[6873] in linux-scsi channel archive
Re: Problem using "scsi add-single-device" >/proc/scsi/scsi
daemon@ATHENA.MIT.EDU (D. Lance Robinson)
Thu Jul 22 17:53:56 1999
Date: Thu, 22 Jul 1999 14:05:53 -0700
From: "D. Lance Robinson" <lancer@airways.com>
Reply-To: lancer@airways.com
To: Gerard Roudier <groudier@club-internet.fr>
CC: linux-scsi <linux-scsi@vger.rutgers.edu>
Gerard,
Gerard Roudier wrote:
>
> > The test against BUS connected was kind of check for safety, but indeed
> > it is not that smart. Will study the situation and let you know.
> > For now, I am not sure if removing it is really safe.
>
> This test is just stupid and must be removed.
> Could you check if removing it fixes the problem?
I put in the code...
if (scntl1 & ISCON && hsts != HS_SEL_TIMEOUT)
goto reset_all;
This seems to work perfectly well. You suggest taking it out altogether?
----MORE----------------
Related to the SEL_TIMEOUT subject, Following is a patch that lets the
upper drivers know about the select timeout. Is this patch correct usage
of the DID_NO_CONNECT state?
--- sym53c8xx.c.orig Thu Jul 22 14:03:04 1999
+++ sym53c8xx.c Thu Jul 22 13:52:27 1999
@@ -6465,8 +6465,14 @@
*/
SetScsiResult(cmd, DID_OK, cp->scsi_status);
- } else if ((cp->host_status == HS_SEL_TIMEOUT)
- || (cp->host_status == HS_TIMEOUT)) {
+ } else if (cp->host_status == HS_SEL_TIMEOUT) {
+
+ /*
+ ** No response
+ */
+ SetScsiResult(cmd, DID_NO_CONNECT, cp->scsi_status);
+
+ } else if (cp->host_status == HS_TIMEOUT) {
/*
** No response
Thanks for your time, <>< Lance.
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
