Subject: port-i386/3384: i386 bootup hangs waiting for response from adaptec 1542
To: None <gnats-bugs@gnats.netbsd.org>
From: None <stephen.ma@jtec.com.au>
List: netbsd-bugs
Date: 03/24/1997 08:41:26
>Number:         3384
>Category:       port-i386
>Synopsis:       bootup hangs in swapconf(), at tsleep() in scsi_exectute_xs()
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    gnats-admin (GNATS administrator)
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sun Mar 23 13:50:02 1997
>Last-Modified:
>Originator:     stephen.ma@jtec.com.au
>Organization:
Why don't we call it an "organisation"?
>Release:        NetBSD-current 970321
>Environment:
 [ this is from an earlier kernel that works ]
NetBSD 1.2C (KHEMLANI) #17: Sat Mar  8 01:04:31 EST 1997
    stephen@khemlani.jtec.com.au:/usr/src/sys/arch/i386/compile/KHEMLANI
cpu0: Intel 486DX (486-class)
real mem  = 16384000
avail mem = 14196736
using 225 buffers containing 921600 bytes of memory
mainbus0 (root)
isa0 at mainbus0
aha0 at isa0 port 0x330-0x333 irq 11 drq 5
aha0: aha_cmd, host not finished(0x0)
aha0: model AHA-unknown, firmware 
aha0: aha_cmd, host not finished(0x0)
aha0: aha_cmd, host not finished(0x0)
aha0: sync, parity
aha0: aha_cmd, host not finished(0x0)
scsibus0 at aha0: 8 targets
sd0 at scsibus0 targ 0 lun 0: <SEAGATE, ST1401, 7338> SCSI2 0/direct fixed
sd0: 322MB, 1113 cyl, 9 head, 66 sec, 512 bytes/sec
sd1 at scsibus0 targ 1 lun 0: <CONNER, CFP1080S, 3939> SCSI2 0/direct fixed
sd1: 1030MB, 3658 cyl, 6 head, 96 sec, 512 bytes/sec
com0 at isa0 port 0x3f8-0x3ff irq 4: ns16550a, working fifo
com1 at isa0 port 0x2f8-0x2ff irq 3: ns16550a, working fifo
lpt0 at isa0 port 0x378-0x37b irq 7
sb0: sbdsp probe failed
npx0 at isa0 port 0xf0-0xff: using exception 16
pc0 at isa0 port 0x60-0x6f irq 1: color
fdc0 at isa0 port 0x3f0-0x3f7 irq 6 drq 2
fd0 at fdc0 drive 0: 1.44MB 80 cyl, 2 head, 18 sec
isapnp0 at isa0 port 0x279: read port 0x207
isapnp0: <Creative SB16 PnP, CTL0031, Audio> port 0x220/16,0x300/2,0x388/4 irq 5 drq 0,5 not configured
isapnp0: <Creative SB16 PnP, CTL2011, IDE> port 0x168/8,0x36e/2 irq 10 not configured
isapnp0: <Creative SB16 PnP, PNPFFFF, Reserved> port 0x100/1 not configured
isapnp0: <Creative SB16 PnP, CTL7001, Game> ignored; resource conflict
biomask 840 netmask 840 ttymask 8da
boot device: sd1
root on sd1a swap on sd1b
root file system type: ffs

System: NetBSD whitewater.jtec.com.au 1.2D NetBSD 1.2D (WHITEWATER) #11: Sat Mar 22 14:40:11 EST 1997 stephen@whitewater.jtec.com.au:/usr/src/sys/arch/i386/compile/WHITEWATER i386

>Description:
During bootup, the device probe/attach routines run with no problems. The
kernel reaches the swapconf() call in autoconf.c and hangs. (This HD
activity LED stays lit after the kernel hangs). A DDB backtrace
indicates the kernel tsleep()ing in scsi_exectute_xs(). This is, I presume
the first drive system call after the device probe/attach has completed.

As far as I can see, aha_scsi_cmd() is called correctly from that
scsi_execute_xs(), but aha_intr() is not subsequently called.

I can see aha_intr() being called during the probe of the scsi bus.
There appears to be a spurious aha_intr() going off at the
spl0() call in configure(). This last call is probably confusing
aha.c

>How-To-Repeat:
Try booting a recent kernel on a machine with an Adaptec 1542B.
>Fix:
dunno
>Audit-Trail:
>Unformatted: