From: Digestifier To: Linux-Development@senator-bedfellow.mit.edu Reply-To: Linux-Development@senator-bedfellow.mit.edu Date: Wed, 17 Nov 93 06:13:24 EST Subject: Linux-Development Digest #231 Linux-Development Digest #231, Volume #1 Wed, 17 Nov 93 06:13:24 EST Contents: Public Domain ORB? (Bryan Sparks) Re: BusLogic SCSI under Linux...compatible with AHA 1740 or not? (David E. Wexelblat) Re: Motif - Pay? BAH! (Bernd Wiserner) Time frame for GCC 2.5.2? (Drinks All The Water) Re: 4.3 BSD sendmsg/recvmsg (Dragon) Linux Modem Server 2.09beta1 available for Beta Testing (Anthony Rumble) Xmosaic + term ? (Harri Pasanen) How many BogoMips on a Pentium? (Harri Pasanen) term: any filesystems over term? (Mark Eichin) Linux Modem Server 2.09beta2 patch (Anthony Rumble) ---------------------------------------------------------------------------- Crossposted-To: comp.object From: bsparks@Novell.COM (Bryan Sparks) Subject: Public Domain ORB? Date: Tue, 16 Nov 1993 21:56:01 GMT Hi all, A few questions and please pardon my ignorance, if I inadvertently show some :-). Is there in the public domain (or GPL) a CORBA compliant ORB (and related technologies)? Is anyone working on one that they intend to place in the public domain? Do you want some help? Is there any ORB-like technology that could be made into a CORBA compliant ORB, that anyone knows of? I might be interested in starting (and/or coordinating) some kind of project in this area, if anyone is interested. I have Linux at home and would probably do this work on it. If a project of this nature were started what would you do and how would you do it? What technologies would be necessary? etc... Any ideas and/or flames welcome. Please respond by e-mail and I will summarize if there is interest... Thanks! Bryan -- Bryan Sparks bsparks@Novell.COM (801) 429-3853 ------------------------------ Crossposted-To: comp.os.linux.misc From: dwex@aib.com (David E. Wexelblat) Subject: Re: BusLogic SCSI under Linux...compatible with AHA 1740 or not? Date: Tue, 16 Nov 1993 20:05:34 GMT In article eric@tantalus.nrl.navy.mil (Eric Youngdale) writes: >In article <2camlp$93@jacobs.jacobs.mn.org> root@jacobs.mn.org (Mike Horwath) writes: >>Bruce (chengb@logic.camp.clarkson.edu) wrote: >>: Hi, I just received my BusLogic BT-747S. So far, I have only >>: gotten it to work under the AHA-1542 driver in the kernel, and it >>: doesn't seem to work when I have only AHA-1740 support compiled in. As >>: far as I know, the driver for 1542 probably isn't taking full advantage >>: of the EISA speed (10-33 Mbits/sec!!!), so is there a possibility to >>: hack the existing driver for one that takes full advantage of BusLogic >>: card? >> >>YOu aren't going to believe this, but I tested one of those babies out for >>a week and a half for one of the bosses so he could determine if it was a >>good card for a unix like system. >> >>All I ever got it working in was 1542 mode, so I called them, asked them >>about 174x mode, they said no, it only has a 1542 mode. I asked if there >>were native drivers, they said no, it just emulates the 1542. >> >>Now about performance, it was alot faster on my system than a 1542. Don't >>really know why, except for it maybe emulates the 1542 command and register >>set, but operates in an EISA like way dmaing around the world at high speed. > > This is basically the result that I would have expected. The 1542 >driver does not really know what kind of bus the card is attached to. It could >be ISA, EISA, VLB or S100 for all it cares. All that it wants is a register >compatible card that uses a compatible mailbox message system. The 1542 >basically uses DMA, so the driver is merely telling the card where to dump the >data. The card obviously knows the details of the bus that it is connected to, >and it handles the details of getting the data across. Therefore a 1542 clone >card that is EISA should get EISA performance, not ISA performance. > >-Eric >-- >"The woods are lovely, dark and deep. But I have promises to keep, >And lines to code before I sleep, And lines to code before I sleep." Too bad that it's the wrong answer. The BusLogic EISA and VLB controllers have an enhanced mode, but it is not the same as the Adaptec 174x enhanced mode. The native mode of the BusLogic controllers is the same as the Adaptec 154x. The BusLogic enhanced mode is a lot nicer, too. You don't have to change EISA config to change modes; your driver simply puts the card into enhanced mode. According to the SCSI driver developer at Microport, adding support for the BusLogic enhanced mode was absolutely trivial, compared to the work involved to get the Adaptec working. Suffice it to say that my Microport SVR4 box boots with Adaptec 1542b code on the boot/install floppy, then reinstalls an enhanced SCSI driver that switches the board into enhanced mode automatically (the enhanced driver, which supports 154x + BusLogic, Adaptec, and DPT enhanced SCSI modes, was too big to fit in the floppy-boot kernel). With an Adaptec 174x, you would have to shut down and reconfigure your hardware to get into enhanced mode. Runs like a bat out of hell, too. -- David Wexelblat (703) 430-9247 Fax: (703) 450-4560 AIB Software, Inc., 46030 Manekin Plaza, Suite 160, Dulles, VA 20166 Formerly Virtual Technologies, Inc. Mail regarding XFree86 should be sent to "Ooh, are you feelin' satisfied? Come on, let us give your mind a ride." -- Boston, "Feelin' Satisfied" ------------------------------ From: wiserner@Informatik.TU-Muenchen.DE (Bernd Wiserner) Subject: Re: Motif - Pay? BAH! Date: Tue, 16 Nov 1993 20:48:03 GMT >But no one needs to care about the internal representation that >OSF/Motif uses for its widgets. The whole idea of a widgets is that >it isn abstracted datum. If we can create a widget that has the sameme >interface as, say, a Motif XmText widget, reacts correctly when an >action is called from a translation table, redraws correctly when it >gets an expose event, basically acts, looks, and feels like an XmText >widget, who cares what OSF did internally to implement theirs. A rose >by any other name and all that good stuff. Don't forget about Motif-derived widgets. Let's say you would subclass XmText to handle multiple fonts. If we don't want to break such things, we must at least name our structures ,internal functions ... like Motif does. > >> > >First of all, the XmText widget is probably the most complicated >widget you are going to encounter from the OSF/Motif widget set. >Those 75 pages that you reference describe the behavior of the XmText >widget as well as a slew of convebience functions that make >programming *with* this widget easier. True, it makes it a little bit >harder to program the widget library, it doesn't make it harder to >program the *widget*. One might argue about that. : e.g. RowColumn has about 300 kB of source. There isn't only comment in that. > > >Fine, but the fact is that OSF/Motif is becoming (arguably, I'll >admit), a de facto standard for application programming under X. Part >of this is because it is a fairly robust useful widget set, and part >of it is (admittedly) flash (managers are always impressed by things >that look good :-). But a lot of software is being written to use >Motif, and a Motif-compatible/compliant free toolkit would be an >extremely useful item. There is just one word about that : COSE > > ><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> >Bill Woodward | wpwood@austin.ibm.com <-- Try this first >AIX Software Support | billw@aixwiz.austin.ibm.com >Graphics Group | 512-838-2834 >My other computer is a 386 running Linux !!! B.Wiserner ------------------------------ From: boutell@netcom.com (Drinks All The Water) Subject: Time frame for GCC 2.5.2? Date: Tue, 16 Nov 1993 19:45:44 GMT Is anyone working on bringing all the necessary libraries, etc. up to a level where it is safe and sane to move to GCC 2.5.2? I'm developing a C++ GUI toolkit that would be greatly improved by the fixes made to templates in 2.5.2, but I gather that it's not particularly safe to build the current libc and libg++ of Linux with 2.5.x. Right now I use an incredibly heinous hack to avoid building an enormous library with a bazillion redundant template expansions. Still, I don't want to move to 2.5.x until others can and will generally do so, because the toolkit is of little use if most people won't have the necessary tools to build it properly. I understand Slackware just moved *back* to 2.4.5 because of the problems mentioned above, so I post this here as a plea to system builders to upgrade the affected subsystems to the necessary level. -T -- i'll be under the floorboards with my face in the sun ------------------------------ From: william@CS.Arizona.EDU (Dragon) Subject: Re: 4.3 BSD sendmsg/recvmsg Date: 16 Nov 1993 13:57:13 -0700 In article <2c5pr1$kvv@jacobs.jacobs.mn.org>, Doug McIntyre wrote: >M.C.Little@newcastle.ac.uk (Mark Little) writes: >>The latest version of Linux which we have (0.99.13q) supports send/sendto >>and recv/recvfrom. Are there any plans to support the other 4.3 BSD >>primitives for sending and receiving scatter/gather buffers, i.e., sendmsg >>and recvmsg? We have some software which uses these. Has anyone gotten sento and recvfrom to work with a SOCK_PACKET socket? I'm having some problems figuring out there interface to the socket and sentto/recvfrom calls for this socket type. I'd appreciate any information at all. Source code that actually uses them would be most appreciated. Thanks, William ==========================================+=================================== "Give peace a chance" -John Lennon | Internet: william@cs.arizona.edu "What's so civil about war?" -Axl Rose | ATTNet : 602.795.7626 | SlowNet : Silver Dragon Linux, The FREE 32-bit operating | 3401 N Columbus #25L system available NOW! | Tucson AZ 85712 ==========================================+=================================== ------------------------------ From: smilie@lsupoz.apana.org.au (Anthony Rumble) Crossposted-To: comp.sources.testers,apana.general,lsupoz.general,apana.linux Subject: Linux Modem Server 2.09beta1 available for Beta Testing Date: 17 Nov 1993 09:49:58 +1100 Linux Modem SERVER version 2.09beta1 Now available for BETA testing. AVAILABLE: yarrina.connect.com.au:/tmp/mdm/mdm-2.09-beta1.tar.gz WHAT IS IT: A replacement for GETTY/UUGETTY. Handles dialins/dialouts on your Unix System with the DUEL TTY system FEATURES: *EASY* config files. Much better logging. Debug modes where you can see all the data being sent/recieved while initialising. Multiple configs, one for each line. Specifically for HAYES compatable intelligent modems. Fully UUCP aware, will allow UUCP or EMSI mailers to dial out correctly, and also allows Termial Packages to use the line. (Eg Xcomm, kermit, minicom, Seyon). DOESN'T NEED or USE LOCK FILES!! Has support for 57600 and 115200 speeds. Can be used with High speed Constant speed modems or with variable speed modems. COPYRIGHTS: Please read COPYING for more details on the licence of this product. MISC: **PLEASE** DON'T send this out with any changes, send me patch files if you fix or modify anything, and I will release it with your patches, and recognition of the patch. It is designed to do what uugetty does not. Compiling and Installation just type make install - To make and install the files make - To make the binary To make the program, and install the files. I use GCC 2.4.3, with 0.99pl12 kernal. Edit the Makefile to select what log file you want to use and if you want device logging. -DDEBUG will send a lot of information to the logfile and the logging device, but is recommended until you are sure everything is working. There are a few config files that are used, the main GLOBAL configuration fileis in /etc/default/modem. It is used for all lines, unless a seperate config file is present which will overide any options. Have a look at this file for details of the options. To make a seperate config for a different tty, just make a tty?? file in the /etc/default directory with the overiding commands in it. an example is in /etc/default/ttyS1 Which is loaded for the modem server on ttyS1 Command line parameters :- /etc/modem tty speed login eg: /etc/modem ttys1 2400 bbs This will Initialise the modem at 2400 baud. (Configuration files can override this option). and will login as BBS when connected. If you leave the login option blank, it will execute login to allow the user to login as normal. TODO: Option to make it shut the speaker up at certain times.. Send a "WAKEUP" to the modem ever half an hour or so.. (Eg AT|) to make sure the modem has not died. adding mtmp for keeping the current speed/rel of a connected tty. dialin server type support (Like a real annex) :) probably plenti of other nice things... If you have any probs, additions, fixes mail me at :- root@lsupoz.apana.org.au -- root@lsupoz.apana.org.au APANA Sydney UUCP regional hub (feeds available) Anthony Rumble Linux Support OZ (02) 418-8750 v.32bis - 3 lines Voice (02) 418-8220 For information on APANA mail info@apana.org.au ------------------------------ From: hpasanen@cs.hut.fi (Harri Pasanen) Subject: Xmosaic + term ? Date: 16 Nov 1993 20:10:34 GMT Hi all, I tried running Xmosaic on a decstation over 9600 baud modem with term. While it worked, it was hopelessly slow. What occurred to me that perhaps I could run Xmosaic locally on my linux box, and it could use the term channel to fetch the data. I'm a newbie on Xmosaic, and have not really done my homework, but it seems to me that at least in principle this should work. I believe that Xmosaic is built on top of motif widgets, so I won't be able to compile it - I'm just asking if anybody has done this thing already. Thanks, Harri PS. Where can I find Xmosaic binary for linux? ------------------------------ From: hpasanen@cs.hut.fi (Harri Pasanen) Subject: How many BogoMips on a Pentium? Date: 16 Nov 1993 20:23:00 GMT Hi, Just to satisfy my curiosity: If you have booted Linux on a Pentium machine, how many BogoMips did you get? For the record, my Intel 386/33 gives 5.99 BogoMips. I hear Cyrix 486DLC40 should give around 13. What about other processors? Thanks, Harri PS. Why would anyone think I have a moment of idle time in my hands :) But still, these things hold a certain fashination for me, must be like the thing how fast a car can go from 0 to 100; important to some people. ------------------------------ From: eichin@cygnus.com (Mark Eichin) Subject: term: any filesystems over term? Date: 16 Nov 1993 18:38:09 -0500 I've been using term extensively (both from my home PC and from my laptop... the PC only dials in to one place [the office] but the laptop connects to various places [the home pc, the office, another office at MIT where I volunteer...] so it's a bit more convenient and reliable than SLIP...) and find it very useful. In particular, termtelnet, termncftp, termxmosaic are very nice to have. (I even run uucp over term, but that's a scary [though secure] kludge that you really don't want to try :-) Anyway, the one thing I find missing is remote filesystem access. tupload is useful, sure, and emacs ftp-find-file probably works ok with term-ftp -- except that I don't usually have an ftpd on the remote machine (and I haven't yet seen RMAIL work with ftp-find-file :-) Has anyone done anything like this before? If you assume that the client need only work on the Linux side, you could use the linux user-process file server support on this end, with another program on the other side. You might want to do some clever caching, but reliable access, even if slow, would be a good start. I probably won't have a chance to implement anything like this before Christmas (by which point I'll probably have an ethernet card anyway :-) so I thought I'd toss out the idea to see if it caught anyone's attention. (I'd certainly be willing to provide some advice and suggestions via email, perhaps even alpha-test.) Any takers? _Mark_ MIT Student Information Processing Board Cygnus Support ------------------------------ From: smilie@lsupoz.apana.org.au (Anthony Rumble) Crossposted-To: comp.sources.testers,apana.general,lsupoz.general,apana.linux Subject: Linux Modem Server 2.09beta2 patch Date: 17 Nov 1993 11:17:26 +1100 Anthony Rumble (smilie@lsupoz.apana.org.au) wrote: : Linux Modem SERVER version 2.09beta1 Oooops.. I made a dreadfull mistake.. 57600 was 56700.. anyway.. heres a patch to fix it.. if you want the full file it's available under yarrina.connect.com.au:/tmp/mdm/mdm-2.06-beta2.tar.gz *** ../modem.old/modem.c Tue Nov 16 22:55:12 1993 --- modem.c Wed Nov 17 10:54:41 1993 *************** eg: *** 176,181 **** --- 176,183 ---- /* Macros */ #define LOG_TTY flog("%u VMIN = %u, VTIME = %u, ICANON = %u, ISIG = %u\n", __LINE__,ti.c_cc[VMIN], ti.c_cc[VTIME], ((ti.c_lflag&ICANON) ? 1 : 0), ((ti.c_lflag&ISIG) ? 1 : 0)) + #define LOG_FLAGS flog("i%u o%u c%u l%u\n", ti.c_iflag, ti.c_oflag, ti.c_cflag, ti.c_lflag); + #define LOG_SPEED flog("ospeed = %u ispeed = %u\n", cfgetospeed(&ti), cfgetispeed(&ti)); /* File scope variables */ *************** long setgetbaud(long s) *** 205,210 **** --- 207,215 ---- { long baudrate = 0; struct termios tt; + #ifdef DEBUG + struct termios ti; + #endif #ifdef linux static struct serial_struct ser_io; #endif *************** if (s != 0) /* If setting baudrate... *** 254,260 **** #endif break; #ifdef linux ! case 56700: baudrate = B38400; ser_io.flags &= ~ASYNC_SPD_MASK; ser_io.flags |= ASYNC_SPD_HI; --- 259,265 ---- #endif break; #ifdef linux ! case 57600: baudrate = B38400; ser_io.flags &= ~ASYNC_SPD_MASK; ser_io.flags |= ASYNC_SPD_HI; *************** if (s != 0) /* If setting baudrate... *** 266,271 **** --- 271,277 ---- break; #endif default: + flog("ERROR: Bad Speed %u\n", s); return (-1); } *************** if (s != 0) /* If setting baudrate... *** 274,279 **** --- 280,290 ---- tcsetattr(0, TCSADRAIN, &tt); /* Set the speed and others */ + #ifdef DEBUG + tcgetattr(0, &ti); + LOG_SPEED; + #endif /* DEBUG */ + #ifdef linux if (baudrate == B38400) if (ioctl(0, TIOCSSERIAL, &ser_io) < 0) /* set special speed */ *************** if (s != 0) /* If setting baudrate... *** 281,286 **** --- 292,302 ---- flog("ERROR: Could not set linux serial info (Old Kernel?)"); return -1; } + #ifdef DEBUG + tcgetattr(0, &ti); + LOG_SPEED; + #endif /* DEBUG */ + #endif /*linux*/ return s; } *************** else *** 315,321 **** return 115200; else if ((ser_io.flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI) ! return 56700; else #endif /*linux*/ return 38400; --- 331,337 ---- return 115200; else if ((ser_io.flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI) ! return 57600; else #endif /*linux*/ return 38400; *************** dup2(0, 2); *** 985,991 **** #ifdef DEBUG tcgetattr(0, &ti); LOG_TTY; ! flog("i%u o%u c%u l%u\n", ti.c_iflag, ti.c_oflag, ti.c_cflag, ti.c_lflag); #endif /* Execute login */ --- 1001,1008 ---- #ifdef DEBUG tcgetattr(0, &ti); LOG_TTY; ! LOG_FLAGS; ! LOG_SPEED; #endif /* Execute login */ -- root@lsupoz.apana.org.au APANA Sydney UUCP regional hub (feeds available) Anthony Rumble Linux Support OZ (02) 418-8750 v.32bis - 3 lines Voice (02) 418-8220 For information on APANA mail info@apana.org.au ------------------------------ ** 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 ******************************