This path transforms `vector.contract(A, B, Acc)` into
`vector.contract(A, B, 0) + Acc` to decouple the contraction computation
from the accumulator update for contractions over the loop.
A PIC data reference forms the symbol's GOT offset with
R_SPARC_GOTDATA_OP_HIX22 and R_SPARC_GOTDATA_OP_LOX10 and loads through
it, with R_SPARC_GOTDATA_OP marking the load:
```
sethi %gdop_hix22(sym), %g1
xor %g1, %gdop_lox10(sym), %g1
ldx [%l7 + %g1], %g1, %gdop(sym)
```
When the symbol is neither preemptible nor absolute, optimize the load
to `add %l7, %g1, %g1` over the symbol's GOT-relative address. An
absolute symbol keeps the GOT load, as it can be arbitrarily far from
.got and we don't want the X86_64::relaxOpt complexity.
Co-authored-by: Kirill A. Korinsky <kirill@korins.ky>
Co-authored-by: LemonBoy <thatlemon@gmail.com>
Co-authored-by: Alex Rønne Petersen <alex@alexrp.com>
C++17 guarantees the postfix-expression naming the called function is
sequenced before evaluation of its arguments, so
L->linkPhase1(std::move(L)) is well-formed without the LTmp indirection.
This should also be implemented by MSVC now. See here (P0145R3 and
P0400R0):
https://learn.microsoft.com/ar-sa/cpp/overview/visual-cpp-language-conformance?view=msvc-170
Disclaimer: As I have no LLVM Windows machine with MSVC at hand, I was
not able to test it myself.
Co-authored-by: Claude <noreply@anthropic.com>
`b05a5d0a` added an arity-trimming step to the shared
`ScriptedPythonInterface::Dispatch`: extensions are now allowed to
define methods with trailing parameters as optional
(`num_children(self)` vs. `num_children(self, max_count)`), so before
calling into a method, `Dispatch` needs to know how many positional
arguments it actually accepts and drop any trailing ones we'd otherwise
pass.
That check calls `PythonCallable::GetArgInfo`, which ran a whole
embedded Python script through `inspect.signature` on every call, since
every scripted-extension dispatch goes through it.
For the common case `GetArgInfo()` actually needs to handle fast (plain
Python functions/methods, classes used as constructors, and callable
instances defining `__call__`, i.e. everything `Dispatch<T>()` and
`CreatePluginObject()` ever pass it), the answer is available as plain
data attributes, with no Python bytecode execution required:
`__func__`/`__self__` to unwrap bound methods, and `__code__`'s
`co_argcount`/`co_flags` for the actual positional-argument count and
varargs bit. For a class, that means `__init__` -- except a class may
customize `__new__` instead and leave `__init__` untouched, in which
case `object.__init__` becomes lenient about extra arguments.
Anything else still lacking `__code__` can fall back to the original
`inspect.signature()`-based implementation, now exposed as
`PythonCallable::GetArgInfoFromInspectSignature()` so it can be used
separately.
rdar://183776556
---------
Signed-off-by: Med Ismail Bennani <ismail@bennani.ma>
Exposes `OperationEquivalence` through the MLIR C API so callers can compare operations structurally rather than by handle identity.
Assisted by: Claude
`ConcatInputSection::foldIdentical()` clears the folded functions
`originalUnwindEntry`.
However, it cleared every symbol: with N members, the repeated clearing
is **O(N²)**.
This is redundant: we should only remove the incoming `copy->symbols`.
This patch moves the removing loop ahead and adds more clear comments on
why we need to skip the first element.
Also added a new `lld/test/MachO/icf-scale-same-class.s` with 500K
identical functions as a stress test, which would've taken minutes to
link, and less than a second with the patch.
Testing on real-world app (IRPGO instrumentation + ICF) find that we
achieved a 19x speed up (1:14:42 -> 3:55)
Defer to ConstantFolding on all-constant operands, in
llvm::simplifyIntrinsic. The patch adds a context-function argument to
ConstantFoldIntrinsic for strictfp information, which is used by a new
canConstantFoldIntrinsic under canConstantFoldCallTo: it was
necessitated by exposing the constant-folder via
llvm::simplifyIntrinsic. The patch only has impact on passes that use
simplifyIntrinsic via InstSimplifyFolder, other than InstCombine, which
already constant-folds intrinsics.
Previously, trying to compile libm with LIBC_COPT_PUBLIC_PACKAGING
defined failed on 32-bit Windows with
error: alias must point to a defined variable or function
This is because 32-bit Windows adds a leading underscore to __cdecl C
functions, making the alias declaration not find its target symbol name.
The same problem exists on Apple platforms, which as solution don't emit
the alias for the C++ LIBC_NAMESPACE:: symbol.
Do the same on 32-bit Windows as on Apple platforms: Emit only
the underscore-prefixed symbol, not the LIBC_NAMESPACE:: alias.
Add a lane-level XeGPU operation that re-distributes a subgroup's
fragments across its lanes without changing the element type.
The op takes a 1D vector — the fragment held by one lane — and returns a
fragment of the same type. A pack/unpack mode selects the direction:
viewing the subgroup as an S x N element grid (S = subgroup size, N =
elements per lane), pack moves element j of lane i from logical position
j*S + i to i*N + j, so a lane's elements end up at consecutive
positions; unpack is the reverse.
This implements xegpu.convert_layout semantics at the lane level when
lane_layout is unchanged but lane_data differs.
assisted-by-claude
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Update accumulateConstantOffset and collectOffset to implicitly truncate
struct field offsets. This fixes crashes in the new tests, where the
field offset does not fit in the index width.
PR: https://github.com/llvm/llvm-project/pull/213436
PPC64PILongBranchThunk adds a relative relocation, which runs after
removeUnusedSyntheticSections. With no other dynamic relocation
.rela.dyn is incorrectly removed.
Extract the .rela.auth.dyn condition (#96496, #195649) to `mayGrowLate`
and add a .branch_lt check.
The reason for the failure is known but is not in the log. Add a helper
function to map the error to a string and log it.
---------
Co-authored-by: Jonas Devlieghere <jonas@devlieghere.com>
Currently, it is unclear if a test that is skipped is either:
- Unsupported because it will never run (macosx tests should not run on
Windows).
- Skipped because it's supposed to work but does not yet.
`@expectedFailure` is not always an option because a test might be
timing out and waiting 600s for it to fail is not acceptable.
This patch introduces 2 new decorators to differentiate those tests:
- `@requireX` meaning the tests require X to run (e.g `@requireDarwin`).
- `@requireNotX` meaning the tests requires anything but `X` to run (e.g
`@requireNotDarwin`).
In the tests results summary, this introduces a new category: `skipped`
tests:
- `@requireX` -> `UNSUPPORTED`
- `@skipIfX` -> `SKIPPED`
Here is how they can and cannot be stacked:
This will never run:
```py3
@requireDarwin
@requireWindows
def test():
...
```
This will not run on Linux, because `@requireWindows` is not satisfied.
```py3
@requireNotDarwin
@requireWindows
def test():
...
```
For "require Darwin or Windows", you need:
```py3
@requirePlatform(["darwin", "windows"])
def test():
...
```
For "require Darwin on ARM64 or Windows on ARM64", you need:
```py3
@requireARM64
@requirePlatform(["darwin", "windows"])
def test():
...
```
This patch also converts the tests in
`lldb/tests/API/{macosx,linux,windows}` to use these decorators. Patches
to convert the rest of the tests will follow.
Share LLVM dialect metadata materialization through `ModuleTranslation`
so named metadata and metadata-as-value lowering use one conversion
path. Resolve metadata symbol references to functions, globals, aliases,
and ifuncs, and diagnose malformed required metadata before lowering.
Rename `MDFuncAttr` to `MDGlobalValueAttr` to reflect its support for
symbol-backed global values.
Reject distinct nodes when converting generic LLVM metadata into LLVM
dialect attributes because `MDNodeAttr` cannot preserve node identity.
Apply the policy recursively so top-level and nested nodes use the
existing unsupported-metadata failure path while finite acyclic uniqued
nodes remain importable.
Fix two bugs in readability-use-std-min-max when the if body has no
braces.
For brace-less if bodies, If->getEndLoc() points to the expression end
rather than the trailing semicolon, truncating replacements and losing
comments. We find the semicolon via findNextToken and extend the range
consistent with compound statement logic.
Before:
if (n < -1) n = -1 ; -> n = std::max(n, -1); ;
if (n > 1) n = 1/*comment*/; -> n = std::min(n, 1);;
After:
if (n < -1) n = -1 ; -> n = std::max(n, -1);
if (n > 1) n = 1/*comment*/; -> n = std::min(n, 1); /*comment*/
Fixes#208708.
AI Usage: This patch is AI-assisted, reviewed and verified by me.
The follow-up #212888 left for the other direction. HasCvtFP8SDWASrcSel
and HasCvtFP8ByteSel replace isGFX940Plus and isGFX11Plus on the f32 ->
fp8/bf8 conversions, which split by the same encoding characteristic, so
the two feature descriptions become direction-neutral. The f16 ->
fp8/bf8 conversions had no feature at all, so add
FeatureF16FP8ConversionInsts, the mirror of the existing
FeatureFP8F16ConversionInsts, and use it in place of isGFX1250Plus.
Assisted-By: Claude Opus 5
Drive-by: Rename `constexpr_noexcept.compile.pass.cpp` to
`ctor.default.pass.cpp` as test coverage is added to it, and compile and
run it in all modes.
Move the VGPR/AGPR pairing out of MFMATable into a generalized
AGPRFormTable class and rename getMFMASrcCVDstAGPROp to getAGPRFormOp.
MFMATable inherits AGPRFormTable, so the instantiation sites and the
generated table are unchanged.
Other instruction families with paired VGPR/AGPR pseudos, such as DS,
can now be tagged with AGPRFormTable.
Related to #168983
OffloadArch was a flat enum that hand-duplicated every AMDGPU and NVPTX
target, plus a few edge cases. This was yet another place that needed
updating every time a new target is added, which should now be avoided.
Replace with a tagged union-like scheme.
Co-authored-by: Claude (Claude-Opus-4.8) <noreply@anthropic.com>
XCheriot was not considered in this function, so RV32E+XCheriot
defaulted
to ilp32e instead of cheriot when no -target-abi was given.
Since this function does not have unit test coverage, this adds new
tests
for the ABI inference functions: RISCVISAInfo::computeDefaultABI() and
RISCVABI::computeTargetABI.
This change was created with the help of AI tools
Noticed when reviewing #175830: MCObjectStreamer::reset frees the
fragments PendingBA and PrevInstPosition point into, but X86AsmBackend
keeps them. Define the hook.
Today, git-clang-format will only format lines which have been modified.
However, in some cases, that's not sufficient to get a "clean" file
which would be unmodified by running `clang-format` manually.
I've got a minimal repro using the default clang-format rules. Setup a
new git repository and create a commit with an empty file:
```
mkdir /tmp/bla
cd /tmp/bla
git init
touch t.cpp
git add t.cpp
git commit -m "V1"
```
Add a line to that file containing a comment:
```
echo "int x = 0; // Comment describing x" > t.cpp
git add t.cpp
git clang-format --staged # Reports "clang-format did not modify any files"
git commit -m "V2"
```
There's nothing wrong with this file yet, and it's working as expected.
Problem arises when we add a new line with a comment which is not
aligned with the comment we've already added:
```
echo "int longerThanX = 0; // Should cause above to become aligned" >> t.cpp
git add t.cpp
git clang-format --staged # Reports "clang-format did not modify any files"
```
git-clang-format didn't change the staged file here as it's only
inspecting the modified lines, but according to the default formatting
rules, this new line actually has an effect on the preceding line. i.e.,
if you run `clang-format t.cpp`, you'll see that whitespace is added
before the "Comment describing x" to align with the "Should cause..."
It's not limited to _just_ the preceding line (e.g. in the case where we
have multiple lines like the first with comments which become
unaligned,) so can't just add more lines of context to the diff. This PR
just adds an option to format the whole file, instead of just the
modified sections.
Only UnwindLLDB honored target.process.thread.max-backtrace-depth, so
UnwindWasm reported every frame the stub sent and a stack that recursed
without end was walked to its end.
Bound the frame count by the depth on every query rather than dropping
the frames that exceed it. A Wasm call stack arrives whole on the first
query, so a depth lowered after that has to still apply, which is when a
user reaches for it. The synthetic call frame addresses stay derived
from the whole stack, so the order of the frames does not depend on the
depth.
A function that saves a callee-saved EGPR (R16-R31) cannot be encoded with V1/V2 unwind info, so it must use V3 even when the module default stays on V1/V2 (e.g. an APX clone created by auto-dispatch alongside a baseline generic clone). The previous code rejected such functions with a recoverable backend diagnostic ("EGPR (R16-R31) requires V3 unwind info on Windows x64") instead of emitting valid V3 unwind info.
Introduce a single shared predicate, requiresWinX64UnwindV3(MF), that returns true when the whole module is in V3 mode, or when the function needs an unwind table and may use EGPR. It is consumed by X86FrameLowering (SEH prolog/epilog layout), the X86WinEHUnwindV2 pass (which skips such functions), and the X86WinEHUnwindV3 pass (which stamps a per-function .seh_unwindversion 3 on every WinEH frame -- the entry block and each funclet). Also widen the SEH_UnwindVersion pseudo operand from i1imm to i8imm since it holds 1, 2 or 3.
Exposes `Value::replaceUsesWithIf` through the MLIR C API, allowing callers to replace only a subset of a value's uses based on a predicate.
Assisted by: Claude
commit fd5c1f9497 (2014) emits %r_disp32 for pc_rel entries in
.gcc_except_table and .eh_frame, but the specifier is only recognized in
instruction operands, so llvm-mc cannot reassemble llc output.
Implement parseDataExpr with a data specifier table holding just
%r_disp32, and reject %r_disp32 outside a .word directive and in
instruction operands.
Supersedes #208933.
This PR fixes a crash in ConvertShardToMPI caused by ignoring a failed partial conversion and continuing to apply folding patterns to partially converted IR.
Assisted-by: codex
---------
Co-authored-by: Maksim Levental <maksim.levental@gmail.com>
IrreducibleGraph is the explicit graph of a region -- a loop, or the
whole function -- that BFI searches for irreducible SCCs. It maintains
predecessor lists only so findIrreducibleHeaders can ask of each node
whether a predecessor lies outside its SCC (an entry), and whether a
non-entry in the same SCC reaches it via a backedge (an extra header).
Answer both from the successor lists instead; IrrNode then needs only a
successor vector. findIrreducibleHeaders is left partitioning the SCC
its sole caller is packaging, so fold it into createIrreducibleLoop.
Headers and members are sorted, so the changed iteration order does not
affect output.
Once BFI uses CycleInfo, a region's irreducible SCCs are its
non-reducible child cycles, which should let IrreducibleGraph go away
entirely. That removal has no predecessor lists to walk and
getResolvedNode has no inverse, so it must sweep successors regardless;
settling the formulation here leaves it changing only the graph.
Aided by Claude Opus 5
PlatformWasm hands the runtime the module to run as a path on the host
it launches the runtime on. It takes that path from the launch info,
whose executable is the name the module goes by on the platform. That
name is whatever a stub reported the module under, which need not be a
path that resolves on this host, so a relaunch runs a file that does not
exist:
```
(lldb) run
error: WebAssembly runtime exited with exit code 255
```
Run the file the target has instead. Only a runtime launched on this
host is affected, since a connection to a remote Wasm platform delegates
the launch to that platform.
Exposes `OpBuilder::saveInsertionPoint` / `restoreInsertionPoint` through the MLIR C API for `MlirRewriterBase`, continuing the buildout of the rewrite/conversion C bindings.
Assisted by: Claude
