Date: Mon, 10 Mar 1997 13:23:39 +0000 (GMT) From: Stephen Roome <steve@visint.co.uk> To: freebsd-hackers@freebsd.org Subject: Re: Hard Link Count too small! Message-ID: <Pine.BSF.3.91.970310131830.23316B-100000@bagpuss.visint.co.uk> In-Reply-To: <199703080434.PAA09045@godzilla.zeta.org.au>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, 8 Mar 1997, Bruce Evans wrote: > >> I have some POSIX performance tests that do this several times. They > >> take too long. They take much too long if the filesystem is not async > >> mounted. > > > >They must not be very POSIX dependent, or they would fail from the > >"shall mark for update"/"shall update" discrepancies introduced by > >the async mount. An async mounted FS is not POSIX compliant. An > > Wrong. > > Bruce > Asynchronous mounted filesystems might not be POSIX compliant, but synchronous is unbeleivable slow at times. I'm not sure how much it is dependant on the actual disk/driver/buffering, but for me rm -rf seems to be at least an order of magnitude faster when the filesystem is mounted asynchronously. Is there not some way that perhaps mount could be more sensible and perhaps do some of it's I/O async, or is it really that unsafe. I know ppl seem to want POSIX compliancy, but for some things (like this) is it really worth it to be losing so much speed.. Also, exactly how 'dangerous' is async, as far as I know linux is async and I don't hear that many people complaining that their filesystems have all been destroyed. (I'm probably wrong about this, as I don't keep up with linux much anymore) Steve.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.91.970310131830.23316B-100000>