Date: Thu, 25 Feb 1999 03:28:54 -0800 (PST) From: Matthew Jacob <mjacob@feral.com> To: Matthew Dillon <dillon@apollo.backplane.com> Cc: freebsd-hackers@FreeBSD.ORG Subject: Re: Panic in FFS/4.0 as of yesterday - update Message-ID: <Pine.LNX.4.04.9902250326570.28232-100000@feral-gw> In-Reply-To: <199902251009.CAA02488@apollo.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Yow! Indeed that would bung things up. I like also what you said about getnewbuf shouldn't be converting to async writes. Sounds like real good progress is happening here. I'll be back tonight and as sooon as folks are happy with some patches, I'll throw somwe big iron on the testing. On Thu, 25 Feb 1999, Matthew Dillon wrote: > Ah. It looks like the numfreebuffers accounting is messed up as well. > I had a lockup with processes sitting in 'newbuf' after I added a hard > check/sleep based on 'numfreebuffers'. > > test2:/home/dillon> sysctl -a | fgrep buffers > vfs.numdirtybuffers: 149 > vfs.lodirtybuffers: 95 > vfs.hidirtybuffers: 191 > vfs.numfreebuffers: 16715 <----- actually, there were none > vfs.lofreebuffers: 81 > vfs.hifreebuffers: 162 > > That could account for quite a bit, actually. It means getblk() wouldn't > block when it should. > > -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?Pine.LNX.4.04.9902250326570.28232-100000>