On 09/19/11 10:11, Eric Blake wrote:
> it was only negative fds where the kernel/glibc issue was leaking through

Yes, more specifically, it was only the case where
fd == AT_FDCWD.

Reply via email to