On 8/4/20 4:48 PM, Nikos Tsipinakis wrote:
Given the hesitation to fstat in order to remove the O_NONBLOCK flag, are there any objections to applying the patches sent in the original email? i.e. re-call read until the full file has been read.
Simply switching from safe_read to full_read would lose an errno value if there's an I/O error, so that patch alone wouldn't suffice for GNU 'tar'.
