This patch adds vector-predicated ("VP") reduction intrinsics corresponding to
each of the existing unpredicated `llvm.vector.reduce.*` versions. Unlike the
unpredicated reductions, all VP reductions have a start value. This start value
is returned when the no vector element is active.

Support for expansion on targets without native vector-predication support is
included.

This patch is based on the ["reduction
slice"](https://reviews.llvm.org/D57504#1732277) of the LLVM-VP reference patch
(https://reviews.llvm.org/D57504).

Reviewed By: craig.topper

Differential Revision: https://reviews.llvm.org/D104308

The current x86_64 implementations do not build on the windows bot
machine. We will enable them back after fixing the problem.

This patch adds an extra option to print the module after running one of
the OpenMPOpt passes if debugging is enabled. This makes it much easier
to inspect the effects of this pass when doing debugging.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D108146

his is a fix for PR43678, and is an alternate patch to D105723.

The basic issue we're running into is that LSR + SCEVExpander are moving the very instruction whose operand we're in the process of expanding. This breaks the subtle and ill-documented invariant which let LSR work. (Full story can be found here: https://reviews.llvm.org/D105723#2878473)

Rather than attempting a fix, this change just removes the optimization entirely. The code is entirely untested, and removing it appears to have no impact I can find.  This code was added back in 2014 by 1e12f856 with a single test which does not seem to actually test the hoisting logic.

From a philosophical standpoint, it also seems very strange to have the expander implementing optimizations which should live in a dedicated transform pass.

Differential Revision: https://reviews.llvm.org/D106178

Summary: Change and replace some functions which IE does not support. This patch is made as a continuation of D92928 revision. Also improve hot keys behavior.

Differential Revision: https://reviews.llvm.org/D107366

Fixes an issue with revision 5c6f748c and ad40cb88.

Adds an mcpu argument to the test command, preventing an invalid default
CPU from being used on some platforms.

Similar to D94907 (llvm-nm -D).

The output will match GNU objdump 2.37.
Older versions don't use ` (version)` for undefined symbols.

Reviewed By: jhenderson

Differential Revision: https://reviews.llvm.org/D108097

Reviewed By: gysit

Differential Revision: https://reviews.llvm.org/D108182

The utility can be reused by llvm-objdump -T.

Reviewed By: jhenderson

Differential Revision: https://reviews.llvm.org/D108096

Differential Revision: https://reviews.llvm.org/D108144

Broadcast is not worse than extract+insert of subvector.
https://godbolt.org/z/aPq98G6Yh

Reviewed By: RKSimon

Differential Revision: https://reviews.llvm.org/D105390

The loop had no side-effect since first committed in 642594ae.
While it is obvious what was intended, the code seems to never trigger.

Enable `-Wformat` again, and fix the offending instances.

Differential Revision: https://reviews.llvm.org/D108168

As explained in http://eel.is/c++draft/range.nonprop.cache#note-1, we
should allow copy and move elision to happen when calling emplace_deref
in non-propagating-cache. Before this change, the only way to emplace
into the non-propagating-cache was to call `__set(*it)`, which materialized
`*it` when binding it to the reference argument of `__set` and disabled
move elision.

As a fly-by change, this also renames `__set` to `__emplace` for consistency
and adds tests for it.

Differential Revision: https://reviews.llvm.org/D107932

Fixes an issue with revision 5c6f748c.

Move the test added in the above commit into the X86 folder, ensuring
that it is only run on targets where its triple is valid.

Some compilers started complaining about the test:

	tsan_trace_test.cpp:128:21: error: missing field 'type' initializer

Fix it by initializing all 5 fields, even though the type field will be
reset in the for loop.

Differential Revision: https://reviews.llvm.org/D108207

Fixes issue: https://bugs.llvm.org/show_bug.cgi?id=47983

The AsmLexer currently has an issue with lexing line comments in files
with CRLF line endings, in which it reads the carriage return as being
part of the line comment. This causes an error for certain valid comment
layouts; this patch fixes this by excluding the carriage return from the
line comment.

Differential Revision: https://reviews.llvm.org/D90234

The corresponding definition was removed on Nov 29, 2016 in commit
5a0a2e64.

AIX may use cat_64 for 64 bit cat, this is just update the lit test to accept the name as well.

Reviewed By: #powerpc, shchenz

Differential Revision: https://reviews.llvm.org/D108149

Removed AArch64 usage of the getMaxVScale interface, replacing it with
the vscale_range(min, max) IR Attribute.

Reviewed By: paulwalker-arm

Differential Revision: https://reviews.llvm.org/D106277

Adjust the names of helper function objects to represent better what
they do, as suggested in the review of D107098.

This enables subreg liveness in the arm backend when MVE is present,
which allows the register allocator to detect when subregister are
alive/dead, compared to only acting on full registers. This can helps
produce better code on MVE with the way MQPR registers are made up of
SPR registers, but is especially helpful for MQQPR and MQQQQPR
registers, where there are very few "registers" available and being able
to split them up into subregs can help produce much better code.

Differential Revision: https://reviews.llvm.org/D107642

As a part of D107642, this adds pseudo instructions for MQQPR and
MQQQQPR register classes, that can spill and reloads entire registers
whilst keeping them combined, not splitting them into multiple D subregs
that a VLDMIA/VSTMIA would use. This can help certain analyses, and
helps to prevent verifier issues with subreg liveness.

If both operands are negated, we can invert the min/max and do
the negation after:
smax (neg nsw X), (neg nsw Y) --> neg nsw (smin X, Y)
smin (neg nsw X), (neg nsw Y) --> neg nsw (smax X, Y)

This is visible as a remaining regression in D98152. I don't see
a way to generalize this for 'unsigned' or adapt Negator to
handle it. This only appears to be safe with 'nsw':
https://alive2.llvm.org/ce/z/GUy1zJ

Differential Revision: https://reviews.llvm.org/D108165

Combine two G_PTR_ADDs, but keep the register bank of the constant.
That way, the combine can be used in post-regbank-select combines.

Introduce two helper methods in CombinerHelper, getRegBank and
setRegBank that get and set an optional register bank to a register.
That way, they can be used before and after register bank selection.

Differential Revision: https://reviews.llvm.org/D103326

LLDB is using LLVM's target-specific disassembler which is only available when
the respective LLVM target has been enabled in the build config.

This patch just skips the test if there is no arm64 target (and its
disassembler) available in the current build config.

Reviewed By: jasonmolenda

Differential Revision: https://reviews.llvm.org/D108145

In current implementation, the instruction to be sunk will be inserted before the target instruction without considering the def-use tree,
which may case Instruction does not dominate all uses error. We need to choose a suitable location to insert according to the use chain

Reviewed By: dmgreen

Differential Revision: https://reviews.llvm.org/D107262

This reapplies 54a61c94, its follow up in 547b7125, which were
reverted 95fe61e6. Original commit message:

VarLoc based LiveDebugValues will abandon variable location propagation if
there are too many blocks and variable assignments in the function. If it
didn't, and we had (say) 1000 blocks and 1000 variables in scope, we'd end
up with 1 million DBG_VALUEs just at the start of blocks.

Instruction-referencing LiveDebugValues should honour this limitation too
(because the same limitation applies to it). Hoist the relevant command
line options into LiveDebugValues.cpp and pass it down into the
implementation classes as an argument to ExtendRanges. I've duplicated all
the run-lines in live-debug-values-cutoffs.mir to have an
instruction-referencing flavour.

Differential Revision: https://reviews.llvm.org/D107823

Improves on the unused variable fix from rG43575620

Add `shl` instruction to the DAG post-dominated by `trunc`, allowing
TruncInstCombine to reduce bitwidth of expressions containing left shifts.

The only thing we need to check is that the target bitwidth must be wider
than the maximal shift amount: https://alive2.llvm.org/ce/z/AwArqu

Part of https://reviews.llvm.org/D107766

Differential Revision: https://reviews.llvm.org/D108091

Precommit test for D107766/D108091. Also move fixed test for PR50555
from SLPVectorizer/X86/ to PhaseOrdering/X86/ subdirectory.

They were previously unconstrained, which allowed them to be reordered
before the shadow memory write.

Reviewed By: eugenis

Differential Revision: https://reviews.llvm.org/D107901

Initial commit test.

Remove duplicate methods used to check iterator types.

Reviewed By: aartbik

Differential Revision: https://reviews.llvm.org/D108102

There is some discussion on the bitcast for vector and x86_amx at https://reviews.llvm.org/D99152. This patch is to introduce a x86 specific cast for vector and x86_amx, so that it can avoid some unnecessary optimization by middle-end. On the other way, we have to optimize the x86 specific cast by ourselves. This patch also optimize the cast operation to eliminate redundant code.

Reviewed By: LuoYuanke

Differential Revision: https://reviews.llvm.org/D107544

Make sure return type is int type.

Differential Revision: https://reviews.llvm.org/D108131

Change-Id: Ic02f07d1234cd51b6ed78c3fecd2cb1d6acd5644

This patch refactors the `FrontendAction` class. It merely moves code
around so that re-using it is easier. No new functionality is
introduced.

1. Three new member methods are introduced: `RunPrescan`, `RunParse`,
`RunSemanticChecks`.
2. The following free functions are re-implemented as member methods:
  * `reportFatalSemanticErrors`
  * `reportFatalScanningErrors`
  * `reportFatalParsingErrors`
  * `reportFatalErrors`
`reportFatalSemanticErrors` is updated to resemble the other error
reporting functions and to make the API more consistent.
3. The `BeginSourceFileAction` methods are simplified and the unused
input argument is deleted.

Differential Revision: https://reviews.llvm.org/D108130