AMDGPUTargetTransformInfo.h needs AMDGPUTargetMachine but relies on a
forward declaration of AMDGPUTargetMachine in AMDGPU.h.  This patch
adds a forward declaration right in AMDGPUTargetTransformInfo.h.

While we are at it, this patch removes the one in
AMDGPU.h, where it is unnecessary.

This demonstrates a missed optimization: the `vmv.x.s` instruction is
used to extract the element from the vector, and this instruction
already sign-extends the value to XLEN.

Fixes https://bugs.llvm.org/show_bug.cgi?id=48882.

If the input file does not exist (or has a reading error), the
following code will crash if there are two or more input addresses.

```
auto ResOrErr = Symbolizer.symbolizeInlinedCode(
  ModuleName, {Offset, object::SectionedAddress::UndefSection});
Printer << (error(ResOrErr) ? DILineInfo() : ResOrErr.get().getFrame(0));
```

For the first address, `symbolizeInlinedCode` returns an error.
For the second address, `symbolizeInlinedCode` returns an empty result
(not an error) and `.getFrame(0)` will crash.

Differential revision: https://reviews.llvm.org/D95609

Differential Revision: https://reviews.llvm.org/D95607

This patch fixes updating MemorySSA if the header contains memory
defs that do not clobber a duplicated instruction. We need to find the
first defining access outside the loop body and use that as defining
access of the duplicated instruction.

This fixes a crash caused by bee48685.

Differential Revision: https://reviews.llvm.org/D95562

D36116 refactored the logic of tests and removed the definition of TARGET_FLAGS, but left one use of it. Restore its definition for that one use, so that an x86_64 test is compiled with -m64.

Reviewed By: vitalybuka

Differential Revision: https://reviews.llvm.org/D93634

This patch adds an option to enable the new pass manager in
LTOCodeGenerator. It also updates a few tests with legacy PM specific
tests, which started failing after 6a59f056 when
LLVM_ENABLE_NEW_PASS_MANAGER=true.

This patch updates LTOCodeGenerator to use the utilities provided by
LTOBackend to run middle-end optimizations and backend code generation.

This is a first step towards unifying the code used by libLTO's C API
and the newer, C++ interface (see PR41541).

The immediate motivation is to allow using the new pass manager when
doing LTO using libLTO's C API, which is used on Darwin, among others.

With the changes, there are no codegen/stats differences when building
MultiSource/SPEC2000/SPEC2006 on Darwin X86 with LTO, compared
to without the patch.

Reviewed By: steven_wu

Differential Revision: https://reviews.llvm.org/D94487

This has been fixed by D89841.

Coalesce all version control options into a group

Differential Revision: https://reviews.llvm.org/D94938

Compact unwind entries have 8 bits for the encoding-table offset:
* offsets 0..126 reference the global commmon-encodings table, while
* offsets 127..255 reference a per-second-level-page table.
This diff teaches `llvm-objdump` to print this per-page encodings table.

Differential Revision: https://reviews.llvm.org/D93265

This is an optimized approach for D94155.

Previous code build the model that tile config register is the user of
each AMX instruction. There is a problem for the tile config register
spill. When across function, the ldtilecfg instruction may be inserted
on each AMX instruction which use tile config register. This cause all
tile data register clobber.

To fix this issue, we remove the model of tile config register. Instead,
we analyze the AMX instructions between one call to another. We will
insert ldtilecfg after the first call if we find any AMX instructions.

Reviewed By: LuoYuanke

Differential Revision: https://reviews.llvm.org/D95136

Source Drift happens when the sources are updated after profiling the binary
but before building the final optimized binary. If the source has changed since
the profiles were obtained, optimizing basic blocks might be sub-optimal. This
only applies to BasicBlockSection::List as it creates clusters of basic blocks
using basic block ids. Source drift can invalidate these groupings leading to
sub-optimal code generation with regards to performance.

PGO source drift for a particular function can be detected using function
metadata added in D95495.

When source drift is deected, disable basic block clusters by default
which can be re-enabled with  -mllvm option
bbsections-detect-source-drift=false.

Differential Revision: https://reviews.llvm.org/D95593

Tuples can occupy quite a lot of space, instead of printing out tuple type
everywhere, just use the type alias if larger (arbitrarily chose a bound for
now).

Differential Revision: https://reviews.llvm.org/D95707

[RISCV] Merge rv32 and rv64 vector fadd/fsub/fmul/fdiv sdnode tests into single tests files with 2 run lines.

The IR and CHECK lines are identical so just keep one copy.

As a fixme notes, both of these directory iterator implementations are
conceptually similar and duplicate the functionality of returning and uniquing
entries across two or more directories. This patch combines them into a single
class 'CombiningDirIterImpl'.

This also drops the 'Redirecting' prefix from RedirectingDirEntry and
RedirectingFileEntry to save horizontal space. There's no loss of clarity as
they already have to be prefixed with 'RedirectingFileSystem::' whenever
they're referenced anyway.

rdar://problem/72485443
Differential Revision: https://reviews.llvm.org/D94857

Update ElementsAttr::isValidIndex to handle ElementsAttr with a scalar. Scalar will have rank 0.

Reviewed By: rriddle

Differential Revision: https://reviews.llvm.org/D95663

Legacy Fortran implementations support an alternative form of the
PARAMETER statement; it differs syntactically from the standard's
PARAMETER statement by lacking parentheses, and semantically by
using the type and shape of the initialization expression to define
the attributes of the named constant.  (GNU Fortran gets that part
wrong; Intel Fortran and nvfortran have full support.)

This patch disables the old style PARAMETER statement by default, as
it is syntactically ambiguous with conforming assignment statements;
adds a new "-falternative-parameter-statement" option to enable it;
and implements it correctly when enabled.

Fixes https://bugs.llvm.org/show_bug.cgi?id=48774, in which a user
tripped over the syntactic ambiguity.

Differential Revision: https://reviews.llvm.org/D95697

v0.10 is tagged in V specification. Update the version to v0.10.

Differential Revision: https://reviews.llvm.org/D95680

Not all combinations of SEW and LMUL we need to support. For example, we
only need to support [M1, M2, M4, M8] for SEW = 64. There is no need to
define pseudos for PseudoVLSE64MF8, PseudoVLSE64MF4, and PseudoVLSE64MF2.

Differential Revision: https://reviews.llvm.org/D95667

A condition "mayLoadOrStore" is too broad for that function.

Differential Revision: https://reviews.llvm.org/D95700

This finishes getting rid of all the avoidable Dominator Tree recalculations
in X86 optimized codegen pipeline.

This doesn't help avoid any Dominator Tree recalculations just yet,
there's one more pass to go..

The previously agreed-upon deprecation period for
VectorType::getNumElements() has passed. This patch removes this method
and completes the refactor proposed in the RFC:
https://lists.llvm.org/pipermail/llvm-dev/2020-March/139811.html

Reviewed By: david-arm, rjmccall

Differential Revision: https://reviews.llvm.org/D95570

This patch emits "instr_prof_hash_mismatch" function annotation metadata if
there is a hash mismatch while applying instrumented profiles.

During the PGO optimized build using instrumented profiles, if the CFG of
the function has changed since generating the profile, a hash mismatch is
encountered. This patch emits this information as annotation metadata. We
plan to use this with Propeller which is done at the machine IR level.
Propeller is usually applied on top of PGO and a hash mismatch during
PGO could be used to detect source drift.

Differential Revision: https://reviews.llvm.org/D95495

LLVMTestingSupport is not part of libLLVM, and therefore can not
be linked to via LLVM_LINK_COMPONENTS.  Instead, it needs to be
specified explicitly to ensure that it is linked explicitly
even if LLVM_LINK_LLVM_DYLIB is used.  This is consistent with handling
in clangd.

Fixes PR#48931

Differential Revision: https://reviews.llvm.org/D95653

Currently, for a scf.parallel (i,j,k) after the loop collapsing to 1D is done, the
IVs would be traversed as for an scf.parallel(k,j,i).

Differential Revision: https://reviews.llvm.org/D95693

There are builders that do not have python 3.6. Revert until this situation can be rectified

This reverts commit 0703b075.

Previously, CMake would find any version of Python3. However, the project
claims to require 3.6 or greater, and 3.6 features are being used.

Reviewed By: yln

Differential Revision: https://reviews.llvm.org/D95635

I think every target will want to remove these in the same way. Rather than
making them all implement the same code, let's just put this in
InstructionSelect.

Differential Revision: https://reviews.llvm.org/D95652