Date: Fri, 18 Dec 2009 05:31:24 -0800 (PST) From: RuiDC <ruidc@yahoo.com> To: freebsd-arm@freebsd.org Subject: Re: fetch data corruption on local fs Message-ID: <26843272.post@talk.nabble.com> In-Reply-To: <200912162143.nBGLhbva038162@casselton.net> References: <26803523.post@talk.nabble.com> <4B28C608.1070802@gmail.com> <4B28CFCD.3000401@semihalf.com> <26811801.post@talk.nabble.com> <200912162143.nBGLhbva038162@casselton.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Mark Tinguely wrote: > > Sounds like you are using preload. > > Looks like sys/dev/md/md.c could use a cpu_flush_dcache() for the > BIO_WRITE cases in mdstart_[malloc|preload] cases. Though from just > mdstart_preload, I don't see an obvious reason the "-o sync" is > helping. > > The pmap_kremove() patch for sf_buf_free() should help for the BIO_WRITE > in the mdstart_swap case. > Thanks for your reply, I'm not quite sure what you mean in your reply, but presume that your suggestions may workaround this problem when performing fetch operations against an md filesystem? if so, then this may be the case, but my real target is UFS filesystem, I was just using md as an alternative to the NFS-mounted filesystem suggested to test problem and workaround. If I have misunderstood, then please forgive me. -- View this message in context: http://old.nabble.com/problem-setting-up-ports-tp26803523p26843272.html Sent from the freebsd-arm mailing list archive at Nabble.com.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?26843272.post>