Changes

[SCEV] Refactor isHighCostExpansionHelper

To enable the cost of constants, the helper function has been
reorganised:
- A struct has been introduced to hold SCEV operand information so
  that we know the user of the operand, as well as the operand index.
  The Worklist now uses instead instead of a bare SCEV.
- The costing of each SCEV, and collection of its operands, is now
  performed in a helper function.

Differential Revision: https://reviews.llvm.org/D86050

[MLIR][Standard] Add `dynamic_tensor_from_elements` operation

With `dynamic_tensor_from_elements` tensor values of dynamic size can be
created. The body of the operation essentially maps the index space to tensor
elements.

Declare SCF operations in the `scf` namespace to avoid name clash with the new
`std.yield` operation. Resolve ambiguities between `linalg/shape/std/scf.yield`
operations.

Differential Revision: https://reviews.llvm.org/D86276

[obj2yaml] Add support for dumping the .debug_str section.

This patch adds support for dumping the .debug_str section to obj2yaml.

Reviewed By: jhenderson

Differential Revision: https://reviews.llvm.org/D86867

[X86][AVX] Use lowerShuffleWithPERMV in shuffle combining to support non-VLX targets

lowerShuffleWithPERMV allows us to use the ZMM variants for 128/256-bit variable shuffles on non-VLX AVX512 targets.

This is another step towards shuffle combining through between vector widths - we still end up with an annoying regression (combine_vpermilvar_vperm2f128_zero_8f32) but we're going in the right direction....

X86AvoidStoreForwardingBlocks.cpp - use unsigned for Opcode values. NFCI.

Fixes clang-tidy cppcoreguidelines-narrowing-conversions warnings.

[llvm-readobj/elf] - Introduce Relocation<ELFT> helper.

It removes templating for Elf_Rel[a] handling that we
introduced earlier and introduces a helper class instead.

It was briefly discussed in D87087, which showed,
why having templates is probably not ideal for the generalization
of dumpers code.

Differential revision: https://reviews.llvm.org/D87141

LegalizeTypes.h - remove orphan SplitVSETCC declaration. NFCI.

The implementation no longer exists

[MLIR][Shape] Merge `shape` to `std`/`scf` lowerings.

Merge the two lowering passes because they are not useful by themselves. The new
pass lowers to `std` and `scf` is considered an auxiliary dialect.

See also
https://llvm.discourse.group/t/conversions-with-multiple-target-dialects/1541/12

Differential Revision: https://reviews.llvm.org/D86779

[MLIR] Fix Win test due to partial order of CHECK directives

Differential Revision: https://reviews.llvm.org/D87230

[InstCombine] give a name to an intermediate value for easier tracking; NFC

As noted in PR47430, we probably want to conditionally include 'nsw'
here anyway, so we are going to need to fill out the optional args.

[InstCombine] add test with more unreachable insts; NFC

Goes with D87149

[SelectionDAG] memcpy expansion of const volatile struct ignores const zero

In getMemcpyLoadsAndStores(), a memcpy where the source is a zero constant is expanded to a MemOp::Set instead of a MemOp::Copy, even when the memcpy is volatile.
This is incorrect.

The fix is to add a check for volatile, and expand to MemOp::Copy in the volatile case.

Reviewed By: chill

Differential Revision: https://reviews.llvm.org/D87134

[X86] Replace EmitX86AddSubSatExpr with EmitX86BinaryIntrinsic generic helper. NFCI.

Feed the Intrinsic::ID value directly instead of via the IsSigned/IsAddition bool flags.

MachineStableHash.h - remove MachineInstr.h include. NFC.

Use forward declarations and move the include to MachineStableHash.cpp

[llvm-readobj/elf] - Generalize the code for printing dynamic relocations. NFCI.

Currently we have 2 large `printDynamicRelocations` methods that
have a very similar code for GNU/LLVM styles.

This patch removes the duplication and renames them to `printDynamicReloc`
for consistency.

Differential revision: https://reviews.llvm.org/D87087

Revert "[MLIR][Shape] Merge `shape` to `std`/`scf` lowerings."

This reverts commit 15acdd75439b402e993ebe0dbf8eb02e9b88bbdc.

Revert "[clang] Prevent that Decl::dump on a CXXRecordDecl deserialises further declarations."

This reverts commit 0478720157f6413fad7595b8eff9c70d2d99b637. This probably
doesn't work when forcing deserialising while dumping (which the ASTDumper
optionally supports).

[gn build] Port 23f700c785a

[X86] Update SSE/AVX ABS intrinsics to emit llvm.abs.* (PR46851)

We're now getting close to having the necessary analysis/combines etc. for the new generic llvm.abs.* intrinsics.

This patch updates the SSE/AVX ABS vector intrinsics to emit the generic equivalents instead of the icmp+sub+select code pattern.

Differential Revision: https://reviews.llvm.org/D87101

[X86][SSE] Move llvm.x86.ssse3.pabs.*.128 intrinsics to ssse3-intrinsics-x86-upgrade.ll

These have been auto upgraded for some time so this is just a tidyup.

[KillTheDoctor/CMake] Add missing keyword PRIVATE in target_link_libraries

Add PRIVATE keyword in target_link_libraries to prevent CMake Error on Windows.

While trying to compile llvm/clang on Windows, the following CMake error occurred. The reason is a missing PUBLIC/PRIVATE/INTERFACE keyword in target_link_libraries.

`
CMake Error at utils/KillTheDoctor/CMakeLists.txt:5 (target_link_libraries):
  The keyword signature for target_link_libraries has already been used with
  the target "KillTheDoctor".  All uses of target_link_libraries with a
  target must be either all-keyword or all-plain.

  The uses of the keyword signature are here:

   * cmake/modules/AddLLVM.cmake:771 (target_link_libraries)
`

Reviewed By: tambre

Differential Revision: https://reviews.llvm.org/D87203

[llvm-readobj] - Remove code duplication when printing dynamic relocations. NFCI.

LLVM style code can be simplified to avoid the duplication of logic
related to printing dynamic relocations.

Differential revision: https://reviews.llvm.org/D87089

[NFC][PowerPC] Add tests in constants-i64.ll.

[mlir] Take ValueRange instead of ArrayRef<Value> in StructuredIndexed

This was likely overlooked when ValueRange was first introduced. There is no
reason why StructuredIndexed needs specifically an ArrayRef so use ValueRange
for better type compatibility with the rest of the APIs.

Reviewed By: nicolasvasilache, mehdi_amini

Differential Revision: https://reviews.llvm.org/D87127

[X86][SSE] Use llvm.abs.* vector intrinsics instead of old (deprecated) SSE/AVX intrinsics for combine tests

This also allows us to extend testing to SSE2+ targets

[X86] Auto upgrade SSE/AVX PABS intrinsics to generic Intrinsic::abs

Minor followup to D87101, we were expanding this to a neg+icmp+select pattern like we were in CGBuiltin

Follow up of rG5f1cad4d296a, slightly reduced test case. NFC.