- Add 'Init Undef Pass', which is target-independent now.

Exit early if MaxVF < MinVF. In that case, the loop body below will
never get entered. Note that this adjusts the condition from MaxVF <=
MinVF. If MaxVF == MinVF, vectorization may still be feasible (and the
loop below gets entered).

PR: https://github.com/llvm/llvm-project/pull/83283

Blank comment line afer \endverbatim malfunctions the doxygen parser.
  Check the "Detailed Description" section on
https://llvm.org/doxygen/classllvm_1_1vfs_1_1RedirectingFileSystem.html


Screenshot is attached below


![image](https://github.com/llvm/llvm-project/assets/22870466/35fc2444-d2f4-4638-990e-241cc14d0986)

Equivalent to "umax(A, B) - umin(A, B)"

This is just an initial correctness implementation, hopefully we can make this optimal in the future.

Shows cases where logical shifts of allsignbits values can be profitably converted to masks

g++ -flto has a diagnostic `-Wodr` about mismatched redeclarations,
which even apply to `enum`.

Fix #83529

Reviewers: thesamesam

Reviewed By: thesamesam

Pull Request: https://github.com/llvm/llvm-project/pull/83604

The help output for `thread backtrace` specifies that you can pass -1 to
`--count` to display all the frames.

```
-c <count> ( --count <count> )
            How many frames to display (-1 for all)
```

However, that doesn't work:

```
(lldb) thread backtrace --count -1
error: invalid integer value for option 'c'
```

The problem is that we store the option value as an unsigned and the
code to parse the string correctly rejects it. There's two ways to fix
this:

1. Make `m_count` a signed value so that it accepts negative values and
appease the parser. The function that prints the frames takes an
unsigned so a negative value will just become a really large positive
value, which is what the current implementation relies on.
2. Keep `m_count` unsigned and instead use 0 the magic value to show all
frames. I don't really see a point in not showing any frames at all,
plus that's already broken (`error: error displaying backtrace for
thread: "0x0001"`).

This patch implements (2) and at the same time improve the error
reporting so that we print the invalid value when we cannot parse it.

rdar://123881767

Summary:
These functions were implemented via builtins that aren't acually
supported. See https://godbolt.org/z/Wq6q6T1za. This caused the build to
crash if they were included. Remove these and replace with correct
implementations.

This commit adds the functionality to broadcast events using the
`Debugger::eBroadcastProgressCategory`
bit (https://github.com/llvm/llvm-project/pull/81169) by keeping track
of these reports with the `ProgressManager`
class (https://github.com/llvm/llvm-project/pull/81319). The new bit is
used in such a way that it will only broadcast the initial and final
progress reports for specific progress categories that are managed by
the progress manager.

This commit also adds a new test to the progress report unit test that
checks that only the initial/final reports are broadcasted when using
the new bit.

Masked gather is very expensive operation and sometimes better to
represent it as a serie of consecutive/strided loads + insertsubvectors
sequences. Patch adds some basic estimation and if loads+insertsubvector
is cheaper, decides to represent it in this way rather than masked
gather.

Reviewers: RKSimon

Reviewed By: RKSimon

Pull Request: https://github.com/llvm/llvm-project/pull/83481

This is in preparation for PR
https://github.com/llvm/llvm-project/pull/82533.

Here we Introduce a test for asan instrumentation where invalid code is
output (see bug https://github.com/llvm/llvm-project/issues/64990)
The `CHECK` lines are generated using `update_test_checks.py` script.
The actual fix PR will udpate this test to highlight the changes in the
generated code.

WaveSizePredicate is copied from pseudo to real

This change adds lowering for HLSL's frac intrinsic to DXIL.

This change should complete #70099

If the graph incudes only single "gather" node with only
extractelements/undefs, which used only in insertelement-based
buildvector sequences, it still might be profitable to vectorize it.
Need to rely on the cost model, not throw this graph away immediately.

Reviewers: RKSimon

Reviewed By: RKSimon

Pull Request: https://github.com/llvm/llvm-project/pull/83581

Similar to PR #83364.

For denormals we have to account for the exponent coming
from the significand.

hlsl_intrinsics.h - add the round  api
DXIL.td add the llvm intrinsic to DXIL lowering mapping 
This change reuses llvm's existing intrinsic
`__builtin_elementwise_round`\ `int_round`
This change implements: #70077

Broke a build bot: https://lab.llvm.org/buildbot/#/builders/124/builds/9846

This reverts commit e7c3cd24.

We can support these by changing the sext promotion to -zext(-C) and
replacing a sgt check with ugt. Reframing the logic in terms of how the
unsigned range are affected. More comments in the patch.

The new cases check isLegalAddImmediate to avoid some
regressions in lit tests.

If they are captured by value.

[Sema] -Wpointer-bool-conversion: suppress lambda function pointer conversion diagnostic during instantiation (#83497)

I have seen two internal pieces of code that uses a template type
parameter to accept any callable type (function pointer, std::function,
closure type, etc). The diagnostic added in #83152 would require
adaptation to the template, which is difficult and also seems
unnecessary. Example:

```cpp
template <typename... Ts>
static bool IsFalse(const Ts&...) { return false; }

template <typename T, typename... Ts,
          typename = typename std::enable_if<std::is_constructible<bool, const T&>::value>::type>
static bool IsFalse(const T& p, const Ts&...) {
  return p ? false : true;
}

template <typename... Args>
void Init(Args&&... args) {
  if (IsFalse(absl::implicit_cast<const typename std::decay<Args>::type&>(
              args)...)) {
    // A callable object convertible to false is either a null pointer or a
    // null functor (e.g., a default-constructed std::func...

Summary:
These currently go to `lib/`, but if the user has `lib/<triple>` they
should go there because it's always searched first.

Emit `__ptrauth`-qualified types as `DIDerivedType` metadata nodes in IR
with tag `DW_TAG_LLVM_ptrauth_type`, baseType referring to the type
which has the qualifier applied, and the following parameters
representing the signing schema:

- `ptrAuthKey` (integer)
- `ptrAuthIsAddressDiscriminated` (boolean)
- `ptrAuthExtraDiscriminator` (integer)
- `ptrAuthIsaPointer` (boolean)
- `ptrAuthAuthenticatesNullValues` (boolean)

Co-authored-by: Ahmed Bougacha <ahmed@bougacha.org>

This patch adds support to sort the symbol table by size. The command
already supports sorting and it already reports sizes. Sorting by size
helps diagnosing size issues.

rdar://123788375

Add an assertion before an access of ValMappings to ensure that it is
within the array bounds.
Silence a static analyzer warning through this.

Summary:
We still use this bitcode library in one case, the NVPTX non-LTO build.
The patch updated the search paths to treat it the same as other
libraries, which unintentionally prioritized system paths over
LIBRARY_PATH which is generally not correct. Also we had a test that
relied on system state so remove that.

Copy constructors can have initialization with side effects, and thus
clang should not emit a warning when -Wunused-variable is used in this
context. Currently however, a warning is emitted.

Now, compilation happens without warnings.

Fixes #79518

This adds the parsing (and unparse) of the compiler drective assume_aligned.

The compiler will issue a warning that the directive is ignored.

The first sentence says:

If a template-parameter of a class template, variable template, or alias
template has a default template-argument, each subsequent
template-parameter shall either have a default template-argument
supplied or be a template parameter pack.

However, we were only testing for "not a function function template",
and referring to an older version of the standard. As far as I can tell,
CWG2032 added the variable-template, and the alias-template pre-dates
the standard on github.

This patch started as a bug fix for #83461 , but ended up fixing a
number of similar cases, so those are validated as well.

getCXXRecordType() on those types does not return the type we need.
Use getPointeeCXXRecordType() instead in those cases.

Like the last few patches, branching in/out of a compute construct is
not valid. This patch implements checking to ensure that a 'case' or
'default' statement cannot jump into a Compute Construct (in the style
of a duff's device!).

Operator calls pass their instance member explicitly, so remove
it from NumParams when calling a variadic function

All three of these need to be non-null.

The warnings or errors are emitted in Sema, but we still need to
do the operation and provide a reasonable result.

This was tripping off STL implementations that check for it (like libc++
with debug checking). The goal of this sort is to cluster operations on
the same values so preserve that property but sort everything else based
on the existing numbering.