mjy@sdti.UUCP (Michael J. Young) (11/01/88)
I have some more information on the console driver bug that I reported
in a previous article (the one where the keyboard seemed to hang after
messing around with CTRL-S/CTRL-Q).
I was able to put an extra terminal on /dev/ttyM0 and poke around after
things got hung. The kernel is still running correctly afterthe keyboard
hangs, and the console process is waiting (WCHAN) at 0x2287a72 (kd_tty)
normally. When I kill the console process, the getty login prompt appears,
but the cursor is blinking funny as though unprinting characters are
being continually written to the screen. Then suddenly "Password:" appears
(getty thinks someone is logging in, even though I haven't touched the
keyboard), and things hang up again. In other words, the keyboard is
frotzed, and the kernel works fine on serial ports.
Resetting the machine doesn't always work. The keyboard is hung. When
this happens, only cycling the power does any good.
I can't be absolutely sure that the keyboard itself is not at fault, although
it seems passing strange that this failure mode appeared precisely when I
installed 2.4. Has anyone else seen problems?
By the way, I'm using an Acer 900 10Mhz AT clone, which I know is supported
by Microport.
--
Mike Young
Software Development Technologies, Inc., Sudbury MA Tel: +1 508 443 5779
Internet: mjy@sdti.sdti.com UUCP: {harvard,mit-eddie}!sdti!mjybrian@cbw1.UUCP (Brian Cuthie) (11/03/88)
In article <321@sdti.UUCP> mjy@sdti.UUCP (Michael J. Young) writes: > >I have some more information on the console driver bug that I reported >in a previous article (the one where the keyboard seemed to hang after >messing around with CTRL-S/CTRL-Q). > >I was able to put an extra terminal on /dev/ttyM0 and poke around after >things got hung. The kernel is still running correctly afterthe keyboard >hangs, and the console process is waiting (WCHAN) at 0x2287a72 (kd_tty) >normally. When I kill the console process, the getty login prompt appears, >but the cursor is blinking funny as though unprinting characters are >being continually written to the screen. Then suddenly "Password:" appears >(getty thinks someone is logging in, even though I haven't touched the >keyboard), and things hang up again. In other words, the keyboard is >frotzed, and the kernel works fine on serial ports. > >Resetting the machine doesn't always work. The keyboard is hung. When >this happens, only cycling the power does any good. > >I can't be absolutely sure that the keyboard itself is not at fault, although >it seems passing strange that this failure mode appeared precisely when I >installed 2.4. Has anyone else seen problems? > >By the way, I'm using an Acer 900 10Mhz AT clone, which I know is supported >by Microport. >-- >Mike Young >Software Development Technologies, Inc., Sudbury MA Tel: +1 508 443 5779 >Internet: mjy@sdti.sdti.com UUCP: {harvard,mit-eddie}!sdti!mjy I have experienced EXACTLY (I mean to the letter) the same problem with the V/386 3.0e release. This is clearly a new bug in the new kd driver. I have found some additional modes which cause it to hang. It is sometimes possible to reset the keyboard by just unplugging it. This, however, will not fix the ctrl-s/ctrl-q hang. -brian (hope this will post...) -- Brian D. Cuthie uunet!umbc3!cbw1!brian Columbia, MD brian@umbc3.umd.edu
kjk@pbhyf.PacBell.COM (Ken Keirnan) (11/03/88)
In article <105@cbw1.UUCP> brian@cbw1.UMD.EDU (Brian Cuthie) writes: >In article <321@sdti.UUCP> mjy@sdti.UUCP (Michael J. Young) writes: >> >>I have some more information on the console driver bug that I reported >>in a previous article (the one where the keyboard seemed to hang after >>messing around with CTRL-S/CTRL-Q). >> [ much stuff deleted ] >>Mike Young > >I have experienced EXACTLY (I mean to the letter) the same problem with the >V/386 3.0e release. This is clearly a new bug in the new kd driver. I have >found some additional modes which cause it to hang. It is sometimes >possible to reset the keyboard by just unplugging it. This, however, >will not fix the ctrl-s/ctrl-q hang. > >-brian I had problems with ctrl-s/ctrl-q using Microport (AT and 386) with my machine (great box, lousy keyboard) so I called Microport and complained. They told me they were trying to clean up the keyboard driver to correct the problem, but also, that some keyboards were more prone to exhibit the problem that others. Being innately cheap, I lived with the problem until I finally receive a beta console/keyboard driver to test. The situation had improved but was still an annoyance so I bit the bullet and bought a new keyboard (a so called "intelligent keyboard" from Northgate). My problems all disappeared. No doubt the keyboard hardware is just faster than the piece of junk I was using before (the new one has a nice tactile feel too :-)). I would also suspect that the software driver, keyboard controller chip and keyboard processor (in the keyboard) all contribute to the problem, and finding the right combination will fix it. Ken Keirnan -- Ken Keirnan - Pacific Bell - {att,bellcore,sun,ames,pyramid}!pacbell!pbhyf!kjk San Ramon, California kjk@pbhyf.PacBell.COM