Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!cs.utexas.edu!sdd.hp.com!ucsd!rutgers!njin!princeton!pucc!AUXPOST From: MATLEVAN@EKU.BITNET (Jerry LeVan) Newsgroups: comp.unix.aux Subject: A/UX 2.0 and X11R4 Message-ID: Date: 2 Sep 90 22:02:00 GMT Sender: VMNNPOST@pucc.Princeton.EDU (Listserv to Netnews Gateway) Organization: Listserv to Netnews Gateway at pucc.Princeton.EDU Lines: 79 Approved: AUXPOST@PUCC Hello netters, I have VERY recently installed A/UX 2.0. Today I felt brave and decided to install X11R4 ( I had been running X with A/UX 1.1). I reinstalled gcc 1-37 that I picked up from apple.com last febuary.(Unfortunately this later lead to problems.) I tryed the compiler on several small programs and it seemed to work ok. I then mounted the disk containing the X source tree, I had just applied some patches a couple of weeks ago and thought that everything should just copy over to the target disk! When I pulled the trigger "make install" much to my chagrin, the compiler kicked in and a large part of X got recompiled.Xlib,the clients and demos where recompiled but Xt and Xaw where not! After several hours the installation completed. It seems to be working OK. The only problem I have encountered so far is that vi seems to have changed. vi no longer "understands" xterm. Even though TERM=xterm and TERMCAP contains the termcap info vi says it does not understand "xterm" and falls into "open" mode. Does vi only examine the terminfo file in this version of A/UX? (there is no xterm entry in terminfo.) Has anyone built a xterm terminfo source description? (I am using TERM=vt100 as a workaround but I don't know yet if there are going to be editing problems.) Another problem I have inflicted upon myself is that one must be root in order to do the "make install" command consequently all of the compiled files are now owned by root. I also neglected to give myself the same uid as I had in v1.1, this means that all files in the X system now belong to someone else sigh... I have discovered how to launch X from the login window, In the folder /mac/lib/sessiontypes there are 3 files (this if from an man with an obviously defective memory so be careful). These are the startup documents (disquised as applications). One starts the old A/UX console mode, the other two start the 24 and 32 bit mac enviroments.I duped the mac32 file and peeked inside with resedit. Turns out that there are 4 (as I recall) strings in the file. One is the comment which appears in the session selection dialog, one is the path to a startup file and another appears to be the name of the startup file with a dot in front of the name. I replaced the path string with /usr/bin/X11/X ( I have X linked to the actual startup script for X) and replaced the .mac32 string with a .X I named the resulting file X11R4. It appeared as a session choice. I also modified the X startup script (aka X11R4) by pitching the "screenrestore" line. The screenrestore command evidently just draws the old A/UX console screen. Well it worked! If I want I can now boot X cleanly from the login screen. ******************* End of X Discussion **************************** I have an eight meg mac, NBUF turns out to have a value of 337, buffer size appears to have doubled since the last release, I was running with NBUF=1000 in release 1.1, should NBUF be increased? Anyone out there have any tuning suggestions? I am a little irritated that CommandShell windows doesn't transmit vt100 escape sequences for the keypad. I have to spend a lot of time editing on a VAX/VMS system. TextEditor takes over 30 seconds to start! What's going on? It would be nice if I could save/modify TextEditor defaults I can't RTFM cuz I don't have the money to get the M. Jerry ----------------------------------------------------------------------------- | Jerry LeVan | Phone:(606)-622-1931 | | Department of Computer Science | | | Eastern Kentucky University | Email:matlevan@eku.bitnet | | Richmond Ky 40475 | | |---------------------------------------------------------------------------| | "The series converges so slowly that it actually diverges." | -----------------------------------------------------------------------------