Date: Sun, 19 Aug 2001 17:06:31 -0700 (PDT) From: Matt Dillon <dillon@earth.backplane.com> To: Rik van Riel <riel@conectiva.com.br> Cc: David Greenman <dg@root.com>, Sergey Babkin <babkin@bellatlantic.net>, <hackers@FreeBSD.ORG>, <murray@FreeBSD.ORG>, <jkh@FreeBSD.ORG> Subject: Re: Recommendation for minor KVM adjustments for the release Message-ID: <200108200006.f7K06VK47176@earth.backplane.com> References: <Pine.LNX.4.33L.0108191749140.5646-100000@imladris.rielhome.conectiva>
next in thread | previous in thread | raw e-mail | index | archive | help
:> Yes, and the buffer cache determines how much dirty file-backed data
:> (via write() or mmap()) the system is allowed to accumulate before
:> it forces it out, which should probably be the greater concern here.
:
:How hard would it be to allow dirty data in the file
:cache, without buffer mappings ?
:
:regards,
:
:Rik
This is already supported via MAP_NOSYNC. The problem is that the
dirty data is not tracked under light memory loads (tracking dirty
data is a function of the buffer cache), so unless something
forces the page out or fsync()s the file explicitly, the dirty data
might not be written out for days. It might look good in a benchmark,
but it would play havoc on system stability in the event of a crash.
-Matt
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?200108200006.f7K06VK47176>