This reverts commit 3e12b2e2.

This reverts commit fb13d027.

At the moment, only the trailing dimensions in the vector type can be
scalable, i.e. this is supported:

    vector<2x[4]xf32>

and this is not allowed:

    vector<[2]x4xf32>

This patch extends the vector type so that arbitrary dimensions can be
scalable. To this end, an array of bool values is added to every vector
type to denote whether the corresponding dimensions are scalable or not.
For example, for this vector:

  vector<[2]x[3]x4xf32>

the following array would be created:

  {true, true, false}.

Additionally, the current syntax:

  vector<[2x3]x4xf32>

is replaced with:

  vector<[2]x[3]x4xf32>

This is primarily to simplify parsing (this way, the parser can easily
process one dimension at a time rather than e.g. tracking whether
"scalable block" has been entered/left).

NOTE: The `isScalableDim` parameter of `VectorType` (introduced in this
patch) makes `numScalableDims` redundant. For the time being,
`numScalableDims` is preserved to facilitate the transition between the
two parameters. `numScalableDims` will be removed in one of the
subsequent patches.

This change is a part of a larger effort to enable scalable
vectorisation in Linalg. See this RFC for more context:
  * https://discourse.llvm.org/t/rfc-scalable-vectorisation-in-linalg/

Differential Revision: https://reviews.llvm.org/D153372

Update num_gangs definition in ACC.td and its lowering
to support up to 3 arguments.

Depends on D153796
Depends on D153801

Reviewed By: razvanlupusoru

Differential Revision: https://reviews.llvm.org/D153806

The clause parser generation was not taking into account the
`isValueList` flag. This patch updates the emitter to generate
the correct code.

Reviewed By: razvanlupusoru

Differential Revision: https://reviews.llvm.org/D153801

In the latest spec, the `num_gangs` clause accepts up to three
arguments. Update the dialect to swicth `numGangs` operands from
optional single operand to a variadic operand. The verifier limits
the number of operands to three as specified in the spec.

Reviewed By: razvanlupusoru

Differential Revision: https://reviews.llvm.org/D153796

`__iterator/readable_traits.h` can't be used by itself, intantiating `iter_value_t` requires `__iterator/iterator_traits.h`. `readable_traits.h` can't include `iterator_traits.h` though because `iterator_traits.h` requires `readable_traits.h`.

Move `iter_value_t` to `__iterator/iterator_traits.h` so that both headers can work standalone.

Reviewed By: Mordante, #libc

Differential Revision: https://reviews.llvm.org/D153828

Now that we have a proper way to deprecate things in the SB API, we
should apply it where we have manually done so with comments.

Differential Revision: https://reviews.llvm.org/D153824

The underlying structures no longer use ConstString so we can remove it
wholesale from PluginManager now.

Differential Revision: https://reviews.llvm.org/D153818

The generic expansion still causes a problem for SSE targets without BLENDV/select node, but we can create a custom lowering until that can be addressed.

The test causes unexpectedly high RSS usage there.

This gives us more meaningful information when
`getAbbreviationDeclarationSet` fails. Right now only
`verifyAbbrevSection` actually uses the error that it returns, but the
other call sites could be rewritten to take advantage of the returned error.

Differential Revision: https://reviews.llvm.org/D153459

The initial implementation of the modernize-use-std-print check was
capable of converting calls to printf (etc.) which used the return value
to calls to std::print which has no return value, thus breaking the
code.

Use code inspired by the implementation of bugprone-unused-return-value
check to ignore cases where the return value is used. Add appropriate
lit test cases and documentation.

Reviewed By: PiotrZSL

Differential Revision: https://reviews.llvm.org/D153860

[X86] Add getBitSelect helper function to create OR(AND(LHS,MASK),AND(RHS,~MASK)) bit select patterns. NFC.

Prefer XOR to ANDNP as its commutative

Adds support for the classic dataflow algorithm that partitions a flow graph
into distinct intervals. C.f. Dragon book, pp. 664-666.

A version of this algorithm exists in LLVM (see llvm/Analysis/Interval.h and
related files), but it is specific to LLVM, is a recursive (vs iterative)
algorithm, and uses many layers of abstraction that seem unnecessary for CFG
purposes.

This patch is part 1 of 2. The next patch will generalize the code to work on
intervals, to support computation of the limit flow graph.

Differential Revision: https://reviews.llvm.org/D152263

/scratch/powerllvm/powerllvm_env/aix-ppc64/clang-ppc64-aix/llvm-project/llvm/unittests/tools/llvm-exegesis/X86/TargetTest.cpp:88:17: error: unused function 'IsMovRegToReg' [-Werror,-Wunused-function]
Matcher<MCInst> IsMovRegToReg(unsigned Opcode, int64_t Reg1, int64_t Reg2) {
1 error generated.

Make Zcf/Zcd depend on Zca.

Reviewed By: asb

Differential Revision: https://reviews.llvm.org/D153826

Bit 6 should be 0.

Reviewed By: asb

Differential Revision: https://reviews.llvm.org/D153793

For more details about the nomination, please see
https://discourse.llvm.org/t/rfc-proposing-changes-to-the-code-owner-for-the-llvm-vectorizers/70267/9

This was broken in D145628; it caused the new header not to be displayed
as a header at all and show part of the markup as text.

This patch is a follow up to D43315, and adds the following new load/store
TLS specific instructions for integer and floating point scalar types:
```
LHAXTLS
LWAXTLS
LHAXTLS_32
LWAXTLS_32
LFSXTLS
LFDXTLS
STFSXTLS
STFDXTLS
```
These instructions can be used to optimized TLS sequences when D-Form
loads/stores follow an ADD_TLS instruction.

Duplicate versions of these instructions are also added within an isAsmParserOnly=1
block (similar to D47382) to allow llvm-mc to assemble these instructions.

Differential Revision: https://reviews.llvm.org/D153645

Use Value::getPointerDereferenceableBytes() instead of hardcoding dereferenceable only for allocas. Allows us to infer inbounds GEPs for other Values like CallInsts and Arguments.

Fixed clang test broken in initial land.

Reviewed By: nikic

Differential Revision: https://reviews.llvm.org/D153815

This patch updates ConstraintElimination to track uses of conditions in
the worklist. This allows simplifying conditions using the context that
holds directly at the condition, instead of where the condition is
defined.

This allows us to catch more cases in practice: there are multiple
code-size changes for CTMark while compile-time remains unchanged:
https://llvm-compile-time-tracker.com/compare.php?from=4b020cca9363bebab4643f89cfa92ab2fcc7976c&to=7a6e84b8f029713c137814cd46eec775d24a54a1&stat=instructions:u

This should help to simplify D151799.

Reviewed By: nikic

Differential Revision: https://reviews.llvm.org/D153660

This reverts commit cd43b19c.

Breaks clang/test/CodeGenOpenCL/builtins-amdgcn.cl.

These conversion functions were using LDBL_MANT_DIG (which is the 80-bit
extended float on x86) instead of the appropriate macro for the 128-bit
floating point type expected by the *tf* softfloat library calls.
This was discovered while testing D98261 (which allows building the *tf*
functions on x86).

This also changes the constants used in the two tests to use 128-bit
floating-point literals instead of long double ones to ensure that the
comparison succeeds on platforms with smaller long double (e.g. x86_64)

Reviewed By: scanon

Differential Revision: https://reviews.llvm.org/D131787

This reverts commit 2325e011.

Reverted due to buildbot failure: llvm-clang-x86_64-expensive-checks-debian

Use Value::getPointerDereferenceableBytes() instead of hardcoding dereferenceable only for allocas. Allows us to infer inbounds GEPs for other Values like CallInsts and Arguments.

Reviewed By: nikic

Differential Revision: https://reviews.llvm.org/D153815

This is a follow on to D152740. The focus of this patch is on actually removing the old TA (unsuffixed) version. I realized we already had plumbing for combined TA/TU pseudos - used by some of the ternary instructions. As such, we can go ahead and fully remove the old TA, and rename the _TU variant to be unsuffixed. (The rename must happen in this patch for the table structure to work out as expected.)

The scheduling difference comes from an omission in D152740. If we selected a _MASK variant - either from manual ISEL or instrincs - we were going through doPeepholeMaskedRVV and still getting the TA variant. The use of the IsCombined flag in the MaskedPseudo table causes us to use the TU (now unsuffixed) variant instead.

Differential Revision: https://reviews.llvm.org/D153155

An .ARM.attributes section is divided into subsections, each labelled
with a vendor name. There is one standardised vendor name, which must
be used for all attributes that affect compatibility. Subsections
labelled with other vendor names can be used for optimisation
purposes, but it has to be safe for an object file consumer to ignore
them if it doesn't recognise the vendor name.

LLD currently terminates parsing of the whole attributes section as
soon as it encounters a subsection with a vendor name it doesn't
recognise (which is anything other than the standard one). This can
prevent it from detecting compatibility issues, if a standard
subsection followed the vendor-specific one.

This patch modifies the attribute parser so that unrecognised vendor
subsections are silently skipped, and the subsections beyond them are
still processed.

(Relanded with no change from the original commit 8f208edd. I
reverted it in 949bb7e4 due to widespread buildbot breakage,
failing to notice that 975f71fa had already fixed the failing
unit test. Also, the *revert* caused at least one buildbot to fail,
because I switched the affected lld test to making %t a directory, and
then the reverted version tried to treat it as a file without cleaning
the output directory first.)

Differential Revision: https://reviews.llvm.org/D153335

This patch makes the display of member function calls more true to the user-written code by making use of the syntactical structure of the function calls.
This patch also changes the display of conventional value-based printing from arrow operator to dot operator.
This avoids the syntactical invalidness in notes previously caused by the display of & operator
(lack of parentheses and reference of rvalue)

Fixes https://github.com/llvm/llvm-project/issues/57081

Reviewed By: cjdb
Differential Revision: https://reviews.llvm.org/D151720

Add a missing check that ensures that ComplexDeinterleaving for reduction
is only analyzed for Real and Imaginary Instructions of the same type.

Differential Revision: https://reviews.llvm.org/D153862

Partial progress towards replacing `CreateElementBitCast`, as it no
longer does what its name suggests. Either replace its uses with
`Address::withElementType()`, or remove them if no longer needed.

Reviewed By: barannikov88, nikic

Differential Revision: https://reviews.llvm.org/D153314

X86's CMOV conversion transforms CMOV instructions into control flow between
blocks, meaning the value is computed by a PHI rather than a "real" machine
instruction. In instruction-referencing mode, we need to transfer the
instruction label between the old CMOV and the new PHI instruction to mark
where the variable value is computed.

There's an extra complication in that memory operands can be unfolded from the
CMOV and sunk into the new blocks -- the test checks both scenarios where the
instruction number has to hop between instructions.

This omission exposed by Dexter testing.

Reviewed By: Orlando

Differential Revision: https://reviews.llvm.org/D145565

Move `AttributeMask` out of `llvm/IR/Attributes.h` to a new file
`llvm/IR/AttributeMask.h`.  After doing this we can remove the
`#include <bitset>` and `#include <set>` directives from `Attributes.h`.
Since there are many headers including `Attributes.h`, but not needing
the definition of `AttributeMask`, this causes unnecessary bloating of
the translation units and slows down compilation.

This commit adds in the include directive for `llvm/IR/AttributeMask.h`
to the handful of source files that need to see the definition.

This reduces the total number of preprocessing tokens across the LLVM
source files in lib from (roughly) 1,917,509,187 to 1,902,982,273 - a
reduction of ~0.76%. This should result in a small improvement in
compilation time.

Differential Revision: https://reviews.llvm.org/D153728