Subject: Linux-Development Digest #529 From: Digestifier To: Linux-Development@senator-bedfellow.MIT.EDU Reply-To: Linux-Development@senator-bedfellow.MIT.EDU Date: Tue, 8 Mar 94 21:13:06 EST Linux-Development Digest #529, Volume #1 Tue, 8 Mar 94 21:13:06 EST Contents: Re: Specialix driver (Jay Denebeim P025) rarpd for Linux ?? (Cheryl Dematteis) Re: AMD 486DX problem (with Linux?) (Wen-Chun Ni) Re: AMD 486DX problem (with Linux?) (Peter Herweijer) More power saving w/ VGA-cards (Hendrik G. Seliger) Re: [Q]: Tool to make X-windows programs?? (R.C.Van-Den-Bergh) Re: Screensaver w/ power save ? (Olaf Titz) Re: Fortran g77 (Craig Burley) Re: Specialix Driver (Olaf Titz) Re: Amiga FileSystem, Anyone? (Kai Henningsen) Re: AMD 486DX problem (with Linux?) (H. Peter Anvin N9ITP) Re: AMD 486DX problem (with Linux?) (Gregory McKesey) term "through" VM Telnet? (Alex Ramos) Re: Amiga FileSystem, Anyone? (Sami-Pekka Hallikas) Adaptec 1542C SCSI with New ROM (Don Holzworth) Is my HD fix sane? (Kenneth Preslan) ---------------------------------------------------------------------------- Crossposted-To: gnu.misc.discuss From: denebeim@bnr.ca (Jay Denebeim P025) Subject: Re: Specialix driver Date: Mon, 7 Mar 1994 20:08:23 GMT In article <2kjoui$js9@travis.csd.harris.com> donh@gcx1.ssd.csd.harris.com (Don Holzworth) writes: > >I'm not sure some people have read the original post, I read it, and I agree with you. Don't listen to these net.laywers, you've got it right. The code that is downloaded to the card could just as easily be ROM. There's no reason that you'd have to disclose your code. Heck, you could even compile it with a compiler other than GCC and just write a intel hex (or motorolla srecord) file to asm .byte converter. I would think that's probably what I'd do if I was doing what you're doing. That would be the easiest way I can think of to do it. Hey, I guess that means you could GPL the code, eh? After all, it would just be an assembly file with a bunch of .byte statements in it :-) It wouldn't even make it any easier to reverse engineer the code. By the way, your product sounds great, E-Mail me with the cost. I'll be more than happy to help with the code and/or testing. Jay -- Jay Denebeim Address: UUCP: duke!wolves!deepthot!jay Internet: jay@deepthot.cary.nc.us BBS:(919)-233-9937 VOICE:(919)-233-0776 ------------------------------ From: cdematt@ucrengr.ucr.edu (Cheryl Dematteis) Subject: rarpd for Linux ?? Date: 7 Mar 1994 18:45:43 GMT -- Has anyone out there ported rarpd to Linux. We have a couple of old Sun3's that we need to boot off of a Linux machine. Any pointers or suggestions would be greatly appreciated. Please respond via email, I will post a summary of responses Thanks ============================================================================== Cheryl K. DeMatteis cdematt@cs.ucr.edu Graduate Student (909)787-3824 Department of Computer Science University of California Riverside, CA 92521 ============================================================================== ------------------------------ From: wcn@cs.brown.edu (Wen-Chun Ni) Subject: Re: AMD 486DX problem (with Linux?) Date: Tue, 8 Mar 1994 12:54:25 GMT In article hrohlan@gwdu03.gwdg.de (Hans Christoph Rohland ) writes: > >This was not true! >I checked only the manual for the main board and asked the person who installed >the board. Now I looked for the cpu and it is an AMD! >The test program was successful but buggy: >The cast in the final test is double to float and should be float to double! >Down the corrected test program. > >Christoph >-------------------------------- cut here ------------------------------------- Yeah, right. I tested the program under Sparc and the program guesses the Sparc is using an AMD chip ;-) because the test fails. I do believe that the code itself is much more problematic. -- Wen-Chun Ni, wcn@cs.brown.edu =================================================================== "Great spirits have always encountered violent opposition from mediocre minds..." -- Albert Einstein ------------------------------ From: pieterh@sci.kun.nl (Peter Herweijer) Subject: Re: AMD 486DX problem (with Linux?) Date: Tue, 8 Mar 1994 15:11:10 GMT wcn@cs.brown.edu (Wen-Chun Ni) writes: >Yeah, right. I tested the program under Sparc and the program guesses >the Sparc is using an AMD chip ;-) because the test fails. The program told me that the Sun I'm working on also contains an AMD486DX chip. Somehow, I doubt that :^) >I do believe that the code itself is much more problematic. No, that conclusion is _way_ too hasty. The problematic part is the test for precise equality, where a LSB rounding error should be considered acceptable (Golden Rule: Never Test Floats For Exact Equality). For example, the Sun's output 1.312500 * 7.999900 =10.499868 1.312500 * 7.999900 =10.499869 Test Failed, this must be an AMD 486DX chip! is entirely acceptable (difference one LSD: 0.000001), but the AMD output as posted in an earlier article 1.312500 * 7.999900 =10.499990 1.312500 * 7.999900 =10.499869 Test Failed, this must be an AMD 486DX chip! is fatally out of spec (difference 0.000121) and signifies a _very_ bad and _real_ bug. I've appended a more sensible (if still rough) version of the test at the end of this article. Please compile this and run it on your computer (be it Intel, AMD or Cyrix). Peter 'Intel DX2-66 :^)' Herweijer pieterh@sci.kun.nl =====8<=====cut here=====amdtest.c=====cut here=====8<===== /* * AMD Floating Point processor fault demo * v0.01pl3 :^) */ main() { float a=1.3125; float b=7.9999; double x= 1.3125; double y= 7.9999; double diff; printf ("%f * %f =", a, b); a *= b; printf ("%f\n", a); printf ("%f * %f =", x, y); x *= y; printf ("%f\n", x); diff = (double) a - x; if ( diff > 0.000001 || diff < -0.000001 ) { printf("Test Failed, this must be an AMD 486DX chip!\n"); } else { printf("Test succeeded!\n"); } } =====8<=====cut here=====amdtest.c=====cut here=====8<===== ------------------------------ From: hank@Blimp.automat.uni-essen.de (Hendrik G. Seliger) Subject: More power saving w/ VGA-cards Date: 7 Mar 1994 16:18:03 GMT Reply-To: hank@automat.uni-essen.de Hi! So on my last posting I got some ideas on how the VESA support for monitor power saving works. Switch either hsync or vsync off and the thing goes into suspend mode (which will switch the monitor into power save mode after a certain time), switch both syncs off and the beast goes into power save at once. (Of course this needs a VESA conforming monitor with this capability, usually something with a digital control). NOW: I don't have any real documentation on the S3 card (that's the one I have at home) or the ET4000 (that's the ones I have at work, which doesn't interest me too much :-). Is there anybody out there with good infos on what to do do which registers to switch the syncs on and off? It should be pretty easy than to put this code into the kernel's console.c to use this feature under linux. Thanks for any info! Hank -- ====================================================================== Hendrik G. Seliger Universitaet Essen hank@automat.uni-essen.de Schuetzenbahn 70 Tel.: +49-201-183-2898 45117 Essen, Germany ====================================================================== "Handling interrupts is simple." (G. Pajari) "Interrupts are an unpleasant fact of life." (A. Tanenbaum) ------------------------------ From: rcv@ukc.ac.uk (R.C.Van-Den-Bergh) Subject: Re: [Q]: Tool to make X-windows programs?? Date: Mon, 07 Mar 94 18:44:03 GMT Christian Moen (christim@ifi.uio.no) wrote: =>Please take a look at ParcPlace's ObjectBuilder and ObjectInterface =>available for free on Linux. Binaries are available from sunsite, =>tsx-11 and funet. Also take a look at InterViews and especially the ibuild tool. I uploaded a copy on Saturday to sunsite. Cedric -- Zhaumer, High Priest of Amalgaer, the dwarven God of Something. ------------------------------ From: uknf@rzstud1.rz.uni-karlsruhe.de (Olaf Titz) Subject: Re: Screensaver w/ power save ? Date: 8 Mar 1994 16:56:16 GMT In article <2lg816$5as@smurf.noris.de>, Matthias Urlichs wrote: > > A kernel config variable would be good IMHO because I don't know what > > ordinary monitors do without sync pulses (snow?). > Ordinary monitors should be blank in that case. After all, no sync pulses > are not distinguishable from a monitor that isn't plugged into the video > card in the first place. This is the reason why some monitors don't go blank if the sync is missing, but light the whole screen half-bright instead. That is a good feature to detect cable or plug problems, or a computer that is turned off with the monitor still running. Olaf -- olaf titz o olaf@bigred.ka.sub.org praetorius@irc comp.sc.student _>\ _ s_titz@ira.uka.de LINUX - the choice karlsruhe germany (_)<(_) uknf@dkauni2.bitnet of a GNU generation what good is a photograph of you? everytime i look at it it makes me feel blue ------------------------------ From: burley@apple-gunkies.gnu.ai.mit.edu (Craig Burley) Subject: Re: Fortran g77 Date: 8 Mar 94 12:02:52 How to get information on GNU Fortran (the .plan file for user fortran at Project GNU): If you can do finger -l, try: finger -l fortran@gate.gnu.ai.mit.edu If that produces output of around 250 lines, then it probably works. Otherwise, if you can do ftp, try: ftp gate.gnu.ai.mit.edu Username: ftp Password: (any) get g77.plan The ftp copy of the file is updated every 3 hours from the live copy. NOTE: Try gate-1.gnu.ai.mit.edu, gate-2.gnu.ai.mit.edu, or gate-3.gnu.ai.mit.edu if you are having problems with gate.gnu.ai.mit.edu. Otherwise, you could try sending email to user fortran@gnu.ai.mit.edu asking to be sent the latest copy of the file, but _please_ make this your last option, because it requires human intervention! -- James Craig Burley, Software Craftsperson burley@gnu.ai.mit.edu Member of the League for Programming Freedom (LPF) lpf@uunet.uu.net ------------------------------ From: uknf@rzstud1.rz.uni-karlsruhe.de (Olaf Titz) Crossposted-To: gnu.misc.discuss Subject: Re: Specialix Driver Date: 8 Mar 1994 17:23:22 GMT They make an extension board with a CPU on it and develop code to run on that CPU. That code is proprietary. Someone who wants that extension supported under Linux writes a Linux driver that can upload the proprietary code to the extension. The uploader program becomes part of the kernel and is GPLed, while the actual upload code - provided as a separate binary file, and not dependent on hardware or OS types - remains proprietary. Also GPLed is the part of the kernel device driver that does operational data transfers to the card based on a published interface. What the hell should be wrong with that? Suppose I have an EPROM burner with software that is GPLed. I use this burner, driven by GPL software, to load the latest update of ZyXEL modem operation software gotten from their FTP server onto two chips which I insert in the modem to operate with the stock Linux 16550 driver (GPLed). Now the chips obviously still carry proprietary and copyrighted software. What should be wrong with this? Should someone complain about ZyXEL in that case (for which reason)? Olaf -- olaf titz o olaf@bigred.ka.sub.org praetorius@irc comp.sc.student _>\ _ s_titz@ira.uka.de LINUX - the choice karlsruhe germany (_)<(_) uknf@dkauni2.bitnet of a GNU generation what good is a photograph of you? everytime i look at it it makes me feel blue ------------------------------ Date: 07 Mar 1994 19:22:00 +0100 From: kai@khms.westfalen.de (Kai Henningsen) Subject: Re: Amiga FileSystem, Anyone? bdwheele@silver.ucs.indiana.edu wrote on 06.03.94 in : > Um.....if the mac doesn't use a 'standard format' how is it that I can > read mac (1.44) disks in my pc? Granted, I had to use a shareware reader The Mac *didn't* - now he does. That's why you can read 1.44M but not 800K disks. Kai -- Internet: kh@ms.maus.de, kai@khms.westfalen.de Bang: major_backbone!{ms.maus.de!kh,khms.westfalen.de!kai} ## CrossPoint v2.93 ## ------------------------------ From: hpa@ahab.eecs.nwu.edu (H. Peter Anvin N9ITP) Subject: Re: AMD 486DX problem (with Linux?) Reply-To: hpa@nwu.edu (H. Peter Anvin) Date: Tue, 8 Mar 1994 21:54:24 GMT In article of comp.os.linux.development, hrohlan@gwdu03.gwdg.de (Hans Christoph Rohland ) writes: > I wrote: > : This is obviously no AMD Problem! I get a similar error with an > : Intel 486DX66 > : (Linux 0.99.15 from Slack 1.1.2). The testprogram for AMD works fine but > > This was not true! > I checked only the manual for the main board and asked the person > who installed > the board. Now I looked for the cpu and it is an AMD! > The test program was successful but buggy: > The cast in the final test is double to float and should be float to double! > Down the corrected test program. > NOT!! This "corrected" test program compares a result computed at float precision with one computed at double precision, at double precision!! It fails on my Intel 486DX/33, whereas the original one doesn't. /hpa -- INTERNET: hpa@nwu.edu FINGER/TALK: hpa@ahab.eecs.nwu.edu IBM MAIL: I0050052 at IBMMAIL HAM RADIO: N9ITP or SM4TKN FIDONET: 1:115/511 or 1:115/512 STORMNET: 181:294/101 Purgamentum init, exit purgamentum. ------------------------------ From: mckesey@imaphics.prior.com (Gregory McKesey) Subject: Re: AMD 486DX problem (with Linux?) Date: 08 Mar 1994 21:55:25 GMT >>>>> "Wen-Chun" == Wen-Chun Ni writes: Wen-Chun> In article Wen-Chun> hrohlan@gwdu03.gwdg.de (Hans Christoph Rohland ) writes: >> This was not true! I checked only the manual for the main >> board and asked the person who installed the board. Now I >> looked for the cpu and it is an AMD! The test program was >> successful but buggy: The cast in the final test is double to >> float and should be float to double! Down the corrected test >> program. >> >> Christoph -------------------------------- cut here >> ------------------------------------- Wen-Chun> Wen-Chun> Yeah, right. I tested the program under Sparc and the Wen-Chun> program guesses the Sparc is using an AMD chip ;-) Wen-Chun> because the test fails. Wen-Chun> I do believe that the code itself is much more Wen-Chun> problematic. You may be right, but the test was not really to test for equality. The test was thrown in as an after thought. I certainly did not mean to have the program run on anything other that 486's. The purpose of program was simply to multiply two numbers and show that my amd486 did not compute the correct value for the float calculation. IE. the output: 1.312500 * 7.999900 =10.499990 1.312500 * 7.999900 =10.499869 is really what concerns me. 1.312500 * 7.999900 does not equal 10.499990 no matter what kind of rounding rules you may use. The rest of the program and output is irrelevant. If boot my system with the no387 option or recompile the program with the msoft-float option, the program gives a more accurate result. Greg. -- ____________________________________________________________________ Gregory McKesey (Software Manager) Gallium Software Inc. Tel: (613)721-0902 ext (431) 303 Moodie Dr., Suite 4000 Fax: (613)721-1278 Nepean, Ontario, Canada. gmckesey@gallium.com K2H-9R4 ==================================================================== ------------------------------ From: ramos@engr.latech.edu (Alex Ramos) Subject: term "through" VM Telnet? Date: 8 Mar 1994 02:23:26 GMT I plan to hack term114 enough to make it work through a VM Telnet (Linux--Modem--Mainframe--Telnet--SunOS) connection. Has anybody done this yet? It shouldn't take more than a couple of hours (famous last words :])... and I wouldn't like to duplicate efforts. Thanks, -- Alex Ramos * This message is copyrighted material! Louisiana Tech University BSEE/Sr * All rights reserved. No warranty, etc ------------------------------ From: semi@dream.nullnet.fi (Sami-Pekka Hallikas) Subject: Re: Amiga FileSystem, Anyone? Date: Mon, 7 Mar 1994 13:13:38 GMT Alan Braggins (armb@setanta.demon.co.uk) wrote: >> Does anyone developing WORKING Amiga filesystem, that you can read amiga >> disks with you Linux machine. I really like to get this FS if anyone is > Most PC disk drives won't read Amiga disks - I doubt an Amiga file Somehow I find that out allready ;-). > "How do I read Amiga disks on a PC? - You don't, use an MS-DOS > filesystem on the Amiga" is definitely an Amiga FAQ. > I'm assuming you are talking about floppies here, not transferring Yeps.. If I have amiga HD, I think I don't have these problems.. > hard disks, or reading them under the Amiga Linux port. Not possible, Because I don't even have Amiga... I just have few amiga floppies... -- +--------------------------+----------+-------------------------------------+ | semi@dream.nullnet.fi | OH1KYL | MAIL MEDIA. Do Not Expose to Flame! | | samip@freeport.uwasa.fi +----------+-------------------------------------| | semi@freenet.hut.fi | Dream World BBS * 358-21-4389843 * 24H * 9600 | +--------------------------+------------------------------------------------+ ------------------------------ From: donh@travis.csd.harris.com (Don Holzworth) Subject: Adaptec 1542C SCSI with New ROM Date: 8 Mar 1994 20:23:21 GMT Reply-To: donh@travis.csd.harris.com (Don Holzworth) Last year there was a post about the (then) current driver not working with the new ROM (which permits more than one meg disks to be used). A friend at work tried to install Slackware 1.1.2 (gotten from ftp.cdrom.com on Feb. 20) and got the following: configuring adaptec @ io: 330, irq 11, dma priority 5 aha1542_out failed (1): aha1542.c: interrupt received, but no mail. aha1542.c: interrupt received but no mail. scsi0 : adaptec 1542 scai : 1 hosts aha1542.c: interrupt received, but no mail. And, of course couldn't load. He's removed all adapters from his system except a TSENG video card, his adaptec 1542C SCSI card, a serial card and the memory. His system is a 486DX/33. Is there a new version of the adaptec SCSI driver for Linux that supports the greater than one meg ROM? Thanks for information. Regards, ============================================================================== donh@travis.csd.harris.com | Don Holzworth All opinions are mine alone. | (305) 977-5563 | "Efficiency is doing things right. Effectiveness is doing the right thing." ============================================================================== ------------------------------ From: preslan@ishmael.ee.tulane.edu (Kenneth Preslan) Subject: Is my HD fix sane? Date: 8 Mar 1994 07:00:28 GMT The Linux kernel (0.99.13) has problems with the hard drive in my new computer. Here's what it prints out when it boots: hd.c: ST-506 interface disk with more than 16 heads detected, probably due to non-standard sector translation. Giving up. (disk 0: cyl=524, sect=63, head=32) Partition check: harddisk I/O error dev 0300, sector 0 unable to read partition table of device 0300 The docs that came with the computer (it's a Gateway 2000 P5-66) tell me the hard drive looks like: cyl=1048, head=16, and sect=63. Can I fix my hard drive problem by telling the kernel what type of hard drive I actually have? I changed the file /usr/src/linux/kernel/blk_drv/hd.c to look like: ... i = NR_HD; while (i-- > 0) { hd[i<<6].nr_sects = 0; /************ This is my HD fix. ***************/ if (i==0) { hd_info[i].head=16; hd_info[i].cyl=1048; } /*************************************************/ if (hd_info[i].head > 16) { printk("hd.c: ST-506 interface disk with more than 16 heads detected,\n"); printk(" probably due to non-standard sector translation. Giving up.\n"); printk(" (disk %d: cyl=%d, sect=%d, head=%d)\n", i, hd_info[i].cyl, hd_info[i].sect, hd_info[i].head); ... This change has stopped the kernel from complaining on boot-up and has let me mount my hard drive. My questions are: Is this fix safe? Will it cause other problems for me later on? Will I be able to read and write my whole hard drive? Thanks, Kenneth Preslan preslan@ee.tulane.edu ------------------------------ ** 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 ******************************