Return to BSD News archive
Path: sserve!manuel.anu.edu.au!munnari.oz.au!news.hawaii.edu!ames!saimiri.primate.wisc.edu!caen!sol.ctr.columbia.edu!ira.uka.de!Germany.EU.net!unidui!du9ds3!veit From: veit@du9ds3.uni-duisburg.de (Holger Veit) Newsgroups: comp.unix.bsd Subject: Re: keycap problems Date: 10 Nov 92 08:49:57 GMT Organization: Uni-Duisburg FB9 Datenverarbeitung Lines: 83 Message-ID: <veit.721385397@du9ds3> References: <1dmnchINNqj7@mailgzrz.TU-Berlin.DE> Reply-To: veit@du9ds3.uni-duisburg.de NNTP-Posting-Host: du9ds3.uni-duisburg.de Keywords: keycap In <1dmnchINNqj7@mailgzrz.TU-Berlin.DE> meikel@marie.physik.tu-berlin.de (Michael Feig) writes: >I installed the recent version of keycap onto our BSD machine by >patching the kernel according to the given diff-Files and supplemental >source files. Everything works fine as long as I don't start X-Windows >(XFree 1.2) using the compiled X386 which is part of the keycap-0.1.1 >distribution. Within X-Windows the mouse works only in horizontal >direction and the mapping of the keyboard is totally out of order >(pressing the 'a', 's', 'd' keys causes numbers to appear, shift right >produces some special character like '=' or '+' - I forgot what it >actually was). Some time later (~30 seconds) the whole BSD-machine >hangs and has to be rebooted. >Just to make things clear: X-Windows worked absolutely all right without any >problems before installing keycap. >As a try to solve the problem, I used the supplied kernel instead of the >one I have build from the sources after applying all necessary patches, >but I discovered that doesn't contain the patch for reserved ports with >NFS which is essential for us, because we have to mount part of the >BSD system from a Sun via NFS. >Has anyone experienced similar problems or has anyone an idea how >I can get things going the way they are supposed to work ? >Thanks for your help, Michael. >-- > Michael Feig INTERNET: meikel@marie.physik.tu-berlin.de > Department of Physics meikel@cs.tu-berlin.de > Technical University of Berlin Thanks for this response. Let's try to figure out what might have happened. I didn't find any of these problems with a valid configured kernel/X386, and some other people already succeeded in getting a working version. 0. What does the kernel say during bootup? Does it say "co0<your video card type>" or "pc0 at .." If it tells you the latter, you should RTFM. The codriver package allows to build a compatibility kernel (which is basically the old pccons with old X11 fixes) and the new codrv. In the latter case you should use the configuration line from the sample config file which reads device co0 at isa? port "IO_KBD" tty irq 1 vector cointr instead of the standard line with pc0. Together with the new XServer the old line will cause many problems. If this does not help: 1. Which type of keyboard do you run? If this is a "converted XT-keyboard", it is likely, that it does not understand AT scan codes. However, you should have found the same effect in text mode already. 2. There is a known problem with ISO-Translation, which is done both in the server and in the kernel. The server code is obsolete. This, however, only affects national keys with codes >=128. 3. Does the /dev/pccons exist, i.e. the device (usually major=12 minor=0) to which /dev/console usually is mapped (or is it called /dev/vga)? Delete it! Also, if you modified the kernel conf.c, you might adjust the /dev entries, which my kernel relys on back to the default (of the fixit disk). 4. Which mouse type do you use? Do you use a busmouse? Or a mouse which is driven through the keyboard controller? There is one change which might afflict the behavior of the mouse in the prebuilt kernel: The prebuilt kernel has cgd's com driver built in, which comes with devices /dev/tty0[01] and /dev/ttyd[01], the latter of which has the right minor number. I didn't get negative responses about the mouse yet, because the codriver doesn't influence the serial lines. The enclosed patch for tty_pty.c is not involved in this behavior. 5. Finally, there might be an address or interrupt conflict, although this will usually make cause problems already in text mode (What happens to keyboard codes when moving the mouse?) If you still have problems, send me your exact hardware and software configuration. Holger -- | | / Dr. Holger Veit | INTERNET: veit@du9ds3.uni-duisburg.de |__| / University of Duisburg | "XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX | | / Dept. of Electr. Eng. | Sorry, the above really good fortune has | |/ Inst. f. Dataprocessing | been CENSORED because of obscenity"