After a little bit of testing (the machine's not in production yet),
commenting out the I2O_EXEC_IOP_CLEAR call seems to work around the
problem.  Someone with some knowledge of the I2O protocols should
take a look at this.  At the very least, this machine's not capable of
soft powerdown, so hopefully won't risk losing any data from powering
off to quickly if there is somehow something still not written out...

--- /var/src.current/sys/dev/i2o/iop.c_v1.26	Mon Oct  7 22:07:11 2002
+++ /var/src.current/sys/dev/i2o/iop.c	Tue Oct 22 04:25:52 2002
@@ -904,8 +904,10 @@
 			continue;
 		iop_simple_cmd(sc, I2O_TID_IOP, I2O_EXEC_SYS_QUIESCE, IOP_ICTX,
 		    0, 5000);
+/*XXX*	Causes SCSI BIOS not to probe during POST after warm boot...
 		iop_simple_cmd(sc, I2O_TID_IOP, I2O_EXEC_IOP_CLEAR, IOP_ICTX,
 		    0, 1000);
+ *XXX*/
 	}
 
 	/* Wait.  Some boards could still be flushing, stupidly enough. */

jdarrow

-- 
John Darrow - Senior Technical Specialist               Office: 630/752-5201
Computing Services, Wheaton College, Wheaton, IL 60187  Fax:    630/752-5968
Pager via email: 6303160707@alphapage.airtouch.com      Pager:  630/316-0707
Email: John.P.Darrow@wheaton.edu (plain text please, no HTML or proprietary)