Add a pass to convert jump tables to switches.
The new pass replaces an indirect call with a switch + direct calls if all the functions in the jump table are smaller than the provided threshold.
The pass is currently disabled by default and can be enabled by -enable-jump-table-to-switch.

Test plan: ninja check-all

I was looking through to check whether Nan was being handled correctly,
and couldn't work out why simple cases were behaving differently than
they should. It turns out the initial limit values was backwards for
minloc/maxloc reductions in general. This fixes that, introduced in
#79469.

I don't think the transient alignment needs to be larger than the
ABI alignment.

We apply custom lowering to 64 bit constants where we use the same logic
as in non-global isel: if materializing in registers is too expensive,
we emit a load from constant pool. Later, during instruction selection,
constant pool address is generated using `selectAddr`.

This makes it easier to run the tests in a containerized environment.

Drop the "Always" prefix to remove the self-contradiction.

… inside modules (#81087)"

This reverts commit 0802596d.

See comments in PR #81087 for a test case that shows why I'm reverting.

Make the type match the printf format.

…fatal error report

This makes the behavior of HWASan matching with ASan: always call user
provided callback function for an error report, but only call
`SetAbortMessage()` on Android when `flags()->halt_on_error` is true.

It claimed to dump all sections by default, but this hasn't been true
since 2017: https://reviews.llvm.org/D37717

EXPORTAS is a new name type in import libraries. It's used by default on ARM64EC,
but it's allowed on other platforms as well.

This reverts commit 01440115.

This PR adds a new attribute to represent the CUDA attribute attached to
procedure. This attribute is attached to the func.func operation during
lowering.
Other procedures information such as `launch_bounds` and `cluster_dims`
will be added separately.

The Ampere1B core is enabled with a new scheduling/pipeline model, as it
provides significant updates over the Ampere1 core; it reduces latencies
on many instructions, has some micro-ops reassigned between the XY and X
units, and provides modelling for the instructions added since Ampere1
and Ampere1A.

Both Ampere1 and Ampere1A support fusion of ADRP+ADD.
This adds the missing feature to enable fusion-aware scheduling for this
case.

Ampere1A introduced the Fusion for A+B+1/A-B-1.
However, the Feature flag to enable that fusion-case never was added to
TuneAmpere1A.  This commit corrects that omission.

The Ampere1B is Ampere's third-generation core implementing a
superscalar, out-of-order microarchitecture with nested virtualization,
speculative side-channel mitigation and architectural support for
defense against ROP/JOP style software attacks.

Ampere1B is an ARMv8.7+ implementation, adding support for the FEAT
WFxT, FEAT CSSC, FEAT PAN3 and FEAT AFP extensions. It also includes all
features of the second-generation Ampere1A, such as the Memory Tagging
Extension and SM3/SM4 cryptography instructions.

Since we commit all changes to the release branch CI to main first, we
need someway to test that these changes to main don't break the CI.

nm already prints sizes for data symbols. Do that for function symbols 
too, and update objdump to also print size information.

Implements item 3 from https://github.com/llvm/llvm-project/issues/76107

The newly introduced `CUDAAttribute` is meant for CUDA attributes
associated with variable. In order to not clash with the future
attribute for function/subroutine, rename `CUDAAttribute` to
`CUDADataAttribute`.

Function annotation, as part of llvm.metadata, is for the function
itself and doesn't apply to its corresponding jump table entry, so with
CFI we shouldn't replace function pointer in function annotation with
pointer to its corresponding jump table entry.

The spurious -Wunused-function warning issue for `target_version` #80227
also applied to `__attribute__((target(...)))` based FMV. #81167 removed
warnings for all `target`-based FMV. This patch restores the warnings
for `__attribute__((target("default")))`.

`extraClassDeclaration` of `CallOpInterface` can be inherited by other
`OpInterfaces` into foreign namespaces, thus types must be fully
qualified to prevent compiler errors, for example:

    def MyCaller : OpInterface<"MyCaller", [CallOpInterface]> {
        let cppNamespace = "::MyNamespace";
    }

Summary:
I neglected the fact that `activemask` is a 6.2 or 6.3 feature, so
building this on older machines is incorrect. Bump this up to 6.3 for
now so it works. In the future we will try to get rid of the N
architecture business.

When a backport request is made, the resulting pull request will have a
title like this:

<release branch>: <First line of HEAD commit for the branch>

And a body that says:

Backport <commit0> <commit1> ..

Requested By: <user>

I was looking for the documentation of that attribute, and the best I
could find was a Stackoverflow answer or the commit message that
originally introduced the attribute. I figured I might as well document
what I find to save everyone time in the future.

Summary:
Some recent support made usage of `__nvvm_reflect` more consistent. We
should expose it as a builtin rather than forcing users to externally
define the function.

Summary:
This test requires at least sm_30 to run, but that is still below the
minimum supported version of sm_52 currently. Just set this to sm_60 so
the tests pass in the future.