drodriguez wrote:

> I need to take a closer look, since at first reading I can't evaluate the 
> consequences, especially if this does change the behaviour when -isysroot is 
> **not** provided.

Small detail that I think you are aware: the `-isysroot` can be provided, but 
it needs to have the `include/c++/v1` directory to be considered.

> And I do not know exactly the use case you are considering. My use case was 
> relatively straightforward, multiple versions of the toolchain are installed 
> in versioned custom folders in user home, and different projects requiring 
> different toolchain versions have symbolic links from the project folder to 
> one of the clang executable.
> 
> If your change does not affect the above use case and also adds more 
> consistency with Apple clang, it should be fine.

I hope my modifications have not changed the behaviour of those versioned 
custom folders and symlinked `clang`. The tests were modified in a couple of 
places to remove a sysroot containing C++ headers in those checks. Before it 
did not matter which sysroot was being used, since it was the last to be 
chosen, but it was now interfering with what the test seemed to try to test.

https://github.com/llvm/llvm-project/pull/80524
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to