Based on the direction of IR Verifier changes in D146845, this documentation
needs to be updated.

Differential Revision: https://reviews.llvm.org/D148138

Add a `StringRef` conversion function to `ConstString`.

This will make using llvm, and other non-ConstString, APIs more convenient.

For demonstration, this updates Module.cpp.

Differential Revision: https://reviews.llvm.org/D148175

The test CHECKs for `-nan`, but printing the sign is optional.

This change removes the sign because it has no meaning.

If we have a >s b and we know that b >s 0 then we also can infer that a >u b.

Reviewed By: fhahn

Differential Revision: https://reviews.llvm.org/D148321

There's a desire to move away from `undef` in LLVM. Currently we want to
have the `addressspace(3)` variables use `poison` instead.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D147719

We recently reverted a patch that automatically set the rpath on OpenMP
executables. This was used because the `libomptarget.so` library is only
expected to work with the same version of compiler that will be using
it. This patch adds some documentation for how to get similar behaviour
as before using a clang configuration file.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D147943

Andrzej proposed the design and led the work for the Flang Driver
over the past couple of years. He is continuing to function as
the primary reviewer and gate-keeper of the Driver.

Reviewed By: sscalpone

Differential Revision: https://reviews.llvm.org/D147573

Another part of Issue #60007

Powf functions are pushed directly to libm. This is problematic for
situations where libm is not available. This patch will decompose the
powf function into log of exponent multiplied by log of base and raise
it to the exp.

Reviewed By: rsuderman

Differential Revision: https://reviews.llvm.org/D148164

This reverts commit de4c038c.

The change that triggered this fix got reverted.

In some cases, breaking large PHIs can very negatively affect
performance (3x more instructions observed in a particular test case).

This patch adds some basic profitability heuristics to help with some of these issues without affecting the "good" cases.
e.g. avoid breaking PHIs if it causes back-and-forth between vector/scalar form for no good reason.

Fixes SWDEV-392803
Fixes SWDEV-393781
Fixes SWDEV-394228

Reviewed By: arsenm

Differential Revision: https://reviews.llvm.org/D147786

Differential Revision: https://reviews.llvm.org/D148333

I was working on legalising vector binary ops to their VP counterparts when I noticed that `vp_copysign` wasn't marked as a binary op. It looks like this might have been an oversight, so this marks it as one.

Reviewed By: craig.topper

Differential Revision: https://reviews.llvm.org/D148241

The test fails when the two generated files have the same timestamp
(lldb uses second granularity).

This preserves NSW flag for AddRecs multiplied by -1 if we can prove
via constant ranges that the AddRec cannot be signed minimum.

An explanation:
Let M be signed minimum. If AddRec's range contains M, then M * (-1) will
stay M and (M + 1) * (-1) will be signed maximum, so we get a signed overflow.
In all other cases if an AddRec didn't signed overflow,
then AddRec * (-1) wouldn't too.

Differential Revision: https://reviews.llvm.org/D148084

Fix review comments from https://reviews.llvm.org/D146252
Merge `WhileRemoveUnusedArgs` pattern with (unused) `WhileUnusedArg`,
use `getConditionOp`, use `SmallPtrSet` and early check, move tests

Differential Revision: https://reviews.llvm.org/D148256

This commit moves the CFGLoopInfo instantiation into the C++ file to
ensure that it is only compiled once. Instantiating the template
explicitly revealed two missing functions that this commit also adds.

Reviewed By: ftynse

Differential Revision: https://reviews.llvm.org/D148219

Instead of using the validity of a brace's source location as a flag
for list initialization, this now uses a PointerIntPair to model it so
we do not increase the size of the AST node to track this information.
This allows us to retain the valid source location information, which
fixes the coverage assertion.

Fixes https://github.com/llvm/llvm-project/issues/62105
Differential Revision: https://reviews.llvm.org/D148245

This fixes a failing check in tests added by 05a142cc.

We often repeat the vector operand in TESTPS(X,X)/TESTPD(X,X) for anyof comparisons - ensure we still only demand the sign bits when the TESTP is the only user of that operand

Zfa provide fli instruction to load a floating point immediate value,
and some of those are not easy to write and read by human, so
hexadecimal floating-point format should be a good alternative way to write:

Consider this case: 1^-16 = 1.52587890625e-05 (decimal) vs 0x1p-16 (hexadecimal)

The hexadecimal format is easier to write for human.

Fortunately hexadecimal floating-point constants already supported in
C99, so actually we don't need to add any extra code to support that.

This patch added test case for demonstrate we support that and also make
sure this will be supported in future.

I also gonna to talk with ISA folks to adding hexadecimal floating-point
to the ISA spec, so that user will know fli accept value in this format.

Reviewed By: asb

Differential Revision: https://reviews.llvm.org/D148077

This reverts commit 75f1f158.
It caused fail on https://lab.llvm.org/buildbot#builders/37/builds/21461

Adjust lowerDotProduct cost estimate to include the cost benefits of:
 * emitting a wide load
 * emitting a wide multiply.

Reviewed By: thegameg

Differential Revision: https://reviews.llvm.org/D147330

[X86] Add TESTPS/TESTPD test coverage showing failure to simplify demanded sign bit when the operands has multiple uses

These case are excluded in https://reviews.llvm.org/D113049.
Now AIX XCOFF 64 integrated-as support improves a lot and all these
cases pass now, so enable them.

Remove C APIs for interacting with PassRegistry and pass
initialization. These are legacy PM concepts, and are no longer
relevant for the new pass manager.

Calls to these initialization functions can simply be dropped.

Differential Revision: https://reviews.llvm.org/D145043

Currently, FunctionAttrs treats landingpads as non-throwing, and
will infer nounwind for functions with landingpads (assuming they
can't unwind in some other way, e.g. via resum). There are two
problems with this:

* Non-cleanup landingpads with catch/filter clauses do not
  necessarily catch all exceptions. Unless there are catch ptr null
  or filter [0 x ptr] zeroinitializer clauses, we should assume
  that we may unwind past this landingpad. This seems like an
  outright bug.
* Cleanup landingpads are skipped during phase one unwinding, so
  we effectively need to support unwinding past them. Marking these
  nounwind is technically correct, but not compatible with how
  unwinding works in reality.

Fixes https://github.com/llvm/llvm-project/issues/61945.

Differential Revision: https://reviews.llvm.org/D147694

Extra tests for D147330.

After performing signed extension, we update the register in MI. We should also update `incr` register which is tracking the register in `MI`.

Fixes https://github.com/llvm/llvm-project/issues/61882.

Reviewed By: #powerpc, shchenz

Differential Revision: https://reviews.llvm.org/D147594

foldAllocaCmp() needs to fold all comparisons of an alloca at the
same time, to ensure that there is a consistent view of the alloca
address. Currently, it folds "all" comparisons by limiting to the
case where there is only one. This patch switches the algorithm to
instead actually collect and fold all comparisons.

Something we need to be careful about here is that there may be
comparisons where both sides of the icmp are based on the alloca.
Such comparisons are comparing offsets of the alloca, and as such
can be ignored here, but shouldn't be folded to false.

Differential Revision: https://reviews.llvm.org/D144492

Embedded systems that do not use an ELF loader locate the
.ARM.exidx exception table via linker defined __exidx_start and
__exidx_end rather than use the PT_ARM_EXIDX program header. This
means that some linker scripts such as the picolibc C library's
linker script, do not have the .ARM.exidx sections at offset 0 in
the OutputSection. For example:

.except_unordered : {
    . = ALIGN(8);
    PROVIDE(__exidx_start = .);
    *(.ARM.exidx*)
    PROVIDE(__exidx_end = .);
} >flash AT>flash :text

This is within the specification of Arm exception tables, and is
handled correctly by ld.bfd.

This patch has 2 parts. The first updates the writing of the data
of the .ARM.exidx SyntheticSection to account for a non-zero
OutputSection offset. The second part makes the PT_ARM_EXIDX program
header generation a special case so that it covers only the
SyntheticSection and not the parent OutputSection. While not strictly
necessary for programs locating the exception tables via the symbols
it may cause ELF utilities that locate the exception tables via
the PT_ARM_EXIDX program header to fail. This does not seem to be the
case for GNU and LLVM readelf which seems to look for the
SHT_ARM_EXIDX section.

Differential Revision: https://reviews.llvm.org/D148033

Differential Revision: https://reviews.llvm.org/D148121