On Sat, Feb 7, 2026 at 3:34 PM Mark Millard <[email protected]> wrote: > > On 2/7/26 14:28, Rick Macklem wrote: > > On Fri, Feb 6, 2026 at 7:57 PM Mark Millard <[email protected]> wrote: > >> > >> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123366#c27 > >> > >> reports that for FreeBSD 15: > >> > >> ) posix_fallocate can and does return EOPNOTSUPP, at least with zfs > >> involved > > ZFS will never know how to do this, as I understand it. (Some hint it > > may be possible > > someday, if the reservation system in ZFS is improved.) > > > >> > >> ) the posix_fallocate man page does not document EOPNOTSUPP (not MFC'd? > >> not in a stable/15 that I checked as well) > > Yea, I suppose this should be in the man page. > > Any volunteers for a man page patch? > > main's posix_fallocate man page has EOPNOTSUPP documented. It may be > that the man page update just was not MFC'd to stable/15 or > releng/15.0 . Yea. Here's the commit message... ZFS has since been changed to use EOPNOTSUPP instead of EINVAL, and fusefs/nfs are following suit. POSIX Issue 8 had also made this move, so it makes sense for us to standardize on EOPNOTSUPP. Note in the HISTORY section where we're diverging from our previous versions to align with the new standard.
So, I'd guess it wasn't MFC'd, since it was waiting until all file systems (ZFS, fuse, etc) have converted to EOPNOTSUPP. I'll add kevans in case he is keeping track of this. rick > > > > > rick > > > >> > >> ) there is at least 1 gcc* for which its cc1plus can apparently get > >> SIGSEGV as a consequence > >> > >> -- > >> === > >> Mark Millard > >> marklmi at yahoo.com > >> > >> > > > > > > > -- > === > Mark Millard > marklmi at yahoo.com
