port-x68k/46293: 6.0_BETA/x68k can't boot from HDD

[isaki%NetBSD.org@localhost]

>Number:         46293
>Category:       port-x68k
>Synopsis:       6.0_BETA/x68k can't boot from HDD
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    port-x68k-maintainer
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Thu Apr 05 01:45:00 +0000 2012
>Originator:     Tetsuya Isaki
>Release:        NetBSD 6.0_BETA/x68k
>Organization:
NetBSD
>Environment:
6.0_BETA/x68k (no uname because it cannot boot)
>Description:
Because mboot which is primary boot loader on HDD is corrupted,
6.0_BETA/x68k can't boot from HDD.

Here is a disassemble source of __asm() of IOCS_S_READEXT()
inline function.  trap#15 with %d0=#245, %d1=#38 is
X68000's SCSI IOCS (service call) _S_READEXT.  For this call
%d4 should contain SCSI ID, but %d4 = %d3 = %d6 = (8>>size)??
in this asm.  Therefore this service call fails and mboot will
exit.

 1ac:   7c08            moveq #8,%d6
 1ae:   e8a6            asrl %d4,%d6          ; %d6 <- (8>>size)
 1b0:   48e7 1c00       moveml %d3-%d5,%sp@-
 1b4:   2400            movel %d0,%d2
 1b6:   2606            movel %d6,%d3
 1b8:   2803            movel %d3,%d4         ; %d4 <- %d3 <- %d6
 1ba:   2a04            movel %d4,%d5
 1bc:   227c 0000 2400  moveal #9216,%a1
 1c2:   7226            moveq #38,%d1
 1c4:   203c 0000 00f5  movel #245,%d0
 1ca:   4e4f            trap #15

The source of the same place of NetBSD 5.1 are as follows:

 13c:   7008            moveq #8,%d0
 13e:   e6a0            asrl %d3,%d0          ; %d0 <- (8>>size)
 140:   48e7 1c00       moveml %d3-%d5,%sp@-
 144:   2404            movel %d4,%d2
 146:   2600            movel %d0,%d3         ; %d3 <- (8>>size)
 148:   2805            movel %d5,%d4         ; %d4 <- %d5
 14a:   2a03            movel %d3,%d5
 14c:   227c 0000 2400  moveal #9216,%a1
 152:   7226            moveq #38,%d1
 154:   203c 0000 00f5  movel #245,%d0
 15a:   4e4f            trap #15

>How-To-Repeat:
Install 6.0_BETA/x68k on 'all zero-cleaned' HDD
and reboot from sysinst.
>Fix:
No, at present.