From nrd1jrb@nrd.ups.com Wed Oct 1 12:05:10 1997 From: Jim Bucciferro Date: Wed Oct 1 12:05:13 PDT 1997 Subject: Debugging C++ programs with Tornado --MimeMultipartBoundary Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I am trying to debug a C++ program with the included debugger but when I attempt to place a breakpoint in the code, it tells me that there is no symbol information for the file. I am compiling through Tornado using a customized build with the following options: CPU=I80486 ADDED_CFLAGS=-g sockets.out sockets.cpp is my source file. When I compile a .c file, the debugger works correctly. What do I need to do to make the debugger work with my C++ file? Thanks. Jim Bucciferro --MimeMultipartBoundary-- From gpbeedu@net.com Wed Oct 1 15:47:20 1997 From: gpbeedu@net.com (Guru Prasad Beedu) Date: Wed Oct 1 15:47:23 PDT 1997 Subject: Password for ftp needed. --MimeMultipartBoundary Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, We are currently using Tornado 5.3 version. This version does not provide the password for ftp. We know that the latest version provide the password. But at present we would like to continue with the same version. Is anybody tried to modify their ftp source code to put the password protection ? May I know how easy or tough to modify ? Thanks, Guru Prasad, Network Equipment Technology, gpbeedu@net.com --MimeMultipartBoundary-- From tmorris@me.umn.edu Wed Oct 1 15:50:47 1997 From: Ted Morris Date: Wed Oct 1 15:50:53 PDT 1997 Subject: PCMCIA socket/hot-sw support --MimeMultipartBoundary Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit greetings, I'd like to get feedback as to anyone's experiences with implimenting VxW on x86 (Dell Pentium 266!) laptops with PCMCIA (possible flavors of i/o cards: Robotics Modem, EtherNet, ComputerBoards ADAC/Descrete I/O). How well is PCMCIA socket level layer supported under VxWORKS? I noticed an earlier posting by someone stating that there exists a utility to boot the kernal from the DOS prompt... Has anyone out there used a PC as the host AND the target? I'm trying to envision how in the world one is supposed to debug :-/ Our initial proposal was to go with PC104 ($cheap$ and available) but they'd prefer that the application be implemented with 'off-the-shelf' Pentium laptops. thanks to anyone in advance, Ted -- Ted Morris tmorris@me.umn.edu 612-625-3520 Center For Advanced Manufacturing Design And Control 612-625-9881 Institute of Technology, U of Minnesota FAX: 612-625-8884 --------------------------------------------------------------------------- "McBride fell 12 stories, hitting the pavement like a Hefty Bag filled with vegetable soup." (Paul Sabourin, Silver Spring) --MimeMultipartBoundary-- From daemon@csg.lbl.gov Thu Oct 2 04:00:34 1997 From: daemon@csg.lbl.gov Date: Thu Oct 2 04:00:42 PDT 1997 Subject: comp.os.vxworks newsdigest Comp.Os.Vxworks Daily Digest Thu Oct 2 04:00:21 PDT 1997 Subject: vxWorks memory map ------------------------------------------------------- Newsgroups: comp.os.vxworks Subject: vxWorks memory map Date: Mon, 29 Sep 1997 16:30:07 -0400 From: dvkan@igs.net Organization: ST Telecom Message-ID: <34300FCF.7FAF@igs.net> Reply-To: dvkan@igs.net Does anybody know how vxWorks' memory map looks like? Where is the available space of memory in IMM(internal memory map)? I need it for device driver buffer descriptors. Thanks, Silviu --------------------------- End of New-News digest ********************** From karstens@chaos.wg.com Thu Oct 2 05:26:31 1997 From: karstens@wg.com (Alexander Karstens) Date: Thu Oct 2 05:26:34 PDT 1997 Subject: Re: Debugging C++ programs with Tornado --MimeMultipartBoundary Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Did you include C++ support in your kernel-configuration (INCLUDE_CPLUS etc..) Hope that helps Regards, Alexander Karstens --MimeMultipartBoundary-- From phil.quiney@gecm.com Thu Oct 2 06:56:09 1997 From: Philip Quiney Date: Thu Oct 2 06:56:12 PDT 1997 Subject: Re: vxWorks memory map Hi there, Apologies if I have missed the point but why not use the rngLib functions to set up buffers for your device driver then plant the RING_ID in your channel descriptor. This is defined statically like a serial driver or you could meke the channel allocation dynamic with a bit of thought. Then you wouldn't need to know about the vxWorks memory map (although the 256 bytes starting at 0x600 - meant for shared memory on a VME bus is spare on our 68k system, we used a few bytes for our use of Applied Microsystems' CodeTEST). > > Does anybody know how vxWorks' memory map looks like? > Where is the available space of memory in IMM(internal memory map)? > I need it for device driver buffer descriptors. > Thanks, > Silviu > Regards Phil Quiney =============================================================================== Phil Quiney _/_/_/_/ _/ _/_/_/_/ email : phil.quiney@gecm.com GEC Marconi _/ _/ _/ _/ _/ tel : +44 1245 353221x3914 Communications Ltd _/_/_/_/ _/ _/ _/ _/_/ fax : +44 1245 275448 Chelmsford _/ _/_/_/_/ _/_/_/_/ England CM1 1PL _/ =============================================================================== From mfischer@qualcomm.com Thu Oct 2 11:55:40 1997 From: Mark Fischer Date: Thu Oct 2 11:55:43 PDT 1997 Subject: ethernet "units" All, I believe (correct me if I'm wrong) that the onboard ethernet interfaces can support more than 1 unit- ie, in addition to eth0 (or dc0, lnPci0, whatever the BSP wants to call it), I should be able to bring up dc1, dc2, and dc3. However, when I try to do a usrNetIfAttach (), on anything higher than 0, (for example, dc1) I get: S_symLib_SYMBOL_NOT_FOUND Network interface dc1 unknown. Anyone know what is going on here? Mark Fischer Qualcomm, Inc. From sgm@ssd.ray.com Thu Oct 2 12:29:55 1997 From: Myerson Date: Thu Oct 2 12:29:58 PDT 1997 Subject: Troubles with 8mm SCSI tape drive I sent the following problem to WRS, but given my past experiences with them, it will be several days before they even acknowledge receipt of my problem, let alone a solution. ----- I am having trouble with an 8mm SCSI tape drive. I've got a Motorola 1604 board. We're running VxWorks 5.3.1. I've got the following options turned on in config.h #define INCLUDE_SCSI #define INCLUDE_TAPEFS #define SCSI_AUTO_CONFIG On powerup, the tape drive is detected. The following output is produced: ID LUN VendorID ProductID Rev. Type Blocks BlkSize pScsiPhysDev -- --- -------- ---------------- ---- ---- -------- ------- ------------ 4 0 EXABYTE EXB-8500-85Qanx0 0415 1R 0 0 0x01bf8ae0 I then try to start my application. The first thing I do is call tapeFsInit. "OK" is returned. THen I make a call to scsiPhysDevCreate (pSysScsiCtrl, 4, 0, 0, NONE, TRUE, 0, 0) and this returns a NULL pointer, indicating failure. errno has the hex value 0x370014, which according to our current source indicates the problem comes from module scsiLib (that makes sense) and the error code is ENOTDIR, not a directory (which doesn't make sense). Where is the problem? I am using the sample code from target/src/config/usrScsi.c Do I not have the most up-to-date sample code, or is there a problem in the configuration? THanks, Steve Myerson (sgm@ssd.ray.com) Raytheon COmpany Portsmouth, RI 401-842-4007 From froeber@bbn.com Thu Oct 2 18:36:33 1997 From: Fred Roeber Date: Thu Oct 2 18:36:36 PDT 1997 Subject: Re: ethernet "units" --MimeMultipartBoundary Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 2 Oct 1997, Mark Fisher wrote: > I believe (correct me if I'm wrong) that the onboard ethernet > interfaces can support more than 1 unit- ie, in addition to eth0 > (or dc0, lnPci0, whatever the BSP wants to call it), I should be able > to bring up dc1, dc2, and dc3. However, when I try to do a > usrNetIfAttach (), on anything higher than 0, (for example, dc1) I get: The usrNetIfAttach function is for connecting the primary interface of a particular type. It assumes you pass it the device name without a unit number and it tacks the unit number 0 on the end. I think you want to call particular device attach routine directly. Note, I don't think all devices necessarily support multiple units. Make sure to check the return code from the device attach routine. BTW, to see how to call the attach routine check the source for usrNetIfAttach. Its delivered in the src/config/usrNetwork.c file. Fred | Fred J Roeber, BBN Systems & Technologies | | 4 John Clarke Road Middletown, RI 02842-5202 | | froeber@bbn.com 401-848-3548 | --MimeMultipartBoundary-- From vbkumar@wipinfo.soft.net Thu Oct 2 20:50:47 1997 From: V B Kumar Date: Thu Oct 2 20:50:56 PDT 1997 Subject: VxWorks Tutorial Online ? --MimeMultipartBoundary Content-Type: TEXT/PLAIN; charset=US-ASCII Halo all, I am new to VxWorks, I am not able to get any published book on VxWorks!. Does anyone know of any VxWorks Tutorial available Online or any material on VxWorks from programmers poit of view available freely on the Web ? Thanks in advance. --MimeMultipartBoundary-- From Serge.Kornfeld@digital.com Fri Oct 3 13:17:53 1997 From: Serge Kornfeld - HLO Date: Fri Oct 3 13:17:57 PDT 1997 Subject: EBSA 110 --MimeMultipartBoundary Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Is ANYBODY working with EBSA 110??? I am having a problem downloading using "ld < " images bigger than 100K! serge --MimeMultipartBoundary-- From Francis.Hallahan@gsfc.nasa.gov Fri Oct 3 15:55:08 1997 From: Francis Hallahan Date: Fri Oct 3 15:55:11 PDT 1997 Subject: Re: Troubles with 8mm SCSI tape drive --MimeMultipartBoundary Content-Type: text/plain; charset="us-ascii" >Submitted-by sgm@ssd.ray.com Thu Oct 2 12:29:55 1997 >Submitted-by: Myerson > >I sent the following problem to WRS, but given my past experiences with them, >it will be several days before they even acknowledge receipt of my problem, >let alone a solution. > Know what you mean! >----- > >I am having trouble with an 8mm SCSI tape drive. > >I've got a Motorola 1604 board. >We're running VxWorks 5.3.1. > >I've got the following options turned on in config.h > #define INCLUDE_SCSI > #define INCLUDE_TAPEFS > #define SCSI_AUTO_CONFIG >. . . > >THen I make a call to > scsiPhysDevCreate (pSysScsiCtrl, 4, 0, 0, NONE, TRUE, 0, 0) >and this returns a NULL pointer, indicating failure. > >errno has the hex value 0x370014, >which according to our current source indicates the problem comes from >module scsiLib (that makes sense) and the error code is ENOTDIR, >not a directory (which doesn't make sense). > >Where is the problem? >I am using the sample code from target/src/config/usrScsi.c > >Do I not have the most up-to-date sample code, >or is there a problem in the configuration? > You need "#define INCLUDE_SCSI2" for SCSI-II Support (a seperate WRS product). If I remember correctly, tapeFs is in SCSI-II Support. I did a quick check of our systems, and found /h/seqIo.h in our scsi2 directories but not in our 5.2 or 5.3 directories. TapeFs is a sequential device and WRS has added another set of sequential scsi calls to support tape. Unfortunately, "scsiPhysDevCreate" is for a Block device, and won't support tape. PS: I hope I'm wrong, but I don't think so. I believe other vendors may have tape support without SCSI-II. Hope this helps, Frank Hallahan /-------------------------------------------------------------------------\ | Frank Hallahan | Hallahan@gsfc.nasa.gov | | Software Environment Leader - Code 521 | fbh@vlsi.gsfc.nasa.gov | | Computer Sciences Corporation | | | at Nasa Goddard Space Flight Center | Phone (301) 286 7064 | \-------------------------------------------------------------------------/ --MimeMultipartBoundary-- From daemon@csg.lbl.gov Sat Oct 4 04:01:25 1997 From: daemon@csg.lbl.gov Date: Sat Oct 4 04:01:31 PDT 1997 Subject: comp.os.vxworks newsdigest Comp.Os.Vxworks Daily Digest Sat Oct 4 04:01:11 PDT 1997 Subject: Re: simple VxWorks database ------------------------------------------------------- Newsgroups: comp.os.vxworks Subject: Re: simple VxWorks database Date: 4 Oct 1997 01:06:28 GMT From: mikei@portal.ca (Michael Imai) Organization: Client of Internet Portal Services Message-ID: <6144qk$p2j@thoth.portal.ca> References: <34351F07.51FBFDE1@ccm.ut.intel.com> Try codebase from sequiter software mike In article <34351F07.51FBFDE1@ccm.ut.intel.com>, Kyle_Unice@ccm.ut.intel.com says... > >Looking for source for a simple database that could be used on VxWorks. > >Any ideas? > >Cheers >Kyle > --------------------------- End of New-News digest ********************** From aweiss@emultek.co.il Sat Oct 4 23:22:50 1997 From: aweiss@emultek.co.il Date: Sat Oct 4 23:22:54 PDT 1997 Subject: unregister --MimeMultipartBoundary Content-type: text/plain; charset=US-ASCII Please unregister me, Thanks, Avi. --MimeMultipartBoundary-- From daemon@csg.lbl.gov Sun Oct 5 04:01:37 1997 From: daemon@csg.lbl.gov Date: Sun Oct 5 04:01:40 PDT 1997 Subject: comp.os.vxworks newsdigest Comp.Os.Vxworks Daily Digest Sun Oct 5 04:01:16 PDT 1997 Subject: Re: GNU alignment question ------------------------------------------------------- Newsgroups: comp.os.vxworks Subject: Re: GNU alignment question Date: 3 Oct 1997 00:04:41 GMT From: don@mars.dgrc.doc.ca (Donald McLachlan) Organization: CRC Message-ID: <611cqp$m1p@crc-news.doc.ca> References: <606tpc$bu3@idiom.com> Reply-To: don@mars.dgrc.doc.ca K&R's "The C Programming Language" 2'nd edition states "Fields may be declared only as ints [...]" Don - -- Donald McLachlan e-mail donald.mclachlan@crc.doc.ca Communications Research Centre / DRX office 613-998-2845 3701 Carling Ave., fax 613-998-9648 Ottawa, Ontario labs 613-998-2423 / 613-998-4118 K2H 8S2 home 613-599-6262 Canada --------------------------- End of New-News digest ********************** From rtp.co.uk!ihw@rtp.co.uk Mon Oct 6 04:40:24 1997 From: Ian Willats Date: Mon Oct 6 04:40:32 PDT 1997 Subject: Troubles with 8mm SCSI tape drive Re: problems seen by Myerson : If it's not too late to chuck in my 2p-worth... > On powerup, the tape drive is detected. > The following output is produced: > ID LUN VendorID ProductID Rev. Type Blocks BlkSize pScsiPhysDev > -- --- -------- ---------------- ---- ---- -------- ------- ------------ > 4 0 EXABYTE EXB-8500-85Qanx0 0415 1R 0 0 0x01bf8ae0 This is the result of the SCSI auto-configuration process. The tape drive has been detected and assigned the SCSI_PHYS_DEV_ID 0x01bf8ae0. > THen I make a call to > scsiPhysDevCreate (pSysScsiCtrl, 4, 0, 0, NONE, TRUE, 0, 0) > and this returns a NULL pointer, indicating failure. > > errno has the hex value 0x370014, > which according to our current source indicates the problem comes from > module scsiLib (that makes sense) and the error code is ENOTDIR, > not a directory (which doesn't make sense). The reason for the failure is that the phys. dev. has already been created during auto-configuration. The errno reflects this (see scsiLi2Lib.h): #define S_scsiLib_DEVICE_EXIST (M_scsiLib | 20) Having got a SCSI phys. dev. ID for the tape drive, the next step should be to create a SCSI_SEQ_DEV followed by initialising a file system on the device. This is described with example code on pp 230-233 of the v5.3.1 Programmer's Guide. Incidentally, Frank Hallahan's comment: > Unfortunately, "scsiPhysDevCreate" is for a Block device, and won't > support tape. is _not_ true. (Fortunately) scsiPhysDevCreate() is used for any SCSI device, be it disc, tape or any other box of tricks. Tape file system support, although not logically dependent on SCSI-2 support, is provided as part of the same upgrade package. Both are standard in Tornado v5.3.1. HTH, Ian ---------------------------------------------------------------- Ian Willats E-mail: ihw@rtp.co.uk Real-Time Products Ltd. Unit 8, Holt Court, Telephone: +44 121 333 6955 Aston Science Park, Fax: +44 121 333 5433 Jennens Road, Birmingham B7 4EJ. England. Web: www.rtp.co.uk ---------------------------------------------------------------- From R.Kirnum@dialogic.com Mon Oct 6 11:45:03 1997 From: "Kirnum, Robert" Date: Mon Oct 6 11:45:06 PDT 1997 Subject: Help! Tornado - cc960 packed structures --MimeMultipartBoundary Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Can someone please explain why I am having trouble with the align and pack pragmas. I did not have a problem with VxWorks 5.2, but the Tornado tools are giving INCONSISTENT results. The following structure has the incorrect offsets as shown. The inconsistency is that the 'a[]' array is apparently padded to 16 bytes while the 'd[]' array is fine at 14 bytes and starts on an odd byte boundary. typedef unsigned char byte; typedef unsigned short word; #pragma align 1 #pragma pack 1 typedef struct { /* Actual Offset Expected Offset */ byte a[0x0F]; /* 0x0000 .. 0x000f 0x0000 .. 0x000e */ word b; /* 0x0010 .. 0x0011 0x000f .. 0x0010 */ byte c; /* 0x0012 0x0011 */ byte d[0x0E]; /* 0x0013 .. 0x0020 0x0012 .. 0x001f */ byte e; /* 0x0021 0x0020 */ byte f; /* 0x0022 0x0021 */ } aStruct; #pragma align 0 #pragma pack 0 Adding the attribute control after the 'a[]' array does nothing, adding it after both 'a[]' and b corrects the packing. #pragma align 1 #pragma pack 1 typedef struct { byte a[0x0F]; __attribute ((packed)); word b; __attribute ((packed)); byte c; byte d[0x0E]; byte e; byte f; } aStruct; #pragma align 0 #pragma pack 0 Any suggestions are welcome. Thanks, Bob ________________________________________________________________________ D I A L Bob Kirnum _|_|_|O Software Engineer _|_|_|G 973-993-3000 Ext. 6589 _|_|_|I 973-993-5916 Fax _|_|_|C mailto:R.Kirnum@dialogic.com Get the Dialogic Edge at http://www.dialogic.com Dialogic Corp. / 1515 Rt. 10 / Parsippany, NJ 07054 ________________________________________________________________________ Did you know . . . that FireWire was developed by Apple? . . . that Apple is 150 in the Fortune 500 and Microsoft is 172? Believe in Macintosh? Check out http://www.evangelist.macaddict.com/ --MimeMultipartBoundary-- From mmcquade@pop.aa.net Mon Oct 6 13:15:08 1997 From: Mike McQuade Date: Mon Oct 6 13:15:11 PDT 1997 Subject: Wind Web Server --MimeMultipartBoundary Content-Type: text/plain; charset="us-ascii" Hello, Is anyone using the Wind River Web Server ? We would like to incorporate a web server into on of our products, and Id like to hear any comments folks might have on the Wind River or other web server for Tornado. Thanks. --MimeMultipartBoundary-- From Brau@dantel.com Mon Oct 6 15:07:00 1997 From: Bodo Rau Date: Mon Oct 6 15:07:03 PDT 1997 Subject: ADS860 and PCMCIA support ? --MimeMultipartBoundary Content-Type: text/plain Hi, Wondering if somebody has gained experience with the ADS860 board from Motorola and VxWorks 5.3.1. The BSP from WindRiver does not support the PCMCIA functionality of the MPC860. I am looking for a driver to support the PCMCIA capabilities of the MPC860. I would appreciate any suggestions. Bodo Rau Dantel, Inc. --MimeMultipartBoundary-- From daemon@csg.lbl.gov Tue Oct 7 04:01:58 1997 From: daemon@csg.lbl.gov Date: Tue Oct 7 04:02:01 PDT 1997 Subject: comp.os.vxworks newsdigest Comp.Os.Vxworks Daily Digest Tue Oct 7 04:01:52 PDT 1997 Subject: Re: Tornado 1.0.1 and the Standard Template Library Subject: Re: Invalid volume label on dosFs ------------------------------------------------------- Newsgroups: comp.os.vxworks Subject: Re: Tornado 1.0.1 and the Standard Template Library Date: Mon, 06 Oct 1997 01:01:28 -0700 From: Ingo Pakleppa Organization: Eagle Creek Systems, Inc. Message-ID: <34389AD6.D2A1EC60@cts.com> References: <01bccde5$aff49d60$b10614ac@evolution> Reply-To: ingop@cts.com You can buy a version of the STL from Rogue Wave that goes with tools.h++ and therefore most likely works under VxWorks. I haven't tried it myself, but you could give RW a call to find out for sure. The other possibility is to try to get the original STL from HP. Ingo Jonathan D.R. Martin wrote: > I am trying to compile a C++ class under Tornado 1.0.1. The class > uses > template classes from the standard template library (stl). Tornado > does > not provide the stl header files. Does anyone know if stl headers for > > VxWorks are available? I tried to get Tornado to compile the class > using > the stl headers from both the Borland and Microsoft Win32 C++ > compilers but > did not have any luck due to modifications made my each vendor. > > The headers that I need in particular are: > > algorith > function > iterator > memory > set > stdexcept > string > tree > vector > > Jonathan Martin > Newnes Machine Ltd. --------------------------- Newsgroups: comp.os.vxworks Subject: Re: Invalid volume label on dosFs Date: 6 Oct 1997 21:56:53 GMT From: "Raymond K.S. Yeung" Organization: AdiCom Wireless Message-ID: <01bcd2a3$1a0ea1c0$df60a9ce@host.hooked.net> References: <01bcd214$be7133c0$41131c26@raymondy> Apparently, to fully utilize the 11 characters, DOS_OPT_LONGNAMES must be used. This is not mentioned in the manual at all. Raymond Raymond Yeung wrote in article <01bcd214$be7133c0$41131c26@raymondy>... > I'm thinking of using the volume label as a sanity > test that a block device (SRAM defined as RAM > disk in my case) has been initialized. > > vxWorks manual says that I can use up to > 11 characters for the label. I found that it > is stored in two places: in boot sector and in > root directory. The boot sector has 11 char. > but the one in root directory only has 8 chars > (i.e. truncated). > > When I read the label using ioctl with FIOLABELGET > during system initialization time, I keep getting > the truncated label. Can anyone shed some light here? > > Thanks, > Raymond > > --------------------------- End of New-News digest ********************** From R.Kirnum@dialogic.com Tue Oct 7 05:46:32 1997 From: "Kirnum, Robert" Date: Tue Oct 7 05:46:35 PDT 1997 Subject: RE: Help! Tornado - cc960 packed structures --MimeMultipartBoundary Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Kevin, thanks for the feedback. I guess what I was looking for was confirmation that the '#pragma pack 1' and '#pragma align 1' were no longer supported. I do not get any compile errors indicating they are unknown pragmas, but their existence in the code has no affect at all. Bob >-----Original Message----- >From: Kevin Bradley [SMTP:kb+@andrew.cmu.edu] >Sent: Monday, October 06, 1997 6:21 PM >To: Kirnum, Robert; vxworks_users@csg.lbl.gov >Subject: Re: Help! Tornado - cc960 packed structures > >Excerpts from mail: 6-Oct-97 Help! Tornado - cc960 packe.. by the U. G. >Exploder@lbl.g >> typedef struct { /* Actual Offset Expected Offset */ >> byte a[0x0F]; /* 0x0000 .. 0x000f 0x0000 .. 0x000e */ >> word b; /* 0x0010 .. 0x0011 0x000f .. 0x0010 */ > > > >> Adding the attribute control after the 'a[]' array does nothing, adding >> it after both 'a[]' and b corrects the packing. > >Although I'm not familiar with cc960, I think I have an idea what's >going on in general with the alignment in gcc. It sees a word, and >wants to align it on a word boundary. I suspect you'd see similar >results with "long", "long long", "float", et al. > >When you put the attribute control on the word, it tells gcc that >you know what's going on and you'll deal with the fact that the >word starts unaligned. > >If you leave off the attribute on "a", and just leave it on "b", >do you also get the right results? Also, do the remaining structures >properly line up, or are they shifted to align themselves properly >on word boundaries after "b" offsets by a byte? > >I would think that you'd need the attribute flags for every element >in the structure, as this would be really brutal on GCC and force >it to behave. > >From the GCC info file (http://nis-www.lanl.gov/info/gcc/gcc_toc.html) > >packed > > The packed attribute specifies that a variable or structure field > should have the smallest possiblealignment--one byte for a variable, > and one bit for a field, unless you specify a larger value with the > alignedattribute. Here is a structure in which the field x is packed, > so that it immediately follows a: > > struct foo > { > char a; > int x[2] __attribute__ ((packed)); > }; > >packed > This attribute, attached to an enum, struct, or union type definition, > specified that the minimum requiredmemory be used to represent the > type. Specifying this attribute for struct and union types is equivalent > tospecifying the packed attribute on each of the structure or union > members. Specifying the `-fshort-enums'flag on the line is equivalent > to specifying the packed attribute on all enum definitions. You may only > specifythis attribute after a closing curly brace on an enum definition, > not in a typedef declaration. > > >So in your example, I think this means you can say: > >struct my_packed_struct { > byte a[0x0F]; > word b; > byte c; > byte d[0x0E]; > byte e; > byte f; >} __attribute__((packed)); > >typedef my_packed_struct aStruct; > > > >which _should_ cause everything to crunch down as much as possible, >giving your structure an overall length of 34 bytes. > >Anybody else have experience with this? > > > > -- Kevin > > --MimeMultipartBoundary-- From mmcquade@pop.aa.net Tue Oct 7 10:16:26 1997 From: Mike McQuade Date: Tue Oct 7 10:16:30 PDT 1997 Subject: free code for mv2600 / mv2300 boards --MimeMultipartBoundary Content-Type: text/plain; charset="us-ascii" For anyone who might be interested, we have written some code that adds functionality to the mv2600 and mv2300 boards, the two modules are: 1. A small "blocking" delay for 2600 or 2300, this delay uses a Raven timer to allow delays down to 100 uSec. 2. An AuxClk driver for the 2300 boards, this code uses a timer in the Raven to generate an AuxClk, this allows you to use the "Spy" tool in Tornado. We offer this code to anyone who mat be interested, we ask that you offer feedback, and or any bug fixes, code is offered "as-is", use it at your own risk. Email me if you want the code. Mike. --MimeMultipartBoundary-- From kb+@andrew.cmu.edu Tue Oct 7 13:58:04 1997 From: Kevin Bradley Date: Tue Oct 7 13:58:08 PDT 1997 Subject: WindView Log File Format? Greetings, all. I've just gotten WindView up & running, and it's really impressive. However, I'm interested in getting worst-case numbers out, and looking at 1,000+ different interrupt events in my 5 second log is proving a bit daunting. Does anybody know the file format for WindView logs? It doesn't appear to be described in my documentation. I'd like to be able to open the log and basically do just what the display part of WindView does, except apply my own grouping / analysis functions rather than doing the graphics. Or, failing that, is there a way to group events of a like kind and perform some kind of analysis on them inside of WindView itself? E.g. write up Tcl code wrappers, etc. What I envision doing is saying "Gimme all the timer interrupts, tell me the worst-case run time for that interrupt level", etc. And before anybody gets exciting about timer granularity, I'm using Pentium cycle-count registers, and I think that it's got about a 4-5 us accuracy what with doing the read and cache and all for a 75 MHz clock... that's about 300-375 clock cycles. Any ideas? I'm mainly concerned about (1) timer interrupt (system overhead in my scheduling models), (2) "user" interrupt (sound-card driver), (3) user-level tasks responding to interrupts, and (4) my applications. I figured I'd start with the really easy, repetitious ones, namely the interrupts. -- Kevin From aguilar@popler.lansce.lanl.gov Tue Oct 7 15:55:32 1997 From: aguilar@popler.lansce.lanl.gov (Ron Aguilar) Date: Tue Oct 7 15:55:35 PDT 1997 Subject: duel ethernet adaptors --MimeMultipartBoundary Content-Type: text/enriched; charset="us-ascii" VxWorks Users, I would like some advice on the selection and implementation of a second ethernet adapter to run a isolated network concurrently on a Windows NT platform. I am interested in communicating with a mvme3604 card over a 10/100mbps ethernet line. I currently use a Intel 82557-based 10/100 Ethernet PCI adapter to communicate with the outside world. The second adapter would ideally run concurrently with the Intel chip and serve as the communication path to the 3604 board and Windows NT running VxWorks 5.3.1. I have tried a 3C900 adapter by 3Com but could not get it to work correctly. Any suggestions on setup and recommendations on adapters would be helpful. Thanks, 0000,0000,8080 Ron Aguilar Los Alamos National Laboratory Manuel Lujan Jr. Neutron Scattering Center LANSCE-12, MS H805 Los Alamos, NM 87545 505-667-8369 505-665-2676 (fax) rgaguilar@lanl.gov --MimeMultipartBoundary-- From 42579788@aol.com Tue Oct 7 16:07:34 1997 From: 42579788@aol.com Date: Tue Oct 7 16:07:38 PDT 1997 Subject: Deliverable Mail 

/////////////////////////////////////////////////////////////////
CREDIT CENTER***CREDIT CENTER***CREDIT CENTER***CREDIT CENTER
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
If you received this electronic mailing by mistake, please pardon the
intrusion.

Dear Friend;

Are you or somebody you know experiencing credit problems?  If the
answer is yes, our special offer might be just the cure you have been
looking for.  No longer will you have to live like a second class
citizen due to a bad credit report.  Bad credit can happen due to a
variety of reasons and can happen to anybody.  There is no profile of a
person with credit problems, it can happen to anybody.  In fact, one in
three Americans live with some form of bad credit.  Fortunately, the
United States Government has passed many laws under the Fair Credit
Reporting Act, all of which protect your rights as a consumer.  Our
manuals will show you how to use these Federal Laws to your advantage to
repair your credit, obtain a new credit report or add good credit to
your credit report!

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
*****  'THE FRESH START PROGRAM II' (TM) *****

Have you recently been denied credit due to a bad credit report?  Have
you seen your credit report lately?  If your answered yes, then you need
to take advantage of the 'FRESH START PROGRAM II' (TM) NOW!  This manual
is unique in that it shows you how to obtain a brand new Social Security
number which will allow you to open a brand new, fresh credit report. 
Using guidelines set forth by the Social Security Administration, this
easy-to-follow manual will guide you through the steps necessary to
obtaining your new Social Security Number.  Regardless of your present
credit, you can open a new credit file free of ANY judgements,
charge-off's, foreclosures, bankruptcies, bad credit, tax liens's,
garnishments, late pays and inquiries.

////////////////////////////////////////////////////////////////

***** THE 'WorldCard' (TM) INSTANT CREDIT Credit Card *****

This program is very simple.  It extends to you a $5,000 instant line of
credit in the exclusive 'WorldCard' (TM) Wholesale Purchasing Club. 
This credit card comes to you with absolutely NO credit checks!

You will receive the 'WorldCard' (TM) Wholesale Purchasing Club catalogs
with over 5,000 items to choose from.  Jewelry, toys, tools, gift items,
housewares and much more are available.  After your first purchase, each
quarter you will receive updated catalogs with brand new items
available.

You will be required to make a small down-payment with your orders
and then make a monthly payment of 10% of the unpaid balance.  It's just
that simple!  Your good credit with WorldCard will be reported to the
credit bureau's, which will show your excellent credit!

As a special BONUS, if you take advantage of this offer within the next
24 hours, you will be given $200 in cash certificates!  $100 may be used
on the first $300 unpaid balance and the other $100 may be used after
you have charged $600 on your new account.

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

***** 'STOP COLLECTION AGENCIES COLD' (TM) *****

If you have ever experienced the humiliation of a collection agency
calling you at home or at work, then this hard-hitting manual will put
an IMMEDIATE stop to those calls!

Millions of collection agency telephone calls and letters are made or
sent each week to those who owe money.  Consumers have been helpless to
stop the harassment of these predators, UNTIL NOW!

This manual will show you how to put an end to these telephone calls and
letters, INSTANTLY!  At the same time, these collection agencies will be
forbidden to EVER call or write you again!!  This manual has the inside
secrets the credit world does not want you to have!

Gain back your self-esteem and piece-of-mind by ordering 'Stop
Collection Agencies Cold' (TM) TODAY!!!

/////////////////////////////////////////////////////////////////

***** 'THE COMPLETE CREDIT REPAIR KIT' (TM) *****

Good credit empowers you to qualify for personal loans, automobile
loans, business loans, credit cards, home loans and other types of
credit you may apply for at the very best interest rate.  As you know,
if you don't have excellent credit, lenders will NOT extend credit to
you!  The Fair Credit Reporting Act will allow you to legally and
permanently ERASE the negative information on your credit bureau file.

Anything damaging to your credit, such as bankruptcies, judgements, tax
liens, slow pays, repossessions, charge off's, foreclosures, late
payments, student loans and inquiries may be permanently and legally
REMOVED from your credit report!

If bad credit on your credit file is standing in the way of your
achieving the American dream, then order 'The Complete Credit Repair
Kit'(TM) NOW!!!

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

IMS
5201 Kingston Pike, Suite 6-216
Knoxville, Tennessee 37919

YES, please rush the manuals checked below:

_____   'The Fresh Start Program II' (TM)         	$34.95

_____   'WorldCard' (TM) Credit Card            	$34.95

_____   'Stop Collection Agencies Cold' (TM)   	 $34.95

_____   'The Complete Credit repair Kit' (TM)   	$34.95

_____   TAKE ALL FOUR (SAVE over $39)   	$99.95


$____________  Sub-Total

$4.95    Shipping & Handling for each manual ordered
$15.00   RUSH Delivery (Optional)

$____________  TOTAL ENCLOSED

I Have Enclosed (Check One):
_____   Cashiers Check
_____   Money Order
_____   Personal Check
_____   FAX ORDER


SHIPPING INFORMATION:

Name:__________________________________________________
Address:________________________________________________
City:_______________________State:________Zip Code:_________
Telephone Number:___________________

Your E-Mail Address (For order confirmation):___________________

////////////////////////////////////////////////////////////////

FREE  -  FREE  -  FREE  -  FREE  -  FREE  -  FREE  -  FREE  -  FREE  

Fax your order within 24 hours and receive a special Kodak film book
entitling you to
100 rolls of Kodak film as well as five 6X10 and ten 4X5 enlargements
FREE!

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

Please tape your signed personal check in the space below and fax to
IMS.  The 24 hour facsimile number is 1-423-681-6520.

******************************************************

FAX 1-423-681-6520

Please TAPE Your Check Here

Please SIGN Your Check

******************************************************

THANK YOU For Your Order!!!








From craig@pacengr.com  Tue Oct  7 16:47:23 1997
From: "Craig A. Humphreys" 
Date: Tue Oct  7 16:47:42 PDT 1997
Subject: PPC 603e instruction/data cache
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit


Has anyone out there experienced difficulties with the PowerPC 603e
processor using both instruction and data caching?  Or, alternately,
are there any success stories/tricks/tips?

I've got a 603e BSP where turning on both the instruction and data
cache causes the SONIC ethernet driver to lose it's mind.  Receive
descriptor management in particular gets messed up.

Any thoughts or ideas would be appreciated.


-- Craig

--MimeMultipartBoundary--


From dundar@venus.aselsan.com.tr  Wed Oct  8 04:56:20 1997
From: Baris DUNDAR 
Date: Wed Oct  8 04:56:34 PDT 1997
--MimeMultipartBoundary
Content-Type: TEXT/PLAIN; charset=US-ASCII

unsubscribe dundar@venus.aselsan.com.tr

--MimeMultipartBoundary--


From Jean-Yves_Patry@Mitel.COM  Wed Oct  8 06:21:48 1997
From: Jean-Yves Patry 
Date: Wed Oct  8 06:21:59 PDT 1997
Subject: task guardian for VxWorks
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi all,
   I am developing a VxWorks BSP for a telecommunication system (PABX)
based on a MC68020 running at 33 Mhz with 32 MB

   During a demo, I ran into a bug whereby the "network" stopped
responding and did not know why? After two days of investigation, I
found out that the tNetTask died with a bus error.

   One thing that is lacking with VxWorks is its ability to recover
important task that die prematurely. Does anybody know of a "guardian"
task for VxWorks that would automatically recover important tasks if
they die prematurely? or can somebody point me to a WEB site where I
could find contributed software that provides this functionality?

Thanks
jean-yves_patry@mitel.com
(613) 592-2122 ext 4992
--MimeMultipartBoundary--


From rich@immrc.eng.mcmaster.ca  Wed Oct  8 06:24:24 1997
From: Richard Teltz 
Date: Wed Oct  8 06:24:26 PDT 1997
Subject: 2 ethernet on 1 VME
--MimeMultipartBoundary
Content-Type: TEXT/PLAIN; charset=US-ASCII


I have a VME cage with two MVME147 MPU's (68030's) and their 
accompanying ethernet/serial/etc breakout panels.

I would like to have one MPU as the system controller, and the
other as simply another CPU.  I would however, like both of  
them to have external ethernet connections (via their respective
breakouts). 

It seems though that the external ethernet functions will 
only work if the MPU board involved is jumpered to be the
system controller.  Neither the Motorola or VxWorks docs
explicitly discuss this.  The VxWorks docs do describe multiple 
target configurations, but only with respect to the "sm" backplane
network.  

Is the backplane network my only option for multiple targets and
VxWorks?

Is the external ethernet/serial/etc functionality really only
available to VME system controller designated boards?

Any help would be appreciated.
Thanks in advance.

Rich

*------------------------------------------------------------------*
|  INTELLIGENT MACHINES AND MANUFACTURING RESEARCH CENTRE (IMMRC)  |
*------------------------------------------------------------------*
|       Richard Teltz,           | Dept. of Mechanical Engineering,|  
|  principal research engineer,  | McMaster University,            |
|   tel: (905) 525-9140, x27351  | Hamilton, Ontario,              |
|   fax: (905) 572-7944          | Canada,                         |
| email: rich@immrc.mcmaster.ca  | L8S 4L7                         |
*------------------------------------------------------------------*

--MimeMultipartBoundary--


From kvernum@spk.hp.com  Wed Oct  8 10:12:38 1997
From: Jeffrey Kvernum 
Date: Wed Oct  8 10:12:42 PDT 1997
Subject: In need of a RTC driver (DS1287 or MC146818A).


Anybody know were I can obtain a driver for either a DS1287 or a MC146818A?

Thanks,
Jeff Kvernum 



--------------------------------------------------------------------------
Jeff Kvernum                            email: kvernum@spk.hp.com
Hewlett-Packard Company                 phone: (509) 921 - 3770
Spokane Division                        fax  : (509) 921 - 3991



From phil.quiney@gecm.com  Thu Oct  9 01:27:30 1997
From: Philip Quiney 
Date: Thu Oct  9 01:27:33 PDT 1997
Subject: Re: 2 ethernet on 1 VME
Hi Rich,

Not sure if this is any help but we have a VME rack with 4 MVME162 boards with
each board having its own physical ethernet connection. The 162 does have its
own AUI connection though but at least it shows that the OS can cope. Also in
our config the boards have no knowledge of each other other than the physical
ethernet (ie we are not using any shared memory stuff).

Of the four boards, the first one is linked to be the 'system controller' but
other than that all the kernels are the same as far as I know. I would guess
that if you tell each board to use the ethernet to boot from (ie boot device =
ei - or whatever the 147 device is called) then the physical connection should
work, if I remember that is all we did to get it to work.

Regards,
Phil Q

> I have a VME cage with two MVME147 MPU's (68030's) and their 
> accompanying ethernet/serial/etc breakout panels.
> 
> I would like to have one MPU as the system controller, and the
> other as simply another CPU.  I would however, like both of  
> them to have external ethernet connections (via their respective
> breakouts). 
> 
> It seems though that the external ethernet functions will 
> only work if the MPU board involved is jumpered to be the
> system controller.  Neither the Motorola or VxWorks docs
> explicitly discuss this.  The VxWorks docs do describe multiple 
> target configurations, but only with respect to the "sm" backplane
> network.  
> 
> Is the backplane network my only option for multiple targets and
> VxWorks?
> 
> Is the external ethernet/serial/etc functionality really only
> available to VME system controller designated boards?
> 
> Any help would be appreciated.
> Thanks in advance.
> 
> Rich
>

===============================================================================
Phil Quiney          _/_/_/_/        _/  _/_/_/_/  email : phil.quiney@gecm.com 
GEC Marconi         _/    _/        _/  _/    _/     tel : +44 1245 353221x3914
Communications Ltd _/_/_/_/  _/    _/  _/  _/_/      fax : +44 1245 275448
Chelmsford        _/        _/_/_/_/  _/_/_/_/
England CM1 1PL                              _/                             
===============================================================================


From lucard@tin.it  Thu Oct  9 03:09:16 1997
From: Luca Rodda 
Date: Thu Oct  9 03:09:22 PDT 1997
Subject: Looking for VxWorks Profibus support
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hello to everyone,

I am looking for products supporting Profibus in VxWorks environment for
VME bus or IP-modules.

Can anyone help me?

Thank you

Luca Rodda

--MimeMultipartBoundary--


From lucard@tin.it  Thu Oct  9 03:21:15 1997
From: Luca Rodda 
Date: Thu Oct  9 03:21:20 PDT 1997
Subject: Looking for VxWorks Sinec H1 support
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hello to everyone,

I am looking for products supporting Sinec H1 in VxWorks environment for
VME bus or IP-modules.

Can anyone help me?

Thank you

Luca Rodda
--MimeMultipartBoundary--


From daemon@csg.lbl.gov  Thu Oct  9 04:00:28 1997
From: daemon@csg.lbl.gov
Date: Thu Oct  9 04:00:36 PDT 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Thu Oct  9 04:00:23 PDT 1997

        Subject: Re: task guardian for VxWorks

-------------------------------------------------------

Newsgroups: comp.os.vxworks
Subject: Re: task guardian for VxWorks
Date: 8 Oct 1997 16:15:48 GMT
From: "Raymond Yeung" 
Organization: EarthLink Network, Inc.
Message-ID: <01bcd405$cf440e00$7a131c26@raymondy>
References: <343B86C1.66A4@Mitel.COM>

I think there's a hangman freeware out there, which allows
you to register critical tasks with "hangman" for monitoring,
possibly restarting.  One potential problem I see is the clean
up of previously allocated resources.  This freeware can be
accessed from WRS ftp site (visit their homepage).

Raymond

Jean-Yves Patry  wrote in article
<343B86C1.66A4@Mitel.COM>...
>    One thing that is lacking with VxWorks is its ability to recover
> important task that die prematurely. Does anybody know of a "guardian"
> task for VxWorks that would automatically recover important tasks if
> they die prematurely? or can somebody point me to a WEB site where I
> could find contributed software that provides this functionality?


---------------------------

End of New-News digest
**********************


From wbrown@csg.lbl.gov  Thu Oct  9 07:29:50 1997
From: wbrown@csg.lbl.gov (Bill Brown)
Date: Thu Oct  9 07:29:53 PDT 1997
Subject: Re: 2 ethernet on 1 VME

We have a system with one mv167 (acting as "system controller" - i.e. it
handles bus arbitration, etc) and 3 Heurikon "Nitro-260" CPUs, all with their
own ethernet interfaces.  These boards use the backplane to communicate with
one Bit3 interface which "talks to a Multibus-I bin (don't ask!).  There is a
5th processor in the bin, a mv162, with it's own ethernet interface which
at the moment "just lives there;" it does not use the backplane for i/o but
"talks" to all its i/o via the net.

The "boot" interface for each cpu is determined by the boot parameters in
that CPUs' NVRAM; in each case the "boot device" parameter is set to the
"ei" ethernet interface.  All are on the same subnet, but that would not
have to be the case.  Of course, each has its' own IP number.  There is no
backplane IP number on any CPU.

The one "gottcha" I had to deal with was that all processors had to be
configured such that its' "processor number" was not equal to "0".
By default (out of the box) vxWorks maps some or all of the onboard ram
in processor 0 onto the backplane.  This overlayed the address space of the
Bit3 card, causing no end of grief and confusion.  Since we are not using
the backplane network there is no CPU 0.

Again; the CPU number is set in NVRAM; the bus controller function is enabled/
disabled by a jumper on the mv167.  I _think_ some processors may use a
software-setable register for this function; as I recall the mv147 uses a
jumper.  Consult the hardware docs.  Having more than one bus controller in
a bin is a NO-NO; having more than one CPU 0 is also a really bad idea.
Having NO CPU 0 works just fine if you are not using the backplane network,
which is not to be confused with using the backplane to talk to i/o devices.
The left-most board in the bin (slot 0) _must_ be the system controller.

Having _no_ system controller is a really bad idea; it can cause all kinds
of problems even if the backplane is not being used for data transfers.
I haven't figured out why this seems to be true but it certainly seems to
be the case with the hardware we're using.

Having made all of errors possible with the above described variables, some
of them more than once, I hope this will get you headed in the right direction.

Disclaimer:  Any opinions are my own and have	    |  -bill
    nothing to do with the official policy or the   |   wlbrown@lbl.gov
    management of L.B.N.L, who probably couldn't    |   Berkeley, CA
    care less about employees who play with trains. |   aka wlbrown@netcom.com


From jrm@mclean.sparta.com  Thu Oct  9 09:41:37 1997
From: "John R. Moore III" 
Date: Thu Oct  9 09:41:40 PDT 1997
Subject: Benchmarks for SPARC
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hello,
	I would like to know if anyone has done, or knows of, any vxWorks
benchmarks on the hyperSPARC.  Such as context switch times,
semGive/semTake etc.

Thanks,

John 

-- 

------------------------------------------------------------ 
|John R. Moore              | Voice -> 703 448 1683 x229   | 
|Principal Engineer         | FAX   -> 703 893 5494        | 
|EMAIL:                     | On the WEB!                  | 
|jrm@mclean.sparta.com      | http://www.mclean.sparta.com | 
|___________________________|______________________________| 
|                                     |                    | 
|     /\\     SPARTA, Inc.            |                    | 
|    ///\\    7926 Jones Branch Drive |                    | 
|   /////\\   Suite 900               |                    | 
|  /////  \\  McLean, VA 22102        |                    | 
| / \\\\\  \\ 703-448-0120            |         	   | 
| \  \\\\\ //                         |                    | 
|  \  //////                          |                    | 
|   \//////   SPARTA                  |                    | 
|    \////    Pride In Performance    |                    | 
|     \//     --------------------    |                    | 
------------------------------------------------------------
--MimeMultipartBoundary--


From R.Kirnum@dialogic.com  Thu Oct  9 11:05:07 1997
From: "Kirnum, Robert" 
Date: Thu Oct  9 11:05:10 PDT 1997
Subject: Boot Device
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit

My application needs to be able to download from a unique device, yet
still have the network available to the downloaded application.  I added
a new boot device which can be specified in the boot string.  Once I use
this, the network (nic) is not properly initialized.  Is there an easy
way to accomplish this without hacking up the bootConfig.c code?  We
need to download the firmware through our unique loader.  If the
firmware detects the presence of a communication board (ethernet and
serial), it will provide the WDB interface for debugging.

Thanks,
Bob

________________________________________________________________________
D I A L	Bob Kirnum
_|_|_|O	Software Engineer
_|_|_|G	973-993-3000 Ext. 6589
_|_|_|I	973-993-5916 Fax
_|_|_|C	mailto:R.Kirnum@dialogic.com
Get the Dialogic Edge at http://www.dialogic.com
Dialogic Corp.  /  1515 Rt. 10  /  Parsippany, NJ 07054
________________________________________________________________________
           Did you know . . . that FireWire was developed by Apple?
       . . . that Apple is 150 in the Fortune 500 and Microsoft is 172?
Believe in Macintosh? Check out http://www.evangelist.macaddict.com/

--MimeMultipartBoundary--


From mea@mclean.sparta.com  Thu Oct  9 11:51:45 1997
From: "Mike Anderson" 
Date: Thu Oct  9 11:51:50 PDT 1997
Subject: Re: Boot Device
--MimeMultipartBoundary
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: base64
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--MimeMultipartBoundary--


From bcrozier@fci.COM  Thu Oct  9 12:59:56 1997
From: bcrozier@fci.COM (Bruce Crozier)
Date: Thu Oct  9 13:00:00 PDT 1997
Subject: Re: Boot Device
Hi Bob,

> From vxwexplo@lbl.gov Thu Oct  9 13:48:43 1997
> 
> Submitted-by R.Kirnum@dialogic.com  Thu Oct  9 11:05:07 1997
> 
> My application needs to be able to download from a unique device, yet
> still have the network available to the downloaded application.  I added
> a new boot device which can be specified in the boot string.  Once I use
> this, the network (nic) is not properly initialized.  Is there an easy
> way to accomplish this without hacking up the bootConfig.c code?  We
> need to download the firmware through our unique loader.  If the
> firmware detects the presence of a communication board (ethernet and
> serial), it will provide the WDB interface for debugging.

Try putting the name of the standard network device into the 'other' field
on your boot parameters.  This will cause it to be properly initialized as
part of the boot-up process, even though you are using a different device
to boot from.

HTH
Bruce

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   | Bruce Crozier                     Voice: (972) 774-1888 x 516 |
   | Customer Engineering                Fax: (972) 774-1889       |
   | FORCE COMPUTERS INC,Dallas Texas                              |
   |                                                               |
   | http://www.forcecomputers.com     Email: bcrozier@fci.com     |
 __|__                                                           __|__
( ___ )~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~( ___ )

Standard Disclaimer: The statements and opinions expressed herein are 
strictly my own, and do not necessarily represent those of FORCE COMPUTERS.



From mea@mclean.sparta.com  Thu Oct  9 13:24:02 1997
From: "Mike Anderson" 
Date: Thu Oct  9 13:24:09 PDT 1997
Subject: Re: Boot Device
--MimeMultipartBoundary
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit


-----Original Message-----
From: the vxWorks Users Group Exploder 
To: vxworks_users@csg.lbl.gov 
Date: Thursday, October 09, 1997 2:34 PM
Subject: Boot Device


>Submitted-by R.Kirnum@dialogic.com  Thu Oct  9 11:05:07 1997
>Submitted-by: "Kirnum, Robert" 
>
>--MimeMultipartBoundary
>Content-Type: text/plain; charset="us-ascii"
>Content-Transfer-Encoding: 7bit
>
>My application needs to be able to download from a unique device, yet
>still have the network available to the downloaded application.  I added
>a new boot device which can be specified in the boot string.  Once I use
>this, the network (nic) is not properly initialized.  Is there an easy
>way to accomplish this without hacking up the bootConfig.c code?  We
>need to download the firmware through our unique loader.  If the
>firmware detects the presence of a communication board (ethernet and
>serial), it will provide the WDB interface for debugging.
>


Simply modify the usrLoadSym.c and usrNetwork.c code (keying on "scsi")
to see how they fixed it for booting from local devices.  You'll also
need to add the device name "nic" to your "other" field in the boot
parameters (NVRAM).
HTH,
===============================================================================
     __       Real-Time System Development, Integration, Training and Services
    //\\
   //  \\     Mike Anderson
  // /\ \\    Chief Engineer              Voice : (703) 448-0210 ext. 235
// /  \ \\   SPARTA, Inc.                FAX   : (703) 893-5494
//   \    \\  7926 Jones Branch Drive     EMAIL : mea@mclean.sparta.com
\\    \   //  Suite 900                   Web   : http://www.mclean.sparta.com
\\ \  / //   McLean, VA 22102
  \\ \/ //                             "Software development is like making
   \\  //                               a baby... You can't make a baby in one
    \\//                                month by impregnating nine women.
     --  "Pride in Performance"         Some things just take time."
===============================================================================


--MimeMultipartBoundary--


From 33205100@aol.com  Thu Oct  9 22:25:39 1997
From: 33205100@aol.com
Date: Thu Oct  9 22:25:42 PDT 1997
Subject: Deliverable Mail


/////////////////////////////////////////////////////////////////
CREDIT CENTER***CREDIT CENTER***CREDIT CENTER***CREDIT CENTER
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
If you received this electronic mailing by mistake, please pardon the
intrusion.

Dear Friend;

Are you or somebody you know experiencing credit problems?  If the
answer is yes, our special offer might be just the cure you have been
looking for.  No longer will you have to live like a second class
citizen due to a bad credit report.  Bad credit can happen due to a
variety of reasons and can happen to anybody.  There is no profile of a
person with credit problems, it can happen to anybody.  In fact, one in
three Americans live with some form of bad credit.  Fortunately, the
United States Government has passed many laws under the Fair Credit
Reporting Act, all of which protect your rights as a consumer.  Our
manuals will show you how to use these Federal Laws to your advantage to
repair your credit, obtain a new credit report or add good credit to
your credit report!

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
*****  'THE FRESH START PROGRAM II' (TM) *****

Have you recently been denied credit due to a bad credit report?  Have
you seen your credit report lately?  If your answered yes, then you need
to take advantage of the 'FRESH START PROGRAM II' (TM) NOW!  This manual
is unique in that it shows you how to obtain a brand new Social Security
number which will allow you to open a brand new, fresh credit report. 
Using guidelines set forth by the Social Security Administration, this
easy-to-follow manual will guide you through the steps necessary to
obtaining your new Social Security Number.  Regardless of your present
credit, you can open a new credit file free of ANY judgements,
charge-off's, foreclosures, bankruptcies, bad credit, tax liens's,
garnishments, late pays and inquiries.

////////////////////////////////////////////////////////////////

***** THE 'WorldCard' (TM) INSTANT CREDIT Credit Card *****

This program is very simple.  It extends to you a $5,000 instant line of
credit in the exclusive 'WorldCard' (TM) Wholesale Purchasing Club. 
This credit card comes to you with absolutely NO credit checks!

You will receive the 'WorldCard' (TM) Wholesale Purchasing Club catalogs
with over 5,000 items to choose from.  Jewelry, toys, tools, gift items,
housewares and much more are available.  After your first purchase, each
quarter you will receive updated catalogs with brand new items
available.

You will be required to make a small down-payment with your orders
and then make a monthly payment of 10% of the unpaid balance.  It's just
that simple!  Your good credit with WorldCard will be reported to the
credit bureau's, which will show your excellent credit!

As a special BONUS, if you take advantage of this offer within the next
24 hours, you will be given $200 in cash certificates!  $100 may be used
on the first $300 unpaid balance and the other $100 may be used after
you have charged $600 on your new account.

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

***** 'STOP COLLECTION AGENCIES COLD' (TM) *****

If you have ever experienced the humiliation of a collection agency
calling you at home or at work, then this hard-hitting manual will put
an IMMEDIATE stop to those calls!

Millions of collection agency telephone calls and letters are made or
sent each week to those who owe money.  Consumers have been helpless to
stop the harassment of these predators, UNTIL NOW!

This manual will show you how to put an end to these telephone calls and
letters, INSTANTLY!  At the same time, these collection agencies will be
forbidden to EVER call or write you again!!  This manual has the inside
secrets the credit world does not want you to have!

Gain back your self-esteem and piece-of-mind by ordering 'Stop
Collection Agencies Cold' (TM) TODAY!!!

/////////////////////////////////////////////////////////////////

***** 'THE COMPLETE CREDIT REPAIR KIT' (TM) *****

Good credit empowers you to qualify for personal loans, automobile
loans, business loans, credit cards, home loans and other types of
credit you may apply for at the very best interest rate.  As you know,
if you don't have excellent credit, lenders will NOT extend credit to
you!  The Fair Credit Reporting Act will allow you to legally and
permanently ERASE the negative information on your credit bureau file.

Anything damaging to your credit, such as bankruptcies, judgements, tax
liens, slow pays, repossessions, charge off's, foreclosures, late
payments, student loans and inquiries may be permanently and legally
REMOVED from your credit report!

If bad credit on your credit file is standing in the way of your
achieving the American dream, then order 'The Complete Credit Repair
Kit'(TM) NOW!!!

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

IMS
5201 Kingston Pike, Suite 6-216
Knoxville, Tennessee 37919

YES, please rush the manuals checked below:

_____   'The Fresh Start Program II' (TM)         	$34.95

_____   'WorldCard' (TM) Credit Card            	$34.95

_____   'Stop Collection Agencies Cold' (TM)   	 $34.95

_____   'The Complete Credit repair Kit' (TM)   	$34.95

_____   TAKE ALL FOUR (SAVE over $39)   	$99.95


$____________  Sub-Total

$4.95    Shipping & Handling for each manual ordered
$15.00   RUSH Delivery (Optional)

$____________  TOTAL ENCLOSED

I Have Enclosed (Check One):
_____   Cashiers Check
_____   Money Order
_____   Personal Check
_____   FAX ORDER


SHIPPING INFORMATION:

Name:__________________________________________________
Address:________________________________________________
City:_______________________State:________Zip Code:_________
Telephone Number:___________________

Your E-Mail Address (For order confirmation):___________________

////////////////////////////////////////////////////////////////

FREE  -  FREE  -  FREE  -  FREE  -  FREE  -  FREE  -  FREE  -  FREE  

Fax your order within 24 hours and receive a special Kodak film book
entitling you to
100 rolls of Kodak film as well as five 6X10 and ten 4X5 enlargements
FREE!

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

Please tape your signed personal check in the space below and fax to
IMS.  The 24 hour facsimile number is 1-423-681-6520.

******************************************************

FAX 1-423-681-6520

Please TAPE Your Check Here

Please SIGN Your Check

******************************************************

THANK YOU For Your Order!!!








From daemon@csg.lbl.gov  Fri Oct 10 04:01:36 1997
From: daemon@csg.lbl.gov
Date: Fri Oct 10 04:01:40 PDT 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Fri Oct 10 04:01:33 PDT 1997

        Subject: Re: Tornado 1.0.1 and the Standard Template Library

-------------------------------------------------------

Newsgroups: comp.os.vxworks
Subject: Re: Tornado 1.0.1 and the Standard Template Library
Date: Thu, 09 Oct 1997 23:12:26 -0700
From: Gregor Overney 
Organization: AT&T WorldNet Services
Message-ID: <343DC74A.7FF15F44@worldnet.att.net>
References: <01bccde5$aff49d60$b10614ac@evolution>

Get SGI's STL implementation: http://www.sgi.com/Technology/STL/. Don't
waste your time with HP's original STL implementation. SGI's version
works much better and is "easy" portable.

Or try Greenhill's compiler which comes with STL support and runs under
VxWorks 5.3.1.

Don't touch Microsoft's STL implementation for VC++ 5.0. It does not
work correctly and does not support exception handling.

gregor

PS: Remember, Tornado and all its tools are fine tuned for Sun
workstations.
Use them on NT and 95 at your own risk ;---()


Jonathan D.R. Martin wrote:

> I am trying to compile a C++ class under Tornado 1.0.1.  The class
> uses
> template classes from the standard template library (stl).  Tornado
> does
> not provide the stl header files.  Does anyone know if stl headers for
>
> VxWorks are available?  I tried to get Tornado to compile the class
> using
> the stl headers from both the Borland and Microsoft Win32 C++
> compilers but
> did not have any luck due to modifications made my each vendor.
>
> The headers that I need in particular are:
>
> algorith
> function
> iterator
> memory
> set
> stdexcept
> string
> tree
> vector
>
> Jonathan Martin
> Newnes Machine Ltd.




---------------------------

End of New-News digest
**********************


From ed.klein@wrs.com  Fri Oct 10 05:52:42 1997
From: Ed Klein 
Date: Fri Oct 10 05:52:45 PDT 1997
Subject: Wind River Seminar
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"




                          WIND RIVER SYSTEMS

                         in Collaboration with
                         Spyglass and JavaSoft

                                Presents

                       INTERNET TECHNOLOGIES FOR
                         EMBEDDED APPLICATIONS


                  **** Wednesday October 22, 1997 ****

				Holiday Inn Express
                		1550 E. Dundee Road
                		Palatine, IL  60067
                


This seminar is intended for Software Engineers and Systems Professionals
who are involved in the design, building and testing of embedded and/or
real-time applications  AND  who need to make their products Internet-ready.

The focus of this seminar will be on connecting embedded devices to the
Internet/Intranet.


        - Learn how the technologies fit into an embedded application:
                - Web Server (HTTP Daemon)
                - Web Browser
                - Java

        - Understand the infrastructure requirements:
                - Scalable, real-time operating system
                - Network protocols
                - Graphics support


SPEED DEVELOPMENT * IMPROVE PORTABILITY * ENHANCE COMPETITIVENESS


A select number of Wind River partners will also be available following
lunch to discuss their products with you.


                       Agenda

8:30 - 9:00     Registration and coffee

9:00 - 12:00    Overview Presentations
                OS Foundations
                Web Server
                Web Browser
                Java Presentation
		  Web Server Demonstration

12:00 - 1:00    Lunch & Videotape

1:00 - 3:00     Vendor Fair
                Q & A


Location:       Holiday Inn Express
                1550 E. Dundee Road
                Palatine, IL  60067
                (847) 934-4900

                (Corner of RT53 and Dundee Road)


This seminar is free, but seating is limited.  PLEASE RSVP by OCTOBER 17th.
Either call (847) 490-6062 or email mbayers@wrs.com (Provide your Name, 
Company/Department, Phone Number and email address). 

We look forward to seeing you on the 22nd!



Ed Klein						
Field Applications Engineer			             847-490-6062		
Wind River System					(fax) 847-490-6511

1901 N. Roselle Rd #800				http://www.wrs.com
Schumburg, Il  60195
--MimeMultipartBoundary--


From koch@tcville.ES.HAC.COM  Fri Oct 10 06:50:28 1997
From: koch@tcville.ES.HAC.COM (Kim Chacon)
Date: Fri Oct 10 06:50:32 PDT 1997
Subject: system() command for VxWorks

VxWorks users,

In an effort to port some code from Sun-Unix to VxWorks-PPC, I
just discovered today that the system() command provided by
VxWorks is not implemented.  It must merely be a stub routine,
since there is indeed a system() symbol to resolve my link,
but when called, the routine does nothing.

I looked in the VxWorks Reference Manual (version 5.3.1) and
confirmed that this routine is indeed provided but it is
also unimplemented.

Has anyone out there implemented a system() call for VxWorks
that they would be willing to share?

I realize that I could create one myself by parsing the
single string that usually gets passed to system() into
individual argument strings needed by a call to sp() or
taskSpawn().  However, I'd rather not do that, if if someone
can provide me with one.

Thank you and regards,
Kim M. Chacon
Hughes Aircraft Company
El Segundo, California   USA
phone: (310) 616-6471
email: chacon@tcville.es.hac.com



From daemon@csg.lbl.gov  Sat Oct 11 04:00:49 1997
From: daemon@csg.lbl.gov
Date: Sat Oct 11 04:00:54 PDT 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Sat Oct 11 04:00:45 PDT 1997

        Subject: Q:debugging w/Tornado

-------------------------------------------------------

Newsgroups: comp.os.vxworks
Subject: Q:debugging w/Tornado
Date: Fri, 10 Oct 1997 09:37:40 -0500
From: Steve Glicker 
Organization: Eaton Semiconductor Equipment Div., Austin, TX
Message-ID: <343E3DB4.2781E494@aus.etn.com>

We would greatly appreciate information on Tornado debugging
software/tools
to address the following problem.

We have a memory overwrite problem which is causing our system to crash,
but it does not always crash in the same way or place.  Our program
consists of about 250kloc of C with about 50 tasks.  Our last
source-code
change was large and there no way to break it up.  (We incorporated some
interdependent software subsystems which were developed at another
site).

We are particularly interested in recording memory changes on a task-
by-task basis along with execution traces.  This information would,
ideally, allow us to backtrack to the cause.

We are considering writing our own tool to do this with the Tornado
tools (API library) and (perhaps) our VME CodeTEST tool.  However we
would prefer to use a tool which is already in place and working.


Regards,

Steve Glicker
Eaton, Austin
sglicker@aus.etn.com

---------------------------

End of New-News digest
**********************


From cgrames@mdc.com  Sat Oct 11 12:27:02 1997
From: Charlie Grames 
Date: Sat Oct 11 12:27:05 PDT 1997
Subject: MVME2604 and Ethernet Switches
All,

I am assisting a project that is using MVME2604 boards in combination with an SMC TigerSwitch
8 Ethernet switch.  Noteworthy about this switch is that it uses a store and forward algorithm
which does not forward "bad" packets.  We have had serious problems with dropped packets with
most (but not all) of the 2604 boards we have tested.  We have NOT had any problems with plain
hubs, switches that do not use store and forward, or the 3Com Super Stack II switch (regardless
of forwarding mode).

Has anyone else run into similar problems, and, if so, how did you resolve them?  Any feedback
would be greatly appreciated.

Charlie Grames
The Boeing Company
(314) 233-1956
Charles.R.Grames@boeing.com


From ugema10@bonsai.de  Sun Oct 12 22:43:13 1997
From: Dirk Hilger 
Date: Sun Oct 12 22:43:16 PDT 1997
Subject: Looking for VxWorks Profibus support
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"

Luca wrote,

>Hello to everyone,

>I am looking for products supporting Profibus in VxWorks environment for
>VME bus or IP-modules.

>Can anyone help me?

>Thank you

>Luca Rodda

We are using a vsbc32 from PEP with a CXM-PFB12 module that supports PROFIBUS
under vxWorks. We communicate with two motor controllers using PROFIBUS DP
and its working fine. The CXM-PFB12 can be used as a master or slave and
it's also supporting FMS.

Best regards
Holger Niebaum

--MimeMultipartBoundary--


From smith.ian@traf.com  Mon Oct 13 06:45:20 1997
From: smith.ian@traf.com
Date: Mon Oct 13 06:45:23 PDT 1997
Subject: expansion memory on Motorola processor
--MimeMultipartBoundary
Content-Type: text/plain; charset=US-ASCII; name="expansion"
Content-Transfer-Encoding: 7bit

Hi

Can anybody confirm if a second memory mezzanine board on Motorola 
processors 167/177/187 is supportable using vxWorks 5.3, and if so what 
do I need to do to acess it.

167bug rom tells me the memory is OK, but.all attempts so far fail to 
see the extra 8 Mb from vxWorks. I adjust LOCAL_MEM_SIZE and 
sysPhysMemDesc [] parameters but still get bus error from m or d 
commands from shell.

 I see pointers to ONLY  the first memory board in sysProcNumSet() and 
sysMemTop() functions and now suspect the second board is not supported 
by vxWorks.

Thanks,
Ian

===================================
Ian D. Smith
(smith.ian@traf.com)
Phone:  +44  (114) 291 4506
Fax:    +44   (114) 291 4026
Kvaerner Metals (Automation)
Prince of Wales Rd, Sheffield. S30 1TY
England
===================================



--MimeMultipartBoundary--


From msa@redifon.com  Tue Oct 14 02:10:09 1997
From: Mark Armstrong 
Date: Tue Oct 14 02:10:12 PDT 1997
Subject: Re: expansion memory on Motorola processor
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Ian Smith wrote:

>  I see pointers to ONLY  the first memory board in sysProcNumSet() 
> and sysMemTop() functions and now suspect the second board is not
>  supported by vxWorks.


	I can't speak from experience of vxWorks 5.3, but the BSP for 
	the MVME167 for vxWorks 5.1.1 had a small fault with respect 
	to the second memory controller on the board.  The BSP 
	completely ignored it and therefore could not support a 
	second memory mezzanine. So historically speaking (1994),
 	vxWorks doesn't support the hardware you have.  Others may 
	know better.  

	To set our board for an extra 16Megs we had to modify the 
	function sysHwInit(void) in config/mv167/sysLib.c to include 
	the following register initialisations.

#if MEGS32
    /*  Change to switch on the second memory controller. */
    *((char *) 0xfff4311c) = cpuSpeed;
    *((char *) 0xfff43114) = 0x1000000 >> 24;	/* Base address */
    *((char *) 0xfff43118) = (MEMC_RCR_RAMEN | MEMC_RCR_SWAIT);
#endif

	We also had to make nasty little changes to get sysMemTop 
	etc. to work correctly that I'm sure you can find your own  
	way round.


|                   |           Mark Armstrong
|   \|/ ____ \|/    |           msa@redifon.com
|    @ / ;; \ @     |           Redifon MEL Ltd.
*---/_(  __  )_\ ---*           United Kingdom.
--MimeMultipartBoundary--


From daemon@csg.lbl.gov  Tue Oct 14 04:01:36 1997
From: daemon@csg.lbl.gov
Date: Tue Oct 14 04:01:40 PDT 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Tue Oct 14 04:01:32 PDT 1997

        Subject: Re: Inconsistent shared message queue structures

-------------------------------------------------------

Newsgroups: comp.os.vxworks
Subject: Re: Inconsistent shared message queue structures
Date: 13 Oct 1997 13:01:02 GMT
From: cchapman@cchapman.gtri.gatech.edu (Charles H. Chapman)
Organization: Georgia Institute of Technology
Message-ID: 
References: 

On Tue, 7 Oct 1997 15:10:32 -0700, Andy Van Pelt  wrote:
>  I have a system where a task calls msgQReceive on a shared memory message
>queue.  At some point, the messages start piling up.  When I do a task trace
>on the task, I can see the call to msgQSmReceive, and I can see the call to
>semSmTake.  When I do a semSmShow on the semaphore address listed in the
>call to semSmTake, it says that there are *no* tasks pending on it!
>
>  Even wierder, the problem is fixed by doing a telnet on a different card
>across shared memory.  I don't even have to type any characters in the
>telnet session, just the making the connection itself clears the problem up.
>
>  This happens with three different task/message queue combinations at the
>same time.  I'm running with VxWorks 5.3 & VxMP on Motorola MV177 (68060)
>cards and a Performance Technologies PT340 card (68EC030).  The problem
>actually shows up on tasks running on the PT340 card.  One theory is that
>the PT340-initiated writes are somehow "failing" to some extent, every now
>and then, leaving the shared memory data structures in an inconsistent
>state.

That sounds like a problem I was having with a similar set-up, in this case
a General Micro Systems 68040 main CPU and a Performance Technologies
PT330 (68020) serial card.  Apparently the mail box interrupts on the PT330
are unreliable, so using the mail box interrupt method for shared memory
communication on these cards sometimes results in a missed interrupt causing
all shared memory objects to be "stuck".  When you telnetted across shared
memory network, this causes -another- interrupt to be generated using the
shared memory network driver (as opposed to the shared memory -object-
driver) which sorta kicks the PT card into action.

One way to solve the problem (if this is the problem in your case) is to 
switch to the polled method of shared memory communication for the PT
card.  Another method  I've tried that seems to work (but is a little kludgie)
is to continue using mail box interrupts but to manually write to the
mail box after every call to msgQSend for messages to the PT card.  That
seems to have worked -- I haven't had any hang-ups due to missed interrupts
since implementing this change.

Hope this helps!

Chuck


---------------------------

End of New-News digest
**********************


From bm417929@muenchen.org  Tue Oct 14 07:52:08 1997
From: bm417929@muenchen.org (Max Seeger)
Date: Tue Oct 14 07:52:12 PDT 1997
Subject: Support for MPC740 and 750
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"

Hi,

does anybody know if and when WRS is going to support
the new 740 and 750 PowerPC processors?

Does anybody know, who offers and a VME board based
on this processor and can deliver?

Thanks,
Max

--MimeMultipartBoundary--


From kheck@cs5.dasd.honeywell.com  Tue Oct 14 14:24:06 1997
From: Kevin Heck 
Date: Tue Oct 14 14:24:09 PDT 1997
Subject: Network interface device driver
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Has anyone written a network interface device driver
using the structure described in the Network Driver
chapter of the VxWorks BSP Porting Kit reference manual?  Specifically,
I am writing a custom driver using
TCP/IP as the communication layers and using 1553 as
the link layer, (which necessarily leads to writing
a custom driver).  I am trying to understand how my 
driver will need to be written to transmit data
across the 1553 bus.  In the Example Network Interface
Driver that they have at the back of the chapter, it
appears that the driver must supply an xxOutput 
service as well as an xxStartOutput service.  It
appears that after the driver has been attached
to the communication layers, that upon an application
layer "Write" command of some kind, the communication
layers will call the xxOutput service in the driver.
In their example driver, this service calls a VxWorks
routine ether_output.  In the book it says that this 
routine is optional.  In my case, I would think that
I wouldn't want to use it since I'm not using 
ethernet.  However, it says that this routine will 
"add the data to be output onto the driver's queue".
After this is done, it looks like xxStartOutput will 
examine the queue head and continue to extract data
off of the queue and move it out to the driver until
the queue is empty. This will not work for my 
application due to a limitation in our 1553 interface.
I will only be able to send one element of the queue 
across the 1553 bus in any given time period.  I will 
then have to reenter this procedure the next time 
period and send the next queue element across the bus.
The questions I have about the use of these routines
are as follows:

1) Can I skip this ether_output routine and examine this
queue of "mbufs" in my own routine somewhere?  Or does
this queue not even get set up unless ether_output is 
called?

2) Is there some other way to set up this queue other
than using ether_output?

If someone can shed some light on the internal workings
of these queues and how the communication layers are 
using them it would be greatly appreciated. 

                       Thank you, 
                              Kevin Heck

e-mail - kheck@cs5.dasd.honeywell.com
--MimeMultipartBoundary--


From raman@net.com  Tue Oct 14 20:08:51 1997
From: raman@net.com (Raj Raman)
Date: Tue Oct 14 20:08:54 PDT 1997
Subject: Re: Network interface device driver
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit


Kevin,
	If I understand you right, here's what can be done:
	
	- you need to implement the 'customized output' function
	that needs to be added to "ptrifnet->if_output" hook
	before calling if_attach
	
	- the vxWorks provides the following argument support to
	this function:
	(struct ifnet *ifp, register struct mbuf *m, struct sockaddr *dst)
	
	The "m" in this case points to the IP packet; you need to leverage
	the values in the "dst" which contains the destination details
	
	You could derive the size of the IP packet from the IP header using
	"mtod" function; Knowing the size of the IP packet, you could
	use "copy_from_mbuf" function to copy it to your buffer. My 
	understanding is that the above function internally frees the
	mbuf(s) after copying it.
	
	
	
% 1) Can I skip this ether_output routine and examine this
% queue of "mbufs" in my own routine somewhere?  Or does
% this queue not even get set up unless ether_output is 
% called?

	You can skip the "ether_output"; The "m" pointer passed
	to your customized output function should have the queue
	of mbuf(s)
	
% 2) Is there some other way to set up this queue other
% than using ether_output?

	This queue of "mbufs" is set up internally by vxWorks,
	before your output function is executed.
	
	Hope this helps,
	
- raj

Raj Raman		(raman@net.com)
Network Equipment Technologies (net)
--MimeMultipartBoundary--


From philweb@sprynet.com  Wed Oct 15 02:06:40 1997
From: Philippe Weber 
Date: Wed Oct 15 02:06:43 PDT 1997
Subject: Re: Support for MPC740 and 750
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

On Tuesday, October 14, 1997 16:52, the vxWorks Users Group Exploder =
wrote:
>=20
> does anybody know if and when WRS is going to support
> the new 740 and 750 PowerPC processors?
>=20
> Does anybody know, who offers and a VME board based
> on this processor and can deliver?
>=20
> Thanks,
> Max
>=20

VI Computer offers a VME board, the power-3e, based on the 740 processor =
at 266 MHz. It supports VxWorks 5.3.1 and is available in production. =
FYI, VI also proposes a PowerPC 760 at 300 Mhz, aka PowerPC604e3 and =
Mach 5, on the Power-4e and -4B.

Regards,

--=20

*************************************
* Philippe WEBER
* THEMIS Computer / V*I Computer
* 1, rue des Essarts
* 38610 GIERES, FRANCE
* Voice : +33 (0)476 59 60 46
* Fax   : +33 (0)476 63 00 30
* E-mail: phil@themis.com
*************************************





--MimeMultipartBoundary--


From business@mail.t-1net.com  Wed Oct 15 05:35:28 1997
From: business@mail.t-1net.com
Date: Wed Oct 15 07:36:53 PDT 1997
Subject: Get Paid To Advertise!
     Hello!

     If you would like to find out how you can get paid to advertise 
your own product, service or business on the Internet, please reply
to  and type 'info' in the
subject line.

     If this has been sent to you in error, please accept my apology
and reply to the forementioned address and type 'remove' in the 
subject line to be taken off my mailing list. 


From daemon@csg.lbl.gov  Wed Oct 15 04:01:35 1997
From: daemon@csg.lbl.gov
Date: Wed Oct 15 07:37:11 PDT 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Wed Oct 15 04:00:30 PDT 1997

        Subject: TCP/IP sockets

-------------------------------------------------------

Newsgroups: comp.os.vxworks
Subject: TCP/IP sockets
Date: 15 Oct 1997 00:47:26 GMT
From: walid@cco.caltech.edu (Walid A. Majid)
Organization: California Institute of Technology, Pasadena
Message-ID: <6213qu$t4b@gap.cco.caltech.edu>

Hello,

In my application I am transferring data from a vxWorks board to a Sun Ultra over
ethernet.  The numbers I get are quite low around 200 KBytes/sec.  If I run the same
package over two Sun machines on the same ethernet line, I get a little over        
1 Mbyte/sec performance.  Are the socket network drivers for vxWorks that slow?  If
anyone has any experience with the performance of sockets on vxWorks, I would
appreciate any suggestions or ideas.

Regards,

Walid

---------------------------

End of New-News digest
**********************


From pat.shaw@canada.cdev.com  Wed Oct 15 08:26:30 1997
From: "Shaw, Pat" 
Date: Wed Oct 15 08:26:46 PDT 1997
Subject: Flash BOOT
This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------ =_NextPart_000_01BCD95C.7CF60D90
Content-Type: text/plain

Hi :
	We have designed and built a Pentium based card with 16Mb of
Flash on board that resides in high memory ( ie: e0000000). Most PC
based flash system using a 64Kb memory window in lower memory to map the
flash into there. We decide to provide a contiguous portion of memory
for flash use. We are trying to figure out how we can get the vxworks to
boot from there.  My initial though was that if we put a DosFs in the
flash and a command line to boot from this file system  The problem is
that we are currently running standalone in the sense that the card is
deeply embedded and does not have ( in it's final configuration ) a
floppy or hard drive connected. So the question how to get a the flash
to boot?. 

We have our BIOS manufacture who will jump to a location in flash after
bootup. This appears to work except that to jump we might need to be in
protected mode. I am not  sure if this would work with Vxworks assuming
it is in real mode at bootup. 

I was wondering if anybody has had some similar experiences with booting
from flash and what they did to overcome the limitations.


Thanks,

Patrick Shaw

------ =_NextPart_000_01BCD95C.7CF60D90
Content-Type: application/ms-tnef
Content-Transfer-Encoding: base64
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------ =_NextPart_000_01BCD95C.7CF60D90--


From nrd1jrb@nrd.ups.com  Wed Oct 15 09:19:29 1997
From: Jim Bucciferro 
Date: Wed Oct 15 09:19:33 PDT 1997
Subject: Memory allocation and realloc
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

I am trying to use realloc to allocate a growing block of memory.  I
have been able to use it with other O/S's , particularly Windows NT 4.0,
but with VxWorks, I get a mempartfree error on the second realloc.

Here is the code:
...
    syncFile *files;
    size = 0;
    count = 0;
    while(readfile)
    {
      size++;
      files = (syncFile *) realloc( files, sizeof(syncFile) * size);
      count++;
    }
...
syncFile is a basic structure.

The idea is to grow the files array with each item read from the file.
On the second realloc I get a memPartRealloc: invalid block xxxx in
partition xxxx.

Is this the proper way to allocate memory with VxWorks or do I need to
create a memory partition first.

Thanks.
JB



--MimeMultipartBoundary--


From phil.quiney@gecm.com  Thu Oct 16 01:40:35 1997
From: Philip Quiney 
Date: Thu Oct 16 01:40:38 PDT 1997
Subject: Re: Memory allocation and realloc
Hi JB,

I tried this bit of code on a vme147 card and it worked fine. I think your
problem could be not setting the files variable to NULL before calling realloc
for the first time. It is the only difference I could spot.


Regards

Phil Q

#include 
#include 

typedef struct {
int fred;
int bill;
char *jim;
} syncFile;


void re (void)
{
    syncFile *files = (syncFile *)NULL;
    int size = 0;
    int count = 0;

    while (count<5)
    {
        size++;
        files = (syncFile *) realloc(files, sizeof(syncFile) * size);
        count++;
        printf ("SyncFile ptr is %x, size is %d\n", (int) files, size);
    }

    printf ("Done\n");
    free (files);
}



> I am trying to use realloc to allocate a growing block of memory.  I
> have been able to use it with other O/S's , particularly Windows NT 4.0,
> but with VxWorks, I get a mempartfree error on the second realloc.
> 
> Here is the code:
> ...
>     syncFile *files;
>     size = 0;
>     count = 0;
>     while(readfile)
>     {
>       size++;
>       files = (syncFile *) realloc( files, sizeof(syncFile) * size);
>       count++;
>     }
> ...
> syncFile is a basic structure.
> 
> The idea is to grow the files array with each item read from the file.
> On the second realloc I get a memPartRealloc: invalid block xxxx in
> partition xxxx.
> 
> Is this the proper way to allocate memory with VxWorks or do I need to
> create a memory partition first.
> 
> Thanks.
> JB
> 
> 

===============================================================================
Phil Quiney          _/_/_/_/        _/  _/_/_/_/  email : phil.quiney@gecm.com 
GEC Marconi         _/    _/        _/  _/    _/     tel : +44 1245 353221x3914
Communications Ltd _/_/_/_/  _/    _/  _/  _/_/      fax : +44 1245 275448
Chelmsford        _/        _/_/_/_/  _/_/_/_/
England CM1 1PL                              _/                             
===============================================================================


From daemon@csg.lbl.gov  Thu Oct 16 04:04:29 1997
From: daemon@csg.lbl.gov
Date: Thu Oct 16 06:06:24 PDT 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Thu Oct 16 04:00:35 PDT 1997

        Subject: Re: Time and Date on x86
        Subject: Time and Date on x86

-------------------------------------------------------

Newsgroups: comp.os.vxworks
Subject: Re: Time and Date on x86
Date: Wed, 15 Oct 1997 14:27:05 +0100
From: Chris Wallett 
Organization: Cranfield Precision
Message-ID: <3444C4A9.F6DC12A9@landis-lund.co.uk>
References: <3444894E.6880@ipk.fhg.de>

Dieter Sorowka wrote:

> Hi,
>
> I am working on Pentium target and have some problems with absolute
> time
> and date. Every file produced by the target has the date 1.1.1980.
>
> We implemeted a driver for reading the CMOS Time and Date and we are
> searching for a way, to pass these data to the vxWorks systems.
> VxWorks
> offers many time functions (ansiTime), but there is no function for
> writing the absolut time to the system. Maybe there is a global data
> structure. Is there anybody, who knows someting about this.
>
> Thank You in advance
>
> Dieter Sorowka

   Try clock_settime()


---------------------------

Newsgroups: comp.os.vxworks
Subject: Time and Date on x86
Date: Wed, 15 Oct 1997 10:13:50 +0100
From: Dieter Sorowka 
Organization: IPK
Message-ID: <3444894E.6880@ipk.fhg.de>
Reply-To: dieter.sorowka@ipk.fhg.de

Hi,

I am working on Pentium target and have some problems with absolute time
and date. Every file produced by the target has the date 1.1.1980.

We implemeted a driver for reading the CMOS Time and Date and we are
searching for a way, to pass these data to the vxWorks systems. VxWorks
offers many time functions (ansiTime), but there is no function for
writing the absolut time to the system. Maybe there is a global data
structure. Is there anybody, who knows someting about this.

Thank You in advance

Dieter Sorowka

---------------------------

End of New-News digest
**********************


From BariF@utrc.utc.com  Thu Oct 16 06:59:16 1997
From: "Bari, Farooq              UTRC" 
Date: Thu Oct 16 06:59:19 PDT 1997
Subject: using m-systems disc on chip and flash disk with vxworks
--MimeMultipartBoundary
Content-type: text/plain
Content-transfer-encoding: 7BIT

Hi,

I am trying to use m-systems "disc on chip" and "flash disk (pc104)" but
have not been able to find any drivers that can be used in the default
bootline parameters. Is anybody aware of drivers for these devices or
has any expeirencing booting from these devices ?

Farooq Bari

--MimeMultipartBoundary--


From eckardt.suzanne@orbital.com  Thu Oct 16 09:22:57 1997
From: "Eckardt.Suzanne"
Date: Thu Oct 16 09:23:00 PDT 1997
Subject: Priority Inheritance with Mutex Semaphores
--MimeMultipartBoundary
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit


     
     Can anyone clarify the behavior of the priority inheritance algorithm 
     used with mutex semaphores?  Does it activate immediately when a 
     higher priority task blocks on the semaphore?  Does the semTake 
     timeout value (discrete timeout versus WAIT_FOREVER) make a 
     difference?  
     
     My scenario: tasks t1, t2, and t3 of high, medium, and low priority 
     respectively.  t3 is running and takes the mutex semaphore.  t3 is 
     then preempted by t1.  t1 runs for awhile and then attempts to take 
     the mutex semaphore using a discrete timeout value.  t1 is now 
     blocked.  t2 is ready.  What happens next??
     
     Suzie
     
     
     -------------------------------------------------------------------
     -                                                                 -
     - Suzie Eckardt, Principal Software Engineer                      -
     - Orbital Sciences Corp.                                          -
     - 21700 Atlantic Blvd.         email: eckardt.suzanne@orbital.com -
     - Dulles, VA  20166            Phone: 703-406-5478                -
     -                                                                 -
     ------------------------------------------------------------------- 


--MimeMultipartBoundary--


From jrm@mclean.sparta.com  Thu Oct 16 09:58:54 1997
From: "John R. Moore III" 
Date: Thu Oct 16 09:58:57 PDT 1997
Subject: Re: Priority Inheritance with Mutex Semaphores
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

If you have created the semaphore with The SEM_INVERSION_SAFE and
SEM_Q_PRIORITY options, t3 will be raised to the priority of t1 and be
allowed to run until it gives the semaphore.  t3 then returns to its
original priority and t1 will run.  t2 won't run until t1 gives up the
CPU.
>      Can anyone clarify the behavior of the priority inheritance
> algorithm
>      used with mutex semaphores?  Does it activate immediately when a
>      higher priority task blocks on the semaphore?  Does the semTake
>      timeout value (discrete timeout versus WAIT_FOREVER) make a
>      difference?
> 
>      My scenario: tasks t1, t2, and t3 of high, medium, and low
> priority
>      respectively.  t3 is running and takes the mutex semaphore.  t3
> is
>      then preempted by t1.  t1 runs for awhile and then attempts to
> take
>      the mutex semaphore using a discrete timeout value.  t1 is now
>      blocked.  t2 is ready.  What happens next??
> 
>      Suzie
 
-- 

------------------------------------------------------------ 
|John R. Moore              | Voice -> 703 448 1683 x229   | 
|Principal Engineer         | FAX   -> 703 893 5494        | 
|EMAIL:                     | On the WEB!                  | 
|jrm@mclean.sparta.com      | http://www.mclean.sparta.com | 
|___________________________|______________________________| 
|                                     |                    | 
|     /\\     SPARTA, Inc.            |                    | 
|    ///\\    7926 Jones Branch Drive |                    | 
|   /////\\   Suite 900               |                    | 
|  /////  \\  McLean, VA 22102        |                    | 
| / \\\\\  \\ 703-448-0120            |         	   | 
| \  \\\\\ //                         |                    | 
|  \  //////                          |                    | 
|   \//////   SPARTA                  |                    | 
|    \////    Pride In Performance    |                    | 
|     \//     --------------------    |                    | 
------------------------------------------------------------
--MimeMultipartBoundary--


From froeber@bbn.com  Thu Oct 16 10:00:34 1997
From: Fred Roeber 
Date: Thu Oct 16 10:00:37 PDT 1997
Subject: Re: Time and Date on x86
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dieter Sorowka wrote:

> I am working on Pentium target and have some problems with absolute time
> and date. Every file produced by the target has the date 1.1.1980.
> 
> We implemeted a driver for reading the CMOS Time and Date and we are
> searching for a way, to pass these data to the vxWorks systems. VxWorks
> offers many time functions (ansiTime), but there is no function for
> writing the absolut time to the system. Maybe there is a global data
> structure. Is there anybody, who knows someting about this.

To get time values set for files created using the DOS file system
you probably want to look at the dosFsDateTimeInstall function.
Allows you to hook in a function for setting file dates right. Fred
-- 
   |     Fred J Roeber,  BBN Systems & Technologies     |
   |   4 John Clarke Road   Middletown, RI  02842-5202  |
   |         froeber@bbn.com     401-848-3548           |
--MimeMultipartBoundary--


From YF.Siu@eng.efi.com  Thu Oct 16 12:17:26 1997
From: "Yuet Fung Siu" 
Date: Thu Oct 16 12:17:29 PDT 1997
Subject: Windows 95 compatible DosFs

I'm just looking for useful information on implementation of newer
DOS file systems under VxWorks (32-bit FAT, Win95 long filenames,...).
The original dosFsLib distributed with VxWorks seems to be outdated
and will encounter compatibility problems with removable file systems
on Zip-drives, etc (created under Win95 for instance).

Thanks,

YF


From SHoward@alhra.af.mil  Thu Oct 16 13:09:40 1997
From: SHoward@alhra.af.mil
Date: Thu Oct 16 13:09:59 PDT 1997
Subject: re:Priority Inheritance with Mutex Semaphores
Suzie,

T3 priority is set to equal t1 and t3 completes it use
of the semaphore.  Once it releases the semaphore its
priority is returned to the original value and is preempted
if t1 or t2 is ready to run.

All this is great as long as t1 uses a wait_forever.  If it uses
a value that is relatively small then it could time out before
any of the inversion algorithm stuff is completes (ie. 1-2 ticks).
Remember every operating system action is done on a tick boundary.


Steve Howard
Hughes



From froeber@bbn.com  Thu Oct 16 13:11:59 1997
From: Fred Roeber 
Date: Thu Oct 16 13:12:02 PDT 1997
Subject: Re: Priority Inheritance with Mutex Semaphores
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Susan Eckhardt asked:

> >      Can anyone clarify the behavior of the priority inheritance
> > algorithm
> >      used with mutex semaphores?  Does it activate immediately when a
> >      higher priority task blocks on the semaphore?  Does the semTake
> >      timeout value (discrete timeout versus WAIT_FOREVER) make a
> >      difference?
> >
> >      My scenario: tasks t1, t2, and t3 of high, medium, and low
> > priority
> >      respectively.  t3 is running and takes the mutex semaphore.  t3
> > is
> >      then preempted by t1.  t1 runs for awhile and then attempts to
> > take
> >      the mutex semaphore using a discrete timeout value.  t1 is now
> >      blocked.  t2 is ready.  What happens next??

John R. Moore answered:

> If you have created the semaphore with The SEM_INVERSION_SAFE and
> SEM_Q_PRIORITY options, t3 will be raised to the priority of t1 and be
> allowed to run until it gives the semaphore.  t3 then returns to its
> original priority and t1 will run.  t2 won't run until t1 gives up the
> CPU.

Having at one time extended the mutex semaphore inheritance to an MP
version of VxWorks we developed based on VxWorks source, I feel I
have a pretty good understanding of how they work. Basically, as
soon as the t1 task tries to take semeaphore t3 has, t3's priority
will be boosted to t1's level (assuming some task higher than t1
isn't already blocked on semaphore). t3's priority will not be reduced
to its original level until it releases all semaphores that it is
holding. Thus t1 won't run until t3 releases any semaphores it is
holding (including the one that t1 was pending on). This behaviour
isn't a strictly correct implementation of priority inheritance
but it is an efficient one. In particular, it is wrong with
respect to the handling of nested semaphores. To be totally
correct, t3's priority should always be the highest level of
any tasks it is blocking. Thus, if the t1 task tried to
get a semaphore with a short timeout (the question asked), t3's
priority should be boosted only until t1 times out and becomes
no longer blocked on the semaphore t3 has. VxWorks doesn't work
this way however; t3 will keep t1's priority and thus block t1
from running until t3 releases all semaphores. It also is incorrect
that it doesn't lower its priority as it releases each semaphore it
may be holding. Trying to get this working "correctly" is horrendous
however. It is typically about the hardest thing to get right in
an OS. Just ask the Solaris developers who struggled with it for
ages.

Hope the info helps.   Fred
-- 
   |     Fred J Roeber,  BBN Systems & Technologies     |
   |   4 John Clarke Road   Middletown, RI  02842-5202  |
   |         froeber@bbn.com     401-848-3548           |
--MimeMultipartBoundary--


From julie.prescott@lmco.com  Thu Oct 16 13:37:40 1997
From: julie.prescott@lmco.com
Date: Thu Oct 16 13:37:43 PDT 1997
Subject: Multiple virtual consoles (one per task)
--MimeMultipartBoundary
Content-Type: text/plain

I have multiple tasks running on VxWorks 5.3 and would like to have a
virtual console for each task.  From looking at the User's Guide, it
looks like I can use WTX_CONSOLE_CREATE command to do this and then use
ioTaskStdSet to attach each task to a virtual console.  I'm a new
VxWorks user, so I haven't figured out how to use WTX yet.

Has anyone else tried this?

Thanks
--MimeMultipartBoundary--


From daemon@csg.lbl.gov  Fri Oct 17 04:02:57 1997
From: daemon@csg.lbl.gov
Date: Fri Oct 17 06:46:15 PDT 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Fri Oct 17 04:01:49 PDT 1997

        Subject: Re: Problems with -O2/-O3 and PowerPC
        Subject: Re: taskDelay question
        Subject: Re: Q:debugging w/Tornado
        Subject: taskDelay question

-------------------------------------------------------

Newsgroups: comp.os.vxworks
Subject: Re: Problems with -O2/-O3 and PowerPC
Date: Thu, 16 Oct 1997 09:37:58 -0600
From: Mike Stimpson 
Organization: Utah Scientific, Inc.
Message-ID: <344634D6.41C6@utsci.com>
References: <624ls4$eia@sdaw04.seinf.abb.se> <624vju$fd$1@ncar.ucar.edu>

Chris Webster wrote:
> 
> I don't even dare use -O.  All prior experience has told me some piece
> of code will get optimized away.
> 
> --Chris
> 
We use -O but also -fvolatile, which seems (so far) to work.
- -- 
Mike Stimpson (mstimp@utsci.com)

Opinions expressed are not necessarily those of the management...

The more I work as an engineer, the more I disbelieve the theory of
evolution.

---------------------------

Newsgroups: comp.os.vxworks
Subject: Re: taskDelay question
Date: Thu, 16 Oct 1997 13:10:48 -0700
From: "jonesboy" 
Organization: Okanagan Internet Junction
Message-ID: <625osr$320$1@news.junction.net>
References: <625n7u$bc6@cnn.motown.lmco.com>



Michael D. Stewart wrote in article <625n7u$bc6@cnn.motown.lmco.com>...

>Hello,
>
>My question is pretty simple:  Does a taskDelay function call with a
parameter
>of zero act as a yield?  In other words,  if I call the taskDelay function
>like this:
>
>...
>taskDelay(0);
>...
>...
>...
>
>Will this give up the running tasks processor time to another runnable
task?

    Yes.  It says so explicitly in the manual.

- -=[Matt]=-



---------------------------

Newsgroups: comp.os.vxworks
Subject: Re: Q:debugging w/Tornado
Date: Thu, 16 Oct 1997 18:23:27 +0200
From: Mikael Ohberg 
Organization: ABB Industrial Systems AB
Message-ID: <34463F7F.480F@seisy.mail.abb.com>
References: <343E3DB4.2781E494@aus.etn.com>

Steve Glicker wrote:
> =

> We would greatly appreciate information on Tornado debugging
> software/tools
> to address the following problem.
> =

> We have a memory overwrite problem which is causing our system to crash=
,
> but it does not always crash in the same way or place.  Our program
> consists of about 250kloc of C with about 50 tasks.  Our last
> source-code
> change was large and there no way to break it up.  (We incorporated som=
e
> interdependent software subsystems which were developed at another
> site).
> =

> We are particularly interested in recording memory changes on a task-
> by-task basis along with execution traces.  This information would,
> ideally, allow us to backtrack to the cause.
> =

> We are considering writing our own tool to do this with the Tornado
> tools (API library) and (perhaps) our VME CodeTEST tool.  However we
> would prefer to use a tool which is already in place and working.
> =

> Regards,
> =

> Steve Glicker
> Eaton, Austin
> sglicker@aus.etn.com

Hello !

We have used a debugger from EST Corp. for doing real-time debugging
through the BDM-interface (MC68360). Their tools are not cheap but very
good but are also available to rent.
Se also http://www.wrs.com/html/visionice.html.

Best regards and good luck

Mikael =D6hberg
ABB Industrial Systems AB

---------------------------

Newsgroups: comp.os.vxworks
Subject: taskDelay question
Date: 16 Oct 1997 18:43:09 GMT
From: mstewart@motown.lmco.com (Michael D. Stewart)
Organization: Lockheed Marietta Corp, Moorestown NJ
Message-ID: <625n7u$bc6@cnn.motown.lmco.com>
Reply-To: mstewart@motown.lmco.com

Hello,

My question is pretty simple:  Does a taskDelay function call with a parameter
of zero act as a yield?  In other words,  if I call the taskDelay function 
like this:

...
taskDelay(0);
...
...
...

Will this give up the running tasks processor time to another runnable task?

Mike Stewart 



---------------------------

End of New-News digest
**********************


From mwbrown@dragon.sandia.gov  Fri Oct 17 08:11:54 1997
From: mwbrown@dragon.sandia.gov (Matthew W. K. Brown)
Date: Fri Oct 17 08:11:58 PDT 1997
Subject: GPIB  interface board
--MimeMultipartBoundary
Content-Type: text/plain
Content-Transfer-Encoding: 7bit

VxWorkers:

Could anyone recommend a GPIB interface board for the VMEbus?

It would be nice if it had a VxWorks driver as well!

Thanks!




-------------------------------------------------------------------------
Matthew Brown, SMTS                             Voice: (505) 844-1160
Sandia National Laboratories                    Fax:   (505) 844-7059
P. O. Box 5800    M/S 0965                      email: mwbrown@sandia.gov
Albuquerque, NM  87185-0965
--MimeMultipartBoundary--


From SHoward@alhra.af.mil  Fri Oct 17 12:10:48 1997
From: SHoward@alhra.af.mil
Date: Fri Oct 17 12:10:51 PDT 1997
Subject: remove from list
please remove me from the exploder distribution list.


From bm417929@muenchen.org  Fri Oct 17 14:05:33 1997
From: bm417929@muenchen.org (Max Seeger)
Date: Fri Oct 17 14:05:36 PDT 1997
Subject: Support for MPC740 and 750
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"

Hi,

I tried to get some answers by when WRS
is going to support the 740 and 750 processors.

From the feedback I got, it seems, that I can
use the existing 603/604 development platform.

Is this true? What compiler should I use, a 603
or a 604 compiler?

I would like top use a 750 board. MCG announced
already one, but I cannot get one. I was told,
that they having problems with the hardware.
I anybody else out there with a 750 based board.

Thanks,
Max


--MimeMultipartBoundary--


From daemon@csg.lbl.gov  Sun Oct 19 04:00:58 1997
From: daemon@csg.lbl.gov
Date: Sun Oct 19 04:01:01 PDT 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Sun Oct 19 04:00:54 PDT 1997

        Subject: Re: task guardian for VxWorks

-------------------------------------------------------

Newsgroups: comp.os.vxworks
Subject: Re: task guardian for VxWorks
Date: Thu, 16 Oct 1997 12:03:51 -0700
From: Anthony Wilson 
Organization: Jet Propulsion Laboratory - Pasadena CA
Message-ID: <34466517.6FA09F46@earthlink.net>
References:  <19971015231600.TAA21349@ladder02.news.aol.com>
Reply-To: wilsonae@earthlink.net

look atftp://ftp.atd.ucar.edu/pub/vxworks/vx/
there is a listing of vxworks files. look at taskmon.shar and deadman.shar
they contain code to monitar the state of task and handle recover of errant task



Work9862 wrote:

> We also have an FT requirement for our systems and since all the modern
>  spacecraft CPUs are supported by VxWorks FT is a big problem for us.  Much
>  insight can be gained from the JPL experience on Mars Pathfinder. A paper was
>  presented on this issue at the last embedded systems conference.  The author
>  was kind enough to send us the presentation charts also.
>
> JPL acheived FT via use of "task hooks" to monitor task state after context
>  switching (I am still trying to find out exactly how they did it") to identify
>  memory corruption and task mortality, and lots of checkpoints in their code
>  connected to what they call a "Fault Monitor" system (Details again unknown).
>
> If anybody has some more information I would be happy to share.
>
> Candice
> TRW




---------------------------

End of New-News digest
**********************


From harvey@wrs.com  Sun Oct 19 19:26:32 1997
From: Harvey Wong 
Date: Sun Oct 19 19:26:36 PDT 1997
Subject: Fujitsu MB86965 driver 
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"

Dear VxWorks users,

I'm looking for an ethernet driver for the Fujitsu MB86965.  Any leads would
be greatly appreciated.   Thanks,


Harvey Wong

***************************************************************
Harvey Wong			Wind River Systems, Inc.
Field Applications Engineer,		1010 Atlantic Avenue
Asia-Pacific Operations		Alameda, CA  94501
e-mail:  harvey@wrs.com		Ph:  510/748-4100
				Fax:  510/814-2011

--MimeMultipartBoundary--


From philweb@sprynet.com  Mon Oct 20 01:12:05 1997
From: Philippe Weber 
Date: Mon Oct 20 01:12:08 PDT 1997
Subject: RE: Support for MPC740 and 750
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

On Friday, October 17, 1997 23:05, the vxWorks Users Group Exploder =
wrote:
> Submitted-by bm417929@muenchen.org  Fri Oct 17 14:05:33 1997
> Submitted-by: bm417929@muenchen.org (Max Seeger)
>=20
> --MimeMultipartBoundary
> Content-Type: text/plain; charset=3D"us-ascii"
>=20
> Hi,
>=20
> I tried to get some answers by when WRS
> is going to support the 740 and 750 processors.
>=20
> >From the feedback I got, it seems, that I can
> use the existing 603/604 development platform.
>=20
> Is this true? What compiler should I use, a 603
> or a 604 compiler?
>=20
> I would like top use a 750 board. MCG announced
> already one, but I cannot get one. I was told,
> that they having problems with the hardware.
> I anybody else out there with a 750 based board.
>=20
> Thanks,
> Max
>=20

Correct. The 750 has still serious hardware problems. I don't think =
anybody can deliver and honestly pretend it's working *reliably*. Rather =
try 740 (like VI's Power3e-266).=20

--=20

*************************************
* Philippe WEBER
* THEMIS Computer / V*I Computer
* 1, rue des Essarts
* 38610 GIERES, FRANCE
* Voice : +33 (0)476 59 60 46
* Fax   : +33 (0)476 63 00 30
* E-mail: phil@themis.com
*************************************


--MimeMultipartBoundary--


From vbkumar@wipinfo.soft.net  Mon Oct 20 02:13:12 1997
From: V B Kumar 
Date: Mon Oct 20 02:13:15 PDT 1997
Subject: Re: Priority Inheritance with Mutex Semaphores
--MimeMultipartBoundary
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Thu, 16 Oct 1997, the vxWorks Users Group Exploder wrote:

>      
>      Can anyone clarify the behavior of the priority inheritance algorithm 
>      used with mutex semaphores? 
       Refer the scenario you have mentioned, What happens if t3 is 
holding the resource and t1 is blocked on that resource. Now t2 can 
preempt t3 because its priority is higher but in the bargain t1 will 
remain blocked for an indefinite period of time.( some other task 
whose priority greater than t2/t3 can preempt either of them or t3 
respectively , before t3 release the resource again !)
       Priority inheritance algorithm makes sure that the highest priority 
task blocked on a resource does not remain in blocked state not more than 
the time it takes for the task holding the resource to be finished 
with the resource. This is ensured by awarding the task holding the 
resource, the priority equal to the highest prioty task which is 
blocked on that resource.(, so that no medium priority tasks preempt it 
before it gives up the resource.)

>       Does it activate immediately when a 
>      higher priority task blocks on the semaphore?  Does the semTake 
>      timeout value (discrete timeout versus WAIT_FOREVER) make a 
>      difference?  
Yes, the priority of the task holding the resource is raised as soon as 
the higher priority task blocks on that resource.
The value of semTake timeout value should not make any difference to this
algorithm or in other words  the higher priority task would still return 
error if the lower task is not finished with the task before timer expires,
neither the timeout value would alter the behaviour of the algorithm.

>      
>      My scenario: tasks t1, t2, and t3 of high, medium, and low priority 
>      respectively.  t3 is running and takes the mutex semaphore.  t3 is 
>      then preempted by t1.  t1 runs for awhile and then attempts to take 
>      the mutex semaphore using a discrete timeout value.  t1 is now 
>      blocked.  t2 is ready.  What happens next??

1) t3 will contnue running till it releases the semophore .( because priority
of t3 is equal to the priority of t1 (because t1 is blocked on that 
resource ) and t3 is holding the resource )

2) As soon as t3 releases the semaphore, t1 takes the semaphore and will be 
scheduled.( because t1 priority is higher than t2 & t3 priority drops to its 
original value )

3) And then t2 will be scheduled if no other higher priority task is ready.
( If t2 is the highest priority task ready )
>      
>      Suzie
>      
>
      
Regards

V Bala Kumar


Disclaimer : Its my personal opinion , I may be wrong !


                             \\\ | /// 
                             \\ - -// 
                             (  @ @  ) 
  +------------------------oOOo-(_)-oOOo---------------------------------+ 
  | V Bala Kumar                                                         |
  | Senior Software Engineer           Phone : 080 - 5588422/523 (Off.)  |
  | Wipro Global R&D                           080 - 8398933 (Res.)      |
  | 88, M.G Road                       Email : vbkumar@wipinfo.soft.net  |
  | Bangalore  - 560 001                       vbkumar@usa.net           |
  | India                                                                | 
  |                                                                      | 
  |                                                                      |  
  |                                    0_                                |          |
  | The world began when I was born   _.>/)_                             |         |
  |                                  (_) \(_) ....The world is mine      |
  |                                                           to win.    |
  |       ( Try out : http://members.tripod.com/~VBKumar/ )              |      
  +------------------------------Oooo------------------------------------+ 
                         oooO   (   ) 
                        (   )    ) / 
                         \ (    (_/ 
                          \_)



--MimeMultipartBoundary--


From pacog@hpbplfs4.bpo.hp.com  Mon Oct 20 03:48:04 1997
From: Paco Guerrero 
Date: Mon Oct 20 03:48:07 PDT 1997
Subject: Re: Priority Inheritance with Mutex Semaphores
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

the vxWorks Users Group Exploder wrote:

Hi all,

One specification about the following:


> >
> >      My scenario: tasks t1, t2, and t3 of high, medium, and low priority
> >      respectively.  t3 is running and takes the mutex semaphore.  t3 is
> >      then preempted by t1.  t1 runs for awhile and then attempts to take
> >      the mutex semaphore using a discrete timeout value.  t1 is now
> >      blocked.  t2 is ready.  What happens next??
> 
> 1) t3 will contnue running till it releases the semophore .( because priority
> of t3 is equal to the priority of t1 (because t1 is blocked on that
> resource ) and t3 is holding the resource )
> 
> 2) As soon as t3 releases the semaphore, t1 takes the semaphore and will be
> scheduled.( because t1 priority is higher than t2 & t3 priority drops to its
> original value )


T3 will return to its normal priority only after relinquishing ALL of
its mutual-exclusion
semaphores that have the inversion-safety option enabled.      ===

This is what I see in the VxWorks Reference Manual, when reading the
section "semMlib", refered
to the mutual exclusion library.

> 
> 3) And then t2 will be scheduled if no other higher priority task is ready.
> ( If t2 is the highest priority task ready )

Francisco Guerrero
Hewlett-Packard Company
Barcelona Division   (BCD) SPAIN
--MimeMultipartBoundary--


From msa@redifon.com  Mon Oct 20 04:35:30 1997
From: Mark Armstrong 
Date: Mon Oct 20 04:35:34 PDT 1997
Subject: Does the MVME172 Live Up To The Hype
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Can anyone confirm to me that the MVME172 will act as a direct
plug in replacement for the MVME162, such that I won't need a new 
vxWorks BSP to upgrade the processor speed of a system. 

Motorola say they're object code compatible but I'd like to hear from
someone who has tried a direct replacement. Did the software fail
or not? Did the processor utilisation drop? Are there any problems 
to watch out for?

There is no substitute for experience!

Mark Armstrong
Redifon MEL
--MimeMultipartBoundary--


From dleviner@HiWAAY.net  Mon Oct 20 07:07:53 1997
From: David Leviner 
Date: Mon Oct 20 07:07:56 PDT 1997
Subject: SCSI 2 library interface???
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

VxWorks Gurus,

I am in the process of writing a scsi driver for a vme host bus adapter =
board. I have completed the scsi 1 interface to the vxworks scsi library =
(scsiLib) for passing through scsi commands. I am currently working on =
the scsi 2 interface. This interface is giving me some problems though, =
especially without the needed documentation, sample code, and support =
from WindRiver. =20

Does anyone have sample code, or even source code for a current scsi =
driver that is interfaced to the scsi 2 library? And, does anyone have =
documentation on the correct way to interface to the scsi 2 library?=20

Thanks,
David Leviner
Real-Time Systems Software Engineer
Chandler/May, Inc.
205-722-0175
dleviner@HiWAAY.net 
--MimeMultipartBoundary--


From al@columbia.mcd.mot.com  Mon Oct 20 07:19:54 1997
From: Al Currano 
Date: Mon Oct 20 07:19:58 PDT 1997
Subject: Re: Does the MVME172 Live Up To The Hype
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

I think there may have been a misunderstanding. The MVME172 is
object-code compatible from an application standpoint, with the
exception of a few differences in the floating point instruction set
(check your compiler to be sure it is compatible with both the '040 and
'060). However a new VXWorks BSP will be required, since the MC68060 has
differences in the supervisory programming model.

If you already have a valid Tornado license from WR, you can get a beta
BSP from Motorola for the MVME172 from Motorola, P/N MRTVWRK172 for the
MVME172LX models, or P/N MRTVWRK172FX for the MVME172FX models.

NOTE: The current beta BSP supports *does not* support versions of the
MVME172 with the MC68LC060 processor.

I hope that clears up  confusion.

Regards.
Al
 
the vxWorks Users Group Exploder wrote:
> 

> Can anyone confirm to me that the MVME172 will act as a direct
> plug in replacement for the MVME162, such that I won't need a new
> vxWorks BSP to upgrade the processor speed of a system.
> 
> Motorola say they're object code compatible but I'd like to hear from
> someone who has tried a direct replacement. Did the software fail
> or not? Did the processor utilisation drop? Are there any problems
> to watch out for?
> 
> There is no substitute for experience!
> 
> Mark Armstrong

-- 
Al Currano                     _/      _/    _/_/_/_/_/    _/_/_/_/_/
al_currano@mcg.mot.com         _/_/  _/_/    _/      _/    _/      _/ 
Motorola Computer Group        _/ _/_/ _/    _/            _/
8945 Guilford Rd., Ste 145     _/  _/  _/    _/            _/
Columbia, MD  21046            _/      _/    _/            _/    _/_/
Phone:   (410)  312-5930       _/      _/    _/      _/    _/      _/
FAX:     (410)  381-4496       _/      _/    _/_/_/_/_/    _/_/_/_/_/
--MimeMultipartBoundary--


From R.Kirnum@dialogic.com  Mon Oct 20 09:14:23 1997
From: "Kirnum, Robert" 
Date: Mon Oct 20 09:14:27 PDT 1997
Subject: Tornado Shell Line Editing
--MimeMultipartBoundary
Content-Type: text/plain

Is the ^D control supposed to work under Tornado?  Under VxWorks it
completes the symbol name when you are in edit mode.  I do not want to
have to include 5.2 compatibility just for this feature.

Thanks,
Bob

________________________________________________________________________
D I A L	Bob Kirnum
_|_|_|O	Software Engineer
_|_|_|G	973-993-3000 Ext. 6589
_|_|_|I	973-993-5916 Fax
_|_|_|C	mailto:R.Kirnum@dialogic.com
Get the Dialogic Edge at http://www.dialogic.com
Dialogic Corp.  /  1515 Rt. 10  /  Parsippany, NJ 07054
________________________________________________________________________
           Did you know . . . that FireWire was developed by Apple?
       . . . that Apple is 150 in the Fortune 500 and Microsoft is 172?
Believe in Macintosh? Check out http://www.evangelist.macaddict.com/

--MimeMultipartBoundary--


From Ricky.D.Graves@jpl.nasa.gov  Mon Oct 20 09:49:53 1997
From: Ricky D Graves 
Date: Mon Oct 20 09:49:56 PDT 1997
Subject: DTI 1553B card compatibility with MVME 2604

I have run into a compatibility problem with Digital Technology 1553B cards (DTI
VME/S) and Motorola PowerPC SBC (MVME 2604).  When performing D16/A16 transfers
to the registers the VME bus hangs.  I looked at DTACK with a scope and the DTI
VME/S card does not release DTACK  after the 2604 releases the DS0 and DS1.  Any
comments or suggestions on how to solve this problem would be greatly
appeciated.

Thanks,

Rick Graves

   


From bberry@nortel.ca  Tue Oct 21 05:40:30 1997
From: "Bo Berry" 
Date: Tue Oct 21 05:40:34 PDT 1997
Subject: WRS on bad sectors - SCSI drives
--3nb8vbC5DHlSmtB7imxoFeeune3ncGwh
Content-type: text/plain; charset="us-ascii"


I'm interested in talking with others that have encountered run-time
problems with scsi drive bad block management.   The problem is that
bad block(s) can be ecountered during run time.  These bad blocks
are not marked as such and may be allocated to a (citical) file.


FYI, below is the response from WRS.  The summary is "Thus, it is 
recommended to do a low-level format (such as scsiFormatUnit)
and put the DOS file system back on the disk."   Problem with this
strategy is that we can not afford to take down the drive for 
formatting and reloading.

In sight from others encountering and solving this short fall 
would be appreciated.

----------------------------------------------------------------
Bo Berry
Nortel - Signaling Server Group     voice: 919-905-3461 ESN: 263
PO Box 13010                          fax: 919-905-3918
Research Triangle Park, NC  27709   email: bberry@nortel.ca
----------------------------------------------------------------


---forwarded-message---->

Oct 17 18:35 1997

Copy to:     'tsrlog@pecos.wrs.com' (INTERNET)

From:        'cokano@wrs.com' (INTERNET)

Subject:     TSR #48653


I did a historical search through our database regarding bad clusters/
sectors and have pretty much determined the following:
   VxWorks does not deal with bad sectors. 

Another customer inquired:

If, during normal dosFs activity (reads/writes), an error which is
related to a bad sector occurs, does the SCSI driver or library map
this bad sector out?

The consistent answer has been the following:
   The SCSI driver and library do not map out bad sectors at any point
   other than the low-level format of a SCSI disk.

Thus, it is recommended to do a low-level format (such as scsiFormatUnit)
and put the DOS file system back on the disk.

However, we pass all zeros to scsiFormatUnit, as can be seen by turning 
scsiDebug on.  We perform the simplest mandatory form of the FORMAT UNIT 
command (with no format data).  Most of the parameters are just packaged
up into a SCSI command and sent off to the unit.  This accomplishes medium 
formatting with little initiator control over defect management.  The target 
implementation determines the degree of defect management that is to be 
performed. 

FYI, the parameters to scsiFormatUnit are described as follows:
    SCSI_PHYS_DEV * pScsiPhysDev,  /* ptr to SCSI physical device */
    BOOL            cmpDefectList, /* whether defect list is complete */

The next 3 parameters are passed directly to the device:
    int             defListFormat, /* defect list format */ 
    int             vendorUnique,  /* vendor unique byte */
    int             interleave,    /* interleave factor */

The following parameter is passed to the device following the SCSI command.
It is generally the list of defect sectors to send.
    char *          buffer,        /* ptr to input data buffer */
    int             bufLength      /* length of buffer in bytes */


Since our dosFs does not map bad sectors into the FAT table for hard drives 
or floppy drives, the following "fix" has been suggested:
   To allow bad clusters to be marked on the disk, one would need a little 
   program which writes data and reads it back from the disk.  
   Any sectors which don't compare correctly would be considered bad.
   Then, dosFsLib would need a new ioctl() function which takes the sector
   number as an argument, calculates the corresponding cluster number, and
   marks it bad in the FAT table.

Of course, this all depends on your application.  Regardless, I hope you 
find this information helpful.

Cheers,
Cyndy

Please feel free to call/email me directly for Nortel support requests.
-------------------------------------------------------------------------
 Cynthia Okano                                     cynthia.okano@wrs.com
 Technical Support Engineer                       direct: 1-510-749-2046
 Major Accounts Program  						 	       800-USA-4-WRS
 Wind River Systems, Inc.                             http://www.wrs.com
-------------------------------------------------------------------------
        When responding via email, please "cc: tsrlog@wrs.com"
                     and list the TSR# in the subject.
-------------------------------------------------------------------------


                                                                                                                        
--3nb8vbC5DHlSmtB7imxoFeeune3ncGwh
Content-type: text/plain; charset="us-ascii"


Received: from bcarsbf5.localhost (actually bcarsbf5.ca.nortel.com) by bcarsca3.ca.nortel.com; Fri, 17 Oct 1997 18:35:00 -0400
Received: from mail.wrs.com (actually unknown-1-11.wrs.com) by bcarsbf5.localhost; Fri, 17 Oct 1997 18:34:46 -0400
Received: from pecos.wrs.com (pecos.wrs.com [147.11.41.5])
	by mail.wrs.com (8.8.6/8.8.5) with SMTP id PAA18372;
	Fri, 17 Oct 1997 15:34:41 -0700 (PDT)
Received: by pecos.wrs.com (SMI-8.6/SMI-SVR4)
	id PAA24177; Fri, 17 Oct 1997 15:34:41 -0700
Date: Fri, 17 Oct 1997 15:34:41 -0700
From: cokano@wrs.com (Cynthia Okano)
Message-Id: <199710172234.PAA24177@pecos.wrs.com>
To: rrevis@nortel.ca
Subject: TSR #48653
Cc: tsrlog@pecos.wrs.com

--3nb8vbC5DHlSmtB7imxoFeeune3ncGwh--


From johill@lanl.gov  Tue Oct 21 07:37:52 1997
From: Jeff Hill 
Date: Tue Oct 21 07:37:55 PDT 1997
Subject: ioctl SIO_MODE_SET=SIO_MODE_POLL
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit


All,

Does anyone understand the SIO_MODE_SET ioctl in vx 5.3.
The i8250Sio driver is implemented following the example of templateSio.c
and I see the ioctl() request for SIO_MODE_SET=SIO_MODE_POLL
arriving in the driver, but the driver's sio poll routines do not ever appear 
to be called by the OS when there are characters to transmit? Unable to find 
anything on this in the doc.

Jeff

______________________________________________________
Jeffrey O. Hill                      Internet     johill@lanl.gov
LANL MS H820                    Voice         505 665 1831
Los Alamos NM 87545 USA   FAX          505 665 5107

--MimeMultipartBoundary--


From nrd1jrb@nrd.ups.com  Tue Oct 21 08:41:46 1997
From: Jim Bucciferro 
Date: Tue Oct 21 08:41:49 PDT 1997
Subject: Using netDrv for FTP servers
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Does anyone have any experience using netDrv with FTP?
I need to synchronize files from a client to a server.
I read that using netDrv makes an FTP drive look like a local drive.

Are there any problems with netDrv?
Will I be able to retrieve file information from the server using a
netDrv drive?

At this point I can use raw FTP LIST to retrieve file info after parsing
each line in the stream.
My goal is to hide the FTP primitives and create a function such as
copyfile("C:/file1","FTP:/file1").
I also need this connection temporarily as the target is mobile.

Any help would be greatly appreciated.
Thanks.
Jim Bucciferro

--MimeMultipartBoundary--


From d.luck@ieee.org  Tue Oct 21 08:59:05 1997
From: Dale Luck 
Date: Tue Oct 21 08:59:08 PDT 1997
Subject: RE: WRS on bad sectors - SCSI drives
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit

> 
> I'm interested in talking with others that have encountered run-time
> problems with scsi drive bad block management.   The problem is that
> bad block(s) can be ecountered during run time.  These bad blocks
> are not marked as such and may be allocated to a (citical) file.
> 
> 
> FYI, below is the response from WRS.  The summary is "Thus, it is 
> recommended to do a low-level format (such as scsiFormatUnit)
> and put the DOS file system back on the disk."   Problem with this
> strategy is that we can not afford to take down the drive for 
> formatting and reloading.
> 
> In sight from others encountering and solving this short fall 
> would be appreciated.
> 

Ya, imagine if they had to do this on Mars.
Dale
--MimeMultipartBoundary--


From gmc@mhs.compd.com  Tue Oct 21 09:25:20 1997
From: gmc@mhs.compd.com
Date: Tue Oct 21 09:25:24 PDT 1997
Subject: VME IP carriers
                               - - Mail - - 
         October 21, 1997      5:22pm    
   FROM: Geoff Collins                   
     TO: Vx Exploder                     
SUBJECT: VME IP carriers                 

Does anyone have any '6U non-intelligent VME Industry pack carrier' board 
recommendations /experiences they could relay to me, especially ones 
relating to Vxworks use.

Thanks
Geoff Collins
Computing Devices Company LTD
Hastings



From ABBOTTD@CEBAF.GOV  Tue Oct 21 10:22:57 1997
From: ABBOTTD@CEBAF.GOV
Date: Tue Oct 21 10:23:01 PDT 1997
Subject: MV2604 and VMEBus Hangs
--MimeMultipartBoundary
Content-type: TEXT/PLAIN; CHARSET=US-ASCII
Content-transfer-encoding: 7BIT

Hello,
  Wind River issued the "collective 3" patch for the MV2604 BSP
in which they state a fix for SPR 8897 (VMEBus Interrupt hangs)
that involves disabling Write Posting when the 2604 is begin accessed
as a VME slave. The unfortunate consequences of this "fix" is that
now my 2604 (and the 2306) is a real DOG as a VME slave (average DTACK
response time is about 850ns/32bit word). 
   Is this fix really necessary? What were the conditions/motivations
that lead to this fix. If anyone out there was experiencing these
"hangs" I would like to hear from you.
                    Thanks, David

                    David Abbott 
                    TJNAF Data Acquisition Group
                    Newport News , VA
                    EMAIL: abbottd@jlab.org
                    TEL: 757 269-7190

  
--MimeMultipartBoundary--


From erik@interspeed.com  Tue Oct 21 10:54:45 1997
From: Erik Sette 
Date: Tue Oct 21 10:54:48 PDT 1997
Subject: A Very Simple Semaphore Question
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi all,

Can someone tell me how to get information on a semaphore.  I.e. the
state (Full or Empty).  I see 2 methods, semInfo and semShow.  However,
semInfo only supplies information on the threads being blocked.  And
semShow is only for the shell.  What I need is a semShow that will give
me an object with all that info.

Thanks, Erik

--
Erik Sette (mailto:erik@interspeed.com)
Brooktrout interSPEED Inc.
Lawrence, MA


--MimeMultipartBoundary--


From jford@sadira.gb.nrao.edu  Tue Oct 21 11:13:48 1997
From: John Ford 
Date: Tue Oct 21 11:13:51 PDT 1997
Subject: Re: VME IP carriers

> Submitted-by: gmc@mhs.compd.com

> 
> Does anyone have any '6U non-intelligent VME Industry pack carrier' board 
> recommendations /experiences they could relay to me, especially ones 
> relating to Vxworks use.
> 
> Thanks
> Geoff Collins
> Computing Devices Company LTD
> Hastings

We have used the Greensprings VIP-616, with no surprises.  We use it
with a z85C30 IP.  

http://www.greenspring.com/

has a catalog.

John


From suong@net.com  Tue Oct 21 11:53:47 1997
From: suong@net.com (Hong-Suong Nguyen)
Date: Tue Oct 21 11:53:50 PDT 1997
Subject: STREAMS from WRS
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi SW developers,

Is anyone using WRS's STREAMS code?

We have problem with the M_PCPROTO messages type.
We use M_PCPROTO msg to send back an ACK.
The problem is that "If an M_PCPROTO messages is already
in the queue when another arrives, the second message is
silently discarded and its messages blocks freed"
(STREAMS MODULES AND DRIVERS)

Does anyone have any work around for this?
Any help/suggestion would be appriciated.

Suong
suong@net.com
--MimeMultipartBoundary--


From david.seltz@wrs.com  Tue Oct 21 12:50:36 1997
From: David Seltz 
Date: Tue Oct 21 12:50:39 PDT 1997
Subject: Technology Rendezvous VHDL Firewire
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"

Hi,
	Is anyone out there using the Technology Rendezvous VHDL Firewire core
with VxWorks?

Dave
--MimeMultipartBoundary--


From mmcquade@pop.aa.net  Tue Oct 21 13:51:48 1997
From: Mike McQuade 
Date: Tue Oct 21 13:51:53 PDT 1997
Subject: Re: MV2604 and VMEBus Hangs
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"

At 10:23 AM 10/21/97 PDT, abbottd@jlab.org wrote:

>  Wind River issued the "collective 3" patch for the MV2604 BSP
>in which they state a fix for SPR 8897 (VMEBus Interrupt hangs)
>that involves disabling Write Posting when the 2604 is begin accessed
>as a VME slave. The unfortunate consequences of this "fix" is that
>now my 2604 (and the 2306) is a real DOG as a VME slave (average DTACK
>response time is about 850ns/32bit word). 


We and at least two other people I have talked to CAN NOT USE the 
"collective 3" patch, adding this patch causes SCSI to stop working
when you are doing VME interrupts.


We had to remove "Collective 3" and drop back to "Jumbo Patch 2".


We never have had and VME interrupt hangs with "Patch #2", so I wonder
why you can't enable write posting.


We are running our instruments (about 10 of them right now) with VME
interrupts, SCSI, LAN you name it, we are running it on MVME-2603 boards,
the instruments are in nearly continous usage without problems.


My theory is that the "604" cpu lockups caused massive confusion in
the PowerPC world, and that these "VME Interrupt hangs" are an artifact
of that whole mess.


I hope Wind River fixes the SCSI lockups in "Collective 3", before they
issue "Collective 4" or the new BSP.


Mike.



--MimeMultipartBoundary--


From bowtie@mclean.mcd.mot.com  Tue Oct 21 14:45:02 1997
From: "Bill Dennen" 
Date: Tue Oct 21 14:45:05 PDT 1997
Subject: Re: Support for MPC740 and 750
--MimeMultipartBoundary
Content-Type: text/plain; charset=US-ASCII

Newsgroups: comp.os.vxworks
Subject: Re: Support for MPC740 and 750
References: <3.0.32.19971014165220.0069c710@mail.muenchen.org>

bm417929@muenchen.org (Max Seeger) writes:

>Hi,
>does anybody know if and when WRS is going to support
>the new 740 and 750 PowerPC processors?
>Does anybody know, who offers and a VME board based
>on this processor and can deliver?
>Thanks,
>Max

The Motorola Computer Group has announced two boards utilizing the 750 (Arthur)
chip, the MCP750 for compact PCI and the MVME2700 for VME.  Both boards are
currently available on an Early Access basis.  Production release is scheduled
for late November.  Datasheets for these boards may be retrieved from
http://www.mot.com/computer/products/boards/index.html.

The BSP for the board is also available on an Early Access basis from MCG.  It
has been delivered to WRS for incorporation into the product following their
normal quality validation cycle.

MCG Engineering is not aware of any "reliability" problems with the MVME2700 
or the MCP750.  Both boards are "up and running". 

Regards
-- 
{>@<}
Bill Dennen		[bowtie@mclean.mcd.mot.com]	| we has met the enemy,
SE, Virginia Sales Territory	CIS: 75126.2272		|  and they is us ...
Motorola Computer Group		(703) 714-0707		|		Pogo
--MimeMultipartBoundary--


From jim@arraycomm.com  Tue Oct 21 18:19:47 1997
From: Jim Oakley 
Date: Tue Oct 21 18:19:50 PDT 1997
Subject: job: VxWorks for San Jose Wireless
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"



QUICK SUMMARY ******************************************************

Wanted: VxWorks programmer for driver and BSP (board support package)
porting work.  Must be familiar with VxWorks internals and PowerPC
development environments.  Full-time and contract positions available.

Detailed information pasted is below...
********************************************************************


ArrayComm, Inc.
{embedded wireless telecom technology}
San Jose, California
http://www.arraycomm.com


We are a 140 person start-up looking for top engineers to 
join our technical staff.  We offer leading-edge signal 
processing projects, competitive pay, and full benefits 
(medical/dental/vision/stock options) in a progressive 
environment.  Our software engineering openings are 
generally targeted at those with industrial real-time C 
programming or DSP assembly programming experience.  We 
are especially interested in finding a strong VxWorks 
programmer for driver and BSP porting projects.

Position: Software Engineer

We are looking for strong C programmers and assembly level DSP
programmers who enjoy working with a superb team of engineers.  If you
have experience with real-time embedded systems programming in C and
perhaps a little assembly language especially on Motorola 56xxx DSPs,
experience with UNIX workstations, and an interest in bringing up new
hardware, we have the job opportunity for you.

Successful SOFTWARE ENGINEERING candidates will have substantial
experience in one or more of the following areas:

  VxWorks or other real-time embedded systems programming
  DSP programming (Motorola 563xx or 560xx)
  communications protocol programming
  scientific/numeric programming
  hands-on experience with basic test equipment such as oscilloscopes
    and logic analyzers

Also one or more of the following is highly desired:

  extensive C/Unix experience
  signal processing and numerical analysis background
  digital circuitry and/or communications system background
  an advanced degree

Please send submit your resume 

to our web page at http://www.arraycomm.com

or via e-mail to Hum.Res@arraycomm.com 
(ascii text, postscript), by fax to (408) 428-9083, 
or by mail to: Human Resources Department, ArrayComm, Inc., 
3141 Zanker Road, San Jose, CA  95134.  

ArrayComm is an equal opportunity employer, and seeks to promote
diversity in the workplace.





--MimeMultipartBoundary--


From andreas.borer@spectraweb.ch  Wed Oct 22 03:44:38 1997
From: "Andreas Borer" 
Date: Wed Oct 22 03:44:42 PDT 1997
Subject: File transmission Protocols
--MimeMultipartBoundary
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit

In order to download files from am PC running NTand a Motorola based
Processor card I am evaluating different "standardized" protocols such as
KERMIT, XMODEM, Y- and ZMODEM.

Does anyone know about existing implementations for VxWorks /CPU32 for the
above protocols?

Thanks

Andreas

--MimeMultipartBoundary--


From page@rocko.mit.edu  Wed Oct 22 05:43:40 1997
From: "Russell J. Page, Sr." 
Date: Wed Oct 22 05:43:44 PDT 1997
Subject: debug help
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi all:

I am having trouble with what SHOULD BE a simple debug command in
vxWorks 5.1 : set breakpoint. My reference material is the 5.1 user
manual and a command reference card for 5.0, both which agree on this
particular subject. What I wish to do is set a breakpoint and have
execution suspend after hitting the breakpoint "n" times. Simple, yes ?

What does work as expected is :

	b symbol_name
	     or
	b address (in hex: 0xf34562)

What I can't get to work is something on the order of:

	b symbol_name, task_name, count
			or
	b symbol_name, ,count	(preferred form)

I've tried every combination I could think of; symbol_names, addresses,
commas, no commas, spaces, double quotes, parentheses, no parentheses --
with no luck.

Is anyone out there able to shed some light. 

Thanks in advance,
Russ
-- 
-----------------------------------------------------------------
|Russell J. Page, Sr.		    Phone (617) 253-9585	|
|MIT Bates Linear Accelerator Ctr.  FAX   (617) 253-9799	|
|PO Box 846			    email :page@rocko.mit.edu	|
|Middleton, MA  01949		    http://mitbates.mit.edu/	|
-----------------------------------------------------------------

--MimeMultipartBoundary--


From cgrames@mdc.com  Wed Oct 22 09:12:58 1997
From: Charlie Grames 
Date: Wed Oct 22 09:13:02 PDT 1997
Subject: A Very Simple Semaphore Question -Reply
--MimeMultipartBoundary
Content-Type: text/plain
Content-Disposition: inline

Erik,

I was also presented with a situation in which I wanted to know whether a semaphore was
available.  In this case, the semaphore being available ("full") indicates that an overrun has
occurred.  What I ended up doing was performing a semTake operation with a zero timeout.  If the
semTake completes without error, the semaphore was full and is now empty.  If the semTake
does not complete without error, the semaphore was already empty, so perform another
semTake to pend.

Following is a code sample:

    if (semTake(waitSemaphore,NO_WAIT) == OK)
	*overrun = TRUE;
    else
    {
	semTake(waitSemaphore,WAIT_FOREVER);
	*overrun = FALSE;
    }

If you wanted to determine whether a semaphore was full amd leave it that way, you could do a
semGive after doing the first semTake.

Hope this helps.

Charlie Grames
The Boeing Company
(314) 233-1956
Charles.R.Grames@boeing.com

>>> the vxWorks Users Group Exploder  10/21/97 12:54pm >>>
Submitted-by erik@interspeed.com  Tue Oct 21 10:54:45 1997
Submitted-by: Erik Sette 

Hi all,

Can someone tell me how to get information on a semaphore.  I.e. the
state (Full or Empty).  I see 2 methods, semInfo and semShow.  However,
semInfo only supplies information on the threads being blocked.  And
semShow is only for the shell.  What I need is a semShow that will give
me an object with all that info.

Thanks, Erik

--MimeMultipartBoundary--


From mea@mclean.sparta.com  Wed Oct 22 12:37:36 1997
From: "Mike Anderson" 
Date: Wed Oct 22 12:37:40 PDT 1997
Subject: Re: A Very Simple Semaphore Question
--MimeMultipartBoundary
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit


-----Original Message-----
Subject: A Very Simple Semaphore Question


>Submitted-by erik@interspeed.com  Tue Oct 21 10:54:45 1997
>
>Hi all,
>
>Can someone tell me how to get information on a semaphore.  I.e. the
>state (Full or Empty).  I see 2 methods, semInfo and semShow.  However,
>semInfo only supplies information on the threads being blocked.  And
>semShow is only for the shell.  What I need is a semShow that will give
>me an object with all that info.
>


Greetings!

  Well, another alternative is to take into consideration that a
SEM_ID is actually a pointer to a structure.  That structure is defined
in /h/private/semLibP.h as follows:

typedef struct semaphore /* SEMAPHORE */
    {
    OBJ_CORE  objCore;  /* 0x00: object management */
    UINT8 semType;  /* 0x04: semaphore type */
    UINT8 options;  /* 0x05: semaphore options */
    UINT16  recurse;  /* 0x06: semaphore recursive take count */
    Q_HEAD  qHead;    /* 0x08: blocked task queue head */
    union
      {
        UINT     count; /* 0x18: current state */
        struct windTcb  *owner; /* 0x18: current state */
      } state;
    } SEMAPHORE;

So, by looking at a live system, we should be able to figure something out:

-> _semId = semBCreate(1,0)
_semId = 0x38674c: value = 3699384 = 0x3872b8 = _semId + 0xb6c

-> show(_semId)
Semaphore Id        : 0x3872b8
Semaphore Type      : BINARY
Task Queuing        : PRIORITY
Pended Tasks        : 0
State               : EMPTY
value = 0 = 0x0

-> d _semId, 32, 2
003872b0:                      000d eedc 0001 0000   *          ......*
003872c0:  0000 0000 0000 0000 0000 0000 000c a9e0   *................*
003872d0:  0000 0000 0038 72b0 0000 0020 002c 9edc   *.....8r.... .,..*
003872e0:  0038 72ec 0038 72fc 0000 0900 5f73 656d   *.8r..8r....._sem*
003872f0:  4964 3100 0038 72d4                       *Id1..8r.........*
value = 21 = 0x15

-> semTake(_semId, -1)
value = 0 = 0x0
-> d _semId, 32, 2
003872b0:                      000d eedc 0001 0000   *          ......*
003872c0:  0000 0000 0000 0000 0000 0000 000c a9e0   *................*
003872d0:  0037 6f5c 0038 72b0 0000 0020 002c 9edc   *.7o\.8r.... .,..*
003872e0:  0038 72ec 0038 72fc 0000 0900 5f73 656d   *.8r..8r....._sem*
003872f0:  4964 3100 0038 72d4                       *Id1..8r.........*
value = 21 = 0x15
->
-> i
  NAME        ENTRY       TID    PRI   STATUS      PC       SP     ERRNO  DELAY
---------- ------------ -------- --- ---------- -------- -------- ------- -----
tExcTask   _excTask       3d4230   0 PEND          99280   3d4190   d0003     0
tLogTask   _logTask       3d191c   0 PEND          99280   3d1878       0     0
tShell     _shell         376f5c   1 READY         2dd98   376bf8       0     0

It seems to have filled in the task ID of the shell task as the semaphore's
"owner", in spite of the fact that this is a binary semaphore!  Now give the
semaphore back:

-> semGive(_semId)
value = 0 = 0x0
-> d _semId, 32, 2
003872b0:                      000d eedc 0001 0000   *          ......*
003872c0:  0000 0000 0000 0000 0000 0000 000c a9e0   *................*
003872d0:  0000 0000 0038 72b0 0000 0020 002c 9edc   *.....8r.... .,..*
003872e0:  0038 72ec 0038 72fc 0000 0900 5f73 656d   *.8r..8r....._sem*
003872f0:  4964 3100 0038 72d4                       *Id1..8r.........*
value = 21 = 0x15
->

And, voila, the "owner" is back to null.  This would lead one to believe that
if the 32bit value at offset 0x18 from the beginning of the semaphore structure
is a NULL, then the semaphore's available, otherwise that is a TID/pointer to
the
TCB of the task who "took" the semaphore.  This seems to be true as long as
the semaphore is a binary or a mutex.  If the semaphore is a counting flavor,
the value at offset 0x18 is the actual count of how many are still available
rather than a TCB pointer.

HTH,

===============================================================================
     __       Real-Time System Development, Integration, Training and Services
    //\\
   //  \\     Mike Anderson
  // /\ \\    Chief Engineer              Voice : (703) 448-0210 ext. 235
 // /  \ \\   SPARTA, Inc.                FAX   : (703) 893-5494
//   \    \\  7926 Jones Branch Drive     EMAIL : mea@mclean.sparta.com
\\    \   //  Suite 900                   Web   : http://www.mclean.sparta.com
 \\ \  / //   McLean, VA 22102
  \\ \/ //                             "Software development is like making
   \\  //                               a baby... You can't make a baby in one
    \\//                                month by impregnating nine women.
     --  "Pride in Performance"         Some things just take time."
===============================================================================





--MimeMultipartBoundary--


From Richard_W_Mattox@ccmail.orl.lmco.com  Wed Oct 22 12:43:37 1997
From: Richard_W_Mattox@ccmail.orl.lmco.com
Date: Wed Oct 22 12:43:40 PDT 1997
Subject: ATM Device Driver
--MimeMultipartBoundary
Content-type: TEXT/PLAIN
Content-transfer-encoding: 7BIT

Hello Fellow VxWorkers,

I am in search of an ATM Device driver for a PMC Sierra ATM Lasar chip set.  The
processor is an i960 processor.  The hardware folks created a custom board based
on a Cyclone iq960rp board design.  This board contains the i960 and Lasar chip 
set.  There is code provided by Trillium to program and initialize the physical 
layer and aal5, but it is by no means a VxWorks type of device driver.  Before I
spend any time developing a device driver to wrap around the Trillium code I was
wondering if anybody has dealt with this situation before?  Did you write your 
own device driver?  Is there another vendor out there that has a VxWorks device 
driver for the platform I described above?

TIA,
Rick Mattox
richard.w.mattox@lmco.com
--MimeMultipartBoundary--


From mwbrown@dragon.sandia.gov  Wed Oct 22 13:39:17 1997
From: mwbrown@dragon.sandia.gov (Matthew W. K. Brown)
Date: Wed Oct 22 13:39:22 PDT 1997
Subject: Re: A Very Simple Semaphore Question
--MimeMultipartBoundary
Content-Type: text/plain
Content-Transfer-Encoding: 7bit

Mike:

Excellent advice ... I must admit that I've always just tried taking the 
semaphore to see if it was available or not.

Thanks for your posting.

Matthew


-------------------------------------------------------------------------
Matthew Brown, SMTS                             Voice: (505) 844-1160
Sandia National Laboratories                    Fax:   (505) 844-7059
P. O. Box 5800    M/S 0965                      email: mwbrown@sandia.gov
Albuquerque, NM  87185-0965
--MimeMultipartBoundary--


From daemon@csg.lbl.gov  Thu Oct 23 04:10:58 1997
From: daemon@csg.lbl.gov
Date: Thu Oct 23 04:11:01 PDT 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Thu Oct 23 04:10:55 PDT 1997

        Subject: SNMP Agent from WindRiver
        Subject: Re: 2000 problem

-------------------------------------------------------

Newsgroups: comp.protocols.snmp,comp.os.vxworks
Subject: SNMP Agent from WindRiver
Date: Wed, 22 Oct 1997 12:12:51 -0500
From: "Ricky W. West" 
Organization: Ericsson Inc.
Message-ID: <344E3413.923432F7@screamer.rtp.ericsson.se>

Has anyone used or knows any information about the SNMP agent that Wind
River is offering for VxWorks 5.3.1?  I think its name is WindNet SNMP.

Thanks,

Ricky W. West
west@rtp.ericsson.se
(919) 472-6534

---------------------------

Newsgroups: comp.os.vxworks,comp.realtime,comp.unix.programmer,comp.lang.c,comp.lang.c++
Subject: Re: 2000 problem
Date: 22 Oct 1997 19:31:03 GMT
From: goudreau@dg-rtp.dg.com (Bob Goudreau)
Organization: Data General Corporation, RTP, NC
Message-ID: <62lk9n$gqf@dg-rtp.dg.com>
References: <62fk8h$mjq@bcarh8ab.bnr.ca> <62ijm7$qmu@bgtnsc02.worldnet.att.net> <344DCDF8.26351658@absyss.fr>
Followup-To: comp.os.vxworks,comp.realtime,comp.unix.programmer,comp.lang.c,comp.lang.c++

Doug Seay (seay@absyss.fr) wrote:

: Years start with 1, not with 0, so there are a few boundry [sic]
: problems. Life would be easier if non-computer people understood
: the usefulness of zero, although a AD0 and a 0BC would be kinda
: confusing too.

I'm continually amazed at how many people profess surprise or
annoyance at the fact that no year numbered "0" exists in our
calendrical system.  Why aren't these people also surprised that
month numbers and day numbers commence with 1 and not with 0 (i.e.,
so that the year begins on 0/0 instead of on 1/1)?  I don't see why
year numbers should be any different; it's quite natural to dub
number the first time unit (day, month, year) as "1" instead of as "0".

- ----------------------------------------------------------------------
Bob Goudreau			Data General Corporation
goudreau@dg-rtp.dg.com		62 Alexander Drive	
+1 919 248 6231			Research Triangle Park, NC  27709, USA

---------------------------

End of New-News digest
**********************


From erik@interspeed.com  Thu Oct 23 06:25:06 1997
From: Erik Sette 
Date: Thu Oct 23 06:25:09 PDT 1997
Subject: Re: A Very Simple Semaphore Question
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Mike,

Thanks... that is exactly what I needed to know.  I was assuming that
SEM_ID was more like a descriptor, not an actual address to the object.

--Erik


--
Erik Sette (mailto:erik@interspeed.com)
Brooktrout interSPEED Inc.
Lawrence, MA


--MimeMultipartBoundary--


From erik@interspeed.com  Thu Oct 23 06:54:00 1997
From: Erik Sette 
Date: Thu Oct 23 06:54:03 PDT 1997
Subject: Re: A Very Simple Semaphore Question
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi all again,

Even though SEM_ID is an actual pointer to a SEMAPHORE object, it should
still be treated as a descriptor.  VxWorks should provide an API to
access certain private object members.  We should not directly access
the elements in a structure due to the fact that the structure may
change.  I have decided to keep my own semCount in my class object until
a better access method is provided.

Thanks again for the all the replies...

--
Erik Sette (mailto:erik@interspeed.com)
Brooktrout interSPEED Inc.
Lawrence, MA


--MimeMultipartBoundary--


From mam@pluto.dspt.com  Thu Oct 23 06:54:15 1997
From: "Mark Menge" 
Date: Thu Oct 23 06:54:18 PDT 1997
Subject: MV2604 dribbling
Every morning I see a couple of messages in my telnet to the MV2604:
dc: dribbling bit
dc: CRC error 

We have thousands of hours on mvme147 and gmsv46s but I have never
seen these errors (I even have a 68040 on the same net)

I checked underneath the mv2604 but found no spit pool of these dribbled bits. 
Has anyone else seen the same thing?

Mark


From Nick_Davis@owa.com.au  Thu Oct 23 08:01:17 1997
From: Nick_Davis@owa.com.au
Date: Thu Oct 23 08:01:21 PDT 1997
Subject: Nick Davis/PlantEng/OWA is on RDO Friday.
--MimeMultipartBoundary
Content-type: text/plain; charset=US-ASCII


I will be out of the office from 23/10/97 until 27/10/97.

I will respond to your message when I return.


--MimeMultipartBoundary--


From cgrames@mdc.com  Thu Oct 23 10:15:53 1997
From: Charlie Grames 
Date: Thu Oct 23 10:15:57 PDT 1997
Subject: MV2604 dribbling -Reply
--MimeMultipartBoundary
Content-Type: text/plain
Content-Disposition: inline

Mark,

It sounds like you are using an older version of the dec21140.obj driver for the 2604.  The latest
version does not display these errors.  I'm not convinced, however, that these errors no longer
exist with the newer driver.

There is a very good chance that Motorola will be providing an Ethernet-related hardware change
for the 2604 in the near future.  Ask your Motorola rep to keep his/her ear to the ground.

Charlie Grames
The Boeing Company
(314) 233-1956
Charles.R.Grames@boeing.com

-----

Submitted-by mam@pluto.dspt.com  Thu Oct 23 06:54:15 1997
Submitted-by: "Mark Menge" 

Every morning I see a couple of messages in my telnet to the MV2604:
dc: dribbling bit
dc: CRC error 

We have thousands of hours on mvme147 and gmsv46s but I have never
seen these errors (I even have a 68040 on the same net)

I checked underneath the mv2604 but found no spit pool of these dribbled bits. 
Has anyone else seen the same thing?

Mark

--MimeMultipartBoundary--


From mmcquade@pop.aa.net  Thu Oct 23 10:15:53 1997
From: Mike McQuade 
Date: Thu Oct 23 10:15:59 PDT 1997
Subject: Re: MV2604 dribbling
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"

At 06:54 AM 10/23/97 PDT, you wrote:
>Submitted-by mam@pluto.dspt.com  Thu Oct 23 06:54:15 1997
>Submitted-by: "Mark Menge" 
>
>Every morning I see a couple of messages in my telnet to the MV2604:
>dc: dribbling bit
>dc: CRC error 
>
>We have thousands of hours on mvme147 and gmsv46s but I have never
>seen these errors (I even have a 68040 on the same net)
>
>I checked underneath the mv2604 but found no spit pool of these dribbled
bits. 
>Has anyone else seen the same thing?
>


Get the latest DEC Chip driver from your Motorola FAE.


This fixed the dribbling bits, plus make the 10/100 performance
much better.


I don't think the patch has made it through the WRS system yet.


Mike.

--MimeMultipartBoundary--


From bmcgonig@mariners.com  Thu Oct 23 10:41:42 1997
From: Brian 
Date: Thu Oct 23 10:41:46 PDT 1997
Subject: Re: A Very Simple Semaphore Question
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"


>Subject: A Very Simple Semaphore Question
>

HI:

Just a word of warning here.  The testing and using of  semaphore needs to be
an indivisible set of instructions otherwise if you test the SEM_ID pointer,
you could become victim to a race condition whereby the semaphore status
is changed btw the time you test it and the time you take it (if your task
is switched out
in that brief period).   Neat problem to debug :)

Brian
--MimeMultipartBoundary--


From cgrames@mdc.com  Thu Oct 23 13:23:08 1997
From: Charlie Grames 
Date: Thu Oct 23 13:23:12 PDT 1997
Subject: Ethernet Performance on MVME2604
All,

There have been several versions of the dec21140.obj Ethernet driver for the MVME2604 that
claim to greatly improve 100BaseT performace, some available from Motorola, and some from
Wind River.  While some of these versions provide nominal performance improvements over
other versions, what REALLY makes the difference is making sure the Ethernet descriptors and
buffers are cached!  By default, they are not.

There are two ways to do this:

1)  Set DC_POOL_ADRS in config.h to (NONE).  This causes the driver to dynamically allocate its
memory pool, which it does from cached memory.

2) Modify the sysPhysMemDesc entries in sysLib.c so that the memory below RAM_LOW_ADRS
is cached.  I'm not sure whether it is a good idea to cache the exception vector tables, so you may
want to leave those uncached.  (Anyone with more knowledge in this area please feel free to
comment!)

Note:  We opted for number 2, because we wanted explicit control over DC_POOL_SIZE, which
determines the number of Ethernet receive and transmit descriptors and buffers.  Since we do a
large amount of Ethernet traffic (UDP/IP) in real time, we need to have a large number of buffers to
avoid message drops.

The last I had heard from Wind River was that the dec21140.obj delivered with the Collective 3
patch was the one that will be included in the next BSP release (I don't know if that is still true or
not).  If you are concerned about version control (I am), you may want to consider using the
Collective 3 version in conjuction with the above-mentioned changes.

ICharlie Grames
The Boeing Company
(314) 233-1956
Charles.R.Grames@boeing.com


From 73009841@aol.com  Thu Oct 23 15:29:45 1997
From: 73009841@aol.com
Date: Thu Oct 23 15:29:48 PDT 1997
Subject: Deliverable Mail


/////////////////////////////////////////////////////////////////
CREDIT CENTER***CREDIT CENTER***CREDIT CENTER***CREDIT CENTER
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
If you received this electronic mailing by mistake, please pardon the
intrusion.

Dear Friend;

Are you or somebody you know experiencing credit problems?  If the
answer is yes, our special offer might be just the cure you have been
looking for.  No longer will you have to live like a second class
citizen due to a bad credit report.  Bad credit can happen due to a
variety of reasons and can happen to anybody.  There is no profile of a
person with credit problems, it can happen to anybody.  In fact, one in
three Americans live with some form of bad credit.  Fortunately, the
United States Government has passed many laws under the Fair Credit
Reporting Act, all of which protect your rights as a consumer.  Our
manuals will show you how to use these Federal Laws to your advantage to
repair your credit, obtain a new credit report or add good credit to
your credit report!

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
*****  'THE FRESH START PROGRAM II' (TM) *****

Have you recently been denied credit due to a bad credit report?  Have
you seen your credit report lately?  If your answered yes, then you need
to take advantage of the 'FRESH START PROGRAM II' (TM) NOW!  This manual
is unique in that it shows you how to obtain a brand new Social Security
number which will allow you to open a brand new, fresh credit report. 
Using guidelines set forth by the Social Security Administration, this
easy-to-follow manual will guide you through the steps necessary to
obtaining your new Social Security Number.  Regardless of your present
credit, you can open a new credit file free of ANY judgements,
charge-off's, foreclosures, bankruptcies, bad credit, tax liens's,
garnishments, late pays and inquiries.

////////////////////////////////////////////////////////////////

***** THE 'WorldCard' (TM) INSTANT CREDIT Credit Card *****

This program is very simple.  It extends to you a $5,000 instant line of
credit in the exclusive 'WorldCard' (TM) Wholesale Purchasing Club. 
This credit card comes to you with absolutely NO credit checks!

You will receive the 'WorldCard' (TM) Wholesale Purchasing Club catalogs
with over 5,000 items to choose from.  Jewelry, toys, tools, gift items,
housewares and much more are available.  After your first purchase, each
quarter you will receive updated catalogs with brand new items
available.

You will be required to make a small down-payment with your orders
and then make a monthly payment of 10% of the unpaid balance.  It's just
that simple!  Your good credit with WorldCard will be reported to the
credit bureau's, which will show your excellent credit!

As a special BONUS, if you take advantage of this offer within the next
24 hours, you will be given $200 in cash certificates!  $100 may be used
on the first $300 unpaid balance and the other $100 may be used after
you have charged $600 on your new account.

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

***** 'STOP COLLECTION AGENCIES COLD' (TM) *****

If you have ever experienced the humiliation of a collection agency
calling you at home or at work, then this hard-hitting manual will put
an IMMEDIATE stop to those calls!

Millions of collection agency telephone calls and letters are made or
sent each week to those who owe money.  Consumers have been helpless to
stop the harassment of these predators, UNTIL NOW!

This manual will show you how to put an end to these telephone calls and
letters, INSTANTLY!  At the same time, these collection agencies will be
forbidden to EVER call or write you again!!  This manual has the inside
secrets the credit world does not want you to have!

Gain back your self-esteem and piece-of-mind by ordering 'Stop
Collection Agencies Cold' (TM) TODAY!!!

/////////////////////////////////////////////////////////////////

***** 'THE COMPLETE CREDIT REPAIR KIT' (TM) *****

Good credit empowers you to qualify for personal loans, automobile
loans, business loans, credit cards, home loans and other types of
credit you may apply for at the very best interest rate.  As you know,
if you don't have excellent credit, lenders will NOT extend credit to
you!  The Fair Credit Reporting Act will allow you to legally and
permanently ERASE the negative information on your credit bureau file.

Anything damaging to your credit, such as bankruptcies, judgements, tax
liens, slow pays, repossessions, charge off's, foreclosures, late
payments, student loans and inquiries may be permanently and legally
REMOVED from your credit report!

If bad credit on your credit file is standing in the way of your
achieving the American dream, then order 'The Complete Credit Repair
Kit'(TM) NOW!!!

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

IMS
5201 Kingston Pike, Suite 6-216
Knoxville, Tennessee 37919

YES, please rush the manuals checked below:

_____   'The Fresh Start Program II' (TM)         	$34.95

_____   'WorldCard' (TM) Credit Card            	$34.95

_____   'Stop Collection Agencies Cold' (TM)   	 $34.95

_____   'The Complete Credit repair Kit' (TM)   	$34.95

_____   TAKE ALL FOUR (SAVE over $39)   	$99.95


$____________  Sub-Total

$4.95    Shipping & Handling for each manual ordered
$15.00   RUSH Delivery (Optional)

$____________  TOTAL ENCLOSED

I Have Enclosed (Check One):
_____   Cashiers Check
_____   Money Order
_____   Personal Check
_____   FAX ORDER


SHIPPING INFORMATION:

Name:__________________________________________________
Address:________________________________________________
City:_______________________State:________Zip Code:_________
Telephone Number:___________________

Your E-Mail Address (For order confirmation):___________________

////////////////////////////////////////////////////////////////

FREE  -  FREE  -  FREE  -  FREE  -  FREE  -  FREE  -  FREE  -  FREE  

Fax your order within 24 hours and receive a special Kodak film book
entitling you to
100 rolls of Kodak film as well as five 6X10 and ten 4X5 enlargements
FREE!

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

Please tape your signed personal check in the space below and fax to
IMS.  The 24 hour facsimile number is 1-423-681-6520.

******************************************************

FAX 1-423-681-6520

Please TAPE Your Check Here

Please SIGN Your Check

******************************************************

THANK YOU For Your Order!!!








From ceselsa@coastalnet.com  Thu Oct 23 17:13:23 1997
From: Rehosting Team 
Date: Thu Oct 23 17:13:27 PDT 1997
Subject: ftpXfer function
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

I'm using the "ftpXfer" function to send/receive files to/from a remote
machine.

Does anybody know how use this function to connect to the other machine
unsing a user who doesn't have password? I tried to write "" in the
password field of the function and it didn't work.

Thanks.


Juanjo Aguado
Indra DTD.
ceselsa@coastalnet.com
--MimeMultipartBoundary--


From Jean-Yves_Patry@Mitel.COM  Thu Oct 23 17:23:12 1997
From: Jean-Yves Patry 
Date: Thu Oct 23 17:23:16 PDT 1997
Subject: loginLib
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi all,
   I am not to familiar with the loginLib used for security in VxWorks. 
The question I have are:

1) The user/login password table seems to be stored in volatile
memory.    So if the system is rebooted, the login username and password
are       lost? It would be  nice if these were stored in NON-VOLATILE
ram 
   (flash) or disk, so that they can be retrieved when the system
is        rebooted. This could mimic to the UNIX /etc/passwd file.
   
2) The encryption utilities are located on the host, so one must
   install its own encryption facilities if a command to allow
   a user to change his/her password is implemented on the target.

3) Has anybody implemented this level of security in VxWorks?

-- 
------------------------------------------------------------------------Jean-Yves
Patry         Jean-Yves.Patry@Mitel.COM          (613) 592-2122 x4992
------------------------------------------------------------------------
--MimeMultipartBoundary--


From blohm@vs.dasa.de  Fri Oct 24 01:48:26 1997
From: "Gundula Blohm" 
Date: Fri Oct 24 01:48:31 PDT 1997
Subject: Re: ftpXfer function
On 23 Oct 97 at 17:13, the vxWorks Users Group Explo wrote:

> I'm using the "ftpXfer" function to send/receive files to/from a remote
> machine.
> 
> Does anybody know how use this function to connect to the other machine
> unsing a user who doesn't have password? I tried to write "" in the
> password field of the function and it didn't work.
> 
> Thanks.
> 
> 
> Juanjo Aguado
> Indra DTD.
> ceselsa@coastalnet.com

FTP is defined to work with more security options than TFTP. Therefore with 
most FTP Servers(like Windows NT) you need to establish a password. So you can 
either set a password for your user or switch to the TFTP protocol. For 
Windows based platforms there is a shareware version available from Walusoft.

Email info@walusoft.demon.co.uk	   Web Site http://www.walusoft.co.uk

HTH, Dula
===================================================
Gundula Blohm
Tel. +49 (731) 392-4731
Fax  +49 (731) 392-4958
mailto:blohm@vs.dasa.de
===================================================


From daemon@csg.lbl.gov  Fri Oct 24 04:00:35 1997
From: daemon@csg.lbl.gov
Date: Fri Oct 24 04:00:39 PDT 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Fri Oct 24 04:00:26 PDT 1997

        Subject: Re: SNMP Agent from WindRiver

-------------------------------------------------------

Newsgroups: comp.protocols.snmp,comp.os.vxworks
Subject: Re: SNMP Agent from WindRiver
Date: 23 Oct 1997 19:24:37 -0600
From: bgeer@xmission.xmission.com (bgeer)
Organization: XMission Internet (801 539 0900)
Message-ID: <62otcl$aan$1@xmission.xmission.com>
References: <344E3413.923432F7@screamer.rtp.ericsson.se>

"Ricky W. West"  writes:

 >Has anyone used or knows any information about the SNMP agent that Wind
 >River is offering for VxWorks 5.3.1?  I think its name is WindNet SNMP.

The book cover sez: WindNet SNMPv1/v2c Component Release Supplement
1.0.  We've had it for a few months now.  As I recall, WRS previous
SNMP dev kit linked SNMP into the kernel making kernel config
management for multiple identical processors some with, some without
SNMP, a pain to manage.  This version is based on making one call to
the agent to "register" (or link) your custom mibs & code.  Fairly
painless & a lot easier to config-manage.

I am not impressed by the documentation & examples:

Documentation is a bit thin for newbies.  Considering that reading
RFC's is about as rewarding as counting the flecks of tree bark in
cheap toilet paper, SNMP could use a good tutorial & reference book -
somethink like "SNMP for Dummies"!  Unfortunately Barnes & Noble
doesn't carry SNMP books [here] & I can't afford "trial & error"
mailordering books to find one that's useful to me.

Examples are trivial (no table examples) &/or didn't work "out of the
box".  The "mibcomp" program has a ton of capability, but the examples
appear to use little of it.

I wrote a custom SNMP agent 4 years ago using Peer Networks Unix code
I ported to VxWorks.  Never did get table access to work reliably, but
all the scalar get/set stuff, what I really needed, worked
flawlessly.

With the new WindNet, it took a bit of study to decide how to best use
"mibcomp" to produce a foundation for building a custom agent.  That
done, implementation was straightforward...except for tables again!

After I got around WRS's stupid "in-house expert" tech support
strategy [uh, is bottleneck a better word?] I received an answer in
the form of another code example.  I tweaked my table access code &
the problems appear solved, tho I'm not totally confident.  Well, it
works, but what I really want to know is:

My mib defines 5 tables in succession.  "Get"s for one table works ok,
but my code ends up walking part way thru the next table before the
WRS code becomes convinced the table is really empty.  Like, lots &
lots of unnecessary, performance-robbing function calls.  I mean,
shouldn't a function like nextproc_no_next() be sufficient to stop the
flood?

My failure to handle tables properly is, I think, my lack of desire to
deal with "tree-walking" theory (the only sentient trees that are
worth reading about are Ents, anyway! :-) & the pedantically abstract
nature of all published SNMP info I've seen.  It's a lot of hoo-hah
considering that when it is finally filtered down to what you really
want to accomplish, throwing together mib & agent code isn't all that
difficult.

I bet that first guy (in the gender neutral sense) that thought up
SNMP was really thinking "Simple [hah!] Network Management Protocol"!

Anyway, I haven't paid attention to WRS price for WindNet nor have any
experience on current competitor offerings, but WindNet is quite
adequate for our needs.

Cheers, Bob

- -- 
<> Robert Geer & Donna Tomky /              *             <>
<>    bgeer@xmission.com    |    _o      *   o *      o   <>
<>   Salt Lake City, Utah   |   -\<,      * <\      
<>          U S A           |   O/ O     __ /__,    />    <>

---------------------------

End of New-News digest
**********************


From bobperry@bellatlantic.net  Fri Oct 24 05:18:34 1997
From: "Robert L. Perry" 
Date: Fri Oct 24 05:18:37 PDT 1997
Subject: Re: ftpXfer function
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"

At 05:13 PM 10/23/97 PDT, you wrote:
>Submitted-by ceselsa@coastalnet.com  Thu Oct 23 17:13:23 1997
>Submitted-by: Rehosting Team 
>
>
>I'm using the "ftpXfer" function to send/receive files to/from a remote
>machine.
>
>Does anybody know how use this function to connect to the other machine
>unsing a user who doesn't have password? 

Every ftp server I've used always required a password.  I don't know if
this is mandated by the ftp protocol but my bet is that this is not a
problem w/ the ftpXfer function.



--MimeMultipartBoundary--


From Thomas.Arand@vs.siemens.de  Fri Oct 24 05:27:50 1997
From: "Arand, Thomas OEN BN EF3" 
Date: Fri Oct 24 05:27:54 PDT 1997
Subject: Galileo system controller
Hi VxWorks World!

We are thinking about using the Galileo system controller GT-64120 in
our next product, when available. Before, we will use GT-64010A.

Has anyone experience with that chip(s)? Are there drivers for use
within VxWorks available? Are there evaluation boards?

Thanks for your answer!

Thomas Arand
Siemens AG
e-mail: thomas.arand@vs.siemens.de


From cpinkard@falcon.kla-tencor.com  Fri Oct 24 08:42:22 1997
From: Christopher Pinkard 
Date: Fri Oct 24 08:42:25 PDT 1997
Subject: Re: ftpXfer function
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

> Does anybody know how use this function to connect to the other machine
> unsing a user who doesn't have password? I tried to write "" in the
> password field of the function and it didn't work.
> 
I believe the prompt says put a period '.' in any field you want 
to clear.  It's always worked for me.
-- 
----------------------------------------------------
Chris Pinkard             Sr. Staff SW Engineer
KLA-Tencor                RAPID Engineering  
160 Rio Robles            C2-3400
San Jose, CA              95161-9055       
cpinkard@falcon.kla.com   408-875-4346
--MimeMultipartBoundary--


From lenny@den.mcd.mot.com  Fri Oct 24 11:39:12 1997
From: lenny_bates 
Date: Fri Oct 24 11:39:16 PDT 1997
Subject: RE: Ethernet Performance on MVME2604
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

There is allot of confusion about this matter so I thought I would try =
to shed some light on the subject.
There has been around 3 different DEC patches that have came out over =
the past several months as well as
several patches for the 260x BSP. Below is an excerpt of what the latest =
collective-3 patch fixes. Notice that
it also states that there are additional fixes that have come out that =
are not included in the collective-3
which is the latest DEC Ethernet patch. The DEC patch is available from =
your local Motorola SE. I listed a
copy of the instructions so you can see what needs to be done but would =
suggest that you get the collective-3
before applying the dec21140.obj file and modifying config.h and =
sysLib.c.

I hope this helps clear up the confusion.
Good luck,
Lenny
MCG - Denver
Registered users of VxWorks should regularly check the WindRiver =
WindSurf  page for updates on =
BSP problems and patches.=20

MVME260x BSP Tornado 1.0.1 Collective-3 Patch
WindRiver now has a patch for the MVME260x BSP which includes the MPIC =
and Extended VME capabilities and fixes for some other problems. It is =
called the Tornado 1.0.1 Collective-3 patch and all customers should =
contact WindRiver to get this patch.=20
There are still some problems with the 2600 BSP that WindRiver does not =
have in the Tornado 1.0.1 Collective-3 patch. These are identified in =
the patch documentation so please make sure your customers review the =
patch documentation

DEC21140 100BaseT Patch, 10/13/97
There is a new patch for the DEC21140 driver which significantly =
increases the 100BaseT performance in BOTH directions!=20
This new driver requires DC_POOL_SIZE in config.h to be set to =
0x00012000.=20
Additional performance increases are available by caching the DEC driver =
buffers. This is done by searching sysLib.c for:=20
PHYS_MEM_DESC sysPhysMemDesc [] =3D
and replacing the first two table entries with the following:=20
    {
    /* Vector Table and Interrupt Stack */

    (void *) LOCAL_MEM_LOCAL_ADRS,
    (void *) LOCAL_MEM_LOCAL_ADRS,
    RAM_LOW_ADRS,
    VM_STATE_MASK_VALID | VM_STATE_MASK_WRITABLE | =
VM_STATE_MASK_CACHEABLE |
    VM_STATE_MASK_MEM_COHERENCY,
    VM_STATE_VALID      | VM_STATE_WRITABLE      | VM_STATE_CACHEABLE |
    VM_STATE_MEM_COHERENCY
    },

    {
    /* Local DRAM - Must be second entry in sysPhysMemDesc for Auto =
Sizing */

    (void *) RAM_LOW_ADRS,
    (void *) RAM_LOW_ADRS,
    LOCAL_MEM_SIZE -  RAM_LOW_ADRS,
    VM_STATE_MASK_VALID | VM_STATE_MASK_WRITABLE | =
VM_STATE_MASK_CACHEABLE |
    VM_STATE_MASK_MEM_COHERENCY,
    VM_STATE_VALID      | VM_STATE_WRITABLE      | VM_STATE_CACHEABLE |
    VM_STATE_MEM_COHERENCY
    },
To install this driver:
Save your existing dec21140.obj file in a different directory.=20
Extract the dec21140.obj file from the tar file.=20
Put the new dec21140.obj file in the BSP directory (mv2604 & mv2603).=20
Rebuild the bootrom and the kernel.=20
Make sure you FLASH the new bootrom binary before using the new kernel.=20

--MimeMultipartBoundary--


From daemon@csg.lbl.gov  Sun Oct 26 04:00:26 1997
From: daemon@csg.lbl.gov
Date: Sun Oct 26 04:00:31 PST 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Sun Oct 26 04:00:21 PST 1997

        Subject: FIC 8232 and FIC 8234
        Subject: ObjectAda for VxWorks on PPC - any good?

-------------------------------------------------------

Newsgroups: comp.os.vxworks
Subject: FIC 8232 and FIC 8234
Date: Fri, 24 Oct 1997 08:17:33 +0100
From: Bruno Di Giandomenico 
Organization: Alenia
Message-ID: <34504B8D.41C67EA6@test.alenia.polito.it>

I happen to have a number of the subject boards, produced by CES, for
which I
am going to use VxWorks. I have been told that the BSP for these boards
have
some bugs.

Could anybody who has used these boards tell me whether this is true or
not and which bugs have been found ?

I have also been told that hte company who ported VxWorks to these
boards is now
out of business. If I wanted to find the source code for the BSP and
modify them
(to correct the bugs, if possible), does anybody have a suggestion on
how to 
proceed ?

Last, but not least, is there a repository for the old messages of this
group somewhere, so that I can maybe find answers to some of my
questions, without bothering the whoel world ?


			Thanks in advance

- -- 
- --------------------------------------------------------------------------------
Bruno Di Giandomenico
Alenia Aeronautica
Dept. SIMT BOX 17
Corso Marche 41 10146 Torino
Tel (39)-11-7181-837
Fax (39)-11-7181-079
E-Mail : bruno@test.alenia.polito.it

---------------------------

Newsgroups: comp.os.vxworks
Subject: ObjectAda for VxWorks on PPC - any good?
Date: 24 Oct 97 02:04:04 GMT
From: steved@pacifier.com (Steve Doiel)
Message-ID: <34500214.0@news.pacifier.com>

Would anyone like to share there experience with ObjectAda from Aonix
targeting VxWorks on the PowerPC architecture?  Good? Bad? or Indifferent?



---------------------------

End of New-News digest
**********************


From jshaw_shade@compuserve.com  Mon Oct 27 03:39:33 1997
From: John Shaw 
Date: Mon Oct 27 03:39:36 PST 1997
Subject: Re: comp.os.vxworks newsdigest
--MimeMultipartBoundary
Content-Type: text/plain; charset=ISO-8859-1
Content-Disposition: inline

Regarding:-

>>From: Bruno Di Giandomenico 

>>Last, but not least, is there a repository for the old messages of this=

>>group somewhere, so that I can maybe find answers to some of my
>>questions.

I think it is an FTP site. The VxWorks exploder archive is at csg.lbl.gov=

in
the directory /pub/vxwexplo. Anonymous login is permitted.


John Shaw,
Shade (Computer Services) Ltd, Calne, Wiltshire, England, UK.

jshaw_shade@compuserve.com
john.shaw@shade.co.uk
--MimeMultipartBoundary--


From daemon@csg.lbl.gov  Mon Oct 27 04:01:15 1997
From: daemon@csg.lbl.gov
Date: Mon Oct 27 04:01:19 PST 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Mon Oct 27 04:01:12 PST 1997

        Subject: Re: 2000 problem

-------------------------------------------------------

Newsgroups: comp.os.vxworks,comp.realtime,comp.unix.programmer,comp.lang.c,comp.lang.c++
Subject: Re: 2000 problem
Date: Fri, 24 Oct 1997 21:05:30 -0400
From: Alicia Carla Longstreet 
Organization: The Computer Solution
Message-ID: <345145DA.5EE9@ici.net>
References: <62fk8h$mjq@bcarh8ab.bnr.ca>
	    <34509979.60FFA516@absyss.fr> <1997Oct24.131957.19653@leeds.ac.uk>
Reply-To: carla@ici.net

R S Haigh wrote:
> 
> In article <34509979.60FFA516@absyss.fr>, Doug Seay  writes:
> > James Youngman wrote:
> > >
> > > >>>>> "R" == R S Haigh  writes:
> > >   R> Well, if I look at the race clock during a race and it says
> > >   R> 1:12:34, that's 1 hr 12 min 34-point-something from the start of
> > >   R> the race.
> > >
> > >   R> 1997/3/14 should be 1997 years, 3 months and 14 days from, er...
> > >   R> 29th Nov 2BC, I make it.
> > >
> > > 2BC didn't _have_ a November, the Romans invented it later -- or was
> > > that just August and October?
> 
> I think the Romans had done mucking about with the calendar some
> time before 2BC
> 
> > July comes from Julius (as in Ceasar) and August comes from Augustus
> > Ceasar.  I think that they both took prexisting months and renamed them,
> > although Augustus stole some days from February so his month would be as
> > long as his adopted father's.
> >
> > I think that September through December are "original" because the first
> > few leters come from numbers.  "November" means the 9th month.  I don't
> > know which months are the two new ones though.  July and August would
> > sound reasonable, but that isn't what I (vaguely) remember.
> 
> I guess there were always 12, but March was reckoned as the first,
> which is also why the leap day is at the end of February

If memory serves, the Roman calendar started with the month of March
(hence the import of the Ides of March, this was the first one of the
year), that would make January and February as the newest months.

February was not shortened to add to August, rather it was shortened by
Pope Gregory to reduce the number of days to 365.

- -- 
Some people are so sensitive that they feel snubbed
if an epidemic overlooks them.
                                           Kin Hubbard
***********************************************************
You have not converted a man because you have silenced him.
                                           Lord John Morley
***********************************************************
Alicia Carla Longstreet     carla@ici.net
READ THE FAQ for more information:
C-FAQ ftp sites: ftp://ftp.eskimo.com
                 ftp://rtfm.mit.edu
Hypertext C-FAQ: http://www.eskimo.com/~scs/C-faq/top.html

---------------------------

End of New-News digest
**********************


From wong@vms.masd.northgrum.com  Mon Oct 27 10:53:19 1997
From: wong@vms.masd.northgrum.com (J. Pierre Wong)
Date: Mon Oct 27 10:53:22 PST 1997
Subject: MVME162-510
Greetings,

We just received some MVME162-510's and I tried to boot the boards using
the WRS supplied bootroms and a MVME712M transition board. Is anyone familiar
with the following error message?

Attaching network interface ei0... 
ei: device did not initialize       
failed: errno = 0x3d0002.

Thanks,


Pierre Wong
wong@vms.masd.northgrum.com


From dillow@ataway.aptec.com  Mon Oct 27 13:35:26 1997
From: dillow@ataway.aptec.com (Mike Dillow (staff))
Date: Mon Oct 27 13:35:29 PST 1997
Subject: Job Opportunity

Applied Technology Associates, a small Albuquerque, NM based R&D company,
is looking for Realtime Programmers who are experienced with VxWorks and
VME.  Experience with hardware integration, TCP/IP, and X11/Motif is a plus.
The new employee would be involved in a diverse range of interesting research
projects.

A similar position also exists at our White Sands Missile Range facility.

If you are interested, please E-mail or FAX your resume.
Sorry, US Citizens Only.

      Send resumes to

         Mike Dillow
         Applied Technology Associates
         1900 Randolph, SE
         Albuquerque, NM  87106

      or E-mail to
         dillow@aptec.com

      or Fax to
         (505) 768-1379

Any questions can be sent via E-mail.
ATA is an equal opportunity employer.



From dropka@aptec.com  Mon Oct 27 14:02:08 1997
From: Tom Dropka 
Date: Mon Oct 27 14:02:23 PST 1997
Subject: Re: MVME162-510
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi,

Check out     
  http://www.mot.com/GSS/MCG/products/boards/vme/mvme162/mvme162.html

The 162-510 doesn't have built-in SCSI or Ethernet.

Tom
--MimeMultipartBoundary--


From daemon@csg.lbl.gov  Tue Oct 28 04:02:51 1997
From: daemon@csg.lbl.gov
Date: Tue Oct 28 04:02:54 PST 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Tue Oct 28 04:02:47 PST 1997

        Subject: Re: 2000 problem
        Subject: Re: 2000 problem
        Subject: Re: 2000 problem
        Subject: Re: ioctl SIO_MODE_SET=SIO_MODE_POLL

-------------------------------------------------------

Newsgroups: comp.os.vxworks,comp.realtime,comp.unix.programmer,comp.lang.c,comp.lang.c++
Subject: Re: 2000 problem
Date: Mon, 27 Oct 1997 10:04:55 -0000
From: "Kornel Lambert" 
Message-ID: 
References: <62fk8h$mjq@bcarh8ab.bnr.ca> <62ijm7$qmu@bgtnsc02.worldnet.att.net> <344DCDF8.26351658@absyss.fr>


Doug Seay wrote in message <344DCDF8.26351658@absyss.fr>...
>Harold Rabbie wrote:
>
>>   A few years ago, assuming that the date was in the 20th century was
>> reasonable.  However, in 26 months, it will be a very unreasonable
>> assumption.
>
>The year AD2000 is still in the 20th century, so it is perfectly
>reasonable and correct assumption.  The 1st century is AD1 to AD100, the
>2nd century is AD101 to AD200 and the 20th century is AD1901 to AD2000.
>The 21st century won't start until AD2001.
>
>Years start with 1, not with 0, so there are a few boundry problems.
>Life would be easier if non-computer people understood the usefulness of
>zero, although a AD0 and a 0BC would be kinda confusing too.
>
>But you are right in that just over 26 months, the year will start with
>a "20" instead of a "19", and this will break a lot of software.
>
>- doug


has anyone thought of this:

while everyone has been busy writing year 2000 fixes there is another
'doomsday' date 10/10/2001. when this date is reached some code might
inadvertantly scramble date records because the date would have 10 digit
length instead of 9 which will be from 1st jan 2000 ie:
9/10/2000 is 9 digits long but 10/10/2000 is 10 long. so if database records
are patched to year 2000 compliance there is a risk unless the 01/02/2000
date format is used which will allow for digit changes.

none of this makes sense, i know. but there you go.



---------------------------

Newsgroups: comp.os.vxworks,comp.realtime,comp.unix.programmer,comp.lang.c,comp.lang.c++
Subject: Re: 2000 problem
Date: 27 Oct 1997 16:20:21 +0000
From: James Youngman 
Organization: VG Gas Analysis Systems
Message-ID: 
References: <62fk8h$mjq@bcarh8ab.bnr.ca> <344cd284.0@pgh.nauticom.net>
	
Sender: james@XENON

>>>>> "Chris" == Chris Hills  writes:

  Chris> RM supplied most of the schools in the UK with their own 186
  Chris> PC.  It never cought on becasue of the 186 interupt system is
  Chris> different to all other "real" PC's

WhileI was at University, one of the PG students had an RM PC running
version 1.02 of Microsoft Windows.  This was in 1996!

---------------------------

Newsgroups: comp.os.vxworks,comp.realtime,comp.unix.programmer,comp.lang.c,comp.lang.c++
Subject: Re: 2000 problem
Date: 24 Oct 97 00:39:22 -0500
From: randy@avian.dars.muohio.edu (Randy Kaelber)
Message-ID: <1997Oct24.003922.117@nntp.muohio.edu>
References: <62fk8h$mjq@bcarh8ab.bnr.ca> <01bcdf9c$71b5bfc0$daac2ac0@pub00005m>
Followup-To: comp.os.vxworks,comp.realtime,comp.unix.programmer,comp.lang.c,comp.lang.c++

> Bob Goudreau  wrote in article

> I'm continually amazed at how many people profess surprise or
> annoyance at the fact that no year numbered "0" exists in our
> calendrical system.  Why aren't these people also surprised that
> month numbers and day numbers commence with 1 and not with 0 (i.e.,
> so that the year begins on 0/0 instead of on 1/1)?  I don't see why
> year numbers should be any different; it's quite natural to dub
> number the first time unit (day, month, year) as "1" instead of as "0".

Given that our years are actually ordinal, the 0th year doesn't make a lot
of sense. This is actually the 1997th year in the current calendar. We've
had 1996 complete years. That's why the century ends a year after the "big
y2k flip", only at the end of the year 2000 will we have marked to passing
of 2000 years.

- --
   Randy Kaelber:  kaelbers@muohio.edu  http://avian.dars.muohio.edu/~randy/
        DARS Programmer/Analyst, Miami University, Oxford, OH 45056 USA 
         ------------------------------------------------------------

---------------------------

Newsgroups: comp.os.vxworks
Subject: Re: ioctl SIO_MODE_SET=SIO_MODE_POLL
Date: Mon, 27 Oct 1997 17:53:45 -0800
From: joel williams 
Organization: Williams Consulting
Message-ID: <345545A9.167EB0E7@williams-consulting.com>
References: <01BCDDFC.A1EC3440.johill@lanl.gov>

Jeff Hill wrote:

> All,
> 
> Does anyone understand the SIO_MODE_SET ioctl in vx 5.3.
> The i8250Sio driver is implemented following the example of templateSio.c
> and I see the ioctl() request for SIO_MODE_SET=SIO_MODE_POLL
> arriving in the driver, but the driver's sio poll routines do not ever appear
> to be called by the OS when there are characters to transmit? Unable to find
> anything on this in the doc.
> 
> Jeff
>
The polled mode can be switched in/out dynamically. Polled mode
is typically used by debuggers that run under the OS. They are 
used to get Tornado to work over a serial port, and alledgedly 
allow multiple tasks to be debugged at once, (something you can not
do over ethernet).

- -- 
  -----------------------------------------------------
  |/\/\ Phone: 408.249.5574 /\ Fax 408.246.5036 /\/\/\|
  -----------------------------------------------------

---------------------------

End of New-News digest
**********************


From Wynn.Rostek-1@kmail.ksc.nasa.gov  Tue Oct 28 08:10:47 1997
From: Rostek-1 Wynn 
Date: Tue Oct 28 08:10:54 PST 1997
Subject: FL Job Opportunity


I'll be brief.  I'm looking to hire a VxWorks expert to work in
America's
space program. 

Full time job, very good future, still time to get in at
the beginning.  For additional information, email:

wynn.rostek-1@ksc.nasa.gov

Wynn Rostek
Lead Engineer, Advanced Development Laboratory
DNX-20
Kennedy Space Center, FL 32899

Voice: (407) 867-6770
>Fax: (407) 867-7435


From Wynn.Rostek-1@kmail.ksc.nasa.gov  Tue Oct 28 08:21:31 1997
From: Rostek-1 Wynn 
Date: Tue Oct 28 08:21:34 PST 1997
Subject: Fujitsu SCSI drive

Has anyone brought up a Fijitsu M2652SA SCSI-2 drive under VxWorks?
We have two of these drives left over from another project and would
like to use them on our current project.  They had been on Suns, but
we would like to use them with our Radstone PPC board.

Any hints greatly appreciated.

Wynn Rostek
Lead Engineer, Advanced Development Laboratory
DNX-20
Kennedy Space Center, FL 32899

wynn.rostek-1@ksc.nasa.gov

Voice: (407) 867-6770
>Fax: (407) 867-7435


From chip@Fiberlane.com  Tue Oct 28 15:02:51 1997
From: Chip Roberson 
Date: Tue Oct 28 15:02:54 PST 1997
Subject: Add keys to the new license manager
This is a multi-part message in MIME format.
--------------31F28334FF3D8D7E25E3021B
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

I'm running on a Solaris host and have a simple question.
We have 6 license and just purchased 5 more.  When I received
the license key from WRS, it says "# of licenses: 5".

What I want to know is this.  When I install this new license
key, will I have a total of 5 seats or 11?  I can't find the
answer anywhere in the documentation, I can't reach anybody
at key administration, and my sales rep and FAE are both
unavailable.

Thanks,
 Chip
-- 
Charles S. Roberson 
This message may contain two attachments:
My Netscape Address Card and my S/MIME Public Key Digital Certificate
--------------31F28334FF3D8D7E25E3021B
Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Charles Roberson
Content-Disposition: attachment; filename="vcard.vcf"

begin:          vcard
fn:             Charles Roberson
n:              Roberson;Charles
org:            Fiberlane Communications, Inc.
adr:            Suite 200;;1318 Redwood Way;Petaluma;CA;94954;USA
email;internet: chip@fiberlane.com
title:          Manager, Software Development
tel;work:       (707) 793-1942
tel;fax:        (707) 793-9057
tel;home:       (707) 765-1675
x-mozilla-cpt:  ;0
x-mozilla-html: TRUE
version:        2.1
end:            vcard


--------------31F28334FF3D8D7E25E3021B--



From warwicke@syd.csa.com.au  Tue Oct 28 18:46:22 1997
From: warwicke@syd.csa.com.au (Warwick Edwards)
Date: Tue Oct 28 18:46:25 PST 1997
Subject: Help upgrading to 32 Mbyte MVME147SC-2
Dear All,

I am having trouble upgrading the from an 8Mbyte 147 card to a 32 Mbyte
MVME147SC-2.  We only need to use 16Mbyte memory at the moment and as we
have previously had trouble with going to 32Mbyte we are just trying to use
16Mbyte.

So far I have compiled and loaded a 16Mbyte version of vxWorks (by just
changing LOCAL_MEM_SIZE in config.h).  I can access the 16Mbytes with a test
program no problems.

The problem is that we have some periferal cards, one of which is a serial
card which is addressed by the short I/O address map.

I can load and configure our executable on the 8Mbyte cards no problems.  If
I use exactly the same executable and vxWorks (the 8Mbyte version) but use
the MVME147SC-2 card instead when I go to configure the image I get an
exception about the serial card being dead.  The 16Mbyte version of vxWorks
gets the same error.  It seems that somehow vxWorks knows its a 32Mbyte card
and does some things (the addressing?) differently as a result.  I would have
thought having the extra memory would not cause a problem until I tried to use
it.

Anyway, from what I can work out I may have to change the LOCAL_MEM_BUS_MULT in
config.h and the PCC slave base address SLAVE_BASE_CTL defined in
sysProcMunSet() of sysLib.c.  Is this so and if so to what?  Are there other
changes necessary for example to sysLocalToBusAdrs() or sysBusToLocalAdrs in
sysLib.c?

Any help would be greatly appreciated as I have no previous experience with
either VME or vxWorks.

Oh yeah, we are using version 5.02 of vxWorks.

Thanks
Warwick Edwards
warwicke@syd.csa.com.au 


From TOM2910@aol.com  Tue Oct 28 20:56:54 1997
From: TOM2910@aol.com
Date: Tue Oct 28 20:56:57 PST 1997
Subject: Benchmark for ADS860.
Hello,

Does anyone have the benchmark information for the ADS860 ?

Thanks in advance.

-Tom


From jayr@protocol.com  Wed Oct 29 07:42:53 1997
From: Jay Rasmussen 
Date: Wed Oct 29 07:42:56 PST 1997
Subject: Re: Benchmark for ADS860.
--MimeMultipartBoundary
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Tom,

We worked with our FAE to develop the necessary AuxClk (Timestamp)
driver.  WindRiver should be able to provide this and the benchmark tool
to you.  How they can sell an RTOS without providing these figures up
front is a real concern.  By the way, the numbers produced are rather
dismal.  The BSP seems to have a number of opportunities for some
optimizations to be applied.

Jay Rasmussen
jayr@protocol.com
----------------------------------------------------------
the vxWorks Users Group Exploder wrote:
> 
> Submitted-by TOM2910@aol.com  Tue Oct 28 20:56:54 1997
> Submitted-by: TOM2910@aol.com
> 
> Hello,
> 
> Does anyone have the benchmark information for the ADS860 ?
> 
> Thanks in advance.
> 
> -Tom
--MimeMultipartBoundary--


From mmcquade@pop.aa.net  Wed Oct 29 11:47:13 1997
From: Mike McQuade 
Date: Wed Oct 29 11:47:20 PST 1997
Subject: Re: MV2604 dribbling -Reply
--MimeMultipartBoundary
Content-Type: text/plain; charset="us-ascii"

At 10:16 AM 10/23/97 PDT, you wrote:

>There is a very good chance that Motorola will be providing an
Ethernet-related hardware change
>for the 2604 in the near future.  Ask your Motorola rep to keep his/her
ear to the ground.
>


Charlie,


we are having some very strange problems with the ethernet on our 2603
boards, we have the latest dec driver, plus have made the changes to sysLib.c,
to cache the buffers.


Can you elude to these possible hardware changes ?


What are they supposed to fix ?


Thanks.


Mike.


--MimeMultipartBoundary--


From cgrames@mdc.com  Wed Oct 29 12:56:33 1997
From: Charlie Grames 
Date: Wed Oct 29 12:56:36 PST 1997
Subject: Add keys to the new license manager -Reply
--MimeMultipartBoundary
Content-Type: text/plain
Content-Disposition: inline

Chip,

I was presented with this problem.  I just mailed the pertinent information (including the fact that I
was *adding* keys) to license@wrs.com.  WIthin a day, I had a new key which included the total
number of licenses.

Hope this helps.

Charlie Grames
The Boeing Company
(314) 233-1956
Charles.R.Grames@boeing.com

>>> the vxWorks Users Group Exploder  10/28/97 05:02pm >>>
Submitted-by chip@Fiberlane.com  Tue Oct 28 15:02:51 1997
Submitted-by: Chip Roberson 

'm running on a Solaris host and have a simple question.
We have 6 license and just purchased 5 more.  When I received
the license key from WRS, it says "# of licenses: 5".

What I want to know is this.  When I install this new license
key, will I have a total of 5 seats or 11?  I can't find the
answer anywhere in the documentation, I can't reach anybody
at key administration, and my sales rep and FAE are both
unavailable.

Thanks,
 Chip

--MimeMultipartBoundary--


From prb@aplexus.jhuapl.edu  Wed Oct 29 14:23:01 1997
From: prb@aplexus.jhuapl.edu (Paul R. Bade)
Date: Wed Oct 29 14:23:04 PST 1997
Subject: Anyone Using C++

Hi,


I am trying to get my feet wet with Wind's GNU C++ toolset.
It looks to me like not much is supported by GNU.
Yeah, I can send "Hello World" to cout. But, basic things such
as support for the string class are missing. Perhaps these are
available with the Rogue classes, but how close does Rogue's
implementation match the C++ std libraries. How much would
still be missing. How much time would I spend re-writing 
my code when the std libs become available?

I am starting to believe that the only 
effective solution is to use GreenHills C++.
But, I think that presents another set of problems.
Since GNU and Greenhills do not mangle names in the same
manner, things like WindSh will probably not work. And
since GNU and greenhills do not match with respect to EABI 
compliance, we would not be able to take advantage of 
products like Look!. 

Can you make any suggestions?


+====================================================================+
|       __     ____         __     __                                |
|      /\ \   /  \ \       /\_\__ /\ \      Johns Hopkins University |
|     /  \_\ | /\ \ \     / / /\_\\ \ \     Applied Physics Lab.     |
|    / /\ | |\ \/  \ \   / / / / / \ \ \                             |
|   /  \/ |_| \  /\ \_\ / / / / /   \ \ \   Paul R. Bade             |
|  / /\__/_/   \ \ \/_// / / / /    / / /   (301)-953-6000 x8681     |
| / / /         \ \_\  \ \/ / /    / / /    prb@aplexus.jhuapl.edu   |
| \/_/           \/_/   \__/_/     \/_/                              |
|               __     ____         __       __                      |
|              /\ \   /  \ \       /\ \     /\ \                     |
|             /  \_\ | /\ \ \     /  \_\   /  \ \                    |
|            / /\ | |\ \/  \ \   / /\ |_| / /\ \ \                   |
|           /  \/ |_| \  /\ \_\ / / / | | \/ /  \ \                  |
|          / /\  / /   \ \ \/_// / / / /    / /\ \_\                 |
|          \ \/ / /     \ \_\  \ \/ / /     \/ / / /                 |
|           \__/_/       \/_/   \__/_/        /_/_/                  |
|                                                                    |
+====================================================================+


From mam@pluto.dspt.com  Wed Oct 29 14:58:35 1997
From: "Mark Menge" 
Date: Wed Oct 29 14:58:39 PST 1997
Subject: FTP Server not allowing login
I am having a problem where the vxWorks sever allows me to connect,
but will not allow a login.

I can't find anywhere in the manuals where vxWorks even talks about
ftp server security.

Any ideas out there?

Thanks in Advance
Mark Menge
mmenge@dspt.com


From cgrames@mdc.com  Wed Oct 29 15:46:27 1997
From: Charlie Grames 
Date: Wed Oct 29 15:46:31 PST 1997
Subject: MVME2604 Ethernet (orig. Re: MV2604 dribbling)
--MimeMultipartBoundary
Content-Type: text/plain
Content-Disposition: inline

All,

>There is a very good chance that Motorola will be providing an Ethernet-related hardware change
>for the 2604 in the near future.  Ask your Motorola rep to keep his/her ear to the ground.

I may have been a little premature making this comment about potential pending MVME2604
Ethernet changes from Motorola.  They thought they had fixed a problem we were having, but they
haven't yet.  The problem search is still in progress.

The problem is this:

We were using 2604 boards in conjunction with an SMC TigerSwitch 8 Ethernet switch, which
uses a store-and-forward switching method.  The Ethernet switch will not forward any packet it
considers to be in error.  Our application uses UDP/IP almost exclusively, and we were seeing a
large number of Ethernet packets never make it to their destination.  TCP/IP was affected as well,
though since the protocol has a built in recovery method, the problems manifested themselves as
slowdowns (sometimes considerable) rather than outright failures.  Such slowdowns impact the
success of loading large objects and doing symbol table synchronization.

My initial finger pointing was at SMC, though after considerable testing with them (not to mention
the fact that I got the same erroneous behavior with a NetGear Ethernet switch), I was led to
believe that the problem was with the 2604 itself.

At this point, I still do not know the source of the problem.  Motorola engineering in Tempe has all
the equipment required to reproduce the problem, and I am hoping they determine the cause of
the problem and its solution in the very near future.

Sorry for the false start.  I will try to keep this forum informed about what I find out.

Charlie Grames
The Boeing Company
(314) 233-1956
Charles.R.Grames@boeing.com

P.S.  We are currently using a 3Com SuperStack II 3000 switch instead of the SMC switch.  We are
NOT having problems with this configuration.
--MimeMultipartBoundary--


From chip@Fiberlane.com  Wed Oct 29 19:20:09 1997
From: Chip Roberson 
Date: Wed Oct 29 19:20:13 PST 1997
Subject: Re: Add keys to the new license manager -Reply
This is a cryptographically signed message in MIME format.

--------------ms7FA3C10FDDB8189E947E6239
Content-Type: multipart/mixed; boundary="------------D1C9565F482C48DEF6941642"

This is a multi-part message in MIME format.
--------------D1C9565F482C48DEF6941642
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Thanks to everyone who explained the WRS license management to me.
We finally got everything straightened out.

Thanks,
 Chip

-- 
Charles S. Roberson 
This message may contain two attachments:
My Netscape Address Card and my S/MIME Public Key Digital Certificate
--------------D1C9565F482C48DEF6941642
Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Charles Roberson
Content-Disposition: attachment; filename="vcard.vcf"

begin:          vcard
fn:             Charles Roberson
n:              Roberson;Charles
org:            Fiberlane Communications, Inc.
adr:            Suite 200;;1318 Redwood Way;Petaluma;CA;94954;USA
email;internet: chip@fiberlane.com
title:          Manager, Software Development
tel;work:       (707) 793-1942
tel;fax:        (707) 793-9057
tel;home:       (707) 765-1675
x-mozilla-cpt:  ;0
x-mozilla-html: TRUE
version:        2.1
end:            vcard


--------------D1C9565F482C48DEF6941642--

--------------ms7FA3C10FDDB8189E947E6239
Content-Type: application/x-pkcs7-signature; name="smime.p7s"
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="smime.p7s"
Content-Description: S/MIME Cryptographic Signature
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--------------ms7FA3C10FDDB8189E947E6239--



From daemon@csg.lbl.gov  Thu Oct 30 04:02:01 1997
From: daemon@csg.lbl.gov
Date: Thu Oct 30 04:02:03 PST 1997
Subject: comp.os.vxworks newsdigest
Comp.Os.Vxworks Daily Digest    Thu Oct 30 04:01:58 PST 1997

        Subject: Re: RAD6000

-------------------------------------------------------

Newsgroups: comp.os.vxworks
Subject: Re: RAD6000
Date: Wed, 29 Oct 1997 12:48:57 -0800
From: "jonesboy" 
Organization: Okanagan Internet Junction
Message-ID: <638geo$3ev$1@news.junction.net>
References: <3454E5CF.1D01@omitron.com>

Dan.Reed@omitron.com wrote in article <3454E5CF.1D01@omitron.com>...

>We are going to be purchasing a RAD6000 card for use as a satellite's
>embedded computer.
>
>Does anyone have any gotcha's with this card? How do you talk to it
>since there is no ethernet?
>
>The documentation we have so far is non-existant.
>

    I guess I have a question: why are you even *considering* buying a card
for something as important as a satellite with no communications and no
documentation?  Just a random thought....

- --
Matt Walters, Embedded Systems @ Waverider (http://www.waverider.ca)
jonesboy@waverider.ca
Advertisments proof-read for CDN$200 per word.  Submission of your ad to my
email address constitutes your acceptance of these terms.
(thx. Howard Chu for the great idea)



begin 666 Matt Walters.vcf
M0D5'24XZ5D-!4D0-"DXZ5V%L=&5R
Organization: Okanagan Internet Junction
Message-ID: <63auec$ap$1@news.junction.net>

     I'm trying to write a network driver for VxWorks 5.3.1 without much
luck.  The books and example code don't tell me what I need to know.  Can
anyone send me some code with no hardware-specific stuff in it that is
basically a bunch of stubs (or point me at some)?  I'm going nuts.  :)


- --
Matt Walters, Embedded Systems @ Waverider (http://www.waverider.ca)
jonesboy@waverider.ca
Advertisments proof-read for CDN$200 per word.  Submission of your ad to my
email address constitutes your acceptance of these terms.
(thx. Howard Chu for the great idea)



begin 666 Matt Walters.vcf
M0D5'24XZ5D-!4D0-"DXZ5V%L=&5R
Date: Fri Oct 31 04:37:41 PST 1997
Subject: FTP Server without RAM copies
--MimeMultipartBoundary
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit

Hi

Trying to implement an FTP server on our VxWorks Target system I
found that FTP copies the entire file to be transferred into RAM. We do not
have enough RAM available to allow this. Is there a possibility to have the
target be an FTP Server and directly store received files on a rawFs - Flash
Memory?

Thanks!

Andreas M. Borer


--MimeMultipartBoundary--


From prb@aplexus.jhuapl.edu  Fri Oct 31 08:41:33 1997
From: prb@aplexus.jhuapl.edu (Paul R. Bade)
Date: Fri Oct 31 08:41:36 PST 1997
Subject: Re: MVME2604 Ethernet (orig. Re: MV2604 dribbling)

Charlie Grimes,

I am curious as to the configuration of your MV2604.
Does the ethernet come out the front panel or via P2?

Are you running the Dec21140 in full duplex or half duplex?

Do you see the problem if two MV2604 boards are wired using a
cross over wire?

We are using a similiar PowerPC 604e board from VI computer.
However, we chose to use 100 BaseT4 because we need all of our
I/O to come out through P2 and we were concerned about running
100 MHZ signals through P2. There were no boards with 100 BaseTX
running through P2 on the market at the time. Although if 100 BaseTX
works reliably through P2, we may want to switch at some point in
the future.

We initially had some problems with lost/corrupted packets due
to a problem with the Broadcom T4 chip. But, we now use a newer
revision of the part and all of the problems are gone.
Our problem usually presented itself as packets lost due to CRC errors.

With our problem, we were fortunate in that we could reproduce the
problem with two boards cross wired to each other.
Then we could count UDP packets using the Show commands.
Our test code sent packets as fast as possible.
We changed the socket buffer size so that packets
were not lost due to overflow. With no lost packets there,
we also wrote code to read CSR8 to be sure that packets were
not lost due to a lack of descriptors. 
It is also important to properly set task priorities relative
to NetTask if you want to drive the interface very hard.



+====================================================================+
|       __     ____         __     __                                |
|      /\ \   /  \ \       /\_\__ /\ \      Johns Hopkins University |
|     /  \_\ | /\ \ \     / / /\_\\ \ \     Applied Physics Lab.     |
|    / /\ | |\ \/  \ \   / / / / / \ \ \                             |
|   /  \/ |_| \  /\ \_\ / / / / /   \ \ \   Paul R. Bade             |
|  / /\__/_/   \ \ \/_// / / / /    / / /   (301)-953-6000 x8681     |
| / / /         \ \_\  \ \/ / /    / / /    prb@aplexus.jhuapl.edu   |
| \/_/           \/_/   \__/_/     \/_/                              |
|               __     ____         __       __                      |
|              /\ \   /  \ \       /\ \     /\ \                     |
|             /  \_\ | /\ \ \     /  \_\   /  \ \                    |
|            / /\ | |\ \/  \ \   / /\ |_| / /\ \ \                   |
|           /  \/ |_| \  /\ \_\ / / / | | \/ /  \ \                  |
|          / /\  / /   \ \ \/_// / / / /    / /\ \_\                 |
|          \ \/ / /     \ \_\  \ \/ / /     \/ / / /                 |
|           \__/_/       \/_/   \__/_/        /_/_/                  |
|                                                                    |
+====================================================================+