This is motivated by the examples and discussion in:
https://llvm.org/PR51245
...and related bugs.

By using vector compares and vector logic, we can convert 2 'set'
instructions into 1 'movd' or 'movmsk' and generally improve
throughput/reduce instructions.

Unfortunately, we don't have a complete vector compare ISA before
AVX, so I left SSE-only out of this patch. Ie, we'd need extra logic
ops to simulate the missing predicates for SSE 'cmpp*', so it's not
as clearly a win.

Differential Revision: https://reviews.llvm.org/D110342

Refactor handleArgs() to use that method.

Differential Revision: https://reviews.llvm.org/D110350

This reverts commit bb9333c3.

This exposes another existing bug that causes an infinite loop as shown in
D110170
...so reverting while I look at another fix.

The stress test does various assorted things
(memory accesses, function calls, atomic operations,
thread creation/join, intercepted libc calls)
in multiple threads just to stress various parts
of the runtime.

Reviewed By: melver

Differential Revision: https://reviews.llvm.org/D110416

It caused compiler crashes, see comment on the code review for repro.

> This is basically D108837 but for jump threading. Free instructions
> should be ignored for the threading decision. JumpThreading already
> skips some free instructions (like pointer bitcasts), but does not
> skip various free intrinsics -- in fact, it currently gives them a
> fairly large cost of 2.
>
> Differential Revision: https://reviews.llvm.org/D110290

This reverts commit 1e3c6fc7.

Add a test for __tsan_flush_memory() and for background
flushing of the runtime memory.

Reviewed By: melver

Differential Revision: https://reviews.llvm.org/D110409

This reverts commit 5ce89279.
Makes clang crash, see comments on https://reviews.llvm.org/D109844

In build_symbolizer.sh we can safely remove the -eu argument from the shebang (which is an unportable construct), as the scripts sets **-e** and **-u** already.

Differential Revision: https://reviews.llvm.org/D110039

Identified in D109359.

Reviewed By: uweigand

Differential Revision: https://reviews.llvm.org/D109598

Refactor Socket::DecodeHostAndPort() to use LLVM API over redundant
LLDB API.  In particular, this means llvm::Regex, llvm::Error return
type and llvm::to_integer().

While at it, change the port type from int32_t to uint16_t.  The method
never returns any value outside this range, and using the correct type
allows us to rely on getAsInteger()'s implicit overflow check.

Differential Revision: https://reviews.llvm.org/D110391

There are several places in the code that are currently broken where
we assume an Instruction is always a member of a BasicBlock that
lives in a Function. This is a problem specifically when
attempting to get the vscale_range attribute. This patch adds checks
that an Instruction's parent also has a parent!

I've added a test for a function-less @llvm.vscale intrinsic call here:

  unittests/Analysis/ValueTrackingTest.cpp

Add support to create unique name for namelist group and be able to
deconstruct them.

This patch is part of the upstreaming effort from fir-dev branch.

Reviewed By: jeanPerier

Differential Revision: https://reviews.llvm.org/D110331



Co-authored-by: Jean Perier <jperier@nvidia.com>
Co-authored-by: Eric Schweitz <eschweitz@nvidia.com>

This reverts commit a6daf992.  It causes
buildbot regressions, I'll investigate.

Refactor Socket::DecodeHostAndPort() to use LLVM API over redundant
LLDB API.  In particular, this means llvm::Regex, llvm::Error return
type and llvm::to_integer().

While at it, change the port type from int32_t to uint16_t.  The method
never returns any value outside this range, and using the correct type
allows us to rely on getAsInteger()'s implicit overflow check.

Differential Revision: https://reviews.llvm.org/D110391

In repairIntervalsInRange, if the new instructions refer to subregs but
the old instructions did not, make sure any existing live interval for
the superreg is updated to have subranges. Also skip repairing any range
that we have recalculated from scratch, partly for efficiency but also
to avoids some cases that repairOldRegInRange can't handle.

The existing test/CodeGen/AMDGPU/twoaddr-regsequence.mir provides some
test coverage for this change: when TwoAddressInstructionPass converts
REG_SEQUENCE into subreg copies, the live intervals will now get
subranges and MachineVerifier will verify that the subranges are
correct. Unfortunately MachineVerifier does not complain if the
subranges are not present, so the test also passed before this patch.

This patch also fixes ~800 of the ~1500 failures in the whole CodeGen
lit test suite when -early-live-intervals is forced on.

Differential Revision: https://reviews.llvm.org/D110328

The fix applied in D23303 "LiveIntervalAnalysis: fix a crash in repairOldRegInRange"
was over-zealous. It would bail out when the end of the range to be
repaired was in the middle of the first segment of the live range of
Reg, which was always the case when the range contained a single def of
Reg.

This patch fixes it as suggested by Matthias Braun in post-commit review
on the original patch, and tests it by adding -early-live-intervals to
a selection of existing lit tests that now pass.

(Note that D23303 was originally applied to fix a crash in
SILoadStoreOptimizer, but that is now moot since D23814 updated
SILoadStoreOptimizer to run before scheduling so it no longer has to
update live intervals.)

Differential Revision: https://reviews.llvm.org/D110238

Unrevert with some changes to the tests:
- Add -verify-machineinstrs to check for remaining problems in live
  interval support in TwoAddressInstructionPass.
- Drop test/CodeGen/AMDGPU/extract-load-i1.ll since it suffers from
  some of those remaining problems.

They create a new vector with the list, so they can be const.

Reviewed By: fhahn

Differential Revision: https://reviews.llvm.org/D110394

Add support for intersecting, subtracting, complementing and checking equality of sets having divisions.

Reviewed By: bondhugula

Differential Revision: https://reviews.llvm.org/D110138

This canonicalization pattern complements the tensor.cast(pad_tensor) one in
propagating constant type information when possible. It contributes to the
feasibility of pad hoisting.

Reviewed By: nicolasvasilache

Differential Revision: https://reviews.llvm.org/D110343

When moving an entire basic block BB before InsertPoint, currently
we check for all instructions whether the operands dominates
InsertPoint, however, this can be improved such that even an
operand does not dominate InsertPoint, as long as it appears as
a previous instruction in the same BB, it is safe to move.

Reviewed By: Whitney

Differential Revision: https://reviews.llvm.org/D110378

Fixes issue found on greendragon buildbot, in which an incorrectly
indented statement following an if block led to entire frames being
dropped instead of simply filtering unneeded watches.

This reverts commit 1f44fa3a.

[analyzer] Retrieve a value from list initialization of constant array declaration in a global scope.

Summary: Fix the point that we didn't take into account array's dimension. Retrieve a value of global constant array by iterating through its initializer list.

Differential Revision: https://reviews.llvm.org/D104285

Fixes: https://bugs.llvm.org/show_bug.cgi?id=50604

This reverts commit 0c2a4548.

This was my mistake. When gdb can find its data directory it'll
import it automatically. If it can't (like when you're using a
version from a build folder) you need to give it the data
directory path.

We're safe to assume gdb is installed for testing purposes
so it'll import it for us.

Add the tail policy argument to Clang builtins. There
are two policies for tail elements. Tail agnostic means users do not
care about the values in the tail elements and tail undisturbed means
the values in the tail elements need to be kept after the operation. In
order to let users control the tail policy, we add an additional
argument at the end of the argument list.

For unmasked operations, we have no maskedoff and the tail policy is
always tail agnostic. If users want to keep tail elements under unmasked
operations, they could use all one mask in the masked operations to do
it. So, we only add the additional argument for masked operations for
most cases. There are exceptions listed below.

In this patch, we do not handle the following cases to reduce the
complexity of the patch. There could be two separate patches for them.

Use dest argument to control tail policy
vmerge.vvm/vmerge.vxm/vmerge.vim (add _t builtins with additional dest
argument)
vfmerge.vfm (add _t builtins with additional dest argument)
vmv.v.v (add _t builtins with additional dest argument)
vmv.v.x (add _t builtins with additional dest argument)
vmv.v.i (add _t builtins with additional dest argument)
vfmv.v.f (add _t builtins with additional dest argument)
vadc.vvm/vadc.vxm/vadc.vim (add _t builtins with additional dest
argument)
vsbc.vvm/vsbc.vxm (add _t builtins with additional dest argument)

Always has tail argument for masked/unmasked intrinsics
Vector Single-Width Integer Multiply-Add Instructions (add _t and _mt
builtins)
Vector Widening Integer Multiply-Add Instructions (add _t and _mt
builtins)
Vector Single-Width Floating-Point Fused Multiply-Add Instructions (add
_t and _mt builtins)
Vector Widening Floating-Point Fused Multiply-Add Instructions (add _t
and _mt builtins)
Vector Reduction Operations (add _t and _mt builtins)
Vector Slideup Instructions (add _t and _mt builtins)
Vector Slidedown Instructions (add _t and _mt builtins)

Discussion: https://github.com/riscv/rvv-intrinsic-doc/pull/101

Differential Revision: https://reviews.llvm.org/D109322

Add the tail policy argument to LLVM IR intrinsics. There are two policies for tail elements. Tail agnostic means users do not care about the values in the tail elements and tail undisturbed means the values in the tail elements need to be kept after the operation. In order to let users control the tail policy, we add an additional argument at the end of the argument list.

For unmasked operations, we have no maskedoff and the tail policy is always tail agnostic. If users want to keep tail elements under unmasked operations, they could use all one mask in the masked operations to do it. So, we only add the additional argument for masked operations for most cases. There are exceptions listed below.

In this patch, we do not handle the following cases to reduce the complexity of the patch. There could be two separate patches for them.

* Use dest argument to control tail policy
vmerge.vvm/vmerge.vxm/vmerge.vim (add _t builtins with additional dest argument)
vfmerge.vfm (add _t builtins with additional dest argument)
vmv.v.v (add _t builtins with additional dest argument)
vmv.v.x (add _t builtins with additional dest argument)
vmv.v.i (add _t builtins with additional dest argument)
vfmv.v.f (add _t builtins with additional dest argument)
vadc.vvm/vadc.vxm/vadc.vim (add _t builtins with additional dest argument)
vsbc.vvm/vsbc.vxm (add _t builtins with additional dest argument)

* Always has tail argument for masked/unmasked intrinsics
Vector Single-Width Integer Multiply-Add Instructions (add _t and _mt builtins)
Vector Widening Integer Multiply-Add Instructions (add _t and _mt builtins)
Vector Single-Width Floating-Point Fused Multiply-Add Instructions (add _t and _mt builtins)
Vector Widening Floating-Point Fused Multiply-Add Instructions (add _t and _mt builtins)
Vector Reduction Operations (add _t and _mt builtins)
Vector Slideup Instructions (add _t and _mt builtins)
Vector Slidedown Instructions (add _t and _mt builtins)

Discussion: https://github.com/riscv/rvv-intrinsic-doc/pull/101

Differential Revision: https://reviews.llvm.org/D105092

Testing on a SLM box suggests these can run on either port, but the throughput is 4cy on either (inc MMX versions). Confirmed with Intel AoM / Agner / InstLatX64.

Record is a SmallVector<uint64_t, 1024> - we really need to avoid passing this by value.

Avoid unnecessary big copies, reported by coverity.

Earlier versions of GDB do not do this automatically.
(from my checks 8.3 does not and 9.2 does)

There are several places in the code that are currently broken as
they assume an Instruction always has a parent Function when
attempting to get the vscale_range attribute. This patch adds checks
that an Instruction has a parent.

I've added a test for a parentless @llvm.vscale intrinsic call here:

  unittests/Analysis/ValueTrackingTest.cpp

Differential Revision: https://reviews.llvm.org/D110158

This change is to keep the help text and command guide of objcopy in
tandem.

- In the help output the options --rename-section and
  --set-section-flags were missing the flag exclude, which is found in
  the command guide.
- In the command guide the alias -G for --keep-global-symbol was
    missing, which is found in the help output.

Differential Revision: https://reviews.llvm.org/D110340

I found this after upgrading from Ubuntu bionic (gdb 8.1.1) to
Focal (gdb 9.2). (where this test fails, but that's for a
different patch)

9.2 allows you to set breakpoint commands from
Python, which was added in 8.3.
(bintutils a913fffbdee21fdd50e8de0596358be425775678
"Allow breakpoint commands to be set from Python")

The reason this test never failed before was because it did so
silently. "source <python file>" doesn't fail even if that script
raises an Exception.

To fix this extend the gdb lit feature to check that:
* gdb exists
* has Python support
* allows you to set breakpoint commands

Reviewed By: #libc, ldionne

Differential Revision: https://reviews.llvm.org/D110334

The type legalizer has by default no method of doing this bitcast other than
storing and reloading the value from stack.

This patch implements a custom lowering of this operation using extractions
of subregs (z13 and earlier using FP128 register pairs), or of vector
elements (with 'vector enhancements 1' using VR128 FP registers).

Review: Ulrich Weigand

Differential Revision: https://reviews.llvm.org/D110346

Instead of relying on many volatile loads/stores in a single function,
rewrite the test to use separate functions as any other test would.

* Do not discard static result type information that cannot be inferred from lower/upper padding.
* Add optional argument to `PadTensorOp::inferResultType` for specifying known result dimensions.

Differential Revision: https://reviews.llvm.org/D110380

When statically linking C++ standard library, we shouldn't add -Bdynamic
after including the library on the link line because that might override
user settings like -static and -static-pie. Rather, we should surround
the library with --push-state/--pop-state to make sure that -Bstatic
only applies to C++ standard library and nothing else. This has been
supported since GNU ld 2.25 (2014) so backwards compatibility should
no longer be a concern.

Differential Revision: https://reviews.llvm.org/D110128