From owner-cvs-all Thu Jan 13 0: 4:32 2000 Delivered-To: cvs-all@freebsd.org Received: from mailman.zeta.org.au (mailman.zeta.org.au [203.26.10.16]) by hub.freebsd.org (Postfix) with ESMTP id 165F214F86; Thu, 13 Jan 2000 00:04:26 -0800 (PST) (envelope-from bde@zeta.org.au) Received: from bde.zeta.org.au (beefcake.zeta.org.au [203.26.10.12]) by mailman.zeta.org.au (8.8.7/8.8.7) with ESMTP id TAA29623; Thu, 13 Jan 2000 19:02:51 +1100 Date: Thu, 13 Jan 2000 19:02:48 +1100 (EST) From: Bruce Evans X-Sender: bde@alphplex.bde.org To: Gerard Roudier Cc: "Rodney W. Grimes" , Matthew Jacob , Peter Wemm , obrien@NUXI.com, "Chris D. Faulhaber" , cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/i386/conf GENERIC LINT In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk On Wed, 12 Jan 2000, Gerard Roudier wrote: > By the way, releng_3 (from last week-end) did crash when I attempted to > write to a write-protected diskette: > > panic: vinvalbuf: dirty buffer > > Has this one been caught? This is well understood. Rev.1.196 of vfs_bio.c arranges to not invalidate buffers in brelse() after a write error, without fixing all the things that depend on them being invalidated. Writing of buffers is now retried forever after an error. Panics ensue for retries after last-close. This is partly fixed in -current. Block devices went away, so the problem is harder for inexperienced users to stumble over. There are still issues for write errors on mounted file systems. It is impossible to unmount some filesystems, because the final sync for unmount retries write errors forever. Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message