Port-sparc64 archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]


Is the disk the only device attached to the SCSI bus concerned? ie: watch out for mixtures of wide and narrow devices on the same SCSI bus.

-- Mike Spooner

--------- Original Message ---------
From: Michael
To: port-sparc64%netbsd.org@localhost
Date: Sun Jun 11 17:33:15 GMT+01:00 2017
Subject: esp vs. MSG_IGN_WIDE_RESIDUE

I keep getting messages like this with my new old toy, and Ultra 2:
sd(0,0,0): unrecognized MESSAGE; sending REJECT

The disk is hooked to an esp/hme SBus card, so wide SCSI. I've been
using those on sparc with all sorts of disks and never saw this. It
seems harmless but seems to be followed by a drive reset /
recalibration every time, by the sound of it.
So I tried one of the known working-with-esp drives - same behaviour.
Some more digging revealed that the message in question is
MSG_IGN_WIDE_RESIDUE, which is handles only by a few drivers, and they
only ACK it and move on.

With this patch esp should do the same ( and prints which message is
being rejected, which is kinda helpful ):

Index: ncr53c9x.c
RCS file: /cvsroot/src/sys/dev/ic/ncr53c9x.c,v
retrieving revision 1.147
diff -u -w -r1.147 ncr53c9x.c
--- ncr53c9x.c 11 Jan 2017 07:16:48 -0000 1.147
+++ ncr53c9x.c 11 Jun 2017 16:29:26 -0000
@@ -1837,12 +1837,15 @@
goto reject;
+ NCR_MSGS(("ignore wide residue "));
+ break;

NCR_MSGS(("ident "));
- printf("%s: unrecognized MESSAGE; sending REJECT\n",
- device_xname(sc->sc_dev));
+ printf("%s: unrecognized MESSAGE (%x); sending REJECT\n",
+ device_xname(sc->sc_dev), sc->sc_imess[0]);

Since I'm not familiar with low level SCSI or esp hardware - any
comments? Is this even sane? It helps here, the resets are gone and
things work normally.

have fun

Home | Main Index | Thread Index | Old Index