On Wed, 7 Jun 2006, Lev A. Melnikovsky wrote:
> No. It has only one SCSI connector and this is why I assumed that I shouldn't
> care about terminators.
Ok, let's assume for now it knows what to do...
Now, below is a patch... That's about all I can say about it:-) The
reselect routine didn't seem right to me, specifically, it checks for "no
active request-block or active request-block is not disconnected" and then
sends an abort to the device, but if there was indeed an active srb, it
gets lost, AFAICS. FWIW, tmscsim does the same... The patch below tries to
complete the command before dropping it on the floor... Any reviews
appreciated.
And there's one more page_address there in dc395x...
Warning, Lev - this patch may explode your scanner, or upload your bank
account info to p2p servers, or send your gpg keys to microsoft, or...
You've been warned:-) You might want to wait for some comments to the
patch...
Thanks
Guennadi
---
Guennadi Liakhovetski
Return the active request for retry to the upper layer in case of
"reselect without disconnect" instead of just losing it.
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@xxxxxx>
diff -u -p a/drivers/scsi/dc395x.c b/drivers/scsi/dc395x.c
--- a/drivers/scsi/dc395x.c 30 Mar 2006 19:19:15 -0000
+++ b/drivers/scsi/dc395x.c 7 Jun 2006 21:07:34 -0000
@@ -3198,9 +3198,13 @@ static void reselect(struct AdapterCtlBl
/*
* abort command
*/
- dprintkl(KERN_DEBUG,
- "reselect: w/o disconnected cmds <%02i-%i>\n",
- dcb->target_id, dcb->target_lun);
+ dprintkl(KERN_WARNING,
+ "reselect: w/o disconnected cmds <%02i-%i> with%s active srb\n",
+ dcb->target_id, dcb->target_lun, srb ? "" : "out");
+ if (srb) {
+ SET_RES_DID(srb->cmd->result, DID_SOFT_ERROR);
+ srb_done(acb, dcb, srb);
+ }
srb = acb->tmp_srb;
srb->state = SRB_UNEXPECT_RESEL;
dcb->active_srb = srb;
-
: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
