Date: Mon, 11 Mar 2002 16:36:41 -0500 (EST) From: John Baldwin <jhb@FreeBSD.org> To: SXren Schmidt <sos@FreeBSD.org> Cc: cvs-all@FreeBSD.org, cvs-committers@FreeBSD.org Subject: RE: cvs commit: src/sys/dev/ata ata-all.c ata-all.h ata-disk.c Message-ID: <XFMail.20020311163641.jhb@FreeBSD.org> In-Reply-To: <200203112104.g2BL4Xm01754@freefall.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 11-Mar-2002 SXren Schmidt wrote: > sos 2002/03/11 13:04:32 PST > > Modified files: > sys/dev/ata ata-all.c ata-all.h ata-disk.c ata-raid.c > atapi-all.c atapi-cd.c atapi-fd.c > atapi-tape.c > Log: > Add new support for locking an ATA channel and use that throughout > the ATA/ATAPI driver. This solves the concurrency problem with > the new GEOM code, and also cuts a good deal of the patch size > in the upcoming MFC. Hmm, could you try using a mutex or sx lock instead of using atomic operations? You aren't using any memory barriers, thus this won't work on SMP sparc64, ia64, Pentium III+, or Alpha EV6+ machines. Also, the use of ATA_FORCELOCK_CH()in ata_reinit() looks very evil as you are just assuming that no threads on any other CPU's are messing with the channel at that point in time. -- John Baldwin <jhb@FreeBSD.org> <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?XFMail.20020311163641.jhb>