[libcxx-commits] [libcxxabi] [libc++abi] Handle null pointer-to-object: Issue #64593 (PR #68076)

[Iain Sandoe via libcxx-commits]

iains wrote:

> > I suppose libc++abi moves slowly, not sure why it does not bite more with libc++ tho - if the assumption of the lit config is that it's the installed libs being tested, but actually we're testing the new ones.
> 
> We're compiling and linking against the new library, but we are running against the installed system one.

For the `apple-system-backdeployment*` that appears to be the case.

but for the `apple-system` case we have:

```
$ ls .../bot-test/arm64/lib
libc++.1.dylib		libc++.999.99.dylib	libc++.a		libc++.dylib		libc++abi.1.dylib	libc++abi.999.99.dylib	libc++abi.a		libc++abi.dylib		libc++experimental.a
```
and the test execution line is (long paths trimmed):
```
.../libcxx/utils/run.py --execdir .../bot-test/arm64/test/Output/catch_null_pointer_to_object_pr64953.pass.cpp.dir --env DYLD_LIBRARY_PATH=.../bot-test/arm64/lib --  .../bot-test/arm64/test/Output/catch_null_pointer_to_object_pr64953.pass.cpp.dir/t.tmp.exe
```

So, in this case, the test is executed with new libraries.

I'll do the `UNSUPPORTED` change later.


https://github.com/llvm/llvm-project/pull/68076