
/*defaultImplementation=*/[{
return detail::convolution_impl::getDilations($_op);
}]>
];
}

Expand Down
Loading
Loading