From: Digestifier To: Linux-Development@senator-bedfellow.mit.edu Reply-To: Linux-Development@senator-bedfellow.mit.edu Date: Thu, 31 Mar 94 20:13:07 EST Subject: Linux-Development Digest #593 Linux-Development Digest #593, Volume #1 Thu, 31 Mar 94 20:13:07 EST Contents: Re: profiling anyone? (Bas de Bakker) Re: Bug in atof (Kai Arstila) Re: Hardware developing X/Motif (Alan Cox) Re: BusLogic BT445S driver? (Victoria Harrington) Re: BusLogic BT445S driver? (J. den Ouden) Re: Bug in atof (Rob Janssen) Re: Announcement of Linux/68k Version 0.08 (Uytterhoeven Geert) pcnfsd and long shadow pwds (Hendrik G. Seliger) Re: TCP/IP-Bug in 1.0 Kernel? (Alan Cox) pdcurses on linux (Bryan Zarnett) Re: BusLogic BT445S driver? (Drew Eckhardt) Re: IDE Performance Package (Mark Lord) Re: IDE Performance Package (Mark Lord) Re: Announcement of Linux/68k Version 0.08 (Hamish Macdonald) Are ADDTRON AE-200LC/JL network cards supported? (Bob Myers) IDE transfer and interrupts, again (Olaf Titz) Re: IDE Performance Package (Daniel Aaron Supernaw-Issen) ---------------------------------------------------------------------------- Subject: Re: profiling anyone? From: bas@phys.uva.nl (Bas de Bakker) Date: 31 Mar 1994 08:25:58 GMT Ivan writes: > !As to profiling tools: gprof is in the binutils package > Yes, except that the output is function count based ... there is no > timing information. That is probably because 'profil(2)' isn't > implemented. 'profil(3)' obviously isn't doing the "same thing", > otherwise gprof would've provided me with timing data. gprof gives me lots of timing data, so I would guess something is wrong with your installation. Due to changes in the signal stack (where the signal handler installed by profil() gets the program counter from) there exist incompatible combinations of kernel/libc which break timing data from gprof. Perhaps you should update either your kernel or (more likely) your C library. Bas. ------------------------------ From: arstila@kl-linux.helsinki.fi (Kai Arstila) Subject: Re: Bug in atof Date: 31 Mar 1994 11:07:49 GMT Sergio Ramirez (sergio@bianca.amd.com) wrote: > I found a bug in the subroutine atof() > the offending code is the following > double mortrate_pc; > while ((c = getopt(argc, argv, "i:f:b:p:m:c:s:t:x:")) != -1) > switch (c) { > ... > case 'm': > mortrate_pc = atof(optarg); > break; > ... > When the program is called with the option -m 6 or any integer > mortrate_pc gets the correct value of 6. > When the program is called with the option -m 7.5 the variable > mortrate_pc gets the correct value of 1. Do you have the line #include in your code. I have had similar action many times, and I always find the same bug in my program. Kai Arstila Universtity of Helsinki arstila@kruuna.helsinki.fi ------------------------------ Crossposted-To: comp.os.linux,comp.unix.questions From: iiitac@uk.ac.swan.pyr (Alan Cox) Subject: Re: Hardware developing X/Motif Date: Wed, 30 Mar 1994 09:35:42 GMT In article <1994Mar29.101553.262@elmrd6.ineab.ikea.se> tsva@elmrd6.ineab.ikea.se (Tom Svaleklev) writes: >Hello there ! > >I'm about to start my own firm consulting and teaching X windows/Motif, >Unix and C. I looking for opinions/experiences on decent hardware >requirements needen to run and develop X/Motif/Unix for personal use. >I do own a PC 486/66, 4Mb/200Mb (IDE) and have thoughts on upgrading it >and install Linux, FreeBSD or something like that. More memory - 8 or for serious work 16Mb, maybe a larger disk. With either Linux or *BSD you'll need to buy Motif (its not free). Several vendors sell it for Linux > >But since I also want DOS/Windows I understand that I have to buy >another disk for my Unix system. Also it's been said that PC's can't You can have both on one disk. OS/2 comes with its own boot selector, otherwise Linux comes with Lilo and at the worst you just switch the active partition around >what system to run. I also believe I need at least 16Mb of memory >to cope with Unix. I do pretty serious development in 8Mb. For serious C++ work you want 16Mb (thats more a reflection on gnu C++ liking to waste memory...) Alan ------------------------------ From: vicki@eskimo.com (Victoria Harrington) Subject: Re: BusLogic BT445S driver? Date: Wed, 30 Mar 1994 13:18:42 GMT pyeatt@CS.ColoState.EDU (Larry Pyeatt) writes: >I have a BusLogic BT445S fast SCSI adapter. Is there a driver for >this board, or should I start writing my own. Any pointers would be >appreciated. I have such a board on order also. There are drivers available for in at tsx-11.mit.edu:/pub/linux/drivers/ALPHA (I think). Check the ls-lR for the exact location. However, ONLY the drivers are provided (buslogic.tar.gz = buslogic.c+buslogic.h). Installation instructions are most notable by their absence. To anyone who has installed this: guidence is requested Steve harringt@netcom.com aka: vicki@eskimo.com ------------------------------ From: denouden@inter.NL.net (J. den Ouden) Subject: Re: BusLogic BT445S driver? Date: Thu, 31 Mar 1994 11:35:01 GMT Victoria Harrington (vicki@eskimo.com) wrote: : pyeatt@CS.ColoState.EDU (Larry Pyeatt) writes: : >I have a BusLogic BT445S fast SCSI adapter. Is there a driver for : >this board, or should I start writing my own. Any pointers would be : >appreciated. : I have such a board on order also. There are drivers available for : in at tsx-11.mit.edu:/pub/linux/drivers/ALPHA (I think). Check the : ls-lR for the exact location. However, ONLY the drivers are : provided (buslogic.tar.gz = buslogic.c+buslogic.h). Installation : instructions are most notable by their absence. : To anyone who has installed this: guidence is requested : Steve : harringt@netcom.com : aka: vicki@eskimo.com For my own convenience I have changed buslogic.c and buslogic.h and some other files a bit so that the buslogic driver can be configured in and out with 'make config'. Mail me if you want the diffs. Cheers, Jan den Ouden denouden@inter.nl.net ------------------------------ From: rob@pe1chl.ampr.org (Rob Janssen) Subject: Re: Bug in atof Date: Thu, 31 Mar 1994 06:26:56 GMT Reply-To: pe1chl@rabo.nl In sergio@bianca.amd.com (Sergio Ramirez) writes: >I found a bug in the subroutine atof() >the offending code is the following > double mortrate_pc; > while ((c = getopt(argc, argv, "i:f:b:p:m:c:s:t:x:")) != -1) > switch (c) { > ... > case 'm': > mortrate_pc = atof(optarg); > break; > ... >When the program is called with the option -m 6 or any integer >mortrate_pc gets the correct value of 6. >When the program is called with the option -m 7.5 the variable >mortrate_pc gets the correct value of 1. Did you include ? This is required on Linux, and isn't on some other systems. Rob -- ========================================================================= | Rob Janssen | AMPRnet: rob@pe1chl.ampr.org | | e-mail: pe1chl@rabo.nl | AX.25 BBS: PE1CHL@PI8UTR.#UTR.NLD.EU | ========================================================================= ------------------------------ Crossposted-To: comp.unix.amiga From: uytterho@cs.kuleuven.ac.be (Uytterhoeven Geert) Subject: Re: Announcement of Linux/68k Version 0.08 Date: Thu, 31 Mar 1994 09:35:50 GMT In article <1994Mar30.215327.3594@ida.liu.se>, t92johtu@odalix.ida.liu.se (Johan 'tuben' Tufvesson) writes: |> I'm a owner of a A4000/040/10, and have been waiting for a unix for my machine |> (the original IDE-drive has been the problem so far). The first thing I did |> after seeing that Linux had adopted the IDE-drives was to test it out. |> I used the new kernel and the new filesystem, and had no luck. Many of the |> programs on in the ramdiskimage don't work, for example mkfs and mount. |> Because of this I haven't been able to test the driver. Could the reason for |> this problem be the difficulties with 040 copybackcache and shared libraries? |> If so, is there a way around this right now, or will I have to wait for a new |> kernel? |> /Waiting for unix on A4000/IDE... On my A4000/040 with linux 0.08 mount works, in contrast with some other programs (like sync...). I didn't tried mkfs since I already had a MINIX file system on my IDE drive (created under 0.07pl4). Geert -- +--------+----------------------------+------------------------------------+ |4000/040| Geert Uytterhoeven | E-mail: uytterho@cs.kuleuven.ac.be | | _ | Stud Master of Engineering | Fax : +32-16-535823 | | _ // | Dept of Computer Science | Snail : Huysmansstraat 12 | | \X/ | KU Leuven | B-3128 BAAL / BELGIUM | +--------+----------------------------+------------------------------------+ | You don't have to sleep to see nightmares - Anne Clark | | (just take a look at MessyWindows) | +--------------------------------------------------------------------------+ ------------------------------ From: hank@Blimp.automat.uni-essen.de (Hendrik G. Seliger) Subject: pcnfsd and long shadow pwds Date: 29 Mar 1994 08:30:20 GMT Reply-To: hank@automat.uni-essen.de Hi! I encountered a problem with pcnfsd if passwords longer than 8 characters are used (we use the DOUBLE-option from the shadow suite). This is a problem in the comparision of the encrypted passwords. Here comes a patch to get this working: *** pcnfsd.c.orig Mon Sep 20 11:23:27 1993 --- pcnfsd.c Mon Mar 28 14:35:49 1994 *************** *** 362,371 **** --- 362,373 ---- struct auth_args *a; { static struct auth_results r; char username[32]; char password[64]; + char pwd_crypt[64]; + char *pwd_tmp; int c1, c2; struct passwd *p; #ifdef SHADOWPWD struct spwd *spwd; #endif *************** *** 390,401 **** if (p->pw_name && p->pw_passwd[0] == '@') { if (pw_auth(p->pw_passwd+1, username, PW_LOGIN)) return ((char *) &r); } else { ! if (!valid(password, p)) ! return ((char *) &r); } #else c1 = strlen(password); c2 = strlen(p->pw_passwd); if ((c1 && !c2) || (c2 && !c1) || --- 392,422 ---- if (p->pw_name && p->pw_passwd[0] == '@') { if (pw_auth(p->pw_passwd+1, username, PW_LOGIN)) return ((char *) &r); } else { ! ! pwd_tmp= crypt( password, p->pw_passwd ); ! strcpy ( pwd_crypt, pwd_tmp); ! ! if (strlen (password) > 8) { ! pwd_tmp = crypt (password + 8, p->pw_passwd); ! strcat (pwd_crypt, pwd_tmp + 2); ! } ! #ifdef DEBUG ! if (buggit){ ! fprintf(stderr, "AUTHPROC crypt=%s\r\n",pwd_crypt); ! fprintf(stderr, "AUTHPROC p->pwpasswd=%s\r\n", p->pw_passwd); ! } ! #endif DEBUG ! ! c1 = strlen(password); ! c2 = strlen(p->pw_passwd); ! if ((c1 && !c2) || (c2 && !c1) || ! (strcmp(p->pw_passwd, pwd_crypt))) { ! return ((char *) &r); ! } } #else c1 = strlen(password); c2 = strlen(p->pw_passwd); if ((c1 && !c2) || (c2 && !c1) || =================== That's it. Ciao, 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: iiitac@uk.ac.swan.pyr (Alan Cox) Subject: Re: TCP/IP-Bug in 1.0 Kernel? Date: Wed, 30 Mar 1994 09:44:52 GMT In article <1994Mar29.075924.2610@pe1chl.ampr.org> pe1chl@rabo.nl writes: >Then I'm afraid this is the first one that is not caused by that... >I have used dip337uri, and have patched it further to solve some problems >with route handling. (I think you have received these patches via the >NET mailing list) >The problem does not seem to be in DIP but in ifconfig. >Or does ifconfig use a new syntax as well? The ifconfig syntax is effectively unchanged. There are two fundamental changes to the system about routing from early pl15 to 1.0 kernels. Firstly the route logic was rewritten by Linus to be readable and logical. Secondly for 1.00-pre the BSD route convention of route add somesite gw localifaceaddr was added to allow BSD type route specification and support the Berkeley routed. Alan ------------------------------ From: bzarnett@jupiter.scs.Ryerson.CA (Bryan Zarnett) Subject: pdcurses on linux Date: 31 Mar 1994 13:22:58 GMT Has anyone tried porting PDCurses to linux? I know its the same as ncurses but..... - bryan ------------------------------ From: drew@kinglear.cs.colorado.edu (Drew Eckhardt) Subject: Re: BusLogic BT445S driver? Date: 30 Mar 1994 19:47:02 GMT In article <1994Mar30.030019.132837@yuma>, Larry Pyeatt wrote: > >I have a BusLogic BT445S fast SCSI adapter. Is there a driver for >this board, or should I start writing my own. Any pointers would be >appreciated. There are two drivers that work with the board, the Adaptec 1540 driver in the standard kernel, and the Buslogic alpha driver available from the usual place (tsx-11.mit.edu:/pub/linux/ALPHA/scsi). You also want to use the cluster patches for improved (perhaps by 100%) throughput. ------------------------------ From: mlord@bnr.ca (Mark Lord) Subject: Re: IDE Performance Package Date: 31 Mar 1994 14:05:03 GMT In article <2ndj10$8gb@levelland.cs.utexas.edu> danielsi@cs.utexas.edu writes: > >I've installed the ide performance package upon linux 1.0 and have found >the following: Whenever I have disk activity, the mouse jumps around under X. The *only* possible way that the performance patches could cause this is if you are using multiple mode *without* allowing the driver to unmask interrupts. On my system and on many others, the exact opposite behaviour is observed when the patches are applied and interrupt unmasking is enabled.. the mouse goes from very unresponsive to more responsive. -- mlord@bnr.ca Mark Lord BNR Ottawa,Canada 613-763-7482 ------------------------------ From: mlord@bnr.ca (Mark Lord) Subject: Re: IDE Performance Package Date: 31 Mar 1994 14:10:11 GMT In article wayne@cse.unl.edu writes: >Just out of curiosity, why are you suspicious of any drive that >supports fewer than 32 sectors? I would think that 16 or even 8 >sectors would go a long way in reducing the overhead, and that you >probably aren't saving much by increasing it to 32. Cutting it down >to 16 (or 8) might also help those serial port overruns. Suspicions are because of problems encountered with non-standard implementations on older drives, and the one common denominator was that all of the troubled drives supported fewer than 32 sectors in multiple mode. However, this restriction is removed in the final version (posted yesterday), which does not enable multiple mode by default for any drive. Instead, a separate program is provided to allow you to tweak the settings on the fly, to find what works best for your system, if anything. On troublesome drives, the error-recovery should still kick in and turn the feature off again if there are problems. Have fun! -- mlord@bnr.ca Mark Lord BNR Ottawa,Canada 613-763-7482 ------------------------------ From: Hamish.Macdonald@bnr.ca (Hamish Macdonald) Crossposted-To: comp.unix.amiga Subject: Re: Announcement of Linux/68k Version 0.08 Date: 31 Mar 1994 14:11:28 GMT In article <1994Mar30.215327.3594@ida.liu.se>, t92johtu@odalix.ida.liu.se (Johan 'tuben' Tufvesson) writes: |> I used the new kernel and the new filesystem, and had no luck. Many |> of the programs on in the ramdiskimage don't work, for example mkfs |> and mount. Because of this I haven't been able to test the |> driver. Could the reason for this problem be the difficulties with |> 040 copybackcache and shared libraries? If so, is there a way |> around this right now, or will I have to wait for a new kernel? |> /Waiting for unix on A4000/IDE... >>>>> uytterho@cs.kuleuven.ac.be (Uytterhoeven Geert) wrote: Uytterhoeven> On my A4000/040 with linux 0.08 mount works, in contrast Uytterhoeven> with some other programs (like sync...). I didn't tried Uytterhoeven> mkfs since I already had a MINIX file system on my IDE Uytterhoeven> drive (created under 0.07pl4). Hmmm... I didn't write the IDE driver, but I just had a look at it while thinking "cache". After reading from the disk, the driver doesn't invalidate the instruction cache for the memory into which the disk sectors were read. The driver should really do this, since the IC doesn't know that there is new information in that memory. ------------------------------ Crossposted-To: comp.os.linux.admin From: bmyers@dseg.ti.com (Bob Myers) Subject: Are ADDTRON AE-200LC/JL network cards supported? Reply-To: bmyers@dseg.ti.com Date: Tue, 29 Mar 1994 20:41:59 GMT I've just seen the ad for getting these cards at $19/29 respectively (limited 2 cards per customer). Would they both work with the .99p15 and above kernels, since they are NE2000 clones? Likewise, since the AE-200JL is software configurable vs the AE-200LC's hardware jumpered, is there a utility to configure it (or others that are software configurable)? thanks bob p.s. ad was in March '94 edition of LAN magazine, page 77 ------------------------------ From: uknf@rzstud1.rz.uni-karlsruhe.de (Olaf Titz) Subject: IDE transfer and interrupts, again Date: 31 Mar 1994 14:42:05 GMT In article <2ndj10$8gb@levelland.cs.utexas.edu>, Daniel Aaron Supernaw-Issen wrote: > the following: Whenever I have disk activity, the mouse jumps around under X. > This has made the system unusable whenever there is any real swapping going > on. Worse yet, I can't manage to control the mouse enough to be able to The problem is that with 32 sectors read at once, you will certainly lose interrupts. The solution is to enable interrupts during disk transfer (not to use a 16550 even for the mouse at 1200bps...) Someone has claimed the following patch doesn't work for all disk drives. (No, I don't know which) Careful testing with a r/o partition first is recommended. The patch is against a 0.99.15something w/o multi-mode and runs stable in my machine for a long time. If a multi-mode hd.c doesn't patch with this, replace every insw() with ninsw() manually, ditto for outsw(). Olaf Patch follows ------------------------------------------------------------- --- drivers/block/hd.c.orig Fri Jan 21 13:50:11 1994 +++ drivers/block/hd.c Tue Feb 8 19:41:32 1994 @@ -30,6 +30,7 @@ #define REALLY_SLOW_IO #include #include +#include #include #define MAJOR_NR HD_MAJOR @@ -333,7 +334,7 @@ do_hd_request(); return; ok_to_read: - insw(HD_DATA,CURRENT->buffer,256); + ninsw(HD_DATA,CURRENT->buffer,256); CURRENT->errors = 0; CURRENT->buffer += 512; CURRENT->sector++; @@ -392,7 +393,7 @@ end_request(1); if (i > 0) { SET_INTR(&write_intr); - outsw(HD_DATA,CURRENT->buffer,256); + noutsw(HD_DATA,CURRENT->buffer,256); sti(); } else { #if (HD_DELAY > 0) @@ -502,7 +503,7 @@ bad_rw_intr(); goto repeat; } - outsw(HD_DATA,CURRENT->buffer,256); + noutsw(HD_DATA,CURRENT->buffer,256); sti(); return; } --- include/asm/ion.h.orig Tue Feb 8 19:41:32 1994 +++ include/asm/ion.h Tue Feb 8 19:41:32 1994 @@ -0,0 +1,40 @@ +#ifndef _ASM_ION_H +#define _ASM_ION_H + +/* + * This contains versions of ins/outs with interrups enabled. -ot + */ + +#ifdef CONFIG_NO_STI_IO + +#define ninsb insb +#define ninsw insw +#define ninsl insl + +#define noutsb outsb +#define noutsw outsw +#define noutsl outsl + +#else + +#define __NINS(s) \ +extern inline void nins##s(unsigned short port, void * addr, unsigned long count) \ +{ __asm__ __volatile__ ("pushf ; sti ; cld ; rep ; ins" #s "; popf" \ +: "=D" (addr), "=c" (count) : "d" (port),"0" (addr),"1" (count)); } + +#define __NOUTS(s) \ +extern inline void nouts##s(unsigned short port, const void * addr, unsigned long count) \ +{ __asm__ __volatile__ ("pushf; sti ; cld ; rep ; outs" #s "; popf" \ +: "=S" (addr), "=c" (count) : "d" (port),"0" (addr),"1" (count)); } + +__NINS(b) +__NINS(w) +__NINS(l) + +__NOUTS(b) +__NOUTS(w) +__NOUTS(l) + +#endif + +#endif -- 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: danielsi@cs.utexas.edu (Daniel Aaron Supernaw-Issen) Subject: Re: IDE Performance Package Date: 31 Mar 1994 08:41:37 -0600 I did not in any way turn off interuupt handling. I applied the patches directly on a linux 1.0 kernel without modification and found my mouse to be completely unusable under X when there is any disk activity. I really hope this doesn't make it into the standard kernel as I would have to forever more use an old kernel (at least until I purchase a scsi disk). Daniel Supernaw-Issen danielsi@cs.utexas.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 ******************************