uplcom broken by MPSAFE TTY commit?

[Ed Schouten]

Hello Christian,

* Christian Weisgerber <naddy at mips.inka.de> wrote:
> Has anybody noticed problems with uplcom after ed's MPSAFE TTY
> commit?
> 
> ucom1: <FTDI USB HS Serial Converter, class 0/0, rev 1.10/2.00, addr 4> on uhub2
> ucom2: <Prolific Technology Inc. USB-Serial Controller, class 0/0, rev 1.10/3.00, addr 5> on uhub2
> 
> I'm seeing incomplete input followed by repetitive spew on ucom2.
> The uftdi-based ucom1 seems fine.
> 
> ucom2 is hooked up to the serial console of an OpenBSD box.  The
> getty there issues a banner like this:
> 
> OpenBSD/alpha (kemoauc.mips.inka.de) (tty00)
> 
> What I now get in cu(1) is a stream of this:
> 
> OpenBSD/alpha (kemoauc.mips.in
> OpenBSD/alpha (kemoauc.mips.in
> OpenBSD/alpha (kemoauc.mips.inka.de) (tty00)
> 
> OpenBSD/alpha (kemoauc.mips.in
> OpenBSD/alpha (kemoauc.mips.inka.de) (tty00)
> 
> OpenBSD/alpha (kemoauc.mips.in
> OpenBSD/alpha (kemoauc.mips.in
> OpenBSD/alpha (kemoauc.mips.inka.de) (tty00)
> 
> OpenBSD/alpha (kemoauc.mips.in
> OpenBSD/alpha (kemoauc.mips.in
> OpenBSD/alpha (kemoauc.mips.inka.de) (tty00)
> 
> [... and it keeps spewing ...]

I think the best thing to test first is if things work if you
enable/disable hardware and software flow control. The semantics of the
new TTY layer are pretty much the same as the old TTY layer, with the
big exception of flow control.

If that doesn't help, could you compile your kernel with USB_DEBUG,
set sysctl hw.usb.ucom.debug=1 and give me a dump of the text that
scrolls by?

Thanks for reporting this issue to the lists!

-- 
 Ed Schouten <ed at 80386.nl>
 WWW: http://80386.nl/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 195 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-current/attachments/20080821/a74609c2/attachment.pgp