The -g flag has been selecting whether to emit dwarf or codeview based
on the target ABI since 2018, so simply aliasing these flags does the
right thing for clang-cl.

This moves some code from Clang::ConstructJob to renderDebugOptions to
make things a little clearer now that we don't need to keep track of
whether we're doing codeview or not in multiple places, and also
combines the duplicate handling of the cl vs clang handling of jmc
flags as a result.

This is mostly NFC, but some -cc1 flags may be rendered in a slightly
different order because of the code that was moved around.

Differential Revision: https://reviews.llvm.org/D157794

Differential Revision: https://reviews.llvm.org/D157554

Test all the different itanium mangled opencl functions that are
interesting to replace with raw intrinsic calls.

https://reviews.llvm.org/D157873

When using debug info correlation, value profiling needs to be switched off.
So, we are only merging counter sections. In that case the existance of data
section is just used to provide an extra check in case of corrupted profile.

This patch performs counter merging by iterating the counter section by counter
size and add them together.

Reviewed By: ellis, MaskRay

Differential Revision: https://reviews.llvm.org/D157632

Use header_information to generate the __std_clang_module header. Instead of using lit_header_restrictions like the manually written header did, make a new header_include_requirements to codify what can be included rather than what can be fully tested.

Reviewed By: Mordante, #libc

Differential Revision: https://reviews.llvm.org/D157364

Top level modules don't need `requires` because they're only built when their headers are included.

Reviewed By: ldionne, Mordante, #libc

Differential Revision: https://reviews.llvm.org/D157363

Resolvers are running before the module is initialised which leads to
crashes due to the santizer is not yet initialised.

Fixes #40287

Reviewed By: hctim

Differential Revision: https://reviews.llvm.org/D150262

The word "Assertion" is not necessarily present. Limit the checks to the
assertion message.

Reviewed By: kadircet

Differential Revision: https://reviews.llvm.org/D157207

A static analysis tool reported a "use after move" error in a place where the
same object was being passed to two different calls using std::forward. While
this seems unlikely to present an actual problem, removing the std::forward
makes the code slightly more robust.

This relates to the comments here: https://reviews.llvm.org/D152891#inline-1522454

Differential Revision: https://reviews.llvm.org/D157453

VPseudoVWMAC_VV_VF_BF got committed around the same time we added
SchedTernary, so we missed having it use SchedTernary.

Differential Revision: https://reviews.llvm.org/D157898

This reverts commit 6da69114.

This commit failed to build.

VPseudoVWMAC_VV_VF_BF got committed around the same time we added
SchedTernary, so we missed having it use SchedTernary.

Differential Revision: https://reviews.llvm.org/D157898

iAdd the acc.declare_action attribute to attach pre/post
allocation/deallocation functions to operations.

Reviewed By: razvanlupusoru

Differential Revision: https://reviews.llvm.org/D157743

Constructs such as inline variables, #line, and #include can create
lexical blocks with a different filename.
GCOVProfiling and llvm-cov gcov currently don't handle such cases (see
GCOVLines::writeOut and GCOVFile::readGCNO) and would incorrectly
attribute the line number to the current file.

For now, ignore such blocks. Missing line execution counts is better
than wrong ones.

---

As a workaround that Apple targets don't use -mconstructor-aliases yet,
allow line execution count 4 on the A::A line (1f34e282).

Reviewed By: reames

Differential Revision: https://reviews.llvm.org/D157818

Handle special case of element-per-element assignments generated
for creating a temp for unlimited polymorphic hlfir.expr.
We currently end up generating an invalid fir.store. We should use
the assignment runtime instead.

Reviewed By: tblah

Differential Revision: https://reviews.llvm.org/D157752

This is needed to support things like null pointers.

Reviewed By: reames

Differential Revision: https://reviews.llvm.org/D157822

The register list in the arg string is declared as `{$rlist}`. This patch removes the wrapping curly brackets because of the following:

- Curly brackets are the syntax for variant selection, e.g. given `X = {v0 | v1}`, the result after `CodeGenInstruction::FlattenAsmStringVariants` should be `X[AsmVariantNo]`.
- ARM also supports the register list, and they do not use the bracket wrapper.
- Parse of curly brackets are handled by `RISCVAsmParser::parseReglist`, the brackets in the td file do not correspond to asm syntax. Thus no testcase is affected.

So the curly brackets here are redundant and will become dangerous if RISCV needs more asm parser variants (took me several hours to figure out some wired assertion failures in a downstream fork)

Reviewed By: craig.topper

Differential Revision: https://reviews.llvm.org/D157472

support DW_FORM_strx in dsymutil.

Differential Revision: https://reviews.llvm.org/D157036

Sorting algorithms require strict weak ordering for comparators, final
fix for cmp instructions comparator.

This allows -mexec-model=reactor -shared produces a library module
with _initialize entrypoint, which is preferrable over __wasm_call_ctors.

This partially reverts https://reviews.llvm.org/D153293

Discussion: https://github.com/dicej/component-linking-demo/issues/3

Reviewed By: sbc100

Differential Revision: https://reviews.llvm.org/D156205

Address post-commit simplification suggestion for 8a56179b: Replace
IsTruncated by conditionally setting TruncResultTy only if truncation
is required.

This reverts commit 1f34e282.

Test fails on x86-64 mac, see comments on https://reviews.llvm.org/rG1f34e282e8066281eb1447e21e44a2a2e9983e79.

As in all the other cpus from D134521, this adds FeatureFuseAdrpAdd to
NeoverseV2 to allow more linker relaxations.

Reviewed By: springerm

Re-land 691a2fab which was incorrectly
reverted.

Differential Revision: https://reviews.llvm.org/D157706

This reverts commit 20966fcb.

Bot is broken https://lab.llvm.org/buildbot/#/builders/61/builds/47577

Generalize D127933 (Fuchsia special case) to other ELF targets. Ensure
that musl, NetBSD, OpenBSD, etc do not get ifunc codegen which is
unsupported in their rtld.

Link: https://discourse.llvm.org/t/does-ifunc-use-from-llvm-require-os-support/67628
Close: https://github.com/llvm/llvm-project/issues/64631

This reverts commit 691a2fab.

The bot is broken: https://lab.llvm.org/buildbot/#/builders/61/builds/47577

In case of unlimited polymorphic type, there is no derived type
spec, so use NoneType explicitly.

Reviewed By: tblah

Differential Revision: https://reviews.llvm.org/D157751

SelectionDAG::getZeroExtendInReg does exactly the same masking.

The zext constant expression was detected by the fold, but then
handled as a sext. Use ZExtOperator instead of ZExtInst to handle
constant expressions.

Fixes https://github.com/llvm/llvm-project/issues/64669.

The comment was out of date, the device libs build does provide all
the pointer overloads. An extremely pedantic interpretation of the
spec would suggest only the flat version exists, but the overloads do
exist in the implementation.

https://reviews.llvm.org/D156720

If we can fit an entire vector of i1 into a single element, e.g. v32i1 ->
v1i32, then we can reverse it via vbrev.v.
We need to handle the case where the vector doesn't exactly fit into the larger
element type, e.g. v4i1 -> v1i8. In this case we shift up the reversed bits
afterwards.

Reviewed By: fakepaper56, 4vtomat

Differential Revision: https://reviews.llvm.org/D157614

https://reviews.llvm.org/D156708

Issue #10894 seems to claim this wasn't working. The test does seem to
work as intended, except the CHECKs added in
3ac4299d aren't doing anything since
it wasn't really using FileCheck.

Match how the generic implementation handles this. We now will leave
behind the dead other user for later passes to deal with.

https://reviews.llvm.org/D156707

This adds a RISC-V special case to ToolChain::GetDefaultDwarfVersion,
affecting Linux/Haiku/RISCVToolChain.

DWARF v5 .debug_loclists/.debug_rnglists's
DW_LLE_offset_pair/DW_RLE_offset_pair entry kinds utilitize `.uleb128 A-B`
directives where A and B reference local labels in code sections.
When A and B are separated by a RISC-V linker-relaxable instruction,
A-B is incorrectly folded without a relocation, causing incorrect debug
information.

```
void ext(void);
int foo(int x) {ext(); return 0;}
// DW_AT_location [DW_FORM_loclistx] of a DW_TAG_formal_parameter references a DW_LLE_offset_pair that can be incorrect after linker relaxation.

int ext(void);
void foo() { {
  int ret = ext();
  if (__builtin_expect(ret, 0))
    ext();
} }
// DW_AT_ranges [DW_FORM_rnglistx] of a DW_TAG_lexical_block references a DW_RLE_offset_pair that can be incorrect after linker relaxation.
```

D157657 will implement R_RISCV_SET_ULEB128/R_RISCV_SUB_ULEB128
relocations, fixing the issue, but the relocation is only supported by
bleeding-edge binutils 2.41 and not by lld/ELF yet.

The goal is to make the emitted DWARF correct after linking.
Many users don't care about the default DWARF version, but a linker
error will be unacceptable. Let's just downgrade the default DWARF
version, before binutils>=2.41 is more widely available.

An alternative compatibility option is to add a toggle to DwarfDebug.cpp,
but that doesn't seem like a good idea.

Reviewed By: asb, kito-cheng

Differential Revision: https://reviews.llvm.org/D157663