dag@atomica.fi (Dag Nygren) (01/12/91)
Anybody out there succeeded in mapping the console to another device in SCO UNIX, for example an to apply to xconsole? It is a bit of a problem with the console messages messing up the X-window screen. Grateful for any help.... (SCO is also invited :-) ) -- Dag Nygren email: dag@atomica.fi Oy Atomica Ab or santra!atomica!dag Box 22 (Ahventie 4A) 02170 Esbo
mschedlb@hawk.ulowell.edu (Martin J. Schedlbauer) (01/13/91)
In article <1991AMSat.12.4222@atomica.fi> dag@atomica.fi (Dag Nygren) writes: > >Anybody out there succeeded in mapping the console to another device >in SCO UNIX, for example an to apply to xconsole? It is a bit of a >problem with the console messages messing up the X-window screen. I think you get a similar effect if you start one of your xterm with the flag -console. ...Martin Martin Schedlbauer (mschedlb@ulowell.edu) Institute for Visualization and Perception Research Department of Computer Science University of Lowell, Lowell, MA 01854 (Tel: (508) 934-3612)
allbery@NCoast.ORG (Brandon S. Allbery KB8JRR) (01/13/91)
As quoted from <1991AMSat.12.4222@atomica.fi> by dag@atomica.fi (Dag Nygren): +--------------- | Anybody out there succeeded in mapping the console to another device | in SCO UNIX, for example an to apply to xconsole? It is a bit of a | problem with the console messages messing up the X-window screen. +--------------- The Xenix port of MGR opens /dev/error and issues an ioctl() on it so that it will get SIGUSR1 when a kernel message is generated; it then reads /dev/error and displays it in the "console" window. I left this as is in my port to SCO UNIX, and it does in fact work. Most likely, this is documented under the "error" device. (Status report on the MGR port: either 95% or 45% complete, depending on whether the occasional core dump I'm getting is my code or a symptom of some apparently lax argument checking in MGR escape sequences, and if the latter, whether I want to fix the generic MGR code or just report it to Bellcore. Display of icons seems to be good at triggering core dumps, depending on the exact position chosen on the display for the bitcopyto fnction --- I suspect something isn't checking for 16-bit alignment.) ++Brandon -- Me: Brandon S. Allbery VHF/UHF: KB8JRR on 220, 2m, 440 Internet: allbery@NCoast.ORG Packet: KB8JRR @ WA8BXN America OnLine: KB8JRR AMPR: KB8JRR.AmPR.ORG [44.70.4.88] uunet!usenet.ins.cwru.edu!ncoast!allbery Delphi: ALLBERY
dag@atomica.fi (Dag Nygren) (01/14/91)
allbery@NCoast.ORG (Brandon S. Allbery KB8JRR) writes: >As quoted from <1991AMSat.12.4222@atomica.fi> by dag@atomica.fi (Dag Nygren): >+--------------- >| Anybody out there succeeded in mapping the console to another device >| in SCO UNIX, for example an to apply to xconsole? It is a bit of a >| problem with the console messages messing up the X-window screen. >+--------------- >The Xenix port of MGR opens /dev/error and issues an ioctl() on it so that it >will get SIGUSR1 when a kernel message is generated; it then reads /dev/error >and displays it in the "console" window. I left this as is in my port to SCO >UNIX, and it does in fact work. Most likely, this is documented under the >"error" device. Thanks, I will try that, but does it really prevent (disable) the writing of the messages to the screen ??...... OK, I will try it.... -- Dag Nygren email: dag@atomica.fi Oy Atomica Ab or santra!atomica!dag Box 22 (Ahventie 4A) 02170 Esbo
dag@atomica.fi (Dag Nygren) (01/14/91)
mschedlb@hawk.ulowell.edu (Martin J. Schedlbauer) writes: >In article <1991AMSat.12.4222@atomica.fi> dag@atomica.fi (Dag Nygren) writes: >> >>Anybody out there succeeded in mapping the console to another device >>in SCO UNIX, for example an to apply to xconsole? It is a bit of a >>problem with the console messages messing up the X-window screen. >I think you get a similar effect if you start one of your xterm with the >flag -console. Thanks, but the real problem is getting the SCO compiled xterm to understand this. That is to find something to replace the missing TIOCCONS ioctl with..... -- Dag Nygren email: dag@atomica.fi Oy Atomica Ab or santra!atomica!dag Box 22 (Ahventie 4A) 02170 Esbo
allbery@NCoast.ORG (Brandon S. Allbery KB8JRR) (01/15/91)
As quoted from <1991PMSun.13.24375@atomica.fi> by dag@atomica.fi (Dag Nygren): +--------------- | allbery@NCoast.ORG (Brandon S. Allbery KB8JRR) writes: | >As quoted from <1991AMSat.12.4222@atomica.fi> by dag@atomica.fi (Dag Nygren): | >+--------------- | >| Anybody out there succeeded in mapping the console to another device | >| in SCO UNIX, for example an to apply to xconsole? It is a bit of a | >| problem with the console messages messing up the X-window screen. | >+--------------- | | >The Xenix port of MGR opens /dev/error and issues an ioctl() on it so that it | >will get SIGUSR1 when a kernel message is generated; it then reads /dev/error | >and displays it in the "console" window. I left this as is in my port to SCO | >UNIX, and it does in fact work. Most likely, this is documented under the | >"error" device. | | Thanks, I will try that, but does it really prevent (disable) the | writing of the messages to the screen ??...... OK, I will try it.... +--------------- I don't know for certain... as I said, I didn't pick the code apart. In fact, my statement above is just what a cursory examination seemed to be saying. In any case, error messages are showing up under MGR in a window whose controlling terminal is a pty... so TIOCCONS would likely not work regardless. I suspect that you could arrange to grab error messages and either bit-bucket them (since they get put in /usr/adm/messages anyway --- not that I recommend this) or redirect them to someplace less annoying than a window-splatter, like a special reserved "error window" like MGR does. ++Brandon -- Me: Brandon S. Allbery VHF/UHF: KB8JRR on 220, 2m, 440 Internet: allbery@NCoast.ORG Packet: KB8JRR @ WA8BXN America OnLine: KB8JRR AMPR: KB8JRR.AmPR.ORG [44.70.4.88] uunet!usenet.ins.cwru.edu!ncoast!allbery Delphi: ALLBERY