pull #1
Open
pull #1
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
Fixes #106846. This is what I learned from GCC. I found that GCC does not duplicate the BB that has indirect jumps with the jump table. I believe GCC has provided a clear explanation here: > Duplicate the blocks containing computed gotos. This basically unfactors computed gotos that were factored early on in the compilation process to speed up edge based data flow. We used to not unfactor them again, which can seriously pessimize code with many computed jumps in the source code, such as interpreters. (cherry picked from commit dd21aac)
Fixes a crash with extract_subvectors in Hexagon backend seen when the source vector is a vector-pair and result vector is not hvx vector size. LLVM Issue: llvm/llvm-project#128775 Fixes #128775 --------- Co-authored-by: aankit-quic <[email protected]> (cherry picked from commit 29d3fc3)
…`isGuaranteedNotToBeUndefOrPoison` (#130111) Fixes (keep it open) #130110. If the incoming value is PHI itself, we can skip this. If we can guarantee that the other incoming values are neither undef nor poison, then we can also guarantee that the value isn't either. If we cannot guarantee that, it makes no sense in calculating it. (cherry picked from commit 462eb7e)
…and 'w' (#129864) - Allow 'u' and 'w' on LASX, LSX or floating point register operands. - Also add missing description in LangRef. Fixes #129863. (cherry picked from commit bae6644)
The default triple of Amazon Linux on AArch64 is aarch64-amazon-linux,
see issue highlighded by PR #109263, somewhat serious linker issues are
encountered if any other triple is being used.
Unfortunately, this makes XFAIL lines like
`XFAIL: target=aarch64{{.*}}-linux-gnu` ineffective, making it
impossible to complete all of the check-cxx on Amazon Linux without
failing.
(cherry picked from commit 8f4ee42)
Summary: These return values are actually signed, meaning that casting will extend it and then all the bits will be one. (cherry picked from commit 4ca8ea8)
…ny extract is waiting to be erased (#129375) If any extract is waiting to be erased, then bail out as this will distort the cost calculation and possibly lead to infinite loops. Fixes #129373 (cherry picked from commit 5ddf40f)
…l(). (#130693) It has found to be quite a slowdown to traverse the users of a function from each call site when it is called many (~70k) times. This patch fixes this for now as long as this verification is disabled by default, but there is still a need to eventually cache the results to avoid recomputation. Fixes #130541 (cherry picked from commit 378739f)
This effectively reverts #108535. The old AA code was looking for the *first* clobber between the load and store and then trying to move all the way up there. The new MSSA based code instead found the *last* clobber. There might still be an earlier clobber that has not been accounted for. Fixes #130632. (cherry picked from commit 5da9044)
…epth (#128704) Backports b8d1f3d. This fixes a potential integer overflow bug that has been around for many versions and was exposed by my patch recently. So we think it warrants a backport
… (#130517) Bug introduced #120995. The LLD code calculates the "size" of the Mach-O headers, and then uses that size to place the segments, but the `__TEXT` section stays at `fileoff` zero. When I wrote the code into llvm-objcopy I calculated the extra space into the initial offset, which moved all the sections back 1 page. Besides the modified test checking for the right `fileoff` values of the sections and the segments, I also manually checked the generated binaries after `llvm-objcopy` using `dyld_info`, as the bug report suggested. Fixes #130472 (cherry picked from commit 8413f4d)
…7) (#129686)
Fix's not rejecting global_load_lds_dwordx3 and x4 on other targets.
The encoded versions of instructions should not touch
SubtargetPredicate,
and only AssemblerPredicate.
(cherry picked from commit f319a65)
…f nodes have been seen (#129934) This mitigates a regression introduced in #87824. The mitigation here is to store pointers the deduplicated AST nodes, rather than copies of the nodes themselves. This allows a pointer-optimized set to be used and saves a lot of memory because `clang::DynTypedNode` is ~5 times larger than a pointer. Fixes #129808. (cherry picked from commit 8c7f0ea)
Alexei reported a bpf selftest failure with recent llvm for bpf prog
file progs/arena_spin_lock.c. The failure only happens when clang is
built with cmake option LLVM_ENABLE_ASSERTIONS=ON.
The error message looks like:
```
clang: /home/yhs/work/yhs/llvm-project/llvm/lib/IR/Instructions.cpp:3460:
llvm::BitCastInst::BitCastInst(Value *, Type *, const Twine &, InsertPosition):
Assertion `castIsValid(getOpcode(), S, Ty) && "Illegal BitCast"' failed.
```
Further investigation shows that the problem is triggered in
BPF/BPFAbstractMemberAccess.cpp
for code
```
auto *BCInst =
new BitCastInst(Base, PointerType::getUnqual(BB->getContext()));
```
For the above BitCastInst, Since 'Base' has non-zero AddrSapce, the
compiler expects the type also has the same AddrSpace. But the above
PointerType::getUnqual(...) does not have AddrSpace and hence causes the
assertion failure.
Providing the proper AddrSpace for the BitCast type fixed the issue.
Co-authored-by: Yonghong Song <[email protected]>
(cherry picked from commit 5686786)
This backports d911085 because it fixes a regression introduced in 19 and we don't want it to persist in 20
…ss. (#129087) In some cases, it is possible for the same underlying object to be accessed via pointers to different address spaces. This could lead to pointers from different address spaces ending up in the same dependency set, which isn't allowed (and triggers an assertion). Update the mapping from underlying object -> last access to also include the accessing address space. Fixes llvm/llvm-project#124759. PR: llvm/llvm-project#129087
…131438) It seems that Apple Clang 17 starts to be used for CI, while it hasn't supported `__builtin_is_virtual_base_of` yet. And thus we need to skip the test for `is_virtual_base_of`. Follows up #131302.
…mber access (#131450) For an expression `nodiscard_function().static_member(), the nodiscard warnings added by #120223, are not useful or actionable, and are disruptive to some library implementations; we just remove them. Fixes #131410 (cherry picked from commit 9a1e390)
The pull request [1] changed bpf default cpu from -mcpu=v1 to -mcpu=v3 in clang20. Recently in [1], Yuval Deutscher suggested to add an entry to clang20 ReleaseNotes so users can easily find the change from documentation. [1] llvm/llvm-project#107008 Co-authored-by: Yonghong Song <[email protected]>
…rToZero` (#131522) llvm/llvm-project#94525 assumes that the loop will be infinite when the stride is zero. However, it doesn't hold when the start value of addrec is also zero. Closes llvm/llvm-project#131465. (cherry picked from commit c5a491e)
This is to fix compile error with explicit Clang modules like
```
../../third_party/libc++/src/include/__vector/vector_bool.h:85:11: error: default argument of '__bit_iterator' must be imported from module 'std.bit_reference_fwd' before it is required
85 | typedef __bit_iterator<vector, false> pointer;
| ^
../../third_party/libc++/src/include/__fwd/bit_reference.h:23:68: note: default argument declared here is not reachable
23 | template <class _Cp, bool _IsConst, typename _Cp::__storage_type = 0>
| ^
```
(cherry picked from commit 672e385)
…824) Address missing changes: - V[,U]COMXSD need to have XD (F3.0F –> F2.0F) - V[,U]COMXS[S,H] need to have XS (F2.[0F,MAP5] -> F3.[0F,MAP5]) - VGETEXPBF16 needs to have T_MAP6 and NP (66.MAP5 -> NP.MAP6) Ref.: https://cdrdv2.intel.com/v1/dl/getContent/828965 (cherry picked from commit 975c208)
"libmsvcrt-os" was added to the list of excluded libs in binutils in 9d9c67b06c1bf4c4550e3de0eb575c2bfbe96df9 in 2017. "libucrt" was added in c4a8df19ba0a82aa8dea88d9f72ed9e63cb1fa84 in 2022. "libucrtapp" isn't in the binutils exclusion list yet, but a patch for adding it has been submitted. Since 0d403d5dd13ce22c07418058f3b640708992890c in mingw-w64 (in 2020), there's such a third variant of the UCRT import library available. Since 18df3e8323dcf9fdfec56b5f12c04a9c723a0931 in 2025, "libpthread" and "libwinpthread" are also excluded. (cherry picked from commit af93db9)
(cherry picked from commit 4c2d1b4)
Without this we can try to generate invalid instructions or create illegal types. This patch generates a SVE fcopysign instead and use its lowering. BF16 is left out of the moment as it doesn't lower successfully (but could use the same code as fp16). (cherry picked from commit d4ab3df)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
20 participants
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.
DO NOT FILE A PULL REQUEST
This repository does not accept pull requests. Please follow http://llvm.org/docs/Contributing.html#how-to-submit-a-patch for contribution to LLVM.
DO NOT FILE A PULL REQUEST