[mlir][bazel] Port https://github.com/llvm/llvm-project/commit/517800e37e8d3a4ee84214bef65e227612c2a98b #127544
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Introduces a `LinalgInterfaces` target so that `TensorDialect` doesn't need to depend on `LinalgDialect`, which would introduce a circular dependency.
chsigg
requested review from
rupprecht,
keith,
aaronmondal,
dcaballe,
nicolasvasilache and
rengolin
as code owners
llvmbot
added
mlir:linalg
mlir
mlir:tensor
bazel
|
@llvm/pr-subscribers-mlir-tensor @llvm/pr-subscribers-mlir-linalg Author: Christian Sigg (chsigg) ChangesIntroduces a Full diff: https://github.com/llvm/llvm-project/pull/127544.diff 4 Files Affected:
diff --git a/mlir/lib/Dialect/Linalg/IR/LinalgInterfaces.cpp b/mlir/lib/Dialect/Linalg/IR/LinalgInterfaces.cpp
index 466a9799295f9..c16c38ea22a5d 100644
--- a/mlir/lib/Dialect/Linalg/IR/LinalgInterfaces.cpp
+++ b/mlir/lib/Dialect/Linalg/IR/LinalgInterfaces.cpp
@@ -13,8 +13,6 @@
#include "mlir/Dialect/Arith/Utils/Utils.h"
#include "mlir/Dialect/Complex/IR/Complex.h"
#include "mlir/Dialect/Linalg/IR/Linalg.h"
-#include "mlir/Dialect/MemRef/IR/MemRef.h"
-#include "mlir/Dialect/Tensor/IR/Tensor.h"
#include "mlir/IR/AffineExpr.h"
#include "mlir/IR/AffineExprVisitor.h"
#include "mlir/IR/AffineMap.h"
diff --git a/mlir/lib/Dialect/Tensor/IR/TensorOps.cpp b/mlir/lib/Dialect/Tensor/IR/TensorOps.cpp
index e741144647043..72e9d4d9c64d9 100644
--- a/mlir/lib/Dialect/Tensor/IR/TensorOps.cpp
+++ b/mlir/lib/Dialect/Tensor/IR/TensorOps.cpp
@@ -10,7 +10,7 @@
#include "mlir/Dialect/Arith/IR/Arith.h"
#include "mlir/Dialect/Arith/Utils/Utils.h"
#include "mlir/Dialect/Complex/IR/Complex.h"
-#include "mlir/Dialect/Linalg/IR/Linalg.h"
+#include "mlir/Dialect/Linalg/IR/LinalgInterfaces.h"
#include "mlir/Dialect/Tensor/IR/Tensor.h"
#include "mlir/Dialect/Tensor/Utils/Utils.h"
#include "mlir/Dialect/Utils/IndexingUtils.h"
diff --git a/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel b/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel
index 92aedac837197..60f66b154e313 100644
--- a/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel
@@ -7763,7 +7763,6 @@ td_library(
name = "TensorOpsTdFiles",
srcs = [
"include/mlir/Dialect/Tensor/IR/TensorBase.td",
- "include/mlir/Dialect/Tensor/IR/TensorInterfaces.td",
"include/mlir/Dialect/Tensor/IR/TensorOps.td",
],
includes = ["include"],
@@ -7813,23 +7812,6 @@ gentbl_cc_library(
deps = [":TensorOpsTdFiles"],
)
-gentbl_cc_library(
- name = "TensorInterfacesIncGen",
- tbl_outs = [
- (
- ["--gen-op-interface-decls"],
- "include/mlir/Dialect/Tensor/IR/TensorInterfaces.h.inc",
- ),
- (
- ["--gen-op-interface-defs"],
- "include/mlir/Dialect/Tensor/IR/TensorInterfaces.cpp.inc",
- ),
- ],
- tblgen = ":mlir-tblgen",
- td_file = "include/mlir/Dialect/Tensor/IR/TensorInterfaces.td",
- deps = [":TensorOpsTdFiles"],
-)
-
cc_library(
name = "TensorDialect",
srcs = [
@@ -7859,13 +7841,13 @@ cc_library(
":InferIntRangeInterface",
":InferTypeOpInterface",
":InliningUtils",
+ ":LinalgInterfaces",
":LoopLikeInterface",
":ParallelCombiningOpInterface",
":ShapedOpInterfaces",
":SideEffectInterfaces",
":SubsetOpInterface",
":Support",
- ":TensorInterfacesIncGen",
":TensorOpsIncGen",
":TilingInterface",
":TransformDialectInterfaces",
@@ -11206,6 +11188,23 @@ gentbl_cc_library(
deps = [":LinalgOpsTdFiles"],
)
+gentbl_cc_library(
+ name = "LinalgRelayoutOpsIncGen",
+ tbl_outs = [
+ (
+ ["-gen-op-decls"],
+ "include/mlir/Dialect/Linalg/IR/LinalgRelayoutOps.h.inc",
+ ),
+ (
+ ["-gen-op-defs"],
+ "include/mlir/Dialect/Linalg/IR/LinalgRelayoutOps.cpp.inc",
+ ),
+ ],
+ tblgen = ":mlir-tblgen",
+ td_file = "include/mlir/Dialect/Linalg/IR/LinalgRelayoutOps.td",
+ deps = [":LinalgOpsTdFiles"],
+)
+
gentbl_cc_library(
name = "LinalgEnumsIncGen",
tbl_outs = [
@@ -11532,10 +11531,50 @@ cc_library(
],
)
+cc_library(
+ name = "LinalgInterfaces",
+ srcs = [
+ "include/mlir/Dialect/Linalg/IR/Linalg.h",
+ "lib/Dialect/Linalg/IR/LinalgInterfaces.cpp",
+ ],
+ hdrs = ["include/mlir/Dialect/Linalg/IR/LinalgInterfaces.h"],
+ includes = ["include"],
+ deps = [
+ ":AffineDialect",
+ ":ArithDialect",
+ ":ArithUtils",
+ ":BytecodeOpInterface",
+ ":ComplexDialect",
+ ":ControlFlowInterfaces",
+ ":CopyOpInterface",
+ ":DestinationStyleOpInterface",
+ ":DialectUtils",
+ ":IR",
+ ":InferTypeOpInterface",
+ ":LinalgEnumsIncGen",
+ ":LinalgInterfacesIncGen",
+ ":LinalgOpsIncGen",
+ ":LinalgRelayoutOpsIncGen",
+ ":LinalgStructuredOpsIncGen",
+ ":SideEffectInterfaces",
+ ":Support",
+ ":TilingInterface",
+ ":ViewLikeInterface",
+ "//third_party/llvm/llvm-project/llvm:Support",
+ ],
+)
+
cc_library(
name = "LinalgDialect",
- srcs = glob(["lib/Dialect/Linalg/IR/*.cpp"]),
- hdrs = glob(["include/mlir/Dialect/Linalg/IR/*.h"]),
+ srcs = [
+ "lib/Dialect/Linalg/IR/LinalgDialect.cpp",
+ "lib/Dialect/Linalg/IR/LinalgOps.cpp",
+ "lib/Dialect/Linalg/IR/ValueBoundsOpInterfaceImpl.cpp",
+ ],
+ hdrs = [
+ "include/mlir/Dialect/Linalg/IR/Linalg.h",
+ "include/mlir/Dialect/Linalg/IR/ValueBoundsOpInterfaceImpl.h",
+ ],
includes = ["include"],
deps = [
":AffineDialect",
@@ -11554,9 +11593,10 @@ cc_library(
":InferTypeOpInterface",
":InliningUtils",
":LinalgEnumsIncGen",
- ":LinalgInterfacesIncGen",
+ ":LinalgInterfaces",
":LinalgNamedStructuredOpsYamlIncGen",
":LinalgOpsIncGen",
+ ":LinalgRelayoutOpsIncGen",
":LinalgStructuredOpsIncGen",
":MathDialect",
":MemRefDialect",
@@ -11568,6 +11608,7 @@ cc_library(
":SubsetOpInterface",
":Support",
":TensorDialect",
+ ":TensorUtils",
":TilingInterface",
":ValueBoundsOpInterface",
":ViewLikeInterface",
@@ -11599,6 +11640,7 @@ cc_library(
":IR",
":IndexDialect",
":LinalgDialect",
+ ":LinalgInterfaces",
":LinalgMatchOpsIncGen",
":LinalgTransformEnumsIncGen",
":LinalgTransformOpsIncGen",
@@ -11710,6 +11752,7 @@ cc_library(
":IR",
":IndexDialect",
":LinalgDialect",
+ ":LinalgInterfaces",
":LinalgPassIncGen",
":LinalgStructuredOpsIncGen",
":LinalgUtils",
diff --git a/utils/bazel/llvm-project-overlay/mlir/test/BUILD.bazel b/utils/bazel/llvm-project-overlay/mlir/test/BUILD.bazel
index 3e6114abfc078..9b005b206a101 100644
--- a/utils/bazel/llvm-project-overlay/mlir/test/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/mlir/test/BUILD.bazel
@@ -425,6 +425,7 @@ cc_library(
"//mlir:LLVMDialect",
"//mlir:LLVMIRToLLVMTranslation",
"//mlir:LinalgDialect",
+ "//mlir:LinalgInterfaces",
"//mlir:LoopLikeInterface",
"//mlir:MemorySlotInterfaces",
"//mlir:Pass",
|
|
Replaces #127533 |
rengolin
approved these changes
Feb 17, 2025
aaronmondal
approved these changes
Feb 17, 2025
banach-space
approved these changes
Feb 17, 2025
There was a problem hiding this comment.
LGTM, thanks for the quick fix!
Let me just note that we should probably revisit the underlying layering issue at some point. This is addressing Bazel build issue specifically, but ideally Tensor should not require anything from Linalg. But that's something that will apply to other dialects as well.
chsigg
added a commit
that referenced
this pull request
Feb 18, 2025
github-actions bot
pushed a commit
to arm/arm-toolchain
that referenced
this pull request
Feb 18, 2025
This was referenced Feb 18, 2025
chsigg
added a commit
that referenced
this pull request
Feb 18, 2025
The BUILD file changes in #127544 adds `LinalgInterfaces` which is incomplete without `LinalgDialect`. For now, just add the `LinalgDialect` as dependency to tests which do not otherwise depend on it (but depend on `LinalgInterfaces` through e.g. `TensorDialect`). This is a temporary solution until the dependency of `TensorDialect` is trimmed to just the `linalg::RelayoutOpInterface`, but not the other linalg interfaces. See #127544 (review).
github-actions bot
pushed a commit
to arm/arm-toolchain
that referenced
this pull request
Feb 18, 2025
The BUILD file changes in llvm/llvm-project#127544 adds `LinalgInterfaces` which is incomplete without `LinalgDialect`. For now, just add the `LinalgDialect` as dependency to tests which do not otherwise depend on it (but depend on `LinalgInterfaces` through e.g. `TensorDialect`). This is a temporary solution until the dependency of `TensorDialect` is trimmed to just the `linalg::RelayoutOpInterface`, but not the other linalg interfaces. See llvm/llvm-project#127544 (review).
|
+1, do we have a tracking issue for this? (Removing of linalg deps tensor side) |
Now we do #127668 😄 |
wldfngrs
pushed a commit
to wldfngrs/llvm-project
that referenced
this pull request
Feb 19, 2025
Introduces a `LinalgInterfaces` target so that `TensorDialect` doesn't need to depend on `LinalgDialect`, which would introduce a circular dependency.
wldfngrs
pushed a commit
to wldfngrs/llvm-project
that referenced
this pull request
Feb 19, 2025
wldfngrs
pushed a commit
to wldfngrs/llvm-project
that referenced
this pull request
Feb 19, 2025
The BUILD file changes in llvm#127544 adds `LinalgInterfaces` which is incomplete without `LinalgDialect`. For now, just add the `LinalgDialect` as dependency to tests which do not otherwise depend on it (but depend on `LinalgInterfaces` through e.g. `TensorDialect`). This is a temporary solution until the dependency of `TensorDialect` is trimmed to just the `linalg::RelayoutOpInterface`, but not the other linalg interfaces. See llvm#127544 (review).
dorde-antic
referenced
this pull request
in dorde-antic/rocMLIR
Mar 13, 2025
…a4db
777317efa4db merge main into amd-staging (#872)
cebd37351b9b Fix missing sync in init (#870)
2559d6ed7f24 merge main into amd-staging
ae1dbe7dc4db [offload][amdgpu] Do small host mem copies synchronously (#859)
f409340cc217 [lldb-dap] Return an llvm::Error instead of calling exit directly (NFC) (#128951)
2761d4ca828a Revert "[MLIR][LLVMIR] Import unregistered intrinsics via llvm.intrinsic_call" (#128973)
d708bfb3c0be AMDGPU: Fix si-fix-sgpr-copies asserting on VReg_1 phi (#128903)
829e2a552618 [libc][hdrgen] Allow to treat hdrgen Python code as a Python module. (#128955)
524711c344b4 [libc][bazel] Add targets for strfrom<float> (#128956)
26ac7429d1d6 [memprof] std::move matchings (NFC) (#128933)
f3b4d94f35ee [compiler-rt][rtsan] truncate/ftruncate interception. (#128904)
effd7f04b678 [gn build] Port 8fb88f568011
ff80bdcf7349 [RISCV] Adding missing P600 sched model test for RVV segmented loads/stores
1559a65efaf3 Revert "[AMDGPU] Handle memcpy()-like ops in LowerBufferFatPointers (#126621)"
d584d1f18855 [MLIR][LLVMIR] Import unregistered intrinsics via llvm.intrinsic_call (#128626)
8fb88f568011 [clang][modules] Separate parsing of modulemaps (#119740)
39bab1de3333 [SLP]Check if the operand for removal is the reduction operand, awaiting for the reduction
02128342d281 Revert "DAG: Preserve range metadata when load is narrowed" (#128948)
147d9d6915cd [WebAssemblyLowerEmscriptenEHSjLj] Avoid setting import_name where possible (#128564)
469757efafeb [AMDGPU] Handle memcpy()-like ops in LowerBufferFatPointers (#126621)
42526d240cc9 [mlir][AMDGPU] Plumb address space 7 through MLIR, add address_space attr. (#125594)
7371f691b979 [MLIR][Vector]: Generalize conversion of `vector.insert` to LLVM in line with `vector.extract` (#128915)
d77520595aa9 merge main into amd-staging (#862)
eb84c1181eee [flang][cuda] Add more math intrinsic interfaces in cudadevice (#128931)
418a9872851e [SLP]Do not use node, if it is a subvector or buildvector node
7f482aa848c5 [clang modules] Setting `DebugCompilationDir` when it is safe to ignore current working directory (#128446)
364b97f23b4d [AMDGPU][True16][CodeGen] 16bit spill support in true16 mode (#128060)
7b6abd827ff2 [LV] Remove stray check lines after be28365ca78.
579ead1a69f2 [libc][bazel] Add targets for stdbit functions (#128934)
177ede2122b8 [mlir][tosa] Rename ReduceProd to ReduceProduct (#128751)
ca5bb238d05e [mlir][tosa] Change zero points of convolution ops to required inputs (#127679)
be28365ca78e [LV] Generate check lines for if-conversion.ll
2c36411ed26e [libcxx] Add LWG4135: The helper lambda of std::erase for list should specify return type as bool (#128358)
1e246704e23e [lldb-dap] Use existing lldb::IOObjectSP for DAP IO (NFC). (#128750)
cfdeca394e8c [gn build] Port 8dd8e5f7d692 (BuiltinTemplates.td)
f6703a4ff569 [ctxprof] don't inline weak symbols after instrumentation (#128811)
1be48fdf8bb2 [mlir][TosaToLinalg] Fix TosaToLinalg to restrict `tosa.cast` types to integer or float (#128859)
c690b3065d58 [Bazel] Port 128541 (#128809)
74306afe87b8 Fix the schedule of vectorizer improvement monthly sync
5a5a9e79369a [Telemetry] Fix a warning
4059faf61355 [TableGen] Update comment for size of NumToSkip field in DecoderEmitter. NFC
159b872b3736 [llvm][telemetry]Change Telemetry-disabling mechanism. (#128534)
317461ed6100 [lldb-dap] Avoid a std::string allocation for the help output (NFC)
870b376f0059 [DirectX] Support the CBufferLoadLegacy operation (#128699)
78b6c3a32ecd merge main into amd-staging
5d501c613797 [RISCV][Docs] RISCV -> RISC-V in RISCVUsage.rst. NFC (#128906)
722c7c0b0f9a [flang][Semantics] Ensure deterministic mod file output (#128655)
7717a549e91c [libc++] Optimize ranges::equal for vector<bool>::iterator (#121084)
7ffeab3121c9 [LLD][ELF] Generically report "address assignment did not converge" (#128774)
c0abae33d6e7 [MLIR][OPENMP] Relax requirement about branches as terminator of private alloc (#128481)
8ffda96dbede [RISCV] Update MicroOpBufferSize in P400 & P600 scheduling models (#128786)
f161b1b5265b [libc++][test] Refactor tests for rotate and rotate_copy (#126458)
8039f8e139aa [RISCV][MC] Add assembler support for XRivosVisni (#128773)
dfda75f2e55a [AMDGPU][True16][CodeGen] fix test for true16 codegen valu op (#128905)
26be07b8511b [libc++][format] Disables narrow string to wide string formatters. (#128355)
a841cf91b3e0 [lib++][print] Don't pad the ostream output. (#128354)
d7b3606f7f86 [libc++] Updates ostream's println LWG status. (#128214)
ffc5d2b5d46f [libc++][test] Refactor tests for ranges::swap_range algorithms (#121138)
6c2e170d043d [NVPTX] Convert vector function nvvm.annotations to attributes (#127736)
14da7d5c1fc6 Match .exe on Windows (#128894)
1d583ed2fb76 [libc++][test] Augment ranges::{fill, fill_n, find} with missing tests (#121209)
1ec1d25f691b [MachineOutliner] Add skipModule call for opt-bisect-limit. (#128836)
15ee9d91fbb5 [lldb] Build the API unittests with -Wdocumentation (#128893)
947108619817 [offload][amdgpu] Do small host mem copies synchronously
a955426a16bc [clang][bytecode] Handle UsingDirectiveDecls (#128888)
bb62af7d14f7 [AMDGPU][True16][CodeGen] true16 codegen for valu op (#124797)
30b021ffa483 [lldb] Deindent UnwindAssemblyInstEmulation (#128874)
eeb8c2085fb9 [X86] Fix a warning
de10e44b6fe7 Thread Safety Analysis: Support warning on passing/returning pointers to guarded variables
3c8c0d4d8d9b Thread Safety Analysis: Handle address-of followed by dereference
5c8e22bb2653 [bazel] Export BuiltinTemplates.inc from clang:basic
fd08b0793fbb [bazel] Port 8dd8e5f7d692cc43f4322f04034f5c472381aa43
ea294e3f1d3c [MLIR][Affine] Make isValidLoopInterchangePermutation efficient (#128863)
ae61c960e183 [offload][amdgpu] Fix warnings in rtl.cpp
35bf925f7ea9 [RISCV] Delete dead COPYs to vmv0 during vmv0 elimination
1b17d1ee6e6c [X86] Allow select(cond,pshufb,pshufb) -> or(pshufb,pshufb) fold to peek through bitcasts (#128876)
08a31b43f358 Add reverted commit to `revert_patches.txt` (#857)
5eeb3e2a0f79 Sets the value for rpc: in LIBOMPTARGET_KERNEL_TRACE (#813)
0f7a60b312f7 Cleanup how to search for ockl.bc & ocml.bc in DeviceRTL/CMakeLists.txt (#814)
4ecc0ee93df8 merge main into amd-staging (#852)
7f332423b090 [MemCpyOpt] Add stack move test with ret-only capture (NFC)
8dd8e5f7d692 [Clang] Add BuiltinTemplates.td to generate code for builtin templates (#123736)
defe43bbffb0 Add unsigned integer overloads for abs (#128257)
2c1df2206189 RegAllocFast: Fix 8634635d689c5a7adfb19cde4a313d7c02e95194 to not trip assertions
56379b29042d Simplify flip() for std::bitset (#120807)
0f6240c4ddc8 [HLSL] Allow EmptyDecl in cbuffer/tbuffer (#128250)
8634635d689c RegAllocFast: Stop reading uninitalized memory
4277c21059a8 [VPlan] Introduce explicit broadcasts for live-ins. (#124644)
15fbdc2b9635 [AArch64][SVE] Lower unpredicated loads/stores as LDR/STR. (#127837)
a5d8b7aeb6b3 [AArch64] Improve urem by constant costs (#122236)
3c4fa5a20aff [MergeFunc] Add tests showing incorrect handling of metadata call args.
83ccab35d4ae PeepholeOpt: Remove pointless check for subregister def (#128850)
e2983ac81147 Add reverted commit to `revert_patches.txt`
0f0d3fb6b59b [AMDGPU] Do not allow M0 as v_readlane_b32 dst (#128867)
aace6a2f9d8b [RISCV] Xqcia 0.4 The spec was recently updated, this changes the name in the TD files associated and increments the Extension number in the clang driver. This is mostly a MC change as there is no other generated code for these instructions yet.
8138d85f6307 [analyzer] Update the undefined assignment checker diagnostics to not use the term 'garbage' (#126596)
a00586171cdf [clang-tidy]improve performance-unnecessary-value-param performance (#128383)
523173632922 [AMDGPU] Do not allow M0 as v_readfirstlane_b32 dst (#128851)
5f4d1f74004d [libclc] Make CLC library warning-free (#128864)
900220d44425 [CostModel] Handle vector struct results and cost `llvm.sincos` (#123210)
178b9e5375dd [MergeFunc] Add linkonce test with discardable functions.
d5038b377448 [libclc] Move __clc_ldexp to CLC library (#126078)
01371d64a91e [LLVM][AArch64] Reduce uses of "undef" in SVE CodeGen tests.
6f2345a20e36 [LLVM][AArch64] Change SVE CodeGen tests to use splat().
575656877f1f [LLVM][AArch64] Reduce uses of "undef" in SVE InstCombine tests.
28cf323e8717 [LLVM] Port a few InstCombine tests to use splat instead of shufflevector.
0ba2000b3cec [mlir][tosa] Enhance the conv2d verifier (#128693)
a98c2940dbc0 [X86] Handle multiple use freeze(undef) in LowerAVXCONCAT_VECTORS as zero vectors (#128830)
5cbff437fadd [InstCombine] Test for trunc to i1 in foldLogOpOfMaskedICmps.
13245cea1105 [lldb] Modernize ABI-based unwind plan creation (#128505)
5aec7b4cdb67 merge main into amd-staging
2d12c9e83f5a [bazel] add missing header for RelayoutOptInterface
ec9c2935e191 [MLIR][Bufferization] Remove `GEN_PASS_DEF_BUFFERIZATIONBUFFERIZE` (#128842)
e264b0e85627 [ProfileData] Avoid repeated hash lookups (NFC) (#128829)
b2c8f66eea81 [Passes] Avoid repeated hash lookups (NFC) (#128828)
67d92cf38416 [ExecutionEngine] Avoid repeated hash lookups (NFC) (#128827)
e49c8d5d3d40 [DebugInfo] Avoid repeated map lookups (NFC) (#128826)
bd9e31ef1ea3 [DWARFLinker] Avoid repeated hash lookups (NFC) (#128825)
b8d1f3d62746 [Clang] Fix an integer overflow issue in computing CTAD's parameter depth (#128704)
d8bcb53780bf DAG: Preserve range metadata when load is narrowed (#128144)
1a114fa302b4 RegAlloc: Use new approach to handling failed allocations (#128469)
e160c35c9ec6 Reapply "RegAlloc: Fix verifier error after failed allocation (#119690)" (#128400)
92d822245b0f [LoongArch] Pre-commit tests for vector sext & zext (#128835)
ae839b02504a [clangd] [C++20] [Modules] Add scanning cache (#125988)
a522c227a1d7 [mlir][vector] Move tests for `rewriteAlignedSubByteInt{Ext|Trunc}` (nfc) (#126416)
cdfcce48d5c2 [Passes] Fix a warning
2015626783aa [Clang] Implement CWG2918 'Consideration of constraints for address of overloaded function' (#127773)
29c5e4289f53 AMDGPU: Add baseline tests for bitcast + readlane intrinsics (#128493)
600c244572a4 merge main into amd-staging (#848)
3f648992bf31 [clang][bytecode] Fix initing incomplete arrays from ImplicitValueIni… (#128729)
8dd609598e49 Support: Do not check if a file exists before executing (#128821)
fe13cb985c77 [CodeGen][NewPM] Port RegAllocGreedy to NPM (#119540)
75aff78f64d2 RegAllocFast: Fix verifier errors after assigning to reserved registers (#128281)
b5dd1fedc5dc VirtRegRewriter: Fix verifier errors after regalloc failures (#128280)
98542a3d6d08 [mlir][Vector] Move vector.extract canonicalizers for DenseElementsAttr to folders (#127995)
e3ece07593b3 Revert "Reland "[AArch64][NPM] Chalk out the CodeGenPassBuilder for NPM (#128…" (#128819)
e927cf6653a9 Reland "[AArch64][NPM] Chalk out the CodeGenPassBuilder for NPM (#128… (#128662)
a63ce5ef445a [NFC] SWDEV-517818 - Switch back static_assert to assert to ensure compatibility with older compiler (#841)
472ea0b7821f [RISCV] Merge some of the Sifive decoder tables. (#128794)
852923822fd0 [AMDGPU][NewPM] Port AMDGPUInsertDelayAlu to NPM (#128003)
31897e651a1a [LTO][Pipelines][Coro] De-duplicate Coro passes (#128654)
8fc8a84e2347 [clang-format] Allow breaking before kw___attribute (#128623)
c53eb93dd7e9 PeepholeOpt: Immediately check if a reg_sequence compose supports a subregister (#128279)
42e70446b1eb merge main into amd-staging
01cc1d13cd0c [RISCV] Use Priv tablegen class for sf.cease instruction.
a565f9eb2997 [RISCV] The test for vp.reduce.fminimum/fmaximum with fixed-length should stay in fixed-vectors-reduction-fp-vp.ll. (NFC)
439de05848b2 [RISCV] Rename function name to start with prefix vpreduce for consistency. (NFC)
da37c76ac621 [mlir][vector] Add a check to ensure input vector rank equals target shape rank (#127706)
b1a735b45dcc [mlir][math] expand-math pass assumes the static shaped type (#128299)
e350485595d0 [flang][cuda] Set alloca block in cuf kernel (#128776)
f3000d7d27fa [flang][cuda] Do not trigger automatic deallocation in main (#128789)
c8b40867d144 [AMDGPU][True16][CodeGen] test fix for uaddsat/usubsat true16 selection (#128784)
2db8386867c5 [HLSL] Implement default constant buffer $Globals (2nd attempt) (#128589)
cd4c30bb224e [HLSL][Sema] Fix Struct Size Calculation containing 16/32 bit scalars (#128086)
345a3087d40a merge main into amd-staging (#842)
09832777d830 [CMake][Release] Statically link ZSTD on all OSes (#128554)
b7060d0183f8 [DirectX] Fix printing of DXIL cbuffer info (#128698)
1b39328d7440 [CodeGen] Fix MachineInstr::isSafeToMove handling of inline asm. (#126807)
fc655b1ae783 [DirectX] Fix printing of DXIL cbuffer info (#128698)
30a7c816ee5c [LTO][Pipelines][NFC] Exctract isLTOPostLink (#128653)
864071dd7e19 Bazel fixes for a778930f85b6d17cf31ff0e15964a7c7116e2a9d (#128783)
789bfdc3e60c [OMPT] Use __tsan_init to detect TSan binaries rather than RunningOnValgrind (#128357)
65cf534139ab [flang][cuda] Add interfaces for __ldcg, __ldca, __ldcs, __ldlu, __ldcv, __stwb, __stcg, __stcs, __stwt (#128766)
40566fd674d1 [WebAssembly] Generate invokes with llvm.wasm.(re)throw (#128105)
e6f6a1e86389 [AMDGPU][True16][CodeGen] uaddsat/usubsat true16 selection in gisel (#128233)
59cee030fb9b Generalize creduce-clang-crash.py script to look for cvise (#128592)
8beec9fc4819 [libc] implement a64l (#128758)
f4a80180f141 [libc] Move fileno and fdopen to fullbuild only (#128762)
44ffeecde265 [DXIL][Analysis] Make sure resource accessors are contiguous (#128696)
ad94af973a76 [CIR] React to breaking change to DataLayoutTypeInterface (#128772)
2646c36a864a [mlir][bufferization] Change OneShotModuleBufferize to not analyze or bufferize nested symbol tables (#127726)
d5e274f4c4e0 [NFC] SWDEV-517818 - Switch back static_assert to assert to ensure compatibility with older compiler
0be3f134c3b0 [DirectX] only allow intrinsics defined in DXIL.td (#128613)
303d7fa86740 [MLIR][Interfaces] Make LoopLikeOpInterface inheritable outside of MLIR (#128743)
c79e867cd2bb [DirectX] Update CBuffer to refer to a `dx.Layout` type (#128697)
6a5dd04013a1 [Support] Try to fix AlignedCharArrayUnion with GCC 7.5
5e4938a9918a Exclude hwasan from thread_create_failure.pass.cpp (#128768)
a52579f9e0ce merge main into amd-staging
f1025e671ef1 [Support] Replace deprecated std::aligned_union, NFCI. (#127417)
eacbcbe47744 [CIR] Upstream type `bool` (#128601)
4357a6603f2c [mlir][DLTI] Make `getPreferredAlignment` default to `getABIAlignment` (#128754)
00f02fed8828 [RISCV] Change the vendor prefix for Rivos from "rv." to "ri." (#128761)
f22291c791c8 [RISCV][NFC] Merge Xqci Decoder Tables (#128140)
c8136da26c56 [RISCV] Correctly Decode Unsigned Immediates with Ranges (#128584)
9102afcd0146 [WebAssembly] Use the same lowerings for f16x8 as other float vectors. (#127897)
7c266756ad2e [gn build] Port 66af4923ce24
66af4923ce24 [lldb-dap] Refactor reverse request response handlers (NFC) (#128594)
ab0e6fcaadf1 [libc][cmake] Clean up dead code in add_gen_header (#128753)
f5675243995d [BOLT] Fix doTrace in BAT mode (#128546)
3968ebd00da8 [BOLT] Keep multi-entry functions simple in aggregation mode (#128253)
a778930f85b6 [mlir][linalg] Create a dedicated target for `LinalgRelayoutInterface` (#128485)
67056c280a71 [WebAssembly] Support shuffle for F16x8 vectors. (#127857)
43999deb3701 [spirv][amdgpu] Set atomic size in the clang target info (#128569)
0a7809c64448 [mlir][tosa] Fix ability to expand ranks with dynamic shape support (#128037)
4f18f3f09a74 [RISCV] Use addiw for or_is_add when or input is sign extended. (#128635)
48db4e8377f8 [mlir][tosa] Change Transpose perms operand to attribute (#128115)
d2d469eb7981 [WebAssembly] Make llvm.wasm.throw invokable (#128104)
041b7f508533 [Wunsafe-buffer-usage] Turn off unsafe-buffer warning for methods annotated with clang::unsafe_buffer_usage attribute (#125671)
24b7759a9dfe [FLANG][OpenMP]Add frontend support for ASSUME and ASSUMES (#120770)
f08824b93543 [AArch64] Add udiv and urem uniform tests. NFC
53b46bb09474 [mlir][tosa] Fix crash on attempt to fold int_div by zero (#128682)
e58f475e8454 [mlir][tosa] Move cond_if and while_loop operations to controlflow extension (#128216)
8bea51103000 [X86] Fold AND(Y, XOR(X, SUB(0, X))) to ANDN(Y, BLSMSK(X)) (#128348)
9889de834b0a [Utils] Avoid repeated hash lookups (NFC) (#128634)
38f8ca1d1817 [DebugInfo] Avoid repeated hash lookups (NFC) (#128632)
43401dd0b5c6 [libc++] Make .verify.cpp tests more robust against changing headers (#128703)
9388e42a3c67 [CodeGen] Avoid repeated hash lookups (NFC) (#128631)
791da3c5c2ef [AsmPrinter] Avoid repeated hash lookups (NFC) (#128630)
7501c9c0e124 [AsmParser] Avoid repeated map lookups (NFC) (#128629)
f876d2baf932 [COMGR] Fix compiling tests on Windows with msvc (#809)
568106c2150f [lldb][NFC] Fix comment in lldb/Core/ModuleList.h (#128602)
99207ae835ef [mlir] Fix a warning
a3af9dc07ad8 merge main into amd-staging (#836)
37559c8401cf [Hexagon] Handle Call Operand vxi1 in Hexagon Backend (#128027)
a821ae284724 [libclc] Move round to CLC library (#128721)
cf3b0368a55c [mlir][tosa][tosa-to-linalg] Add NaN Mode Lowering (#125668)
83c6b1a88852 [lldb] Add fzf_history command to examples (#128571)
f10e0f7321b3 [MergeFuncs] Don't introduce calls to (linkonce,weak)_odr functions. (#125050)
5fd188833c4c [mlir][spirv] Refactor image operations (#128552)
85eb7259d9e1 [clang] Fix use-after-scope when diagnosting __attribute__((format_matches))
148111fdcf0e [CMake][Release] Enable bolt optimization for clang on Linux (#128090)
1affadb7c662 AMDGPU: Drop legacy r600.read.global.size intrinsics from amdgcn (#128700)
a12ca57c1cb0 [SLP][REVEC] Add getScalarizationOverhead helper function to reduce error when REVEC is enabled. (#128530)
4f7d8948d9d9 [Hexagon] Add a case to BitTracker for new register class (#128580)
352c48f278c8 [SelectionDAG] Utilizing target hook convertSelectOfConstantsToMath for SelectwithConstant (#127599)
af68927a831c Do not treat llvm.fake.use as a debug instruction (#128684)
1e0e4169dd00 [libclc][NFC] Remove unused intrinsics helpers (#128708)
522b05afb636 [VPlan] Construct immutable VPIRBBs for exit blocks at construction(NFC) (#128374)
d23da7d6300e [InstCombine] Increase recursion limit to 3 in `simplifyNonNullOperand` (#128695)
44d1dbd24c20 [X86][DAGCombiner] Skip x87 fp80 values in `combineFMulOrFDivWithIntPow2` (#128618)
d21b2e619a5e [SPIR-V] Fix generation of gMIR vs. SPIR-V code from utility methods (#128159)
089f988f46d7 [libc] Fix defaulting the full build
4b29c285645e [ConstraintElim] Preserve analyses when IR is unchanged. (#128588)
e47cd4694851 [X86] combineX86ShuffleChain - pass IsMaskedShuffle flag as argument from combineX86ShufflesRecursively instead of computing it internally. NFC.
745a387de39b [Flang]OpenMP]respond to merge comment on flang/lib/Lower/OpenMP/OpenMP.cpp
a93cda47ad97 [X86] combineX86ShuffleChain - pull out repeated getOpcode() calls. NFC.
26273d51b029 merge main into amd-staging
11766a40972f [libc++] Don't try to wait on a thread that hasn't started in std::async (#125433)
0f9720a61b1d [bazel] port 29e14958090cb01150bda068f721a09d4bb1c36b
70de57edcad0 [clang] Add alternative email for steakhal (#128558)
3035dd876617 merge main into amd-staging (#823)
dff2ca424c20 [clang][bytecode] Add special case for anonymous unions (#128681)
f8948d3c4754 [libclc] Move log/log2/log10 to CLC library (#128540)
6aeec5eabfe1 [ConstraintElim] Test for #128588
b57e63b07a7b libclc: Stop using asm declarations for r600 on amdgcn for get_global_size (#128692)
f95ad44068e4 AMDGPU: Mark v_mov_b64_pseudo as a VOP1 instruction (#128677)
a9eec8ca2661 [flang][OMPIRBuilder] Improve debug info for OpenMP target region. (#576)
820aa438a6ec [mlir][vector] Update tests for xfer permutation lowering (4/N) (#127624)
dfa3af9255fd [clang][bytecode] Expand subscript base if of pointer type (#128511)
25c19eb1178a [Flang][OpenMP] Allow host evaluation of loop bounds for distribute (#127822)
29e14958090c [MLIR][OpenMP] Support target SPMD (#127821)
56975b4ecd18 [OpenMPIRBuilder] Split calculation of canonical loop trip count, NFC (#127820)
48397fe41ee6 [AArch64] Add cost model for REV shuffles. (#128498)
446899e7bed5 [MLIR][OpenMP] Host lowering of distribute-parallel-do/for (#127819)
9fc2f7869345 [X86][NFC] Added/Updated Trigonometric functions testcases (#127094)
88163ca79cab [OpenMPIRBuilder] Add support for distribute-parallel-for/do constructs (#127818)
7ff87af533a7 [MLIR][OpenMP] Host lowering of standalone distribute (#127817)
2a0946bc0dff [GVN][NFC] Match coding standards (#128683)
85cf95876c4b [AArch64] Improve codegen for some fixed-width partial reductions (#126529)
547a8bc2365d [clang][x86] Support -masm=intel in cpuid.h (#127331)
b36a18df96f9 [AArch64][Build Attributes] Improve Parsing and Formatting (#126530)
a4656bbc5958 [mlir][tosa] Allow conv ops zero point to be variable (#128533)
87dc245f3e65 [analyzer] Partial revert of #127017 (#128642)
275eeb56ddc4 [gn build] Port 5114b9b386ca
60cc3af0d93e [mlir][xegpu] Fix bazel build failure (#128595)
d7211693af27 [bazel] port e5ce0304335dc1cae6856c880d1d4e14dcf8265d
674dbcfe8f40 [libc++][NFC] Use TEST_STD_VER instead of _LIBCPP_STD_VER in re.tokiter.comp/equal.pass.cpp
49f60b4e0984 [AMDGPU][NFC] Added test for live-in CSR SGPR used partially giving MachineVerifier error (#126696)
0087523e1a27 [MLIR][Affine] Add missing check on fusion compute tolerance on a path (#128454)
ea4e19df53ab [SDAG] Add missing ppc_fp128 ExpandFloatRes for sincos[pi] (#128514)
5114b9b386ca [ORC][llvm-jitlink] Extend weak-linking emulation to real dylibs.
d254fa877f41 [flang-rt] Make `FLANG_RT_INSTALL_RESOURCE_PATH` configurable (#128561)
5088e1b435fd [lldb] Avoid Function::GetAddressRange in ThreadPlanStepRange::InSymbol (#128515)
3083aea44414 [GlobalISel] Avoid repeated hash lookups (NFC) (#128633)
6c17380ea896 [lldb] Fix TestDiagnoseDereferenceFunctionReturn on linux (#128512)
3872503d6eb3 [lldb] Don't hand out UnwindPlan::Row shared_ptrs (#128181)
d3dae841c05c [NVPTX] Switch to imm offset variants for LDG and LDU (#128270)
f5d80c335d79 [lldb] Avoid Function::GetAddressRange in SymbolFileCTF (#128517)
66b2a0ceeb25 merge main into amd-staging
d7903c9f28bd AMDGPU: Add more codegen tests for readfirstlane
7cadba3c97e6 merge main into amd-staging (#821)
9b298a1d3d22 [WebAssembly] Use Register instead of unsigned. NFC
e67cd152cf4d [llvm-size] Initialize Radix to correct value (#128447)
f58fde585775 Exclude RedirectingFileSystem with null OverlayFileDir in VFSUsage (#128267)
6e3b47597fab Reland "[mlir] Silence -Wdangling-assignment-gsl in OperationSupport.h"
ecc7e6ce4cd5 [Clang] Handle instantiating captures in addInstantiatedCapturesToScope() (#128478)
83ddb43cad3e Revert "[AArch64][NPM] Chalk out the CodeGenPassBuilder for NPM (#128471)"
b9cf684d7c39 [mlir][scf] Fix typo of square brackets(NFC) (#128455)
d85685eb8636 [AArch64][NPM] Chalk out the CodeGenPassBuilder for NPM (#128471)
5deb2aa9eb45 AMDGPU: Make is.shared and is.private propagate poison (#128617)
eab6f2d7a90c [MLIR][Affine] Fix fusion in the presence of cyclic deps in source nests (#128397)
b3c51db292f0 [InstrRef] Preserve debug instr num in aarch64-expand-pseudo LOADgot expansion (#128081)
366daddfad9a [Serialization] Update DECL_LAST
06f30792353a AMDGPU: More consistently use the fold list instead of direct mutation (#127612)
3d77fd56105b [COMGR] Fix compiling tests on Windows
862595cab67b [MachineBasicBlock][NFC] Decouple SplitCriticalEdges from pass manager (#128151)
2c7780a96d24 [gn build] Port 253e11695ba8
aa902a0380c1 [ORC] Add dependence on TextAPI to reflect changes in 253e11695ba.
8009c1fd81ad [LV][VPlan] Prevent calculate cost for skiped instructions in precomputeCosts(). (#127966)
cfdb15ae6b31 merge main into amd-staging (#811)
c7101188fb3f [clang] Implement __attribute__((format_matches)) (#116708)
253e11695ba8 [ORC][llvm-jitlink] Add support for emulating ld64 -weak-lx / -weak_library.
688064498a01 [ORC] Remove unused header. NFC.
28002dd50fb7 AMDGPU: Replace some undef pointer uses in test
0e8964e6e7c1 merge main into amd-staging
a60e8a2c2579 [mlir] Python: write bytecode to a file path (#127118)
62ec7b8de97a [LLDB][NFC]Renaming functions to be consistent with LLDB naming style (#128574)
724b91b46783 [flang-rt] Fixed freestanding memmove. (#128604)
305d2738944f Reland "[HLSL] Implement the reflect HLSL function" (#125599)
17d3551c2072 clang: Remove upstream diff in read_exec builtins (#804)
49c31201278a [MachineSched] Add a first valid reason [nfc]
ccbb8882ac75 [lldb] do not show misleading error when there is no frame (#119103)
fc09550bf498 [MLGO] Refactored verbosity flag in mlgo-utils to common location (#128541)
1824bb47c2b5 [Offload][OpenMP] Fix check-prefix (#128599)
f6a30021249c [lldb-dap] Remove unused headers (NFC)
162eb32e7478 [lldb-dap] Add 'source' references to stack frames without source files. (#128268)
6d0cfbc9c0e2 [PPC] Implement `areInlineCompatible` (#126562)
d46e7ff7539d [Comgr] Explicitly describe data object mutation restrictions (#746)
911e94c65169 [lldb-dap] Finish refactoring the request handlers (NFC) (#128553)
e8f1623a223a [lldb-dap] Addressing the order of events during disconnect to flush output. (#128583)
00a0b0be4b18 Reland [mlir] Fix integration test when %host_cc path contains spaces (#128542)
6b444271a011 Reland "[llvm-mt] Use XmlDeleter to free xmlFreeDoc"" (#128579)
36b339b84a98 [flang][cuda] Relax assertion for atomicexch (#128582)
594919c26312 [mlir][spirv] Split conditional basic blocks during deserialization (#127639)
0caa8f42be0b Reland "[flang] Set LLVM specific attributes to fir.call's of Fortran runtime. (#128093)"
d6ec32c8f259 [profile] runtime counter relocation is needed on windows-msvc targets (#127858)
31915248c06d [mlir] NFC: fix typos and improve commentary regarding generation of MLIRConfig.cmake (#127712)
b248817ad609 [gn build] Port 38d7cf1a8143
38d7cf1a8143 [lldb-dap] Refactor remaining request handlers (NFC)Remaining request handlers (#128551)
e5ce0304335d [mlir][xegpu] Improve XeGPU op verification logic for SIMT flavor and update tests. (#127920)
eabe2eb93355 [libc] Remove special full build handling for GPU (#128572)
baa77e30f0f2 [LV] Remove some redundant casts (NFC).
df14dbd8750f [DirectX] Fix build breaks (#128556)
a018788687c8 [gn build] Port d36231940444
d36231940444 [lldb-dap] Refactor breakpoint related request handlers (NFC) (#128550)
1b15a89a23c6 Revert "[Offload] Fix assumptions on symbols after #124846 (#126238)"
47822c80c1b2 [LangRef] Clarify that the pointer after an object must be valid. (#127892)
bef4e5204ada [flang][cuda] Fix type mismatch in atomiccas (#128548)
9bfe48695ed8 [libc] Initial support for parsing SPIR-V for GPU libc (#128570)
1e85e5abb327 [clang][WebAssembly] Always have `-pthread` imply `--shared-memory` linker flag (#127939)
6c61c557569a [BOLT] pacret-scanner: fix regression test failure (#128576)
e063365a9732 Revert "[llvm-mt] Use XmlDeleter to free xmlFreeDoc" (#128578)
23aca2f88dd5 Revert "Symbolize line zero as if no source info is available (#124846)"
53c08dfc1816 [llvm-mt] Use XmlDeleter to free xmlFreeDoc (#128472)
f63c5f36c3e3 Revert "[mlir] Fix integration test when `%host_cc` path contains spaces" (#128573)
b335d5a83032 [gn build] Port 988480323d5e
55c76ea39122 [BOLT] pacret-scanner: fix regression tests... (#128565)
cec35077025d [alpha.webkit.UnretainedLocalVarsChecker] Add a checker for local variables to NS and CF types. (#127554)
988480323d5e [lldb-dap] Refactor custom & testing related request handlers (NFC) (#128549)
8ce17c15577d [clang][NFC] Remove CXXRecordDecl::lookupDependentName() and its helpers (#128392)
914a8aca95f8 Cleanup how to search for ockl.bc & ocml.bc in DeviceRTL/CMakeLists.txt
fc4c76c9388a Sets the value for rpc: in LIBOMPTARGET_KERNEL_TRACE to 1 if rpc is used for the kernels in the image
e55f1a7ef8d3 [SLP] Add test for getSpillCost fix
9fac59a0db7c [HLSL] Allow arrays to be returned by value in HLSL (#127896)
0182b23fe144 [gn build] Port 82264d23a1cc
82264d23a1cc [lldb-dap] Refactor stepping related request handlers (NFC) (#128453)
4c9e14b3ad64 [AMDGPU] Update PromoteAlloca to handle GEPs with variable offset. (#122342)
607a1f2ace77 [CIR] Add cir-opt tool to exercise CIR dialect parsing (#128254)
be5c66d97d79 [clang] Improve module out of date error message (#128103)
4ac43b541c06 [LV] Restrict widest induction type to be IntegerType (NFC) (#128173)
806baca15044 merge main into amd-staging
ab9cd53b86e8 [Wunsafe-buffer-usage] False positives for & expression indexing constant size array (arr[anything & 0]) (#112284)
69cc16fb5508 Revert "[flang] Set LLVM specific attributes to fir.call's of Fortran runtime. (#128093)"
4d536169010e Revert "[flang] Workaround build failure."
823a597d2ad0 [flang] Workaround build failure.
5ea58a4bddfe [Flang][MLIR][OMPIRBuilder] Upstream target SPMD conflict resolution (#803)
41cd6d2b1d63 [gn build] Port 850b49297615
d9b0571b5c40 [gn build] Port 363bfd6090b0
51ce6c437fec [lldb-dap] Fix error C2065: 'PATH_MAX': undeclared identifier
9638d08af96c [NVPTX] Support for memory orderings for cmpxchg (#126159)
d9d1f241f27b [gn build] Port d0e37d972331 (llvm-dap tweaks)
e298fc2da971 Add DISubrangeType (#126772)
902b8fd0037f Replace hostexec with upstream rpc (#802)
8dbc393e4472 [flang][cuda][NFC] Remove shared alloc addr space (#128535)
571b787b83cb [CodeGen] Change copyPhysReg interface to use Register instead of MCRegister. (#128473)
664cbd1b5db1 [lldb-dap] skip TestDAP_server on windows to unblock CI. (#128278)
a0be17dc91d0 [RISCV] Remove unnecessary entries from RISCVVInversePseudosTable. NFC (#128376)
7f6f186736c8 [SPIRV] Stop including SPIRVInstrInfo.h in MCTargetDesc. NFC (#128443)
a6abbe03094a [test] Remove `-emit-llvm` from the `and-errors.hlsl` test to avoid writing to a potentially write-protected directory (#128047)
36fdeb2aded0 [flang] Set LLVM specific attributes to fir.call's of Fortran runtime. (#128093)
cc7f22ee6ccb [object][WebAssembly] Add support for RUNTIME_PATH to yaml2obj and obj2yaml (#126080)
f1252f539ca2 [PPC][MC] Restore support for case-insensitive register names (#128525)
20fd7f0a76ae Remove floating-point working group meeting (#128258)
63af27190be7 [NFC] [test] assert padding in memtag-globals test (#128259)
e89cd500b252 [bazel] Port 7a4cb9bac50c8c19ec0d4ab7f186ef086064a549
aab07d8ca644 fixed #95641 pointless string copy (#127325)
ff7790e6dde7 [MLIR][OpenMP] Simplify definition of the BlockArgOpenMPOpInterface, NFC (#128198)
7a4cb9bac50c [clang-tidy][NFC] Expose HeuristicResolver::lookupDependentName() and use it in StandaloneEmptyCheck (#128391)
4defac91dbdf [mlir][GPUToNVVM] Add `benefit` to `populate` functions (#128484)
5bddadf783c1 [CodingStandard] Rework anonymous namespace section to cover visibility more broadly (#126775)
cebb8f72b793 [OpenMPIRBuilder] Add support for distribute constructs (#127816)
538b898a836a [RISCV] Add Qualcomm uC Xqcilia (Large Immediate Arithmetic) extension (#124706)
b66ec64b5b63 RegAllocGreedy: Remove unnecessary null register class check (#128487)
17ccaf4fa82e [NFC][Offload] Fix typo to output architecture (#128527)
529b3d16daf2 [RISCV][TTI] Remove SK_Select from manual splitting in getShuffleCost
16f9c5da45b8 [SPIR-V] Stop generating StorageImageReadWithoutFormat and StorageImageWriteWithoutFormat for the Unknown image format in the OpenCL environment (#128497)
0b52aa1bdbc7 [flang] Unsupported rounding modes (#128240)
3a6108bcac26 [SLP][REVEC] Fix scalar mask is passed to getScalarizationOverhead but the type is vector. (#128476)
2dfb29a9b2f6 [libclc] Move nan to the CLC library (#128521)
a88167a60d0b [MLIR] Improve error handling when parsing dense attributes (#128523)
f47b06a6998b clang: Remove upstream diff in read_exec builtins
7262a1ea3136 [flang] Set compile definitions for flang-rt build on AIX (#127919)
eb14d2a1d48f [SLP]Fix check for matched gather node, if it is a subvector node
22a5bb32b787 [analyzer] Limit Store by region-store-binding-limit (#127602)
6ac4900fe0c6 [Flang][MLIR][OMPIRBuilder] Upstream target SPMD conflict resolution
96c87a16a507 [MLIR][NFC] Fix comment in `Region.cpp`
aa770857eead [MergeFunc] Add tests for (merging weak|linkonce)_odr functions.
845edee678a1 [HeterogeneousDwarf] Thread-focus vector registers in AsmPrinter (#765)
363bfd6090b0 [libc++] Use the new locale base API on Linux (#128007)
de473fcdea70 [libc++] Synchronize Docker container SHA in run-buildbot-container (#128225)
8199c2d427fc [flang][NFC] update some old tests to HLFIR lowering (#127230)
4dda51261a6e Replace hostexec with upstream rpc
7d4d3ca9f2bd remove flang-legacy from amdflang* (#790)
4ba3ebef642e On Windows, remove the UCRT libraries from the release script (#128378)
d5148f000a92 [X86] Fix arithmetic error in extractVector (#128052)
c83bdc7c111f [MLIR][OpenMP] Normalize lowering of omp.loop_nest (#127217)
a5319ce4a085 merge main into amd-staging (#795)
72768d9bb8ad [ConstraintElim] Teach checkAndReplaceCondition about samesign (#128168)
0f34b656f01a InstCombine: Remove a check for pointer bitcasts (#128491)
aca8a5cb2325 [libclc] Remove clspv-specific clc conversions (#128500)
971fc4225470 TargetTransformInfo: Add missing consts to a couple of methods (#128492)
1ed359e68265 [Dexter] Remove clang-opt-bisect analyzer tool (#128165)
c80b99d98ad0 [Sema][NFC] Move two misplaced uninit tests to clang/test/SemaCXX (#128013)
e1cc5e9f9ab6 [clang-tidy] Fix build after 3dc159431be7a8c5f1a26a8bd57794f1c7008969
6aea6308d125 AMDGPU: Fix creating illegally typed readfirstlane in atomic optimizer (#128388)
9b52d9e18662 [MLIR][OpenMP] Prevent loop wrapper translation crashes (#115475)
db40592de60f MachineFunction: Remove null check on TargetRegisterInfo (#128480)
7de64925da9f [DAG] shouldReduceLoadWidth - hasOneUse should check just the loaded value - not the chain (#128167)
9cbdcfcafdc0 [CaptureTracking] Remove StoreCaptures parameter (NFC)
61fb95410953 [mlir][Vector] Improve support for vector.extract(broadcast) (#116234)
5a2bee04d088 [llvm-objdump]Correct .dynstr finding of getDynamicStrTab() (#127975)
3dc159431be7 [analyzer] Clean up slightly the messed up ownership model of the analyzer (#128368)
e7ad07ffb846 [libclc] Move fma to the CLC library (#126052)
3532651b6fc1 RegAllocGreedy: Add braces
35c90bfdda0d [RISCV] Remove virtual from getOutliningTypeImpl override. NFC (#128482)
c5f40bf024ee [InstCombine] Fold `X!=Y ? ctz(X^Y, true) : BW -> ctz(X^Y, false)` (#128483)
25948fbc2d22 merge main into amd-staging
6c2d418027be [libclc] Fix int<->float conversion builtins (#126905)
769ee86d51c1 merge main into amd-staging
7a9f53cecf57 [X86] combineBROADCAST_LOAD - merge across chains (REAPPLIED) (#128209)
301fe4797191 [AArch64][ELF] Section alignment of 4 for AArch64 instruction (#114031)
229dcf9d3456 [CodeGen][NPM] Port MachineLateInstrsCleanup to NPM (#128160)
47656dc765aa [X86] Avoid repeated hash lookups (NFC) (#128464)
6ad55f15176d [DebugInfo] Avoid repeated hash lookups (NFC) (#128459)
5c6453da8db5 [AArch64] Add BF16 REV costs.
19a39e98ffdd [clang][bytecode] Handle non-primitive array index expressions (#128479)
7cda365e12f7 [MLIR][NVVM] Add support for f32 in redux.sync Op (#128137)
0770afb88ec1 [bolt] Remove unnecessary include.
89824547ecdc [AArch64] Add rev codegen tests for bfloat16. NFC
1dc8578df700 [AArch64] Add vrev shuffle cost tests. NFC
abcb66d18e38 [Scalar] Avoid repeated hash lookups (NFC) (#128463)
ea248365fa15 [IPO] Avoid repeated map lookups (NFC) (#128462)
71b7b1f79f8e [GlobalISel] Avoid repeated hash lookups (NFC) (#128461)
4db54e9f1ac6 [AMDGPU] Avoid repeated hash lookups (NFC) (#128458)
f89a986153a5 [mlir][emitc][NFC] Add an example to the description of the emitc.verbatim operation (#128004)
6b5bde697bb6 [analyzer] Refine TimeTrace name for dispatchWorkItem (#128352)
11fdeadbdff2 [mlir][emitc][NFC] Add an example to the description of the emitc.literal operation (#128005)
850b49297615 [BOLT][binary-analysis] Add initial pac-ret gadget scanner (#122304)
86cb0bd3a237 [MLIR][NVVM] [NFC] Update test cmd-lines and doc links (#128207)
9d191056013d [lldb][bazel] Port https://github.com/llvm/llvm-project/commit/d0e37d9723314199e08e3bb8a61f9b7e033b1ff4
a3093e56fc09 [NFC][llvm-mt] Move XmlDeleter declaration
58035b5ef5bb Revert "[compiler-rt][rtsan] stat api interception." (#128465)
eb6e50f61f60 merge main into amd-staging
6053ca004a58 [MC][CodeGen] Move FirstStackSlot and VirtualRegFlag from MCRegister to Register. NFC (#128444)
8b1d38480b9d [sanitizer] Support "bB" printf GLIBC extension (#128449)
3aef599d072b [MLIR][Affine] NFC. Drop redundant fusion- suffix from fusion pass options (#128405)
8c917f3ccdef [NFC][sanitizer] Add test for length sub-specifier "z" (#128448)
d0e37d972331 [lldb-dap] Refactor request handlers (NFC) (#128262)
12039bf4c562 remove flang-legacy from amdflang*
1581072676b5 merge main into amd-staging (#789)
4d928d5b5858 [compiler-rt][rtsan] stat api interception. (#128430)
1794dfbb0fcc [mlir] Fix integration test when `%host_cc` path contains spaces (#128439)
03610af3d58d [RISCV] Remove Inst bits from Pseudo tablegen class. NFC
b11406288265 merge main into amd-staging
40b0619a5323 [FatLTO] Detect LLD linker more reliably (#128285)
a8bcad790a46 merge main into amd-staging (#788)
14f33c6bc130 [llvm-objcopy][mach-o] Fix section finding logic for object files (#127604)
78bac7f0a675 [MC] Remove unneeded getMemtagAttr()
0b84d72c5209 [Driver] Replace /usr/lib/../$OSLibDir with /usr/$OSLibDir (#128428)
3e284554a80a [libc][math][c23] Add acosf16() function (#127731)
8ea6b735a64d [libc] Fix alignment issue for HermeticTestUtils.cpp. (#128426)
65e44b4301eb [LV] Add tests with deref assumptions and non-constant sizes.
34387fc63b5b [AsmPrinter] Simplify $local after D131429. NFC
15860446a8c3 [libc++] Fix basic_string not allowing max_size() elements to be stored (#125423)
303825d2ab07 [Analysis] Avoid repeated hash lookups (NFC) (#128394)
9d1fbbd2b9a0 [SROA][NFC] Remove Unused Parameter in `promoteAllocas()` (#128382)
cafad2b75a6e [AMDGPU] Add verification for amdgcn.init.exec.from.input (#128172)
72791fef6d6c [VPlan] Remove unused buildPlainCFG from unit tests (NFC).
1d22fba4dcf0 merge main into amd-staging
9e82ee5274ec [Support] Remove AlignedCharArrayUnion from Expected and ErrorOr, NFCI. (#127407)
06cc9681b008 [bazel] Add missing dependency for 2ff3b18554b115b17d5085b9a4cd779eeafd278a
b72bbfc293f4 [VPlan] Remove fixHeaderPhis (NFC).
8102fec00bb9 [clang][bytecode] Reject calls to pure virtual functions (#128412)
0f8769086e8e [clang-tidy][NFC][docs] Use single tick for 'false' and 'true' values in options (#128362)
c38befd94ffe [clang][bytecode] Fix delete[] dtor order (#128411)
2071ea24a27c [LVI] Skip self loops in `solveBlockValueNonLocal` (#127763)
96c723374a28 [llvm] Remove `br i1 undef` from some `llvm/test/CodeGen` tests (#128272)
2ebc69a52120 [InstCombine] Add support for GEPs in `simplifyNonNullOperand` (#128365)
a8fb2d0c1c0d [compiler-rt][rtsan] adding unlink/unlinkat interception. (#128292)
929d70a38dad [llvm-jitlink] Avoid repeated hash lookups (NFC) (#128399)
fb19bddbc5d5 [SPIRV] Avoid repeated hash lookups (NFC) (#128398)
9dd8c1444371 [IR] Avoid repeated map lookups (NFC) (#128396)
b0d1c51a1778 [DebugInfo] Avoid repeated hash lookups (NFC) (#128395)
2ff3b18554b1 Allow option to ignore module load errors in ScriptedProcess (#127153)
0968df9c3a55 [clang-format] Add the C language instead of treating it like C++ (#128287)
8ff4d27fe6b3 [MLIR][Affine] NFC. Fix outdated comment in affine analysis utils (#128406)
33f2686beda0 [llvm-jitlink] Only use candidate library extensions during library search.
21939c4c2eb5 [ORC] Fix type name in comments. NFC.
4efad490af2b [WebAssembly] Use Register instead of unsigned. NFC
b38d47205578 [ARM] Prevent converting a virtual register to MCRegister. NFC
228dbd254a5a [RegAllocGreedy] Use MCRegister instead of Register for functions that return a physical register.
57b445862db1 [LiveRegUnits] Use MCRegister instead of MCPhysReg in interface. NFC (#128232)
a26fd0b75f3c [LivePhysReg] Fix off by 1 error in an assert. NFC (#128379)
0963f0d6459c [AMDGPU] Avoid repeated hash lookups (NFC) (#128393)
dbd219aef4be [DAGCombiner][X86] Correctly clean up high bits in `combinei64TruncSrlAdd` (#128353)
e57eb6770073 merge main into amd-staging (#785)
ccad5e77442f AMDGPU: Respect amdgpu-no-agpr in functions and with calls (#128147)
6e611264c6d7 [mlir] AMDGPUToROCDL: handle 1-element vectors (#128266)
0bd66c4194d0 [RegAllocGreedy] Remove unnecessary conversion from MCRegister to Register. NFC
6fe780ce63ac [RegAllocGreedy] Use Register() instead of 0 for invalid Register. NFC
50b0669e8468 Revert "[X86] combineBROADCAST_LOAD - merge across chains" (#128380)
bac6e7b65178 [RISCV][VLOpt] Put vmclr/vmset back in the RISCVVPseudo table. (#128293)
a7d578dcce30 [RISCV][TTI] Avoid complicated fallthrough in getShuffleCost [nfc]
f5f55ade3aa7 [RISCV][TTI] Common a check in getShufleCost [nfc]
9b7bf1fdd3e7 [RISCV] Set the SEW field in the VPseudoTiedBinaryRoundingMode.
0859df4e42ba [VPlan] Use operands from initial VPInstructions directly (NFC).
d2616cc3926e [analyzer] Fix use after scope after #123003 (#128372)
6db96c9ecc78 [clang][bytecode] Always reject ctors of invalid parent decls (#128295)
f404047ab2ab [DirectX][OpLowering] Simplify named struct handling (#128247)
ff75f371be61 merge main into amd-staging
75bb25b68252 [Xtensa] Use asMCReg after #128095
7e10b6f17577 [mlir][linalg][nfc] Rename test files for linalg.{un}pack
559cb6ba6313 [libc] Fix scanf dependencies on the GPU build
b9e25f65a83e [Github][CI] Name Premerge Jobs Experimental (#128227)
9e8d11d2df14 [X86] Check that the type is integer before calling isUnsignedIntSetCC in combineExtSetcc. (#128263)
62c78919c678 [CMake] Fix some breakages when using ninja multi config (#65451)
30f44c96277e [VPlan] Set values for non-header phis at construction. (NFC)
bb6a273d9ab9 [lldb] Fix manual CURSES_LIBRARIES tinfo finding (#128245)
ae28a375d0d7 [lldb] Document <run-args> behavior in process launch help (#128215)
29cbf9cf90e0 merge main into amd-staging (#783)
e21a1737f352 [X86] combineBROADCAST_LOAD - merge across chains (#128209)
91c0aa5c1038 [mlir][tensor][nfc] Clarify comments for `createPadHighOp`
31824b2a11a8 [libc++] Fix shrink_to_fit to swap buffer only when capacity is strictly smaller (#127321)
eb9215739930 [MemProf] Add ability to export or highlight only a portion of graph (#128255)
9d6f2647ded4 [MemProf] Print internal context id when reporting bytes hinted (#128188)
268853e5689b [libcxx] Enable localization for the GPU in the config files (#128224)
fa64a210b801 [X86][FP16] Adding lowerings for FP16 ISD::LRINT and ISD::LLRINT (#127382)
8db03e8fa4ec merge main into amd-staging
7981c2997c37 [X86] Allow using the lzcnt intrinsics for non-LZCNT targets (#128284)
aa20d536ecdc add sentinel value to flang/EnableFlangBuild to disable classic in build_openmp_extras.sh (#779)
4e44bd027bf4 [mlir][Parser] Fix crash after block parsing failure (#128011)
e23ab73335c8 [VPlan] Don't convert widen recipes to VP intrinsics in EVL transform (#127180)
cbd3801acfa4 [analyzer] Allow overriding Unknown memspaces using a ProgramState trait (#123003)
c0d140353ec9 [libc++][docs][NFC] Update supported OpenXL version (#127072)
b74413bf91bd [VPlan] Use VPSingleDef instead of VPValue in HCFG builder (NFC).
f5d153ef26a9 [VectorCombine] Fold binary op of reductions. (#121567)
8d5c1e61c60e [mlir][vector] Skip uniform vectorization for non scalar type (#128294)
db836edf47f3 Revert "[analyzer] Delay the checker constructions after parsing" (#128350)
26afa2deea12 [VPlan] Create VPInstructions after setting preds in HCFG builder (NFC)
93b263a01cf8 [SimplifyCFG] Drop unused `LockstepReverseIterator` class (NFC)
34172bba115d [CodeGen] Avoid repeated hash lookups (NFC) (#128300)
48a6df36040d Reapply "[Utils] Consolidate `LockstepReverseIterator` into own header (NFC)"
b09dfbd6995b [LLD][COFF] Add support for x86_64 archives on ARM64X (#128241)
3747ba48477a [SPIRV] Avoid repeated hash lookups (NFC) (#128303)
48bccee6b3ac [Hexagon] Avoid repeated hash lookups (NFC) (#128302)
94b810661a39 [DebugInfo] Avoid repeated hash lookups (NFC) (#128301)
7c7e11b0fd96 [AMDGPU] Avoid repeated hash lookups (NFC) (#128298)
8f7e34b0af5f [clang-offload-packager] Avoid repeated hash lookups (NFC) (#128297)
f0088ee87cec [analyzer] Delay the checker constructions after parsing (#127409)
f4e8f6da41a5 [Reassociate] Use a reference to DataLayout instead of copying the underlying string data (NFC) (#128269)
bac417107339 Remove xbegin and _xend (#126952)
c32cb4af68c9 merge main into amd-staging (#778)
876174ffd753 [InstCombine] Test for ephemeral nonnull load. (NFC)
126016b6620e [InstCombine] Simplify nonnull pointers (#128111)
60f3fdde8807 [RISCV][Outliner] Use MCRegister. NFC (#128291)
2c00b3b859d9 [clang][bytecode] Silently reject ctors of invalid decls (#128290)
c645981f2d6b [mlir][bazel] Port https://github.com/llvm/llvm-project/commit/ab166d4d10f5b23262acb7448eb32c056595a0cd
86ef031b64fa [MLIR][Affine] Fix affine memory op verifiers to check valid dim/symbols properly (#127375)
39d1bd85862a [RISCV] Remove unused VPseudo tablegen class. NFC
dc72a93d2b4c [MLIR][Affine] Add check for 'affine.for' Bound Map Results (#127105)
444859634f36 [MachineVerifier] Use Register instead of unsigned for DenseSet key. NFC (#128234)
ffc61dc393e4 [clang-format] Allow specifying the language for `.h` files (#128122)
434d86ac929a [lldb] Fix warnings
646e4f2eede9 [DAGCombiner] visitFREEZE: Early exit when N is deleted (#128161)
3ec83f5774f3 [X86][DAGCombiner] Fix assertion failure in `combinei64TruncSrlAdd` (#128194)
58be69406b55 add sentinel value to flang/EnableFlangBuild to disable classic
9abe0cc1e0ec merge main into amd-staging
998b28f19672 [lldb-dap] Refactoring lldb-dap port listening mode to allow multiple connections. (#116392)
1bb43068f187 PeepholeOpt: Allow introducing subregister uses on reg_sequence (#127052)
93b2e47f1264 [flang][cuda] Avoid assign element mismatch when doing data transfer from a constant (#128252)
dadc147bd244 reland 4 [MLIR][OpenMP] patches (#757)
f13ef072d6b5 Manual update of LLVM_MAIN_REVISION to 528058 (#772)
91192921e8f9 merge main into amd-staging (#768)
5f8da7e7738f [DA] remove wrap-around check from affine definition (#116632)
d175aed4fdde [Tosa] Add necessary `unordered_map` header file (#128248)
7cfab6c4eb93 [HLSL] add parentheses in assertion to eliminate -Wparentheses warning (#128246)
aef63c506be7 [RISCV] Assembler support for XRivosVizip (#127694)
5d0c5c638ad2 [MLIR][ARITH] Adds missing foldings for truncf (#128096)
6038fd44d400 [flang][cuda] Lower atomiccas, atomicxor and atomicexch (#128242)
8ffdc3b2078e [SLP]Fix a crash when checking a scalar in a reordered buildvector node
92e02ad9dc5e [MemProf] Display backedges with dotted line in dot graphs (#128235)
79261d4aab4f [NVPTX][InferAS] assume alloca instructions are in local AS (#121710)
ca0406dd1c77 DeveloperPolicy: Update commit access requirements (#127006)
6b3e7657cfed [libc] Enable 'strftime' for the GPU targets (#128220)
7f5a2cbe7652 [libc][bazel] add targets to build the scanf family (#128082)
61c6e0061cfb [AMDGPU][True16][CodeGen] flat/global/scratch load/store pseudo for true16 (#127945)
b0638623ceb8 Manual update of LLVM_MAIN_REVISION to 528058
2b06ceebb5b7 [mlir][tosa] Align operator descriptions with the spec (#128046)
74084ae73196 [FatLTO] Allow -fno-fat-lto-objects to override -ffat-lto-objects (#128157)
374c470cd193 [Flang] Give a more specific error message for expressions where an IO Unit is expected (#126970)
236fa506d4f1 Revert "[Utils] Consolidate `LockstepReverseIterator` into own header (NFC) (#116657)"
c3d5070086d6 [MemProf] Refactor backedge computation and invoke earlier (#128226)
438b9102534a [Xtensa] Implement Xtensa Boolean Option. (#126022)
210ecfdf35ed [NFC] test interaction between array-bounds and hwaddress (#128218)
3fad640f0698 [libc] Skip scanf internals when no file available (#128097)
c0a6f7acf8b6 Use precise types in DWARF BestForm methods (#126309)
123dca9b56e1 [Utils] Consolidate `LockstepReverseIterator` into own header (NFC) (#116657)
e65d3882af6f [Support] Ensure complete type DelimitedScope (#127459)
b0f0ac3cad2d Add overload of DIBuilder::createArrayType (#125229)
45ca39d8e1a2 [Support] Fix for non-constexpr `__PRETTY_FUNCTION__` on older gcc (#128212)
741f923fac33 [MemProf] Minor fixes to dot graph printing (#128217)
0f13195308c9 Update amd/comgr/include/amd_comgr.h.in
b11e1baf22a4 [llvm-readtapi] Avoid repeated hash lookups (NFC) (#128131)
8a58f83b0416 [X86] Avoid repeated hash lookups (NFC) (#128130)
8c1207bee3c6 [SPIRV] Avoid repeated map lookups (NFC) (#128129)
34cebaf73ad6 [Instrumentation] Avoid repeated hash lookups (NFC) (#128128)
f964377df76f [DebugInfo] Avoid repeated hash lookups (NFC) (#128127)
b04435070142 [CodeGen] Avoid repeated hash lookups (NFC) (#128126)
c896f7bdaa09 [AMDGPU][True16][CodeGen] build_vector pattern in true16 (#118904)
ac757ff6f646 merge main into amd-staging
449f84fea652 [flang] fix AArch64 PCS for struct following pointer (#127802)
00637b7dfb15 [libc] Add strftime_l (#127767)
7e4fef69f9f9 [flang-rt][NFC] Fix grammar
ef6e3ed14ebc [mlir][bazel] Port https://github.com/llvm/llvm-project/commit/ab166d4d10f5b23262acb7448eb32c056595a0cd
976641062e77 [flang][NFC] AliasAnalysis: Use Indirect not Unknown for LoadOp (#127845)
386a45c6bd63 [mlir] Fix warnings
7c1f679c387a [lldb][cmake] Use STATUS instead of explicit '--' in message logging (#128196)
4bb04d417669 [clang][modules] Fix local submodule visibility of macros from transitive import (#122955)
9738f20bb03c [gn build] Port cf50936b23ac
affbb4d9cf6f [libc++] std::atomic primary template should not have a `difference_type` member type (#123236)
45798bd48416 [RISCV] Simplify the debug messages in the disassembler. (#128102)
2d38be5fd4da [LV] Strip redundant casts (NFC) (#128177)
13ca6050b07e Revert "default clause replaced by otherwise clause for metadirective in OpenMP 5.2 (#125648)"
fb25216209c4 [AMDGPU] Enhance verification of amdgcn.cs.chain intrinsic (#128162)
9546afe86c69 [MLIR][XeGPU]Add MemoryEffect to xegpu.atomic_rmw op (#128076)
0c5005482079 Revert "RegAlloc: Fix verifier error after failed allocation (#119690)"
7c7fb9442f9f Revert "Reapply "[Analyzer][CFG] Correctly handle rebuilt default arg and default init expression" (#127338)" (#128205)
115672f57edb [RISCV] Remove now redundant isElementRotate shuffle lowering [NFCI] (#128064)
ab098a7ebf07 [CGSCC] Add statistic on largest SCC visited (#128073)
209252f3d5bf [BOLT] Introduce skip-inline flag (#128135)
e94594323fae [HeterogeneousDwarf] Thread-focus vector registers in AsmPrinter
a66376b0dc3b [lld,CMake] Include Version.inc when LLVM_DISTRIBUTION_COMPONENTS contains lld-headers (#127946)
ab166d4d10f5 [mlir][mpi] Lowering Mpi To LLVM (#127053)
506deb0cce3f [lldb] Fix GCC's `-Wreturn-type` warnings (#127974)
6e457c20016a [LLD][ELF][AArch64] Add support for SHF_AARCH64_PURECODE ELF section flag (3/3) (#125689)
cf50936b23ac [libc++][chrono] implements GPS clock. (#125921)
5a7ee431fda7 [clang] Add `__builtin_sincospi` that lowers to `llvm.sincospi.*` (#127065)
d12a4d45dd43 [mlir][tosa] Remove section numbers in operator tablegen (#128048)
aa6d3ff80d93 [libc++] Re-enable the FreeBSD CI job (#127687)
0181af2f32e5 [MLIR] Imporve location tracking for `parseElementsLiteralType` (#127992)
ad0c7da95373 [libc] Add missing stdc_first_trailing_zero_ to stdbit.yaml (#128101)
6bfedfa0ba31 [mlir] Fix warnings
8a39214b7e6e [Support] Apply `constexpr` to `getTypeName` (#127893)
bd034ab1113c [X86] combineX86ShuffleChain - always combine to a new VPERMV node if the root shuffle was a VPERMV node (#128183)
a1163d8fd350 [NVPTX][NFC] Rename hasAAFeatures() (#127990)
50fcb743ecb9 [Clang] Add support for -rpath on AIX (#89279)
ce5c702d8443 [mlir][linalg][nfc] Fix formatting for linalg.elementwise
6dca33ce2069 Check for mutability better (#127843)
d2d1f143e508 [z/OS] Add option to target older versions of LE on z/OS (#123399)
481e1eba3a3d [NFC] add a pre-commit test case for patch #127121 that hoists xxsplitib out of loop (#127701)
3ce2a7dc32a6 [flang][OpenMP] Support `parallel loop` construct. (#127588)
34167f99668c RegAlloc: Fix verifier error after failed allocation (#119690)
d578dbf9fdb4 [mlir][emitc][NFC] Add type clarification to the description for the emitc.global op (#127997)
d2b391200269 [lldb] Remove commented out declaration in DWARFExpressionList
b732e14cb568 [Clang] Default the warning for chained comparison to an error. (#128145)
bff6b926e2a4 [flang][OpenMP] Map `teams loop` to `teams distribute` when required. (#127489)
4d6167ed96d3 [Clang] [Tests] Canonicalise CLANG_ENABLE_OBJC_REWRITER properly (#125117)
5e26fb169936 [libc++] Qualify calls to nullary functions like __throw_foo (#122465)
e11ca593a29d [libc++] Shuffle the order of pre-commit CI jobs a bit (#128008)
d88b8e6508fb merge main into amd-staging (#754)
884b79a478c3 [X86] Relax vbroadcast(vector load X) -> vbroadcast_load(X) to all types (#128039)
917ed99d815a [lldb] Fix "in function" detection in "thread until" (#123622)
894935cb5146 [SLP]Represent SLP graph as a tree
71af48fafdb6 [lldb] Fixing edge cases in "source list" (#126526)
8616c873350a [LLD][COFF] Support alternate names in both symbol tables on ARM64X (#127619)
02c804dd70d7 [mlir][linalg] Silence unused variable warning
a5f759ed9d11 [lldb][AIX] Added PlatformAIX plugin (#121273)
c78cb3028363 [MLIR][ROCDL] Fix accessedOperands for Rocdl_RawPtrBufferLoadOp (#127983)
fc2f31c9494c [MLIR][ROCDL] Add op for raw.ptr.buffer.load.lds (#127988)
8a3222d8da50 [mlir][Parser] Fix crash when resolving invalid operands with missing location (#128163)
25e12726f74c [mlir][bazel] Move OpAsmOpInterface.h.inc from hdrs to srcs
41437a6067b2 [LoopSimplifyCFG] Fix SCEV invalidation after removing dead exit (#127536)
6de5d1e46d18 [mlir][linalg] Extend elementwise (#124661)
b9622e84b481 [TTI][AArch64] Detect OperandInfo from scalable splats. (#122469)
bd16a87d05d3 [AMDGPU][NewPM] Port SIPostRABundler to NPM (#123717)
6bd88bb3ac82 [MLIR][NVVM] Add Ops for tcgen05 cp and shift (#127798)
73ad78cc57e1 default clause replaced by otherwise clause for metadirective in OpenMP 5.2 (#125648)
6f6fcc5ddb80 Revert "Revert "[flang][mlir] disable mlir threading"" (#751)
6d5ba79c6604 [compiler-rt][asan] Re-enable ManyThreadsTest on AArch64 (#127795)
d6858dd9bd4a Rename TosaComplianceData.h to TosaComplianceData.h.inc (#128153)
a51e6218f293 [MLIR][OpenMP] patches reland : update reverted_patches.txt
2b5ef6c12643 [MLIR][OpenMP] Add LLVM translation support for OpenMP UserDefinedMappers (#124746)
8ea3392e935a [MLIR][OpenMP] Add conversion support from FIR to LLVM Dialect for OMP DeclareMapper (#121005)
092a3d796efc [MLIR][OpenMP] Add Lowering support for OpenMP custom mappers in map clause (#121001)
d607ed777ed7 [MLIR][OpenMP] Add OMP Mapper field to MapInfoOp (#120994)
aad74dc971f7 [compiler-rt] FuzzedDataProvider: modernize outdated trait patterns (#127811)
0a913b5e3a23 [X86] Fold some (truncate (srl (add X, C1), C2)) patterns to (add (truncate (srl X, C2)), C1') (#126448)
db9876760f22 [AArch64][GlobalISel] Add some gisel test coverage for existing select tests. NFC
30ae485835fb [libcxx] Work around picolibc argv handling in tests. (#127662)
ad3f7d2c71b0 [clang][bytecode][test] Qualify a std::move call
97ed2019c41f [clang][bytecode] Use ExtendingDecl mechanism for primitives as well (#128141)
af64f0a6c2e2 [FrameLowering] Use MCRegister instead of Register in CalleeSavedInfo. NFC (#128095)
581599096e8a [AMDGPU] Expand IR Attribute table to handle longer names (NFC)
d46902e31b29 [Hexagon] Simplify an array of physical register ids. NFC (#128066)
6757cf4e6f1c [RISCV] [MachineOutliner] Analyze all candidates (#127659)
3302bef5b41a [X86] Combine FRINT + FP_TO_SINT to LRINT (#126477)
74fca0b09ac9 merge main into amd-staging
43d71baae36c [mlir][bazel] Fix build after @11468c3, @f0134e6.
36d304d2146a [RegAllocFast][NPM] Make RegAllocFastPassOptions a nested class (#127984)
6c90f87b7445 [ORC-RT] Add -num-threads=0 to testcase.
e3c8408593f8 [llvm-jitlink] Apply symbol scope modifiers explicitly for -hidden-lx.
fd5d1cbb75e4 [clang][Index] Use HeuristicResolver in IndexTypeSourceInfo as well (#128106)
f0134e6d312e [MLIR][Math] Add lowering for isnan and isfinite (#128125)
cc675c635bf0 [utils] Use stricter SSA regexp for CHECK-SAME. (#128083)
cc46d00a86f8 AMDGPU: Form v2f16 minimum3/maximum3 on gfx950 (#128123)
e729dc759d05 AMDGPU: Widen f16 minimum/maximum to v2f16 on gfx950 (#128121)
35d7bf21b6d6 [mlir] Remove unused outer loop (NFC) (#127998)
22f526811bdc [mlir][tosa] Add infer shape test of scalar mul op (#128091)
204dcafec0ec [clang-format] Fix a bug in BCIS_AfterColon and `ColumnLimit: 0` (#127964)
89e80abbc58b [lldb][LoongArch] Complete register alias name in `AugmentRegisterInfo`
f27081ba6a12 [FIR] Avoid generating llvm.undef for dummy scoping info (#128098)
24c06a19be7b [C++20] [Modules] handling selectAny attribute for vardecl
6e5f26bba87b Revert "[HLSL] Implement default constant buffer `$Globals`" (#128112)
7c2ebe5dbb4d AMDGPU: Restrict src0 to VGPRs only for certain cvt scale opcodes. (#127464)
776cddacb1ab [HLSL] Implement default constant buffer `$Globals` (#125807)
9fa77c18548a [BOLT][Linker][NFC] Remove lookupSymbol() in favor of lookupSymbolInfo() (#128070)
3c46debe6b01 [MLIR] Fix 0-dimensional case of conversion of vector ops to GPU (#128075)
4d92975b5cf0 [SandboxVec][Scheduler] Don't allow rescheduling of already scheduled (#128050)
f6156bc8722b Revert "Revert "[flang][mlir] disable mlir threading""
62c209b728d4 [mlir][tosa] add test of scalar mul op (#128088)
2ff80d2448f8 [SandboxVec][Scheduler] Fix reassignment of SchedBundle to DGNode
1ca93b15482d [llvm-dlltool] Add a missing dependency
a1a10732e417 [SandboxVec][DAG] Remove all matching nodes from bundle
f83ef281b5d3 [NVPTX] Remove redundant addressing mode instrs (#128044)
dcc08a17c781 [llvm-dlltool] Implement the --identify option (#127465)
43e637c9b173 [Offload][AMDGPU] Added support for runtime tuning (#190)
0b3002052bbc merge main into amd-staging (#742)
74c6111631ce [lld][WebAssembly] Skip BSS when combining data segments (#127735)
08c69b2ef6ee Revert "[CodeGen] Remove static member function Register::isVirtualRegister. NFC (#127968)"
309e3ca08130 Revert "[CodeGen] Remove static member function Register::isPhysicalRegister. NFC"
a67566b185c5 Allow do concurrent inside cuf kernel directive (#127693)
5981335d7541 [libc] Fix scanf cmake for targets without FILE (#128056)
84c8848f816f [flang][cuda] Lower match_any_sync functions to nvvm intrinsics (#127942)
d1dde17ab079 [libc] Fix failing NVPTX tests due to removed internal LLVM option
87bf23216fc1 [libc] Add missing dependencies on `getc` and `ungetc` for GPU (#128069)
ee252b6cee32 [LiveDebugValues] Disable spill tracking on AMDGPU (#561)
cc45365ab906 [clang][NFC] Update isAuxBuiltinID comment (#128038)
6a52276edaa5 [Comgr] Explicitly describe data object mutation restrictions
df9d3c20fa0a Reverting commits 5bc51611446ee3a9fc353 and df427992da4492a664b6 (#128057)
11468c3b072c [mlir][tosa] Add profile-based operation validation (#126992)
5bf37484c6b0 [NFC][hlsl][Sema] Simplify CBuffer Legacy Size Calculation Control Flow (#127921)
f34f21a1f61b [bazel] Fix mlir python build after a72616de18c0814ad37b5748d6bdc60b825dd889
df427992da44 Adding dependency to FIRCodeGen to fix buildbots (#128053)
2e5ec1cc5b8e [flang] Add FLANG_PARALLEL_COMPILE_JOBS option (#127364)
0ba391a85f03 [BOLT] Improve constant island disassembly (#127971)
78d82d3ae7ac [lldb] Store StreamAsynchronousIO in a unique_ptr (NFC) (#127961)
6d84fae60ed2 [gn build] Port 19af8581d51b
29361b326bd8 [Clang] Fix failing clang-tidy test (#128051)
5bc51611446e [flang] Add support to fir::cg in alias analysis (#127827)
b10ddfafcacf [libc] Fix missing includes in GPU scanf reader (#128049)
5fadb3d68090 [CodeGen] Remove static member function Register::isPhysicalRegister. NFC
b0210fee94bc [Clang] [NFC] Fix more `-Wreturn-type` warnings in tests everywhere (#123470)
6e7da07c73c1 [Github][CI] Enable New Premerge on PRs (#127894)
19af8581d51b [HLSL] Constant Buffers CodeGen (#124886)
2a7d3f055d4e [NFC] Add release notes for -Wunsafe-buffer-usage-in-libc-call (#126975)
10b99e97ff86 [SandboxVec][BottomUpVec] Separate vectorization decisions from code generation (#127727)
9ba438d32175 [RISCV] Remove some unnecessary casts from int64_t to uint64_t. NFC
00f5aaf841cb [libc]: Clean up unnecessary function pointers in scanf (#121215)
2fab6db72811 [VectorCombine] foldSelectShuffle - remove extra adds of old shuffles to worklist (#127999)
52ded6724921 [LAA] Always require non-wrapping pointers for runtime checks. (#127543)
f7a10f0e8d46 [Index] Un-XFAIL test with msan
726c4b9f7786 [flang][cuda] Lower match_all_sync functions to nvvm intrinsics (#127940)
02e8fd7a3090 [ARM,AArch64] Fix ABI bugs with over-sized bitfields (#126774)
be51ef4518ad [WebAssembly] Avoid repeated hash lookups (NFC) (#127960)
4a8f41456515 [Utils] Avoid repeated hash lookups (NFC) (#127959)
c612f7961219 [ObjectYAML] Avoid repeated hash lookups (NFC) (#127958)
506b31ec3674 [IPO] Avoid repeated hash lookups (NFC) (#127957)
c0c172213b61 [Analysis] Avoid repeated hash lookups (NFC) (#127955)
e83ad8169928 [NFC][libc++][TZDB] Fixes formatting.
1e1bf7971b1b [mlir][emitc] Add an option to cast array type to ptr type (#126385)
ff99af7ea03b [CodeGen] Remove static member function Register::isVirtualRegister. NFC (#127968)
609732cc2e65 [RISCV] Fix warnings
81bc28d0bcac [lldb] Update PlatformDarwin list of libraries with thread functions (#127922)
c48e0c182c5a [lldb][RISC-V] Extended if conditions to support alias names for registers (#124475)
81ed48531d51 [CMake] Fix variable name (#127967)
a72616de18c0 [mlir][python] fix linalg.pack/unpack (#127729)
60c620272db2 [ADT] Add set_intersects to check if there is any intersection (#127907)
404af37175a5 [VPlan] Remove stale assertion in HCFG builder.
c2e51428bf86 [MLIR][LLVM] Fold extract of constant (#127927)
1b610e69d962 [MLIR][Math] Add floating point value folders (#127947)
1b78ff697278 [InstCombine] Simplify the pointer operand of store if writing to null is UB (#127979)
43f2968a028f [RISCV] Recognize VLA shift pairs from shuffle masks (#127710)
04b5c63ddfe9 [LV] Add inbounds to interleave test.
3b9f9645e017 [LV] Regen a couple of tests with UTC (#127785)
ae5785460dfd [libclc] Define macros for users of gentype.inc (#128012)
6eba2775e22f [LAA] Scale strides using type-size (NFC) (#124529)
9ffab5637c80 [SPIR-V] Initial implementation of SPV_INTEL_long_composites (#126545)
24cd933d02b8 [MemCpyOpt] Add test for call slot opt with ret-only capture (NFC)
1c4e9863fa1c [OpenMP][NFC] Remove unused debug lock (#127928)
41cece8c8639 [flang] Add -f[no-]vectorize flags (#119718)
a03f064b60f0 [X86] combineX86ShufflesRecursively - peek through one use bitcasts to find additional (free) extract_subvector nodes
f00b32e2d0ee [libc++] Reduce the dependency of the locale base API on the base system from the headers (#117764)
8feb5bac3279 [libc++] Add benchmarks for copy algorithms (#127328)
0d46d0063838 merge main into amd-staging
12f8ed58a039 [Clang] Mark P1061 (Structured Bindings can introduce a Pack) as implemented (#127980)
0c809ea33614 [Mips] Reserve hardware register HWR2 (#127775)
0a8341fdb2f5 [RISCV] Avoid VMNOT by swapping VMERGE operands for mask extensions (#126751)
58571c82ef93 [lldb] Un-XFAIL TestDeadStrip.py on windows
70ed381b1693 [GlobalISel][AArch64] Fix fptoi.sat lowering. (#127901)
3df8911d0445 Ensure pointer is copied to target when USM is active (#450)
a42a22f845a7 merge main into amd-staging (#737)
f4f7c71c55da [RISCV][VLOPT] Move mayReadPastVL check into getMinimumVLForUser. NFC (#127972)
6361a8a1b763 [CaptureTracking] Check for equality predicate for null comparisons
54f14d96bb27 [FunctionAttrs] Add test for incorrect predicate in capture analysis (NFC)
458b1e9d2c61 [TBAA] Refine pointer-tbaa for void pointers by pointer depth (#126047)
9855d761f36f [AMDGPU][NewPM] Port SIOptimizeExecMaskingPreRA to NPM (#125351)
505d35aad3aa [X86] getFauxShuffleMask - relax one use limit for insert_subvector concat splat pattern (#127981)
92a3192a963c [X86] vector-shuffle-v192.ll - regenerate VPTERNLOG comments
66cf2a88a4ca [X86] sext-vsetcc.ll - regenerate VPTERNLOG comments
0948fc85aa99 [clang] print correct context for diagnostics suppressed by deduction (#125453)
d7784a649e80 [flang][OpenMP] Catch threadprivate common block vars that appear in equivalence (#127642)
367ecc63338d [lldb] Renaissance LineTable sequences (#127800)
544a161c9a16 [NFC][SPIR-V] Remove unused variable (#127985)
97872409125b [RISCV][test] Add tests for extending negated mask
747588dc642b [flang][Lower][OpenMP] Don't read moldarg for static sized array (#127838)
d595fc91aeb3 Revert "[clang] Lower modf builtin using `llvm.modf` intrinsic" (#127987)
dc326d0b01f6 [Clang][doc] -ffp-contract options and standard compliance (#127621)
e09fc394b7c4 [llvm] Make target triple in TLII const reference (#127718)
6e047a5ab426 Add Wasm, RISC-V, BPF, and NVPTX targets back to Windows release packaging (#127794)
a60675d22723 [CodeGen][NFC] Move test builtin_signbit.cpp to CodeGen (#127814)
44feae869570 [RISCV][VLOPT] Mark some methods + arguments as const. NFC
55fa2fa34809 [SPIR-V] Add SPV_INTEL_bindless_images extension (#127737)
e264317b4516 [lldb] Fix RangeDataVector::CombineConsecutiveEntriesWithEqualData (#127059)
2c022e3617ec [lldb] Replace LineTable::upper_bound with GetLineEntryIndexRange (#127806)
62d77f…
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Introduces a
LinalgInterfacestarget so thatTensorDialectdoesn't need to depend onLinalgDialect, which would introduce a circular dependency.