From: Digestifier To: Linux-Development@senator-bedfellow.mit.edu Reply-To: Linux-Development@senator-bedfellow.mit.edu Date: Thu, 31 Mar 94 13:13:11 EST Subject: Linux-Development Digest #592 Linux-Development Digest #592, Volume #1 Thu, 31 Mar 94 13:13:11 EST Contents: Re: IDE Performance Package (Byron Thomas Faber) telnet in/out problems (Dave Williams) Bug in atof (Sergio Ramirez) Re: IDE Performance Package (Jussi Mika Antero Lahtinen) [RFD] Alternative to automonitoring (Dave Gardner) Re: IDE Performance Package (Mark Lord) BUG: Linux 1.0 doesn't reboot (Andrew Lih) Re: OPTI 495SLC+486DLC+Linux+Internal Cache? (Superuser) profiling anyone? (Ivan) Re: PC as C64 file server (Roloc) Re: Async I/O (Matthew Donadio) Re: Announcement of Linux/68k Version 0.08 (Johan 'tuben' Tufvesson) Announcement of Linux/68k Version 0.08 (Hamish Macdonald) ---------------------------------------------------------------------------- From: bf11620@eehpx16 (Byron Thomas Faber) Subject: Re: IDE Performance Package Date: 30 Mar 1994 17:53:18 GMT jmalahti@karhu.Helsinki.FI (Jussi Mika Antero Lahtinen) writes: >>I had a Connor 170, which wasn't supported either. Hph. Older drive my ass! Sorry for my lack of respect. Its been a long day with one to many exams. Which I didn't do well on. :) Byron -- PGP 2.3 key available (in plan file) at: Support public code: b-faber@uiuc.edu Use GNU software and others. other accts at: btf57346@sumter.cso.uiuc.edu & bf11620@coewl.cen.uiuc.edu ------------------------------ From: williams@dewey.nl.nuwc.navy.mil (Dave Williams) Subject: telnet in/out problems Date: 30 Mar 1994 16:57:56 GMT Greetings. I have a newly installed SlackWare 1.2.0 Linux 1.0 system installed here and have been expereincing problems with telnet and ftp. Machine is configured wtih NE2000 ether card. From Linux: telnet to a host within my domain initiates the connection and reports the escape character back (^]) but then it hangs and the login proompt is not displayed. ftp says connected, but doesn't go any further. ping works fine on hosts within the domain, but outisde our domain (through our router) see fail rate >90%. Ping from toher machien to outside the domain succeeds fine INstallation was done via NFS, so I'm guessing that that was working well (I havne't tried it recently). From other machine: telnet: Unable to connect to remote host: Connection timed out ftp times out as well. ping doesn't show Linux as alive Replies prefered via e-mail as I can't get here to check the group often enough and might miss a posted reply. Thanks Dave Willams williams@etc.atinc.com ------------------------------ From: sergio@bianca.amd.com (Sergio Ramirez) Subject: Bug in atof Date: Wed, 30 Mar 1994 17:07:58 GMT 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. The program was compiled using gcc under linux. Yggdrasil distribution from 3 months ago. I've seen this bug previously in the early versions of SUN'OS for sparc stations. Can somebody point me to where atof is in the gcc compiler so that I can fix it or if someone has already solved point to the location of the patch. Thanks for your help. Sergio P.S A workaround this code is using sscanf(optarg," %lf%",&mortrate_pc); but sounds like an overkill. -- ------------------------------ From: jmalahti@karhu.Helsinki.FI (Jussi Mika Antero Lahtinen) Subject: Re: IDE Performance Package Date: 30 Mar 1994 18:53:03 +0300 In <2nc2kj$90f@vixen.cso.uiuc.edu> bf11620@ehsn3.cen.uiuc.edu (Byron Thomas Faber) writes: >jcgreen@iastate.edu (Jon Green) writes: >>I just installed the patch to enable Multimode with my IDE drives, and >>got the following message on bootup: >>hda: WDC AC2340H (325MB IDE w/128KB Cache) >>hda: older drive, multiple mode not enabled >>hda: hda1 hda2 >>hdb: st3144AT (124MB IDE w/32KB Cache) >>hdb: older drive, multiple mode not enabled >>hdb: hdb1 >>I would have thought the Western Digital drive would have this feature, as >>it is practically brand new. Oh well, it matches the rest of my system. :) >>My question is this: Since multimode is not available, can I expect to see >>any performance increase by installing this patch? >>-- >>* Jon Green * Still searching for the * Friley 5646 Lorch-Russell * >>* jcgreen@iastate.edu * queen of my double-wide * Ames, Iowa 50012-0001 * >>* Jon2@irc * trailer :) * Phone (515) 296-0648 * >I had a Connor 170, which wasn't supported either. Hph. Older drive my ass! >As far as I understand, the package won't speed anything up if it isn't >running in multimode. The driver must have at least 32 sector multiple mode, but WD2340 has only 16 sect. multiple mode so it is not enabled. You can tweak hd.c to allow multiple mode on smaller sector counts. I have done this with my WD2340 and had no problems. Jussi.Lahtinen@Helsinki.FI >Byron >-- >PGP 2.3 key available (in plan file) at: Support public code: >b-faber@uiuc.edu Use GNU software and others. >other accts at: btf57346@sumter.cso.uiuc.edu & bf11620@coewl.cen.uiuc.edu ------------------------------ From: dgardner@netcom.com (Dave Gardner) Subject: [RFD] Alternative to automonitoring Date: Wed, 30 Mar 1994 19:06:45 GMT As a possible alternative to automonitoring proposed by Ian Jackson, I have posted an IRFD (informal request for discussion) in news.groups called 'IRFD: comp.os.linux.beginners.' If you have the time or inclination to discuss this alternative, please check it out. Thanks. -- ============================================================================== Dave Gardner | This space reserved for a snappy, witty saying dgardner@netcom.com | that is quite likely under construction right now. ============================================================================== ------------------------------ From: mlord@bnr.ca (Mark Lord) Subject: Re: IDE Performance Package Date: 30 Mar 1994 17:22:18 GMT In article jcgreen@iastate.edu writes: >I just installed the patch to enable Multimode with my IDE drives, and >got the following message on bootup: > >hda: WDC AC2340H (325MB IDE w/128KB Cache) >hda: older drive, multiple mode not enabled >hda: hda1 hda2 >hdb: st3144AT (124MB IDE w/32KB Cache) >hdb: older drive, multiple mode not enabled >hdb: hdb1 .. Hi. The patches are deliberately suspicious of any drive which supports fewer than 32 sectors in multiple mode. Your drives support only about 16, so it skips them by default. To get around that, just look for the comparism in my patches (look for 32) and change it to 16. There should be no problem with either drive. The problem is that the message it outputs "older drive" is misleading.. ooops. Look elsewhere in this group for the latest version (posted today) that has been submitted to Linus for inclusion in future kernels. It includes a program to change the settings on the fly, letting you try out different values and measure the effects on performance. Cheers, -ml -- mlord@bnr.ca Mark Lord BNR Ottawa,Canada 613-763-7482 ------------------------------ From: lih@news.cs.columbia.edu (Andrew Lih) Subject: BUG: Linux 1.0 doesn't reboot Date: 26 Mar 1994 23:32:06 -0500 I believe I've found a bug which prevents the kernel from automatically rebooting a machine. If there is a "tail -f" of some files going on when you issue the reboot command (or CTRL-ALT-DEL), then things hang right after it prints "done." at the bottom of the screen. Has anyone else run into this problem? -fuz -- `''' Andrew "Fuz" Lih Columbia University c @@ lih@cs.columbia.edu CRF Tech Staff \ - "A witty saying proves nothing." -- Voltaire ------------------------------ Crossposted-To: comp.os.linux.help,comp.sys.ibm.pc.hardware.chips From: root@fusion.cuc.ab.ca (Superuser) Subject: Re: OPTI 495SLC+486DLC+Linux+Internal Cache? Date: Wed, 30 Mar 1994 00:49:02 GMT pa@tekla.fi (Harri Pasanen) writes: > > Hello good people, > > I noticed that my OPTI 495SLC mother board + TI486DLC40 does not > enable 486DLC's internal cache when running Linux. In AMI bios settings I > have both external (128K) cache and internal cache set enabled. AMI (like always) has messed up again... My DLC test machine is a Phoenix BIOS DLC-aware board and works fine without the patch. Needless to say, my DRx2 machine (with the patch) also works (and it's also a Phoenix BIOS). Now that I've got your attention, I think I'll do my annual I-HATE-AMI spiel: I dislike AMI BIOS so much that I'll willingly pay a premium to get a machine that uses some other BIOS (my favorite, as you can probably guess, is Phoenix). In all my dealings with machines containing AMI BIOSes, I've had nothing but trouble- yes, after fiddling with all the settings, you can get it running fine, but put it side-by-side with a Phoenix or other BIOS, and the AMI machine feels much slower whenever anything has to be done with the BIOS. AMI also has several annoying traits- first, during the memory test, instead of completely aborting the test when a key is pressed like other "real" BIOSes do, AMI just starts counting faster, but still slow enough to be annoying. Then, there's the "Wait..." message.. What the *#%@)*^% is it waiting for?? Godin?? :-) Then there's that god-awful color scheme in the setup.. what where the programmers smoking? So why the hell do so many machines use AMI in spite of these bugs and annoyances? Who knows... My guess is that AMI probably undercut all the other players with low licensing fees.. (about 5-6 years ago, I called Phoenix about customizing a BIOS for a machine.. they said it would cost around $25000 to customize and license a copy.. Anyone know what current costs for various BIOS suppliers are?).. Anyways, that's about the end of my anti-AMI rant.. no replies to the above please.. I just wanted to express my opinions on the shortcomings of AMI BIOS and the last thing I want to start is a BIOS war.. > I tried enabling/disabling the internal cache, and sure enough, under > does all the benchmarks showed the expected results. > > How do I know it is not enabled under Linux? Well, I get 7.98 > Bogomips on linux. I tried CxPatch, and the Bogomips jumped to 13.21. > Unfortunately the system then died to a message along the lines: > > Aha1542.c got the interrupt but no mail. > > Sorry, can't remember the exact wording. However, seems that DMA has > not updated the cache. > > Has anybody successfully licked this problem? The source of the problem is that I made a "mistake" in the patch- a bit that gained another 1% speed on a DRx2 should not have been used in the DLC setting. With this bit on, some motherboards will work (like my Phoenix.. sorry, I seem to have slipped back into my AMI-sux mode :-) ), but most *won't*. To be on the safe side, you shouldn't use the default setting unless you're really sure it will work. The fix- change the CCR0 value 0x22 to 0x23. This should fix most machines, and your BogoMips value will drop from 13.21 to 13.10, which is not a big deal (like I said, about 1%) If you all can wait until the end of the week, I'll be releasing CxPatch 0.30, which is Linux 1.0 compatible, and among other things, allows use of the full patch for greater safety (at the expense of OAK VGA detection, which nobody has anyways.. (if you do have an OAK, tough.. it's time to get a real VGA card)). A patch to change the DRAM refresh rate is also included in CxPatch 0.30 for people whose motherboards don't have the hidden refresh option or can't use it for some reason or other. > Thanks, > > Harri > > -- > ------------------------------------------------------ > Harri Pasanen pa@tekla.fi > phone home: 802 5350 work: 887 9436 c4 -- Christopher Lau- "Mr. Unix" | / Fusion: Playing With Fire! StarBright Research | / / H + H -> He + 24 MeV -- | /_/_/_ "Bring back Trudeau!" root,lauc@fusion.cuc.ab.ca |____________ "This space for rent" ------------------------------ From: ivan@djomolungma.Eng.Sun.COM (Ivan) Subject: profiling anyone? Date: 29 Mar 1994 01:56:34 GMT Reply-To: ivan@djomolungma.Eng.Sun.COM Linux doesn't have facilities for helping with sampling style profiling. I was hoping 1.0 would have something but no such luck. Soooo, I"m going to try and add it in myself. Has anyone else tried it? succeeded? what problems they ran into? ------------------------------ Crossposted-To: comp.sys.cbm From: mparson@nyx10.cs.du.edu (Roloc) Subject: Re: PC as C64 file server Date: Thu, 31 Mar 94 00:52:00 GMT Ok, I think you guys are making this entirely too complicated. When I started this thread a few weeks back,I was talking about turning my XT, which is serving as a dust holder right now, into a simple file-server or hard-drive controller for the 64/128 which I use all the time. I dont have, and dont think it would work if I did have it, Linux. I was just thinking that it would be nifty if you could write an app for the dos environment that would take a disk image (.d64?), then you could type LOAD "$",8 on the C= and boom, you would get a dir. Then you could load and play the games off the XT's HD like it was a 15X1 drive. All this stuff would be entirely transparent to the C=, I think this is important as to aloow for maximum compatibility. any thoughts?? ========================================================================== Michael Parson University of Texas at Brownsville Internet Development Academic Computing Email: 840083@utb.edu Phone: 210-982-0280 IRC: Roloc Using Commodores since 1984 ========================================================================== ------------------------------ From: donadio@mxd120.rh.psu.edu (Matthew Donadio) Subject: Re: Async I/O Date: 30 Mar 1994 23:26:36 GMT David F. Carlson (dave@valhalla.ee.rochester.edu) wrote: : Since Linux is already 1003.1 compliant, getting the pieces to 1003.4 : in place seems like the "Portable" thing to do. I everything required by 1003.1 actually in linux now? I thought linux is still considered a subset of 1003.1. If everthing is there then work on 1003.1b (the ratified 1003.4 draft) would be a good idea. -- Beaker aka Matt Donadio | Life is short, --- __ o __~o __ o donadio@mxd120.rh.psu.edu | ride like ---- _`\<, _`\<, _`\<, --- Penn State Cycling ---| the wind. --- ( )/( ) ( )/( ) ( )/( ) ------------------------------ Crossposted-To: comp.unix.amiga From: t92johtu@odalix.ida.liu.se (Johan 'tuben' Tufvesson) Subject: Re: Announcement of Linux/68k Version 0.08 Date: Wed, 30 Mar 1994 21:53:27 GMT 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... ------------------------------ From: Hamish.Macdonald@bnr.ca (Hamish Macdonald) Crossposted-To: comp.unix.amiga Subject: Announcement of Linux/68k Version 0.08 Date: 30 Mar 1994 00:53:32 GMT This message announces the availability of version 0.08 of Linux/68k. It can be ftped from directory /pub/linux/680x0 at tsx-11.mit.edu. A precompiled kernel executable and the Amiga "bootstrap" program can be found in kern-0.08.tar.gz in the "kernel" subdirectory. The kernel source can be found in linux-0.08.tar.gz in the "src" subdirectory. Patches against 0.07pl4 can be found in linux-0.08.diffs.gz in the "src" subdirectory. A new, 1024K ramdisk filesystem image has been put in the "filesys" subdirectory in the "new-filesys.gz" file. This ramdisk filesystem is an "ext2" filesystem containing new shared libraries and a slew of programs which are necessary to build/rebuild hard disk filesystems. You can replace the libraries on your hard disk partitions with the new versions on the ramdisk image. You can also replace some of your statically linked executables with executables from the ramdisk image (see below (#)). There is a bug in binutils-1.9l.1 when linking dynamically. If the program you are linking overrides a shared library function, then the program will get a SIGSEGV when it tries to execute that function. I've put a patch to binutils-1.9l.1 in "ld.diffs.shlib" in the "tools" subdirectory. I'll be distributing a new "usr.tar.gz" archive later which will contain the new "ld", and will get rid of the statically linked executables found in /usr/bin, replacing them with dynamically linked executables. The changes in this release against 0.07pl4 include: *) A change in the way that the return value from system calls are returned to user programs. Unfortunately, this breaks the existing 4.5.19 shared library and any existing statically linked applications. This change was unavoidable. The "new-filesys.gz" file in the "filesys" directory contains the new shared libraries and dynamically linked executables (#). *) A number of bug fixes. *) Changes from Martin Apel which allow use of the copyback cache on 68040 processors. Martin says that he thinks that there may be problems with dynamically linked executables/shared libraries with the copyback cache. *) The swapping mechanism has been ported. You should be able to use swap partitions and swap files using "swapon" now (note that the kernel prints out some debugging messages whenever a page is swapped in or out; these will eventually be removed). *) Unix domain socket support has been added. *) The Amiga bootstrap has been changed so that it does not need to be loaded into CHIP RAM anymore. This means that we don't require the "BLINK" program when building it. *) Amiga CHIP RAM now has an allocator. Existing users of chipram have been changed to use this new allocator. *) Amiga keyboard driver has auto-repeat now. *) Includes a driver for the Amiga 4000/1200 IDE hard disk controller. Thanks to Torsten Ebeling, Michael Rausch and Geert Uytterhoeven for separately implementing *3* IDE drivers. It was hard to decide which one to include. I ended up including the one from Torsten since it seemed to fit best into the current source. I'm hoping that Michael and Geert will send in improvements if they have any to make. *) Amiga Mouse driver from Michael Rausch. Nothing uses this yet. *) Amiga Parallel Port printer driver from Michael Rausch. *) VTxxx Terminal Emulation on the console from Arno Griffioen. *) Bug fixes to the Amiga Fast File System code to allow it to work on partitions consisting of an odd number of sectors. *) Patches from Geert Uytterhoeven to the Amiga Fast File System code to allow it to work with the Amiga MultiUser filesystem *) Support for core files. *) The "ext2" filesystem has been ported to Linux/68k. I'm not certain of the correctness of this port on largish (>10) filesystems yet. Use at your own risk. *) The "proc" filesystem has been ported to Linux/68k. *) The "ptrace" support has been ported to Linux/68k. This allows the use of "gdb" to source-level debug programs (this has been very useful to me already in tracking down various bugs). This release still contains only support for the Amiga. Hopefully the people working on MacIntosh and Atari support will have some sources for inclusion soon. To boot the kernel on an Amiga, use the supplied "bootstrap" command. To boot with the ram disk image, uncompress the image and type: bootstrap -r new-filesys ro The "ro" option tells the kernel to mount the ramdisk as read-only. The "/etc/rc" file on the ramdisk image will "check" the ramdisk image and remount the root (ramdisk) filesystem read/write. The ramdisk image contains an "/sbin/shutdown" script which will unmount all filesystems, mount the ramdisk read-only, sync all dirty buffers, and then print a message indicating that it is safe to reboot. You can boot from an existing Linux hard disk partition by supplying the device name to the bootstrap program: bootstrap root=/dev/[sh]d[a-f][1-16] You may want to provide the "ro" option to mount the partition read-only. The major number for SCSI disks is "0x08", and the minor number depends on the disk and partition. linux/68k searches for SCSI disks from target 0 to target 7, and for Logical Units 0 through 7 on each target. The minor number can be calculated by (disk_number)*16 + partition_number. The first disk found is disk 0. Partition 0 is the whole disk. Partition 1 is the first partition found in the RigidDiskBlock partition table on the Amiga hard disk. Thus 0x0801 is the first partition on the first disk found. 0x0812 is the second partition on the second hard disk found. For example, I have two SCSI hard disks. The first is at target 5, LUN 0 and the second at target 6, LUN 0. The first has three partitions (used for Linux) and the second has 4 partitions used for AmigaDOS. Thus I have: devnum linux device name ------ ------------------------------------ 0x0800 sda (the entire disk at target 5 : BE CAREFUL) 0x0801 sda1 (1st partition on disk at target 5) 0x0802 sda2 (2nd partition on disk at target 5) 0x0803 sda3 (3rd partition on disk at target 5) 0x0810 sdb (the entire disk at target 6 : BE CAREFUL) 0x0811 sdb1 (1st partition on disk at target 6) 0x0812 sdb2 (2nd partition on disk at target 6) 0x0813 sdb3 (3rd partition on disk at target 6) 0x0814 sdb4 (4th partition on disk at target 6) *NOTE* The target numbers above are examples; these are what I get on my system, since the first disk is at target 5 and the second at target 6. If your first disk is at target 0, your sda will *still* be 0x0800 (/dev/sda). My Linux root partition is on the 1st partition of my first drive, so I boot with: bootstrap root=/dev/sda1 After booting from one of the above methods, if the kernel supports your SCSI driver, you should be able to create a minix file system on one of your hard disk partitions if you wish. Determine the size of your partition in 1K blocks (take the number of 512 byte sectors from HDToolBox and divide by two), and determine which special file to use in /dev (see above). *DOUBLE CHECK* that the major/minor numbers for the special device (ls -l /dev/xxx) are correct. If they are incorrect or the device special file doesn't exist, use mknod to change or create the device special file. Then execute: /sbin/mkfs.minix /dev/xxxx size This will create a minix file system on the hard disk partition. You can then mount this partition under /mnt and copy files to it: mount /dev/xxxx /mnt When finished copying, unmount the partition: umount /mnt sync a few times, run "/sbin/shutdown" and then reboot. You can then boot the kernel by providing "bootstrap" with the device name to boot from. Again, you do any mucking around with hard disks at your OWN RISK. Note that the above can be used on IDE hard disks now also, except that the major/minor numbers for IDE hard disks are different, and the special devices in /dev are named hd[ab][1-64]. ------------------------------ ** 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 ******************************