This patch reapplies #90500, addressing a bug which caused binary
operators with dependent operands to be incorrectly rebuilt by
`TreeTransform`.

Summary:
A previous patch moved the code here and accidentally overrwrote the
include path that the LSP interface used. This caused incorrect errors
when using clangd with the offload project. This patch removes the
unnecessary header and makes sure we include the correct folder.

The znver3/znver4 scheduler models are outliers, specifying very large
LoopMicroOpBufferSizes at 512, while typical values for other subtargets
are on the order of ~50. Even if this information is
micro-architecturally correct (*), this does not mean that we want to
runtime unroll all loops to a size that completely fills the loop
buffer. Unless this is the single hot loop in the entire application,
the massive code size increase will bust the micro-op and instruction
caches.

Protect against this by clamping to the default PartialThreshold of 150,
which is the same as the default full-unroll threshold and half the
aggressive full-unroll threshold. Allowing more partial unrolling than
full unrolling certainly does not make sense.

(*) I strongly doubt that this is actually correct -- I believe this may
derive from an incorrect reading of Agner Fog's micro-architecture
guide. The number 4096 that was originally used here is the size of the
general micro-op cache, not that of a loop buffer. A separate loop
buffer is not listed for the Zen microarchitecture. Comparing this to
the listing for Skylake, it has a 1536 micro-op buffer, but only a 64
micro-op loopback buffer, with a note that it's rarely fully utilized.
Our scheduling model specifies LoopMicroOpBufferSize of 50 in that case.

Co-authored-by: Ryosuke Niwa <rniwa@apple.com>

Make sure to merge the sourcelocation of the Dominating Instruction that
is hoisted in a basic block in the CSEMIRBuilder in the legalizer pass.

If this is not done, we can have a incorrect line table entry that makes
the instruction pointer jump around.

For example the line table without this patch looks like:

```
Address            Line   Column File   ISA Discriminator OpIndex Flags
------------------ ------ ------ ------ --- ------------- ------- -------------
0x0000000000000000      0      0      1   0             0       0  is_stmt
0x0000000000000010     11     14      1   0             0       0  is_stmt prologue_end
0x0000000000000028     12      1      1   0             0       0  is_stmt
0x000000000000002c     12     15      1   0             0       0
0x000000000000004c     12     13      1   0             0       0
0x000000000000005c     13      1      1   0             0       0  is_stmt
0x0000000000000064     12     13      1   0             0       0  is_stmt
0x000000000000007c     13      7      1   0             0       0  is_stmt
0x00000000000000c8     13      1      1   0             0       0
0x00000000000000e8     13      1      1   0             0       0  epilogue_begin
0x00000000000000f8     13      1      1   0             0       0  end_sequence
```

The line table entry for 0x000000000000005c should be 0

After this patch, the line table looks like:

```
Address            Line   Column File   ISA Discriminator OpIndex Flags
------------------ ------ ------ ------ --- ------------- ------- -------------
0x0000000000000000      0      0      1   0             0       0  is_stmt
0x0000000000000010     11     14      1   0             0       0  is_stmt prologue_end
0x0000000000000028     12      1      1   0             0       0  is_stmt
0x000000000000002c     12     15      1   0             0       0 
0x000000000000004c     12     13      1   0             0       0 
0x000000000000005c      0      0      1   0             0       0 
0x0000000000000064     12     13      1   0             0       0 
0x000000000000007c     13      7      1   0             0       0  is_stmt
0x00000000000000c8     13      1      1   0             0       0 
0x00000000000000e8     13      1      1   0             0       0  epilogue_begin
0x00000000000000f8     13      1      1   0             0       0  end_sequence
```

This reverts commit 03c53c69.

This causes very large compile-time regressions in some cases,
e.g. sqlite3 at O0 regresses by 5%.

CallStackIdConverter sets LastUnmappedId when a mapping failure
occurs.  Now, since toMemProfRecord takes an instance of
CallStackIdConverter by value, namely std::function, the caller of
toMemProfRecord never receives the mapping failure that occurs inside
toMemProfRecord.  The same problem applies to FrameIdConverter.

The patch fixes the problem by passing FrameIdConverter and
CallStackIdConverter by reference, namely llvm::function_ref.

While I am it, this patch deletes the copy constructor and copy
assignment operator to avoid accidental copies.

```
.irp foo,1
nop
.endr
nop
```

expands to an excess EOL between two nop lines. Remove the excess EOL.

The code paths for mte enabled and disabled were interleaving and which
increases the difficulty of reading each path in both source level and
assembly level. In this change, we move the parts that they have
different logic into functions and minor refactors on the code
structure.

This was an oversight in #91859. Using the subblock ID mechanism
other places that use the bitstream APIs (e.g. `BitstreamRemarkSerializer`) use.

The `--temporal-profile-max-trace-length=0` flag in the `llvm-profdata
merge` command is used to remove traces from a profile. There was a bug
where traces would not be cleared if the profile was already sampled.
This patch fixes that.

When the MOLD= argument's type is polymorphic, the type of the result
cannot be known at compilation time, so the call cannot be folded even
when the SOURCE= is constant.

Fixes https://github.com/llvm/llvm-project/issues/92264.

A !$CUF KERNEL DO directive is allowed to have advisory REDUCE clauses
similar to those in OpenACC and DO CONCURRENT. Parse and represent them.
Semantic validation will follow.

When a BIND(C) interface or subprogram has a dummy argument whose
derived type is not BIND(C) but meets the constraints and requirements
of a BIND(C) type, accept it with a warning.

When a procedure is defined with a subprogram but never referenced in a
compilation unit, it may not be characterized until lowering, and any
errors in characterization then may crash the compiler. So always ensure
that procedure definitions are characterizable in declaration checking.

Fixes https://github.com/llvm/llvm-project/issues/91845.

This allows the warning to be disabled in isolation, as it helps when
treating them as errors.

ELEMENTAL internal subprograms are pure unless explicitly IMPURE.

[Clang][Sema] Do not mark template parameters in the exception specification as used during partial ordering (#91534)

We do not deduce template arguments from the exception specification
when determining the primary template of a function template
specialization or when taking the address of a function template.
Therefore, this patch changes `isAtLeastAsSpecializedAs` such that we do
not mark template parameters in the exception specification as 'used'
during partial ordering (per [temp.deduct.partial]
p12) to prevent the following from being ambiguous:

```
template<typename T, typename U>
void f(U) noexcept(noexcept(T())); // #1

template<typename T>
void f(T*) noexcept; // #2

template<>
void f<int>(int*) noexcept; // currently ambiguous, selects #2 with this patch applied 
```

Although there is no corresponding wording in the standard (see core issue filed here
https://github.com/cplusplus/CWG/issues/537), this seems
to be the intended behavior given the definition of _deduction
substitution loci_ in [temp.deduct.general] p7 (and EDG does the same thing).

This option is a compilation action that parses a source file and
performs semantic analysis on it, like the existing -fdebug-unparse
option does. Its output, however, is preceded by the effective contents
of all of the non-intrinsic modules on which it depends but does not
define, transitively preceded by the closure of all of those modules'
dependencies.

The output from this option is therefore the analyzed parse tree for a
source file encapsulated with all of its non-intrinsic module
dependencies. This output may be useful for extracting code from large
applications for use as an attachment to a bug report, or as input to a
test case reduction tool for problem isolation.

Update the publisher and add a publish script that we can use from
Github actions.

Most violations are stale and should be removed while a few can be
adjusted.

Reported at #92238

`ArrayInitLoopExpr` AST node has two occurences of its as-written
initializing expression in its subexpressions through a non-unique
`OpaqueValueExpr`. It causes double-visiting of the initializing
expression if not handled explicitly, as discussed in #85837.

This avoids visiting `co_await` or `co_yield` operand 5 times (it is
repeated under transformed awaiter subexpression, and under
`await_ready`, `await_suspend`, and `await_resume` generated call
subexpressions).

This also fixes building ... on Linux. Seems like target_compatible_with
isn't enough but you also need a manual tag.

Reverts llvm/llvm-project#91334

This broke the gcc7 build.
I suspect the issue is a mismatch on user-defined move constructor on
the return: `return PreservedGVs;` does not match the return type of the
function.

These are allowed for Darwin and use the same ABI.

Re-apply https://github.com/llvm/llvm-project/pull/81196, with a fix that handles the 
absence of llvm formatting: 
https://github.com/llvm/llvm-project/pull/91868/commits/3ba650e91eded3543764f37921dcce3b
b47d425f

Just as we do for the arrays of extension names.

Summary:
These are lots of random warnings due to inconsistent initialization or
signedness.

They *are* still accepted by the HW but have a conservative effect.

Leave them untouched since handling them would complicate the logic a
bit, and developers who code to such a low level really need to revisit
what they're doing anyway.

This PR adds a string interface to `InternalizePass`' `MustPreserveGV`
option, which is a callback function to indicate if a GV is not to be
internalized. This is for use in LLVM.jl, the Julia wrapper for LLVM,
which uses the C API and is thus required to use the PassBuilder string
API for building NewPM pipelines.

Assume in fewer places that the analysis is of a `FunctionDecl`, and
initialize the `Environment` properly for `Stmt`s.

Moves constructors for `Environment` to header to make it more obvious
that there are only minor differences between them and very little
initialization in the constructors.

Tested with check-clang-tooling.

As discussed in the last sync-up call, because these profiles are not
yet finalised they shouldn't be exposed to users unless they opt-in to
them (much like experimental extensions). We may later want to add a
more specific flag, but reusing `-menable-experimental-extensions`
solves the immediate problem.

This is implemented using the new support for marking profiles s
experimental added in #91993 to move the unratified profiles to
RISCVExperimentalProfile and making the necessary changes to logic in
RISCVISAInfo to handle this.

This reverts commit 2c54bf49.

Fixed gcc-7 issue.

Fixes #92193.

Summary:
If the user does not have the selected backend enabled, we should still
be able to build the LLVM-IR an ddistribute it. This patch makes logic
to suppress tests if the backend can't build it, as well as removing a
flag for the building that's only present int he NVPTX backend.