When we fold `(shift (shift C0, x), C1)` we can propagate flags that
are common to both shifts.

Proofs: https://alive2.llvm.org/ce/z/LkEzXD

Closes #94872

readNext has two variants:

- readNext<uint64_t, endian>(ptr)
- readNext<uint64_t>(ptr, endian)

This patch uses the latter to simplify readBinaryIdsInternal.  Both
forms default to unaligned.

Now that FOR exit and resume value creation is explicitly modeled in
VPlan (05e1b534, 07b33013) it doesn't depend on the first
order recurrence splice being preserved and it can now be marked as not
having side-effects. This allows removal of first-order-recurrence-splce
if the FOR is only used in the exit or as scalar ph resume value.

This patch adds initial support to the half type on RISC-V.

Apply the onlyFirstPartUsed logic generally to all per-part
VPInstructions. Note that the test changes remove the second part
of an unsued first-order recurrence splice.

Added remquof128 function. Closes #94312

This is an implementation of floating point multiplication:

It will consist of 
   - `double x double -> float`

Cppcheck recommends using a const reference for range variables in a
for-each loop.
This avoids unnecessary copying of elements, improving performance.

Caught by cppcheck -
lldb/source/API/SBBreakpoint.cpp:717:22: performance: Range variable
'name' should be declared as const reference. [iterateByValue]
lldb/source/API/SBTarget.cpp:1150:15: performance: Range variable 'name'
should be declared as const reference. [iterateByValue]
lldb/source/Breakpoint/Breakpoint.cpp:888:26: performance: Range
variable 'name' should be declared as const reference. [iterateByValue]
lldb/source/Breakpoint/BreakpointIDList.cpp:262:26: performance: Range
variable 'name' should be declared as const reference. [iterateByValue]

Fix #91213
Fix #91217
Fix #91219
Fix #91220

Identified with modernize-use-default-member-init.

This broadly follows how in almost all places, we accept `(<reg>)` to
mean `0(<reg>)`, but I think these are the first like this for Jumps
rather than Loads/Stores. These are accepted by binutils but not by
LLVM: https://godbolt.org/z/GK7MGE7q7

This depends on enabling the use of extensions.

'override' makes 'virtual' redundant.

Identified with modernize-use-override.

While I am at it, this patch adds const to a couple of places.

Since we can bitcast and then do the same thing sub does in the table
section above, I figured it was trivial to add fsub, fmul, and fdiv.

This patch relands #91469 and uses `uint64_t` for repeat count to avoid
a miscompilation caused by overflow
https://github.com/llvm/llvm-project/pull/91469#discussion_r1623925158.

The pr description in #94008 mismatches with the code.
> + When VT is smaller than ShiftVT, it is safe to use trunc.
> + When VT is larger than ShiftVT, it is safe to use zext iff
`is_zero_poison` is true (i.e., `opcode == ISD::CTTZ_ZERO_UNDEF`). See
also the counterexample `src_shl_cttz2 -> tgt_shl_cttz2` in the alive2
    proofs.

Closes #94824.

[SimplifyCFG] Don't use a mask for lookup tables generated from switches with an unreachable default case (#94468)

When transforming a switch with holes into a lookup table, we currently
use a mask to check if the current index is handled by the switch or if
it is a hole. If it is a hole, we skip loading from the lookup table.
Normally, if the switch's default case is unreachable this has no
impact, as the mask test gets optimized away by subsequent passes.
However, if the switch is large enough that the number of lookup table
entries exceeds the target's register width, we won't be able to fit all
the cases into a mask and the switch won't get transformed into a lookup
table. If we know that the switch's default case is unreachable, we know
that the mask is unnecessary and can skip constructing it entirely,
which allows us to transform the switch into a lookup table.

[Example](https://godbolt.org/z/7x7qfx8M1)

In the future, it might be interesting to consider allowing lookup table
masks to be more than one register large (e.g. using a constant array of
bit flags, similar to `std::bitset`).

This patch removes LLDB from a list of projects that are excluded from
building and testing on pre-merge CI on Linux.

Windows environment needs to be prepared in order to test LLDB
(https://github.com/llvm/llvm-project/pull/94208#issuecomment-2146256857),
but we don't have enough maintenance resources to do that at the moment.

Because LLDB has been in the list of projects that need to be tested on
Clang changes, this PR make this happen on Linux. This seems to be the
consensus in the discussion of this PR.

2e1788f8 reverted #94843. It was creating `%t` as a directory and
causes an error in incremental builds.

`std::aligned_storage` is deprecated with C++23, see
[here](https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2021/p1413r3.pdf).

This replaces the usages of `std::aligned_storage` within compiler-rt
with an aligned `std::byte` array.
I will provide patches for other subcomponents as well.

In some case, constant can survive early constant folding optimization
because they are hidden behind several layers of type changes.

E.g., consider the following sequence (extracted from the arm test that
this commit changes):
```
    t2: v1f16 = BUILD_VECTOR ConstantFP:f16<APFloat(0)>
    t4: v1f16 = insert_vector_elt t2, ConstantFP:f16<APFloat(0)>, Constant:i32<0>
  t5: f16 = bitcast t4
t6: f32 = fp_extend t5
```

Because the constant (APFloat(0)) is hidden behind a <1 x ty> type, all
the constant folding that normally happen for scalar nodes when using
`SelectionDAG::getNode` are blocked.

As a result the constant manages to survive as an actual conversion
instruction down to the select phase:
```
t11: f32 = fp16_to_fp Constant:i32<0>
```

With the change in this patch, we try to do constant folding one more
time during dag combine, which in the motivating example result in the
much better sequence:
```
t7: ch = CopyToReg t0, Register:f32 %0, ConstantFP:f32<0.000000e+00>
```

Note: I'm sure we have this problem in a lot of other places. Generally
speaking I believe SDISel is not that good with <1 x ty> compared to
pure scalar. However, I only changed what I could easily test.

Use handleFloatFloatBinOp to properly diagnose NaN results and divisions
by zero.

Fixes #84871

[mlir][Transforms][NFC] `GreedyPatternRewriteDriver`: Use composition instead of inheritance (#92785)

This commit simplifies the design of the `GreedyPatternRewriterDriver`
class. This class used to inherit from both `PatternRewriter` and
`RewriterBase::Listener` and then attached itself as a listener.

In the new design, the class has a `PatternRewriter` field instead of
inheriting from `PatternRewriter`, which is generally perferred in
object-oriented programming.

---------

Co-authored-by: Markus Böck <markus.boeck02@gmail.com>

Some Ubuntu builds were broken after 20d497c2 "[Driver] Remove unneeded
*-linux-gnu after D158183".

This patch by Fangrui Song fixes this with a handling in config.guess.

`BlockT *LoopBase<BlockT, LoopT>::getLoopPreheader()` was changed in
72436078 to use `llvm::size` rather than
the checking that `child_begin() + 1 == child_end()`. `llvm::size`
requires that `std::distance` be O(1) and hence that clients support
random access. Use `llvm::hasSingleElement` instead.

Reverts llvm/llvm-project#87635

On some corner cases, lld generated an object file with an empty REL
section with `sh_info` set to 0. This file triggers an lld error when
used as its input. See
https://github.com/llvm/llvm-project/pull/87635#issuecomment-2155318065
for details.

Passing the result of c_str() to a stream is slow and redundant. This
change removes unnecessary c_str() calls and uses the string object
directly.

Caught by cppcheck -
lldb/tools/debugserver/source/JSON.cpp:398:19: performance: Passing the
result of c_str() to a stream is slow and redundant. [stlcstrStream]
lldb/tools/debugserver/source/JSON.cpp:408:64: performance: Passing the
result of c_str() to a stream is slow and redundant. [stlcstrStream]
lldb/tools/debugserver/source/JSON.cpp:420:54: performance: Passing the
result of c_str() to a stream is slow and redundant. [stlcstrStream]
lldb/tools/debugserver/source/JSON.cpp:46:13: performance: Passing the
result of c_str() to a stream is slow and redundant. [stlcstrStream]

Fix #91212

Also remove some unused Constraint paramters that appeared before
the sew parameter.

These usually have a single value that is always used. We can just
hardcode into the class body.

* generate Clang configuration file with provided target sysroot
(TOOLCHAIN_TARGET_SYSROOTFS)
* explicitly pass provided target sysroot into the compiler-rt tests
configuration.
* added ability to configure a type of the build libraries -- shared or
static (TOOLCHAIN_SHARED_LIBS, default OFF)

In behalf of: #94284

`HLSLShaderAttr::ShaderType` enum is a subset of
`llvm::Triple::EnvironmentType`. We can use
`llvm::Triple::EnvironmentType` directly and avoid converting one enum
to another.

DFSan's sscanf is incorrect
(https://github.com/llvm/llvm-project/issues/94769), which results in
erroneous matches when scraping RSS from /proc/maps. This patch works
around the issue by using strstr as a secondary check.

It also adds a loose validity check for the initial RSS measurement, to
guard against regressions in get_rss_kb().

Fixes https://github.com/llvm/llvm-project/issues/91287

[mlir] Handle the newly-added "Reserved" FramePointerKind for 1a523925 (NFC)

/llvm-project/mlir/lib/Target/LLVMIR/ModuleImport.cpp:48:
tools/mlir/include/mlir/Dialect/LLVMIR/LLVMConversionEnumsFromLLVM.inc:158:11:
error: enumeration value 'Reserved' not handled in switch [-Werror,-Wswitch]
  switch (value) {
          ^~~~~
1 error generated.