Re: [PATCH] ubsan: disable UBSAN_DIV_ZERO for clang

[Nick Desaulniers]

On Sat, Jul 16, 2022 at 10:34 AM Linus Torvalds
<torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
>
> I decided to make a non-kernel test-case so that clang people can look
> at it without having to worry about any kernel code issues or the
> details of the SMAP rules, and it really shows how clang generates
> horribly pointless and wrong code.

Perfect, thank you.  I was having a hard time reducing this one.

>
> I'm not sure what the right thing to do is to get this sorted out, but
> I created an "issue" on github for this:
>
>     https://github.com/llvm/llvm-project/issues/56568
>
> in the hope that this can get fixed. Because it's very clearly a clang
> misfeature, where clang basically generates insane code that violates
> the rules we try to enforce with objdump.
>
> Nick / Nathan / clang-built-linux people - if there are other better
> ways than that github issue thing to make people aware of this, that
> would be lovely, and please forward that issue to the right people.

Perfect, thank you.

LLVM moved from bugzilla to github for issue tracking last year.  It's
best to file issues there, or in our issue tracker (also github) which
is more specific to issues pertaining to the kernel:
https://github.com/ClangBuiltLinux/linux/issues

You can cc Nathan or myself in bugreports via mentioning @nathanchance
and @nickdesaulniers explicitly, or just forward us the link and we
can subscribe.

Putting out some other fires/regressions at the moment; it looks like
Yuanfang has already started chewing on your bug report.  If they get
stuck, I'll dive in once these regressions are sorted.  Should be an
optimization opportunity regardless of UBSAN when we know only certain
bits are set.
-- 
Thanks,
~Nick Desaulniers