Date: Wed, 5 Apr 2000 15:02:03 +0930 From: Greg Lehey <grog@lemis.com> To: Poul-Henning Kamp <phk@critter.freebsd.dk> Cc: Brad Knowles <blk@skynet.be>, Jeroen Ruigrok/Asmodai <asmodai@wxs.nl>, Matthew Dillon <dillon@apollo.backplane.com>, current@FreeBSD.ORG Subject: Vinum breakage (was: 2nd call for reviews and tests: buf->bio patch) Message-ID: <20000405150203.B66569@freebie.lemis.com> In-Reply-To: <27235.954855223@critter.freebsd.dk> References: <v04220803b50f8d254208@[195.238.1.121]> <27235.954855223@critter.freebsd.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tuesday, 4 April 2000 at 15:33:43 +0200, Poul-Henning Kamp wrote: > In message <v04220803b50f8d254208@[195.238.1.121]>, Brad Knowles writes: >> At 8:23 AM +0200 2000/4/4, Jeroen Ruigrok/Asmodai wrote: >> >>> Like I said in another mail, this is CURRENT, things are >>> expected to break. You want stability of API's, go 3-STABLE. You want >>> a somewhat stable environment, go 4-STABLE. 5 is bleeding edge, have >>> your bandages ready. >> >> Like I said in a previous message, Poul broke vinum (among other >> things) under 4.0-STABLE, and this needs to be fixed ASAP. If Poul >> had kept his changes unique to -CURRENT, then 4.0-STABLE wouldn't >> have been affected. But he didn't, and it was. > > Brad, > > You are confused. > > I have not made one single commit to 4.0-STABLE which even comes > close to vinum. > > The problem in vinum/releng4 is the problem greg has been blaming > on CAM since at least FreeBSDcon and indications seems to be that > it is actually a malloc/free gottcha in vinum. No, this is not correct. The problem you're talking about I blamed elsewhere, not specifically CAM. To quote from http://www.lemis.com/vinum/bugs.html: Technical explanation: A buffer header gets corrupted between the time the top half of the driver issues the request to the disk driver, and when the I/O completes. Currently, the evidence is pointing towards the disk driver, but the corruption is of such an unusual nature that it's difficult to guess what's going on. This bug went away without me making any change to Vinum, so I feel that my opinion is correct. This isn't the bug which Søren has been seeing, and indeed your last lot of commits made it impossible for the bug to ever show up again in this form (since there is no longer a B_CALL in the buffer header). Greg -- Linuxcare - BSD support for the revolution. See complete headers for address and phone numbers To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000405150203.B66569>