The discussion on forum:
https://llvm.discourse.group/t/bug-in-partial-dialect-conversion/4115

The `applyPartialConversion` didn't handle the operations, that were
marked as illegal inside dynamic legality callback.
Instead of reporting error, if such operation was not converted to legal set,
the method just added it to `unconvertedSet` in the same way as unknown operations.

This patch fixes that and handle dynamically illegal operations as well.

The patch includes 2 fixes for existing passes:

* `tensor-bufferize` - explicitly mark `std.return` as legal.
* `convert-parallel-loops-to-gpu` - ugly fix with marking visited operations
  to avoid recursive legality checks.

Reviewed By: rriddle

Differential Revision: https://reviews.llvm.org/D108505

Reviewed By: lattner, ftynse

Differential Revision: https://reviews.llvm.org/D109223

These have been here since r215992, guarding the calls to HostInfo, but
their purpose unclear -- HostInfoLinux provides these functions and they
work fine.

All transforms of IndVars have prerequisite requirement of LCSSA and LoopSimplify
form and rely on it. Added test that shows that this actually stands.

This reverts commit 6fec6552.

The patch was reverted on incorrect claim that this patch may break LCSSA form
when the loop is not in a simplify form. All IndVars' transform insure that
the loop is in simplify and LCSSA form, so if it wasn't broken before this
transform, it will also not be broken after it.

Reviewed By: michaelrj

Differential Revision: https://reviews.llvm.org/D109952

There is a piece of logic that uses the fact that signed and unsigned
versions of the same predicate are equivalent when both values are
non-negative. It's also true when both of them are negative.

Differential Revision: https://reviews.llvm.org/D109957
Reviewed By: nikic

This should probably be rendered as "std::nullptr_t" but for now clang
uses the unqualified name (which is ambiguous with possible user defined
name in the global namespace), so match that here.

Helps with the regression noted on D109065 - don't truncate a broadcast source if the source has multiple uses.

The scev-based salvaging for LSR can sometimes produce unnecessarily
verbose expressions. This patch adds logic to detect when the value to
be recovered and the induction variable differ by only a constant
offset. Then, the expression to derive the current iteration count can
be omitted from the dbg.value in favour of the offset.

Reviewed by: aprantl

Differential Revision: https://reviews.llvm.org/D109044

llvm-dwarfdump: Pretty printing types including a space between const and parenthesized references/pointers to arrays

Unlike psadbw, mpsadbw is not commutable because of how it operates
on blocks. We already marked as not commutable for MachineIR, but
had it commutable for the tablegened isel patterns.

Fixes PR51908.

DWARFDie: Improve type printing for function and array types - with qualifiers (cv/reference) and pointers to them

Both ports are required in most cases. Update the uops counts + port usage based off the most recent llvm-exegesis captures (PR36895) and what Intel AoM / Agner / InstLatX64 reports as well.

Noticed while trying to improve fp costs for vectorization via the D103695 helper script.

We can combine unary shuffles into either of SHUFPS's inputs and adjust the shuffle mask accordingly.

Unlike general shuffle combining, we can be more aggressive and handle multiuse cases as we're not going to accidentally create additional shuffles.

Move most type tests to a pre-generated assembly file to make it easier
to add more weird cases without having to hand craft more DWARF.

Move the novel array types that aren't reachable via clang-generated
DWARF to a separate file for easy maintenance.

Based off a mixture of llvm-exegesis captures (PR36895) and Intel AoM / Agner / InstLatX64 reports.

[X86][TLI] SimplifyDemandedVectorEltsForTargetNode(): don't break apart broadcasts from which not just the 0'th elt is demanded

Apparently this has no test coverage before D108382,
but D108382 itself shows a few regressions that this fixes.

It doesn't seem worthwhile breaking apart broadcasts,
assuming we want the broadcasted value to be preset in several elements,
not just the 0'th one.

Reviewed By: RKSimon

Differential Revision: https://reviews.llvm.org/D108411

[X86] lowerShuffleAsDecomposedShuffleMerge(): if both inputs are broadcastable/identities, canonicalize broadcasts as such

Split off from D108253.
Broadcast is simpler than any other shuffle we might produce
to do what we want to do here, so prefer it.

Reviewed By: RKSimon

Differential Revision: https://reviews.llvm.org/D108382

[X86] combineX86ShufflesRecursively(): call SimplifyMultipleUseDemandedVectorElts() on after finishing recursing

This was suggested in https://reviews.llvm.org/D108382#inline-1039018,
and it avoids regressions in that patch.

Reviewed By: RKSimon

Differential Revision: https://reviews.llvm.org/D109065

If we transform these, we have to propagate no-wrap/undef carefully.

These were apparently missing, having no pattern that could convert a
VGETLANEu of a v4f16 to an i32. Added bf16 whilst here, following the
same code.

1. Add missing indent in CondBranchOp
2. Remove indent in block label

Differential Revision: https://reviews.llvm.org/D109805

Both ports are required, for reg and mem variants - we can also use the WriteFComX class directly and remove the unnecessary InstRW overrides. Matches what Intel AoM / Agner / InstLatX64 report as well.

Optimize (add (shl x, c0), (shl y, c1)) ->
         (SLLI (SH*ADD x, y), c1), if c0-c1 == 1/2/3.

Reviewed By: craig.topper, luismarques

Differential Revision: https://reviews.llvm.org/D108916

Should avoid some regressions in D109065

Reviewed By: RKSimon

Differential Revision: https://reviews.llvm.org/D109989

Missed this one in 80110aaf. This
is another test mixing up alias scopes and alias scope lists.

Mostly this fixes cases where !noalias or !alias.scope were passed
a scope rather than a scope list. In some cases I opted to drop
the metadata entirely instead, because it is not really relevant
to the test.