Bruno Haible <[email protected]> writes:

> I agree with Pádraig, and I disagree with Collin. Even if we can't fix the
> problem once and for all, for all possible new errno values that will be
> added the future, it's worth the effort to handle this one in Gnulib.
> It reduces the time window in which users may see programs crash, even
> if it can't eliminate them.

I think the error only happens with openat2 with OPENAT2_REGULAR, which
we don't use (at least in coreutils, though, I'd be surprised if any
Gnulib using projects have either). So I don't think it can happen. But,
I could always be wrong. So thanks for making that change.

Collin



Reply via email to