Date: Thu, 28 Feb 2008 16:30:13 -0700 (MST) From: "M. Warner Losh" <imp@bsdimp.com> To: henrik@gulbra.net Cc: freebsd-usb@FreeBSD.org Subject: Re: kernel panic when unmount removed usbdrive Message-ID: <20080228.163013.1387160175.imp@bsdimp.com> In-Reply-To: <1204240017.82302.119.camel@Particle> References: <c0f49ea0802280727o5b27038cx426cf77ddfbfa9ec@mail.gmail.com> <20080228170826.1fa1cadd.torfinn.ingolfsen@broadpark.no> <1204240017.82302.119.camel@Particle>
next in thread | previous in thread | raw e-mail | index | archive | help
In message: <1204240017.82302.119.camel@Particle>
Henrik Gulbrandsen <henrik@gulbra.net> writes:
: Right now, your best bet is to take Warner's advice and try 8.0-CURRENT,
: since none of the known patches have been merged to the 7.0 track yet.
: However, I honestly don't know how he can get anything stable without
: the final scsi_da.c patch, since I had something like 100% reproducible
: crashes when I updated my system from late 7.0 to early 8.0. You might
: want to apply patches from http://www.gulbra.net/freebsd-usb/ yourself,
: but keep in mind that CURRENT is a moving target, so I can't guarantee
: that the patches are still working perfectly. Also, please remember to
: use -o sync when mounting the drive. The other case is less tested...
Henrik,
I've not forgotten your patches, but I haven't had the time to
integrate them into the system. I hope to have them ready by 8.0, and
if I'm lucky, in time to MFC for 7.1. Many of them are MFCable,
although some of the underlying issues they expose can only be fixed
properly in head since they may require ABI changes. The changes
should really be done at the fs level, since the current patches will
leak resources and potentially have other bad effects.
I'd also like to have something between full sync and async when
mounting removable file systems. You want synchronous metadata, but
async data blocks for performance often times. I'm still exploring
what this means...
Finally, I'd like to thank you again publicly for taking the time and
initiative to see through the difficulties of this problem and to
persist in presenting your solution to the developers. It made me
lose my fear of the buffer cache code, which ultimately will be a good
thing.
Warner
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080228.163013.1387160175.imp>