Date: Thu, 7 Mar 2002 08:42:59 -0800 From: Alfred Perlstein <bright@mu.org> To: Jan Grant <Jan.Grant@bristol.ac.uk> Cc: freebsd-hackers@freebsd.org Subject: Re: Bug? still looking, yet to knock up small test case. Suggestions solicited. Message-ID: <20020307164259.GE26621@elvis.mu.org> In-Reply-To: <Pine.GSO.4.44.0203071601370.13361-100000@mail.ilrt.bris.ac.uk> References: <Pine.GSO.4.44.0203071601370.13361-100000@mail.ilrt.bris.ac.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
* Jan Grant <Jan.Grant@bristol.ac.uk> [020307 08:12] wrote: > Something odd seems to be happening; I'd appreciate "look here" > suggestions. I suspect mmapped pages aren't being flushed but gawd alone > knows why. > > Situation: vmware2, with a "fake disk", files in the /external FS > (/external/vmware1/nt1.*). > > FBSD-stable. Behaviour's been like this for quite some time (since > around 4.3-release, I think) [snip] > Note: this doesn't result in ultimately broken behaviour, because those > pages are eventually flushed on orderly shutdown. I'm just curious as to > why sync(2) isn't forcing this. Afaik this is a "feature" of Linux that the emulator tries to emulate. From what I've _heard_ (not witnessed), Linux doesn't have the syncer flush mmap(2)'d data on a regular basis, so in FreeBSD we emulate this behaviour and you get the behaviour you've been seeing. Since, sync(2) schedules the syncer to sync out the buffers which are marked for ignore by the syncer you don't see anything being done. :) -Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020307164259.GE26621>