Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!samsung!dali.cs.montana.edu!milton!newton!lgy From: lgy@phys.washington.edu (Laurence G. Yaffe) Newsgroups: comp.sys.mips Subject: Re: X11.R4 (& xdm difficulties) Message-ID: Date: 8 Dec 90 18:43:30 GMT References: <1990Dec06.193150.10328@groucho> Sender: news@milton.u.washington.edu Lines: 41 pbickers@groucho writes: >I'd like to hear from anybody who has ported this to a Mips box. Was it >trivial/easy/hard? Any peculiar problems merging it with what Mips have >done with the R3 directory structure and configuration files? >When can we expect a release of RISCwindows based on the R4 release. >I'm running R4 on some NCD16s and the mips supplied xdm is just one of several >reasons that I'd like to upgrade everything to R4. Hear, hear. Does someone have the R4 version of xdm running under RISC/os (and would be willing to make the binary available)? While I can wait for most of R4, I'd really like to get a version of xdm that talks to my X-terminals better. On a related note, the latest MIPS distributed version of xdm (3.21) has the nasty habit of entirely giving up on a display for which its initial attempt to XOpenDisplay fails (contrary to the behavior implied by the man pages). This is an issue every time someone turns off an X-terminal (or for people sometimes running X-servers on PCs, everytime they temporarily switch back to DOSland). After running some tests with xdm's debug level turned up, it appears that the problem relates to the message "Display exited with UNMANAGE_DISPLAY" which appears after xdm's first attempt to open the display times out. (In contrast, when an xdm session terminates normally, with debugging on, one sees a message "Display exited with REMANAGE_DISPLAY".) This seems to be new behavior (over the 3.11 version of xdm). Previously, running a fake local "server" like Xfake or xping was sufficient to cope with almost all instances an X-terminal only sporadically coming "live". Has anyone else seen this problem? Or found a better way to cope with it? As I understand it, the R4 xdm should fix these problems when dealing with NCD X-terminals (or other servers which broadcast XDMCP messages), but for dealing with various lousy PC X-servers, is there any way to tell the R4 xdm "never give up on this display even though it can't properly talk XDMCP"? -- -------------------------------------------------------------------------- Laurence G. Yaffe Internet: lgy@newton.phys.washington.edu University of Washington Bitnet: yaffe@uwaphast.bitnet