Re: kernel panic with 2.5.67-ac1

Patrick Mansfield (patmans@us.ibm.com)
Fri, 18 Apr 2003 10:30:31 -0700


On Fri, Apr 18, 2003 at 10:18:06AM -0700, Christian Staudenmayer wrote:
> Note: this does not happen with 2.4.20, 2.4.21-pre7, 2.4.21-pre7-ac1 or 2.5.67-bk8
> It does, however happen with 2.5.67-ac2, but the error message is some lines longer
> and some of the addresses have changed.
>
> I'd be really grateful for any insight on this problem.

We were plugging a queue that was about to be freed during scsi scan.

This is fixed in bk8, here is a snippit of part of the patch to scsi_lib.c
that fixed the problem, or look at the end of scsi_prep_fn, (plus the
corresponding call to blk_plug_device was removed from scsi_request_fn):

+ defer:
+ /* If we defer, the elv_next_request() returns NULL, but the
+ * queue must be restarted, so we plug here if no returning
+ * command will automatically do that. */
+ if (sdev->device_busy == 0)
+ blk_plug_device(q);
+ return BLKPREP_DEFER;
+}

-- Patrick Mansfield
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/