This reverts commit d3bf0bb1.
This causes compilation in certain cases to fail.
Reproducer TBD.

Reviewed By: ftynse, herhut

Differential Revision: https://reviews.llvm.org/D93752

Introduce a translation of OpenMP workshare loop construct to LLVM IR. This is
a minimalist version to enable the pipeline and currently only supports static
loop schedule (default in the specification) on non-collapsed loops. Other
features will be added on per-need basis.

Reviewed By: kiranchandramohan

Differential Revision: https://reviews.llvm.org/D92055

[mlir] NFC: Remove ConvertToLLVMPattern::getDataPtr(). All call sites use getStridedElementPtr() now.

Reviewed By: ftynse

Differential Revision: https://reviews.llvm.org/D93751

This was long overdue. Replace the outdated type syntax with the new syntax,
and update the description of how memref load/stores are handled to reflect the
latest changes in the implementation.

Reviewed By: herhut

Differential Revision: https://reviews.llvm.org/D93555

Use getPtrBaseWithConstantOffset in selectFlatOffsetImpl to fold more
vgpr+constant addresses.

Differential Revision: https://reviews.llvm.org/D93692

This patch defines vfwmacc, vfwnmacc, vfwmsc, vfwnmsac intrinsics
and lower to V instructions.
We work with @rogfer01 from BSC to come out this patch.

Authored-by: Roger Ferrer Ibanez <rofirrim@gmail.com>
Co-Authored-by: ShihPo Hung <shihpo.hung@sifive.com>

Differential Revision: https://reviews.llvm.org/D93693

Currently llvm-readelf might print "OS Specific/Processor Specific/<unknown>"
hint when dumping the ELF file type. The patch teaches llvm-readobj to do the same.

This fixes https://bugs.llvm.org/show_bug.cgi?id=40868

I am removing `Object/elf-unknown-type.test` test because it is not in the right place,
it is outdated and very limited.
The `readobj/ELF/file-types.test` checks the functionality much better.

Differential revision: https://reviews.llvm.org/D93689

Define vmerge/vfmerge intrinsics and lower to V instructions.

Include support for vector-vector vfmerge by vmerge.vvm.

We work with @rogfer01 from BSC to come out this patch.

Reviewed By: craig.topper

Differential Revision: https://reviews.llvm.org/D93674

Define the vfmin, vfmax IR intrinsics for the respective V instructions.

Authored-by: Roger Ferrer Ibanez <rofirrim@gmail.com>
Co-Authored-by: Evandro Menezes <evandro.menezes@sifive.com>

Differential Revision: https://reviews.llvm.org/D93673

The last use of the function was removed on Sep 30, 2017 in commit
9b926e90.

Introduce `Vec` records, each bundling all information related to a single SIMD
lane interpretation. This lets TableGen definitions take a single Vec parameter
from which they can extract information rather than taking multiple redundant
parameters. This commit refactors all of the SIMD load and store instruction
definitions to use the new `Vec`s. Subsequent commits will similarly refactor
additional instruction definitions.

Differential Revision: https://reviews.llvm.org/D93660

Returning int64_t was arbitrarily limiting for wide integer types, and
the functions should handle the full generality of the IR.

Also changes the full form which returns the originally defined
vreg. Add another wrapper for the common case of just immediately
converting to int64_t (arguably this would be useful for the full
return value case as well).

One possible issue with this change is some of the existing uses did
break without conversion to getConstantVRegSExtVal, and it's possible
some without adequate test coverage are now broken.

It's an extension to ld64, but all the other ports have it, and
someone asked for it in PR43721.

While here, change the COFF help text to match the other ports.

Differential Revision: https://reviews.llvm.org/D93491

This patch defines vfmadd/vfnmacc, vfmsac/vfnmsac, vfmadd/vfnmadd,
and vfmsub/vfnmsub lower to V instructions.

Authored-by: Roger Ferrer Ibanez <rofirrim@gmail.com>
Co-Authored-by: ShihPo Hung <shihpo.hung@sifive.com>

Differential Revision: https://reviews.llvm.org/D93691

This patch defines vwmacc[u|su|us] intrinsics and lower to V instructions.

We work with @rogfer01 from BSC to come out this patch.

Authored-by: Roger Ferrer Ibanez <rofirrim@gmail.com>
Co-Authored-by: ShihPo Hung <shihpo.hung@sifive.com>

Differential Revision: https://reviews.llvm.org/D93675

This patch adds intrinsics for vslide1up, vslide1down, vfslide1up, vfslide1down.

Authored-by: Roger Ferrer Ibanez <rofirrim@gmail.com>
Co-Authored-by: ShihPo Hung <shihpo.hung@sifive.com>

Differential Revision: https://reviews.llvm.org/D93608

Differential Revision: https://reviews.llvm.org/D76572

The primary and secondary allocators will need to share this bit,
so move the management of the bit to the combined allocator and
make useMemoryTagging() a free function.

Differential Revision: https://reviews.llvm.org/D93730

This will allow the secondary allocator to access the
MaySupportMemoryTagging bool.

Differential Revision: https://reviews.llvm.org/D93729

It does not seem to fold offsets but this is not specific
to the flat scratch as getPtrBaseWithConstantOffset() does
not return the split for these tests unlike its SDag
counterpart.

Differential Revision: https://reviews.llvm.org/D93670

Adjust SITargetLowering::allowsMisalignedMemoryAccessesImpl for
unaligned flat scratch support. Mostly needed for global isel.

Differential Revision: https://reviews.llvm.org/D93669

This will allow the primary and secondary allocators to share
the MaySupportMemoryTagging bool.

Differential Revision: https://reviews.llvm.org/D93728

One common situation is to create a lot of IR at a well known location,
e.g. when doing a big rewrite from one dialect to another where you're expanding
ops out into lots of other ops.

For these sorts of situations, it is annoying to pass the location into
every create call.  As we discused in a few threads on the forum, a way to help
with this is to produce a new sort of builder that holds a location and provides
it to each of the create<> calls automatically.

This patch implements an ImplicitLocOpBuilder class that does this.  We've had
good experience with this in the CIRCT project, and it makes sense to upstream to
MLIR.

I picked a random pass to adopt it to show the impact, but I don't think there is
any particular need to force adopt it in the codebase.

Differential Revision: https://reviews.llvm.org/D93717

Add support to the OpenMP web pages for environment variables supported
by Libomptarget and their usage.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D93723

LLVMType contains multiple instance methods that were introduced initially for
compatibility with LLVM API. These methods boil down to `cast` followed by
type-specific call. Arguably, they are mostly used in an LLVM cast-follows-isa
anti-pattern. This doesn't connect nicely to the rest of the MLIR
infrastructure and actively prevents it from making the LLVM dialect type
system more open, e.g., reusing built-in types when appropriate. Remove such
instance methods and replaces their uses with apporpriate casts and methods on
derived classes. In some cases, the result may look slightly more verbose, but
most cases should actually use a stricter subtype of LLVMType anyway and avoid
the isa/cast.

Reviewed By: mehdi_amini

Differential Revision: https://reviews.llvm.org/D93680

These instructions previously used prefixes like v8x16 to signify that they were
agnostic between float and int interpretations. We renamed these instructions to
remove this form of prefix in https://github.com/WebAssembly/simd/issues/297 and
https://github.com/WebAssembly/simd/issues/316 and this commit brings the names
in LLVM up to date.

Differential Revision: https://reviews.llvm.org/D93722

Reviewed By: herhut

Differential Revision: https://reviews.llvm.org/D93204

Instead of always locking/unlocking a contended mutex, we now do one atomic read
in the common case, and one read + one exchange if the timer has expried.

Also use this for memory profiling which has similar/compatible requirements.

Differential Revision: https://reviews.llvm.org/D93726

I think this is NFC currently, but the bug would be exposed
when we allow binary intrinsics (maxnum, etc) as candidates
for reductions.

The code in matchAssociativeReduction() is using
OperationData::getNumberOfOperands() when comparing whether
the "EdgeToVisit" iterator is in-bounds, so this code must
use the same (potentially offset) operand value to set
the "EdgeToVisit".

Before this, a hello world program would contain many many unnecessary
entries in its string table.

No behavior change, just makes the string table in the output smaller
and more like ld64's.

Differential Revision: https://reviews.llvm.org/D93711