Date: Mon, 15 Oct 2001 19:30:17 -0700 (PDT) From: Matthew Dillon <dillon@apollo.backplane.com> To: Warner Losh <imp@harmony.village.org> Cc: mi@aldan.algebra.com, DougB@yahoo-inc.com, mjacob@feral.com, peter@wemm.org, cvs-committers@FreeBSD.org, freebsd-arch@FreeBSD.org Subject: Re: cvs commit: src/sbin/newfs newfs.8 newfs.c Message-ID: <200110160230.f9G2UHw28982@apollo.backplane.com> References: <200110152328.f9FNSmm74347@misha.privatelabs.com> <200110160153.f9G1rn727634@harmony.village.org>
next in thread | previous in thread | raw e-mail | index | archive | help
:
:In message <200110152328.f9FNSmm74347@misha.privatelabs.com> mi@aldan.algebra.com writes:
:: I don't like the "vaguesness" of this. Do other ratios work (even if
:: slower), or do the trigger bugs nobody wants to chase? Just MHO, of
:: course...
:
:other ratios do seem to work on systems that don't have a high disk
:load. we've shipped systems with 4k/1k and haven't had problems with
:that.
:
:Warner
Well, this is why the bug hasn't been nailed down yet. For a long time
(3.x, 4.0, 4.1 days) the buffer cache and VM system had serious breakage
when it came to oddly sized buffers and we spent a long time fixing that.
So far nobody has been able to determine whether the fragment ratio
breakage (when it can be reproduced at all) is due to bugs in UFS or
if there are still bugs in the buffer cache.
-Matt
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-arch" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200110160230.f9G2UHw28982>