This recommits a87fccb3 with a fix to mark the destination operand
of the marker instruction as def, to fix a machine verifier failure.

This reverts the revert commit c0f2cea7.

Followup to D87604, having confirmed on PR47506 that we can use the llvm codegen expansion for fadd/fmul as well.

Differential Revision: https://reviews.llvm.org/D92940

Default expansion leads to repeated extensions/truncations to/from vXi16 which shuffle combining and demanded elts can't completely unravel.

Better just to promote (any_extend) the input and perform a vXi16 reduction.

We'll be able to remove a lot of this if we ever get decent legalization support for reduction intrinsics in SelectionDAG.

This avoids dumping the module post emitting a reproducer, which results in
many MB logs where a reproducer has already been neatly generated.

Differential Revision: https://reviews.llvm.org/D93165

Adds support for 3 ternary ops from SPIR-V extended instructions for
GLSL. Namely, adds support for FClamp, UClamp, and SClamp.

Reviewed By: antiagainst

Differential Revision: https://reviews.llvm.org/D92859

Try to merge AVX1/AVX2/AVX512 codegen checks where possible

Lines[i] is already trimmed 3 lines before

Differential Revision: https://reviews.llvm.org/D91996

BasicAA currently handles cases like Scale*V0 + (-Scale)*V1 where
V0 != V1, but does not handle the simpler case of Scale*V with
V != 0. Add it based on an isKnownNonZero() call.

I'm not passing a context instruction for now, because the existing
approach of always using GEP1 for context could result in symmetry
issues.

Differential Revision: https://reviews.llvm.org/D93162

Optimize toolchain regression test for VE by removing not a useful test
(-fuse-init-array test) and merge several tests to one test which checks
default behavior of driver.  Also add sysroot to reduce conflicts.

These are suggested in https://reviews.llvm.org/D92996.
Thank you so much.

Reviewed By: MaskRay

Differential Revision: https://reviews.llvm.org/D93084

This is a preparation step to remove those methods from OpState.

Reviewed By: mehdi_amini

Differential Revision: https://reviews.llvm.org/D93098

Zero bit integer types are supported by IntegerType for consistency,
but the asmparser never got updated. Allow them to be parsed, as
required to fix CIRCT issue #316

Differential Revision: https://reviews.llvm.org/D93089

Reviewed By: mehdi_amini

Differential Revision: https://reviews.llvm.org/D93004

On macOS/arm, signature verification has kill semantics by default.
Signature verification is cached with a file's inode (actually, vnode),
and if a new executable is copied over an existing file (which reuses
the inode), the cache isn't invalidated. So when the new executable
is executed, the kernel still has the old content's signature cached
and the kills the executable because the old signatue doesn't match
the new contents (https://openradar.appspot.com/FB8914243).

As workaround, rm the desitnation files first, to ensure they have
a fresh vnode (and hence no stale cached signature) after the copy.

Part of PR46647. See also e0e334a9 for a similar change.

https://reviews.llvm.org/D92631

We should fix them, but let's XFAIL them for now so that we can start
running check-llvm on bots and lock in the passing tests.

Part of PR46647.

The original tests have unneeded symbols and copy-relocation-zero-abs-addr.s
does not actually test anything.

Rewrite them and add copy-relocation-zero-addr.s instead.

Add --soname=b so that the address 0x203400 will be stable.  (When linking an
executable with %t.so, the path %t.so will be recorded in the DT_NEEDED entry if
%t.so doesn't have DT_SONAME. .dynstr will have varying lengths on different
systems.)

macOS/arm is w^x, so these tests don't work. Fixes these failures:

  LLVM-Unit :: Support/./SupportTests/AllocationTests/MappedMemoryTest.AllocAndRelease/5
  LLVM-Unit :: Support/./SupportTests/AllocationTests/MappedMemoryTest.AllocAndReleaseHuge/5
  LLVM-Unit :: Support/./SupportTests/AllocationTests/MappedMemoryTest.BasicWrite/5
  LLVM-Unit :: Support/./SupportTests/AllocationTests/MappedMemoryTest.DuplicateNear/5
  LLVM-Unit :: Support/./SupportTests/AllocationTests/MappedMemoryTest.EnabledWrite/3
  LLVM-Unit :: Support/./SupportTests/AllocationTests/MappedMemoryTest.EnabledWrite/4
  LLVM-Unit :: Support/./SupportTests/AllocationTests/MappedMemoryTest.EnabledWrite/5
  LLVM-Unit :: Support/./SupportTests/AllocationTests/MappedMemoryTest.MultipleAllocAndRelease/5
  LLVM-Unit :: Support/./SupportTests/AllocationTests/MappedMemoryTest.MultipleWrite/5
  LLVM-Unit :: Support/./SupportTests/AllocationTests/MappedMemoryTest.SuccessiveNear/5
  LLVM-Unit :: Support/./SupportTests/AllocationTests/MappedMemoryTest.UnalignedNear/5
  LLVM-Unit :: Support/./SupportTests/AllocationTests/MappedMemoryTest.ZeroNear/5
  LLVM-Unit :: Support/./SupportTests/AllocationTests/MappedMemoryTest.ZeroSizeNear/5

Part of PR46647.

[[GlobalISel][IRTranslator] Fix a crash when the use of an extractvalue is a non-dominated metadata use.

We don't expect uses to come before defs in the CFG, so allocateVRegs() asserted.

Fixes PR48211

[SimplifyCFG] FoldBranchToCommonDest(): bonus instrns must only be used by PHI nodes in successors (PR48450)

In particular, if the successor block, which is about to get a new
predecessor block, currently only has a single predecessor,
then the bonus instructions will be directly used within said successor,
which is fine, since the block with bonus instructions dominates that
successor. But once there's a new predecessor, the IR is no longer valid,
and we don't fix it, because we only update PHI nodes.

Which means, the live-out bonus instructions must be exclusively used
by the PHI nodes in successor blocks. So we have to form trivial PHI nodes.
which will then be successfully updated to recieve cloned bonus instns.

This all works fine, except for the fact that we don't have access to
the dominator tree, and we don't ignore unreachable code,
so we sometimes do end up having to deal with some weird IR.

Fixes https://bugs.llvm.org/show_bug.cgi?id=48450

Some of the pattern matching in PPCInstrVSX.td and node lowering involving vectors assumes 64bit mode.  This patch disables some of the unsafe pattern matching and lowering of BUILD_VECTOR in 32bit mode.

Reviewed By: Xiangling_L

Differential Revision: https://reviews.llvm.org/D92789

Differential Revision: https://reviews.llvm.org/D92782

Part of PR46644

CVP currently handles switches by checking an equality predicate
on all edges from predecessor blocks. Of course, this can only
work if the value being switched over is defined in a different block.

Replace this implementation with a call to getPredicateAt(), which
also does the predecessor edge predicate check (if not defined in
the same block), but can also do quite a bit more: It can reason
about phi-nodes by checking edge predicates for incoming values,
it can reason about assumes, and it can reason about block values.

As such, this makes the implementation both simpler and more
powerful. The compile-time impact on CTMark is in the noise.

We should fix them, but let's XFAIL them for now so that we can start
running check-clang on bots and lock in the passing tests.

Part of 46644.

These cover cases handled by getPredicateAt(), but not by the
current implementation:

 * Assumes based on context instruction.
 * Value from phi node in same block (using per-pred reasoning).
 * Value from non-phi node in same block (using block-val reasoning).

The getPayload/getMask/getPassThrough functions should return values
that could be composed into a masked load/store without any additional
type casts. The previous fix violated that.
Instead, convert scalar mask to a vector right before rescaling.

Differential Revision: https://reviews.llvm.org/D93018

Differential Revision: https://reviews.llvm.org/D93017

- Document which processors are supported by which runtimes.
- Add missing mappings for code object V2 note records

Differential Revision: https://reviews.llvm.org/D93016

When printing verification errors for ops with the incorrect number of
operand segments, print the required number as well as the actual
number. Split off from D93005.

Differential Revision: https://reviews.llvm.org/D93145

The last use of isLoop was removed on Apr 29, 2002 in commit
09bbb5c0 as part of an effort to
remove "old induction varaible cannonicalization pass built on top of
interval analysis".

AlignVectors treats all loaded/stored values as vectors of bytes,
and masks as corresponding vectors of booleans, so make getMask
produce a 1-element vector for scalars from the start.

This makes it possible to use update_llc_test_checks to manage tests
that check for incorrect x86 stack offsets. It does not yet modify any
test to make use of this new option.

The ABI demands a data16 prefix for lea in 64-bit LP64 mode, but not in
64-bit ILP32 mode. In both modes this prefix would ordinarily be
ignored, but the instructions may be changed by the linker to
instructions that are affected by the prefix.

Reviewed By: RKSimon

Differential Revision: https://reviews.llvm.org/D93157

This adds some basic MVE masked load/store costs, notably changing the
cost of legal loads/stores to the MVECostFactor and the cost of
scalarized instructions to 8*NumElts.

Differential Revision: https://reviews.llvm.org/D86538