From: Digestifier To: Linux-Development@senator-bedfellow.mit.edu Reply-To: Linux-Development@senator-bedfellow.mit.edu Date: Thu, 7 Oct 93 19:13:18 EDT Subject: Linux-Development Digest #147 Linux-Development Digest #147, Volume #1 Thu, 7 Oct 93 19:13:18 EDT Contents: Is a memory map for Linux available? (Louis J. LaBash Jr.) How to program the Trident 8900CL? (Dan Everhart) Re: CFC/CFI: XSysadmin (Michael Will) Re: linux scheduler alternatives??? - MY IDEA (Eric Youngdale) Re: CFC/CFI: XSysadmin (Michael Will) Re: CFC/CFI: XSysadmin (Alan Cox) Re: CFC/CFI: XSysadmin (Warner Losh) g++ compile times and #pragma interface (Warner Losh) Re: CFC/CFI: XSysadmin (Fergus James HENDERSON) Re: Elite16 software setup details? (Adrian Ho) Shared Memory and Include file questions (Andy Reynolds) Use of gdb for kernel debugging: an idea (Frank Lofaro) Re: Xfree vs. BIOS (Kelly Murray) debuggers & core inspectors (Yousuf Khan) 586 -- Linux on Pentium (Complementos de Robotica) Re: Xfree vs. BIOS (Steve Tinney) GCC and structure field alignment (Jawaid Bazyar) Re: Page fault (Howlin' Bob) Shm extension to X (Michael Stokes) ---------------------------------------------------------------------------- From: lou@minuet.siue.edu (Louis J. LaBash Jr. ) Subject: Is a memory map for Linux available? Date: 7 Oct 1993 11:40:26 -0400 Reply-To: lou@minuet.siue.edu (Louis J. LaBash Jr. ) Hi, I'm looking for Linux's memory map. Thanks Louis-ljl- ------------------------------ Crossposted-To: comp.windows.x.i386unix From: dan@dyndata.uucp (Dan Everhart) Subject: How to program the Trident 8900CL? Reply-To: dan@dyndata.com Date: Tue, 5 Oct 1993 21:32:42 GMT I need to find out the programming details of the interface to the Trident 8900CL video board. See, I'm trying to interface the card to an old monitor that maxes out at 62 Hz vertical - and the card comes up at 70 Hz. All I want to do is to write a little boot-time utility to click the dot-clock down one notch, which should solve the problem. Or maybe there are other, simpler solutions I haven't thought of? If somebody can either provide me with this info, or point me at it (source files are fine) I would appreciate it. Please respond in email - probably not a subject of wide interest. -- _ / \_ Dan Everhart dan@dyndata.com \_/ \____________________________ 206-743-6982, 742-8604 (fax) / \_/ 7107 179th St SW \_/ Dyndata Engineering Edmonds, WA 98026, USA "C++ is the Swiss Army knife of programming languages." --Frederick Kuhl ------------------------------ From: will@linus (Michael Will) Crossposted-To: comp.os.linux,comp.os.linux.admin Subject: Re: CFC/CFI: XSysadmin Date: 7 Oct 1993 15:30:04 GMT Reply-To: will@peanuts.informatik.uni-tuebingen.de Michael A. Irons (mirons@icarus.ci.net) wrote: : I have been thinking about just this thing. I would use OB/OI : as it seems fairly easy to use and it would be good to show some : support for the product. Exactly my intent - besides creating a nice tool and lerning a lot about ObjectorientedProgramming(tm). : Each of the sections could be written and then merged into a : main utility. If it's designed carfully, if should be extendable as : well. As for the compile time stuff. Once the gui is set, it could be : compiled into a *.o files and then linked with the code for : maintaining the files. Yes, but there is still one major problem. We have to separate the functionality from the user-interface, because we cannot use X alone. To many users do not use X because of lack of memory - these are most likely to be happy to use such a tool like we want to write. I think we have to provide a curses-interface as well - how this can be done we should discuss. My thinking is like: use virtual-functions (late binding) to have the objects communicate with the user with the apropriate methods for the selected user-interface. ParcPlace have done this with Motif/Openlook - why should we not add a curses-layer? But we have to avoid blowing up our task to unsolvable complexity, this is what makes me uneasy :) : I think it would be good to have it very flexable. For : example, with UUCP automatically figuring out which type of config : files you are using and offering a conversion option to one of the : others. That way the user is offered a list of systems, they pick a : system (or add a new one) and get options to either clear/delete it, : or edit it's 'capabilities'. Yes, this is what I wanted to do with the usermaintanement-stuff which should use shadow-passwd-files and add the gui-fields if shadow is detected - but as well work without and not bother about shadow-features. Maybe we should create a mail-channel for this topic? If enough users and developers are interested... Please let me know... Cheers, Michael Will ------------------------------ From: eric@tantalus.nrl.navy.mil (Eric Youngdale) Subject: Re: linux scheduler alternatives??? - MY IDEA Date: Thu, 7 Oct 1993 15:52:52 GMT In article <28v68g$8kr@penchiss10.ee.pdx.edu> bairds@penchiss10.ee.pdx.edu (Scarrow) writes: >ken@PSUEDVAX.PSU.EDU (Kenneth J. Hoover) writes: >> Can anyone who knows about/uses Windoze NoT send me email to let me know if >>WNT does this? NT uses a *lot* of VMSisms. > > VMS > + 111 > ----- > WNT > >Scary, huh? WNT + 111 ----- XOU Even scarier :-). The next time people discuss a new name for linux, perhaps XOU should be considered :-). -Eric -- "When Gregor Samsa woke up one morning from unsettling dreams, he found himself changed in his bed into a lawyer." ------------------------------ From: will@linus (Michael Will) Crossposted-To: comp.os.linux,comp.os.linux.admin Subject: Re: CFC/CFI: XSysadmin Date: 7 Oct 1993 16:19:39 GMT Reply-To: will@peanuts.informatik.uni-tuebingen.de Mhm, now that I read just another load of follow-ups, I think it would be good to create a language as someone proposed. This would describe screens, and actions, acting on curses and X in the same script... The advantage is clear: many users can contribute their configuration- scripts and do not have to bother to much with the userinterface. Only - this I think I cannot do - it is to big a task. If someone can do such a great thing, I think I would retreat to using the language creating some scripts - if not, I think we should stick to the OI/uib-stuff. While xview is nice, I think OI is much more real ObjectOriented - and it is more easy to provide two userinterfaces using the same functionalitycode with true objects... Cheers, Michael Will - away for a week - sorry. ------------------------------ From: iiitac@swan.pyr (Alan Cox) Subject: Re: CFC/CFI: XSysadmin Date: Thu, 7 Oct 1993 16:25:15 GMT In article zmbenhal@netcom.com (Zeyd M. Ben-Halim) writes: > >I know it is not really xview's fault, but any program that includes a lot >of initialized structures (bitmaps, cursors) will spend forever while gcc >handles it. I compiled xcalentool and it took the better part of the night. >Don'y get me wrong, I LIKE xview. I can't stand all that psuedo-oop Xt stuff. >For development work, I guess one needs 8MB to feel comfortable. The binaries >run ok with 4MB. It's a gcc 'feature'. 2.4.5 is a bit better about it. Since you should be loading bitmaps and cursors not compiling them in I think it serves you right 8-) > >I'm not that familiar with the internals of tcl/tk but it is a very good >way of quick prototyping, and looks pretty good. Trying to build secure setuid tcl/tk programs is not fun nor do I consider it viable. I may be wrong however. >It is true that not all of /usr/openwin is required for running binaries. Just libxview and libolgx (about 1.5Mb total) > Alan ------------------------------ From: imp@boulder.parcplace.com (Warner Losh) Subject: Re: CFC/CFI: XSysadmin Date: Tue, 5 Oct 1993 19:18:05 GMT In article zmbenhal@netcom.com (Zeyd M. Ben-Halim) writes: >The problem with OI is speed (or the lack thereof) even with 8MB. OI/OB is a little slow with 8M of memory, granted, but there are some simple things that can be done to improve compile times. First, you can compile things -O rather than -g. Odd as it sounds, this causes much smaller binaries to be produced and is a net win in terms of time. Second, if you just include those .H files from the classes you need it will also improve compile time, since cpp will spend less time spinning its wheels. Finally, if you run fewer, smaller X things it will be faster because g++ tends to like large images. One problem is that uib generates files that include all of OI (OI/oi.H) by default. If you hack these, it takes quite a bit less time to compile the stuff. Many times you don't even need to bother to recompile anything. If it is an interface change, nine times out of ten, you just save the .cf file and you are ready to roll. Sounds like a big win to me. I'm in the process of uploading oimem.tar.gz to tsx-11 packages/OI/contrib right now to show what I mean. This is a quick and dirty memory monitor that I wrote to show how quickly you can turn these things out (my brother asked for it one morning, and he had it by that afternoon). I've found this entire discussion very ironic, since the first thing that I did when I started at Solbourne in the OI group was to start work on some system administration tools. I'd love to see what you folks have in mind. If someone could drop me a .cf file, I'd be happy to take a look at it and give my feedback. Warner -- Warner Losh imp@boulder.parcplace.COM ParcPlace Boulder I've almost finished my brute force solution to subtlety. ------------------------------ From: imp@boulder.parcplace.com (Warner Losh) Subject: g++ compile times and #pragma interface Date: Tue, 5 Oct 1993 19:45:15 GMT I'm looking for ways to reduce the compile times of the OI library when using g++ on Linux. I noticed the #pragma interface in the g++ man page and was wondering if that would help. Have people out there tried the #pragma interface/#pragma implementation stuff with g++ and Linux? How much faster are compiles? How much smaller are binaries? Is it generally worth the effort? What sort of % decreases in size and speed of compile should I expect? Warner -- Warner Losh imp@boulder.parcplace.COM ParcPlace Boulder I've almost finished my brute force solution to subtlety. ------------------------------ From: fjh@munta.cs.mu.OZ.AU (Fergus James HENDERSON) Subject: Re: CFC/CFI: XSysadmin Date: Thu, 7 Oct 1993 18:17:30 GMT iiitac@swan.pyr (Alan Cox) writes: >zmbenhal@netcom.com (Zeyd M. Ben-Halim) writes: > >>How about tcl/Tk? > >What about the security issues. Setuid functions within a tcl/tk program >are a bit tricky. It's for things just like this that I wrote a patch for Linux to implement secure setuid shell scripts. The `tclsh' shell doesn't appear to have any security problems (although I've only read the man page, not the source), so secure tcl programs don't sound too hard to me. [I don't have enough experience with tk to comment about the security of tk programs.] -- Fergus Henderson fjh@munta.cs.mu.OZ.AU ------------------------------ Crossposted-To: comp.os.linux.admin From: scornd9@solomon.technet.sg (Adrian Ho) Subject: Re: Elite16 software setup details? Date: Thu, 7 Oct 1993 15:28:04 GMT I asked: : Does anyone have the details of software config for the Elite 16? : Better still, has anyone written a config program under Linux? Thanks to jjctc@chevron.com (James C. Tsiao) for pointing out that SLS comes with the program "wdsetup", which emulates SMC's own setup program. If you can't find it, it's also available at: sunsite.unc.edu:/pub/Linux/system/Network/drivers/wdsetup-0.6.tar.z -- - Adrian Ho scornd9@solomon.technet.sg ------------------------------ From: Andy Reynolds Subject: Shared Memory and Include file questions Date: 7 Oct 1993 11:58:15 -0700 Reply-To: reynolds@biner.horizon.com We are running Linux SLS 1.03 on a 486 PC and are having the following problems: 1) Shared memory in linux Does anyone know if shared memory works under linux? I am making calls to shmget(2) and get errno=13 which reads "Permission denied". I do not have the "normal" shared memory tools (eg ipcs) available under SunOS which would help me out if it is a permission thingy. 2) Undefines in /usr/include/bsd include files I am including a file /usr/include/bsd/sgtty.h, and it has the define.. #define ioctl bsd_ioctl However the compiler complains that bsd_ioctl is an undefined symbol. I cannot find where it is defined anywhere in /usr/include. In fact, the ONLY reference to it is in /usr/include/bsd/sgtty.h (see above). Any help on the above problems would be greatly appreciated. Thanks. Andy Reynolds reynolds@biner.horizon.com ------------------------------ From: ftlofaro@unlv.edu (Frank Lofaro) Subject: Use of gdb for kernel debugging: an idea Date: Thu, 7 Oct 93 19:28:33 GMT I had posted before that I have used: gdb /usr/src/linux/tools/system /dev/core on a kernel compiled with -g in order to examine kernel variables. However it couldn't modify them; i.e. if I did "set foo=5" it would say "Cannot access memory at 0xbar", where bar equals &foo. Now could one patch gdb to instead of doing a direct write, instead lseek on /dev/kmem (is this right, or should one use /dev/mem), and do a write(fd, &value, sizeof(value)), where value is the value being assigned to the variable? Would this be really hard? ------------------------------ From: kem@prl.ufl.edu (Kelly Murray) Crossposted-To: comp.os.linux.help,comp.os.linux.misc Subject: Re: Xfree vs. BIOS Date: 7 Oct 1993 19:45:46 GMT In article , "Alex R.N. Wetmore" writes: |> Excerpts from netnews.comp.os.linux.development: 6-Oct-93 Re: Xfree vs. |> BIOS by Jerry Whelan@camelot.bra |> > What the first author is suggesting would actually have no |> > portability impact on XFree86. He is asking for a utility that |> > runs during boot that sucks out the video timing values from the |> > bios. That information could easily be written to in an Xconfig |> > format file. That file can then be used by an XFree86 server with |> > absolutely no modifications to the server source. |> |> This is what OS/2 actually seems to do. Is has you run a command called |> "svga on" in a dos window, where it traps the writes to ports and keeps |> track of what it did to go in and out of all of the vesa modes. It then |> uses this information to set its svga modes itself. |> |> Of course I may be totally wrong (I just figured this out by playing |> with it), but it seems to be what it is doing. |> |> This would allow someone to write a driver that would read similar files |> and output the same commands to the same ports to change video modes. |> Might even be a way to pick up ideas on new cards. |> Hmm, maybe you should try using dosemu to run the DOS setmode program before starting up X? A kludge, but if it works.. -Kelly Murray ------------------------------ From: ykhan@gandalf.ca (Yousuf Khan) Subject: debuggers & core inspectors Date: Thu, 7 Oct 1993 19:35:11 GMT Hi, what sorts of debuggers are available for Linux, besides gdb? Anything that is perhaps menu-driven, perhaps Xwindows-based? If not, no sweat. Are there any docs besides the man-pages and internal help of gdb, on how to use gdb, something like a FAQ? What sorts of tools are there for inspecting the insides of core files. Anything that reads the core file through, and prints out a summary report of any kind about what this core file is and what it was doing when it got created? Please send e-mail, the newsfeed around here is four days behind current on average. Besides, it's hard to follow every newsgroup, even if the feed were uptodate. Yousuf Khan -- Gandalf Technologies Inc. Creators of the ultimate network segment technologies in the whole south-western quadrant of the Virgo Galactic Supercluster. Yousuf J. Khan, Automated LAN testing systems designer. GTI would like to dissociate itself from any opinions I may express in the past, present, ------------------------------ From: cr@fct.unl.pt (Complementos de Robotica) Subject: 586 -- Linux on Pentium Date: Thu, 7 Oct 1993 16:34:54 GMT Hi, Is anyone out there running linux on a 586 ?? If you are, Are there any compatibility problems ?? Everything works fine ?? Thanks in advance, Florbela Aires -- Alunos de Comp. de Robotica | BITNET/Internet: cr@fctunl.rccn.pt Universidade Nova de Lisboa | FCT - Dpto. de Informatica | Fax: (+351) (1) 295-5641/4461 2825 Monte Caparica, PORTUGAL | Phone: (+351) (1) 295-4464 ------------------------------ From: sjt@enlil.museum.upenn.edu (Steve Tinney) Crossposted-To: comp.os.linux.help,comp.os.linux.misc Subject: Re: Xfree vs. BIOS Date: 7 Oct 93 19:55:32 GMT Bartosz Blacha (bart+@CMU.EDU) wrote: [stuff to three newsgroups] Please don't post to multiple groups like this. Steve ------------------------------ From: bazyar@mrcnext.cso.uiuc.edu (Jawaid Bazyar) Subject: GCC and structure field alignment Date: 7 Oct 1993 21:20:46 GMT I'm trying to create the following structure: struct foo { unsigned char a; // 1 byte unsigned b : 4; // 4 bits unsigned c : 4; // 4 bits unsigned long d; // 4 bytes } For a total of 6 bytes. Unfortunately, gcc is insisting on aligning the various fields to at least two bytes (a and b/c are being aligned on word boundaries, I believe). This wouldn't be a big deal if THE GCC MANPAGE LISTED THE 80X86 OPTIONS FOR GCC!! I refuse to declare it as a character array because the bloody compiler is supposed to do this kind of thing for me. So, could someone who knows please email me the appropriate option or suggestions? Thanks. -- Jawaid Bazyar | Like UNIX? Like your Apple IIGS? Then ask Procyon, Inc. | me about GNO/ME for the Apple IIgs! bazyar@cs.uiuc.edu | P.O Box 620334 --Apple II Forever!-- | Littleton, CO 80162-0334 (303) 781-3273 ------------------------------ From: gt8134b@prism.gatech.EDU (Howlin' Bob) Subject: Re: Page fault Date: 7 Oct 93 04:33:57 GMT In <1993Oct6.233556.4030@kullmar.se> pand@kullmar.se (Peter Andersson) writes: > I'm currently writing a program that needs the page fault > handler (SIGSYSV) to obtain some speed. But the program needs Okay, that's strike one. SIGSEGV is the "Segmentation Violation" signal, meaning you poked your nose where it doesn't belong. This isn't necessarily a page fault; it could be a literal segment violation (limit or permissions, like writing to CS). If you dereference NULL in a QMAGIC binary, or write to memory outside your brk, *then* it'll be caused by a page fault, but it won't be a page fault handler. Page faults are serviced by the kernel, with no intervention from the user process. > some extra data like where it was trying to write or read and > if it is possible, set the single step trap. If you were in the kernel, you would read CR2. Since you're not, it's not a valid question. -- Robert Sanders Georgia Institute of Technology, Atlanta Georgia, 30332 uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!gt8134b Internet: gt8134b@prism.gatech.edu ------------------------------ From: stokes@Erc.MsState.Edu (Michael Stokes) Subject: Shm extension to X Date: 7 Oct 93 17:06:53 The version I have has it (at least the header data is there in /usr/include/X11/extensions). I haven't tried to link to it, but I bet it works. -- All men have to understand their limitations...C.Eastwood ------------------------------ ** FOR YOUR REFERENCE ** The service address, to which questions about the list itself and requests to be added to or deleted from it should be directed, is: Internet: Linux-Development-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.development) via: Internet: Linux-Development@NEWS-DIGESTS.MIT.EDU Linux may be obtained via one of these FTP sites: nic.funet.fi pub/OS/Linux tsx-11.mit.edu pub/linux sunsite.unc.edu pub/Linux End of Linux-Development Digest ******************************