Avoids unused variable warnings when asserts are disabled.

Following the discussion from https://reviews.llvm.org/D141277 and in
particular Ralf Jung's comment at
https://reviews.llvm.org/D141277#inline-1365148, replacing an unused `atomicrmw
xchg` into an `atomic store` is illegal even for release ordering.

Quoting Connor Horman from the rust lang discussion linked in that comment:
"An acquire operation A only synchronizes-with a release operation R if it
takes its value from R, or any store in the release sequence headed by R, which
is R, followed by the longest continuous sequence of read-modify-write
operations.
A regular store following R in the modification order would break the release
sequence, and if an acquire operation reads that store or something later, then
it loses any synchronization it might have already had."

This fixes https://github.com/llvm/llvm-project/issues/60418

Differential Revision: https://reviews.llvm.org/D142097

Turning idempotent `atomicrmw`s into `load atomic` is perfectly legal
with respect to how the loading happens, but it may not be legal for the
whole program semantic.

Indeed, this optimization removes a store that may have some effects on
the legality of other optimizations.
Essentially, we lose some information and depending on the backend
it may or may not produce incorrect code, so don't do it!

This fixes llvm.org/PR56450.

Differential Revision: https://reviews.llvm.org/D141277

Preserve the alignment of the original atomicrmw, rather than using
the ABI alignment.

The same problem exists for loads, but that code is being removed
in D141277 anyway.

This patch adds the support for `fmax` and `fmin` operations in `atomicrmw`
instruction. For now (at least in this patch), the instruction will be expanded
to CAS loop. There are already a couple of targets supporting the feature. I'll
create another patch(es) to enable them accordingly.

Reviewed By: arsenm

Differential Revision: https://reviews.llvm.org/D127041

Estimated impact on preprocessor output line:
before: 1062981579
after:  1062494547

Discourse thread: https://discourse.llvm.org/t/include-what-you-use-include-cleanup
Differential Revision: https://reviews.llvm.org/D120817

Fixes some clang-tidy llvm-namespace-comment warnings.

For a long time, the InstCombine pass handled target specific
intrinsics. Having target specific code in general passes was noted as
an area for improvement for a long time.

D81728 moves most target specific code out of the InstCombine pass.
Applying the target specific combinations in an extra pass would
probably result in inferior optimizations compared to the current
fixed-point iteration, therefore the InstCombine pass resorts to newly
introduced functions in the TargetTransformInfo when it encounters
unknown intrinsics.
The patch should not have any effect on generated code (under the
assumption that code never uses intrinsics from a foreign target).

This introduces three new functions:
TargetTransformInfo::instCombineIntrinsic
TargetTransformInfo::simplifyDemandedUseBitsIntrinsic
TargetTransformInfo::simplifyDemandedVectorEltsIntrinsic

A few target specific parts are left in the InstCombine folder, where
it makes sense to share code. The largest left-over part in
InstCombineCalls.cpp is the code shared between arm and aarch64.

This allows to move about 3000 lines out from InstCombine to the targets.

Differential Revision: https://reviews.llvm.org/D81728

Use the current getAlign() APIs where it makes sense, and use Align
instead of MaybeAlign when we know the value is non-zero.

This is a followup to D73803, which uses the replaceOperand()
helper in more places.

This should be NFC apart from changes to worklist order.

Differential Revision: https://reviews.llvm.org/D73919