[Bug 256] GSM Modem doesn't seem to work on some devices

bugzilla-daemon at bugzilla.openmoko.org bugzilla-daemon at bugzilla.openmoko.org
Fri Mar 16 05:25:06 CET 2007


------- Additional Comments From werner at openmoko.org  2007-03-16 05:25 -------
I just got a "dead" ("undead" fit better, though) GSM modem to work on a
bv2 device by issuing, from a parallel session
stty -F /dev/ttySAC0 crtscts 

System setup:
- latest u-boot (local build, full of ugly hacks)
- uImage-2.6-200703022008-fic-gta01.bin
- openmoko-devel-image-fic-gta01-20070309001017.rootfs.jffs2
- "cu" from openmoko-devel-image-fic-gta01-20070302115654.rootfs.jffs2

The same system worked fine, without needing stty, if using

In order to avoid as much external influence as possible, I booted,
in both cases, with init=/noinit, /noinit being the following script:

#!/bin/sh -x
mount -t proc none /proc
mount -t sysfs none /sys
mkdir -p /dev/pts
mount -t devpts none /dev/pts
mknod /dev/null c 1 3
mknod /dev/zero c 1 5
mknod /dev/random c 1 8
mknod /dev/urandom c 1 9
mknod /dev/tty c 5 0
mknod /dev/console c 5 1
mknod /dev/ptmx c 5 2
mknod /dev/ttySAC0 c 204 64
ifconfig lo up
ifconfig usb0 up
dropbear -r /etc/dropbear/dropbear_rsa_host_key -p 22
sleep 3600

Thus, given that hardware, u-boot, kernel, and the "cu" binary were
identical, and there were no daemons running at all, this leaves either
glibc or some arbitrary initial condition. I've checked the strace output
of "cu" on both configurations, and they're equivalent, except for the
terminal settings. Interestingly, TCGETS had 0x0002 set in lflags on the
"new" rootfs (i.e., the one where the GSM modem didn't work), which should
be ICANON. On TCSETS[W], the "new" system lacked CRTSCTS and also CLOCAL.
(Note: I just rebooted the regular system and checked with stty -a, and
CLOCAL was set. CRTSCTS was still missing, though.)


------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

More information about the buglog mailing list