On Thu, 31 May 2018 12:25:49 -0400 Keno Fischer <k...@juliacomputing.com> wrote:
> >> + errno = EINVAL; > >> + return -1; > > > > ... I'm more concerned about this new error path. How can this happen ? > > > > As far as I can tell, the flags come from the client without any > intermediate error > checking. Indeed :-\ > Since the Darwin constants do not match the Linux constants (which > have the same numerical values as the 9p constants), we need to perform this > checking/translation somewhere to ensure correct behavior. > Is there a more appropriate place to put this check? The right thing to do would be to check and translate the flag from P9_DOTL_AT_REMOVEDIR to AT_REMOVEDIR in the core 9p server code.
