x x x * REEKE x x x * RRRKE x * ERE x OR RE * x * Ree OK « + & * * RR E x * xxx eT net eee * x * XK x ** * *x * RX RE * KE x * EKERK * *x * * RERRK * * ¥RRR Number 3 April 30, 1976 Page 1 RRR RRR RARE RK RE RK ERT EEK RRR REE ERK * * NOTICE * x This document may contnin information covered * by one or more licenses, copyrichts, and * non-disclosure agreements, Circulntion of * this document is restricted to holders of a * license for the UNIX software csystein from x Western Electric. Such license holders may * reproduce this document for uses in conformi- * ty with the Unix license. * * * * * i All other circulation or reproduction is prohibited. HREM HK HR HOH i TR RRR TOR RRR OR ER SOR IR RR ER A I ROR GE a IE: MEETINGS A report on the meeting in the East will be contained in the next issue. Priefly, it was decided to have the next eastern meeting at Harvard on October I-3. The mecting will start at 2:00 PM on friday and continue through sunday. The invitation to have our firat “globai" meeting in Urbana, Illinois during the early summer of 1977 was extended. More detnils on these meetings will be forthcoming. MANUALS The Science Center at Harvard is willing to undertoke the task of reproducing and distributing the mannale for UNIX. The enclosed order form may be returned no later than May 28, 1976. All orders received by that date will be part of the firet printing. Closing dates for subsequent printings, if any, will be announced. ‘ For logistical rensons, each of the manuals will be available only as indicated below, with 3 hole punching and no binding. Orders should be batched, one order per instnollation ond will be mniled to a sin- gle address which must be at the users’ group's address af record for the instaolintion. The “UNIX PROGRANMER’S MANUAL" Sixth Edition dated May 1975 will be reproduced in ite entirety. Most installations will want to remove severol pages which most users need not know about. "DOCUMENTS FOR USE WITH THE UNIX TIME-SHARING SYSTEM" Sixth Edition will be reproduced omitting sec tions 1 (Setting Up UNIX) and 13 (On the Security of Unix). The: price of each manual will be dependent upou the total number of orders for that manunl and will range hetween 66.00, if 100 copies ore printed, and 612.00, if as few as 10 copies are printed. The order form may be sent in advence of, or with purchase orders or checks. Assume the $12.00 figure. Correct in- voices and refunds will be sent. QUESTIONAIRE The attnched questionsire is the first step in the establishment of nm data base on UNIX installations. When we get the replies, we will make the results available to the users’ group vin this newsletter and will establish a data base on the “hotline“ UNIX system. : BUGS A number of bugs were recorded at the Harvard meeting. A copy of that file was given to D. Ritchie and his fixes, replies, and comments will appear shortly. A bug in pipe.s has been reported by Bill Mayhew at The Children's Museum, Use of a pipe destroys re- @ister R2, the third regiater variable. The fix is to edit, reasscmble and planes in the proper Library. ed source/s2/pipe.s ; mov r2,-Cep) ea - sa sella 9-w0-76 ee, Cupdt r2 / qe /ehr/a : as source/s2/pipe.s my a.out pipe.o ar r /lib/lihe.m pipe.o Yow to Fix your POP-11/4n’ 9 Statte Electricity Problens = or 49 Cents (Plus tox) | R411 tayhew The Children’s tescim, Roston, tA Maren, 1976 Ngere of Migttal’s PNr-1l serias conputers {other than the recent PEN ee been plagued for sone tine with vartous petplexing software/hardware pro! ee thit. are relisted to the influence of static dincharzes on an otherwise-conten machine, The problens have been knaim Co stake several forns: 1. Sudden disablenent of s single terninal on a tinesharing systen; 2. Abrupt failure of the systen as a vnole for no apparent reason; 3. Sudden failure of oll terminals except the console tarminal; & Inexpliceble halts and/or “trpossible” traps. dealing with these (and relnted) The following are sone sungested renedies for es eee Nee DP-11 folklore, robles. Sona of thea are part of the standard PU i : s belteved to de nev. ALL have been used uith succeeds; the author’ s poril/4o ere, ten has now run for nearly four yeeks nonstop, prior to vaich 4t had ctashe virtually datly. L. DLLL/terniaal-related problens retatss oo rom the system without warning, it { the Interrupt Enable bit in one terninal’s associated control unit \hen a single tamiaal appears to disappear f: fa cypically due to the sudden diddpyearanes ° f the tim device status registers in the cub. Nose frequently, these problens are insripated vhen a person scith a eee buildup of static electricity on his hody or clothing amie Le NA . 2 Ler bunped asainst the netal 8 terminal, or hea metal legged chairs are ; oe exe Th in to take place in spite of an terminal. This phenomenon has been baowm ? . s her bizarre capability « stinn, and has oecasionally dononstrated the tathe Seen ate? ae eerminal, many feet distant fron the terninal inte vbich the static was discharged! The easiest method to deel with this ey pee sp Be eres on netal~lezeed chairs in the terminal roon wi th wooden a: 4 Lee ees Roered to relieve uptards of 90% of thea difficultics. Ai BECSRE OPH eae te contact Pigital ici Service; there isa device, the 700 s ee ee ah atalled on each PULL, where the cable fron the terminal plugs «! e Ppa ere ail at (note that this is applicable to 20-nA Sunray ee ees nais only, but Lid terminals do not generclly exhitic this nee : alec anwway). According to represantatives at the Uslthan, HA Rigital © eey a es Breet the H7NG? {e covered under the service contract on your nach te Coe suming you have one); there has, hovever, beon controversy en this p rae in Sher’ field Series offices. If you encounter difficulty, pleage co auahor so that sn attempt can pe nade to get a definitive ruling. Yot a third approach is via software. llany Mgital oneea kath myeteat Sy " second (as determina y ce sal- porated modifications so that, every i “gerne tied 0 eee gece k), the systen cycles through ail of the relevant , gt pasa dntetcupescntle bits on. This nay be occessary to adapt te yeur sygtem 1f you are expertencing a severe case, TL. Kili-assoctated problens, Static discharges through the cabinct of the system, the “poumr™ lock switch on tho front panel, ete. have frequently been known ta cause the Interrvpt Enable bit on the MIL line. clock to drop inexplicably. This typically vill cause all “terninals (except poasibly the console) to “disappear fron the system, even though a glance at the front pancl lights indicates that the systen is behaving "normally". Agide from the general guidelinca (for which see below), the dest vay to deal uith this problem is via software. In MIX, when the systen {is not busy, it executes a subroutine called idle(), which nay be Found in the source file n40.s (or n4S.8) under the nane " idle”, Prinarily, chis routine executes the "wait" hardware instruction. Innediately before the “wait”, one cam insert « single ingtruction to force the elock’a interrupt=enable bit on: “bis $100,* Ike" (note that Ike oust be -nlobl’ed; it {8 sect to point to the systen clock” in main.c). Thus, every time the systen enters an idle state, it will forca the clock back on. IIL. Front-panel problens The final clasa of gtatic problens typically oceurs through the front panel. It may exhibit itself in several vays, either through inexplicable halta, traps to bizarre locations ("panic: trap" on UNIX), or other strange behavior. There are two hardware fixes for this problen, Prinarily, on PDPLI/40’s at least, it resulta fron the fact that when the panel lock ia enabled, the TIL gates served by the various front-panel otltches are allowed to norely float st a logical "high" level (through e 1K pull-up resistor). In practice, this design just doesn’t provide sufficient noise immunity: if the hale switch hap- pens to be dotm when a static discharge arrives, the noise pulse(s) are suffi- cfent to fool the machine inte thinking it’s really been halted, Similar things can happen when any of the other switches sre depresoed, producing equally bizarre reoults, and virtually eliminating the utility of the "panel Lock” feature, First, the physical keyswitch lock is not connected to a good ground when you receive the machine. You chould connect a leazth.of heavy stranded wire (the heavier the better; 1f all you have {s lanp cord, use several parallel lengrhs) between one of the gscrevs on the mounting bracket for the keyswitch and one of the screws that aecurea the front-panel PC board to the chassia (1.e., ground). Second, if you have an experienced electronics technician around, inetall a lacge capacitor to filter out the noise pulse from the statie discharge. The accompanying diagrans show how this ts done, both elecerfeally (Fig. 1) and mechanically (Fig. 2). As it happens, the 11/40 HY¥11~0 console already coacs with a few convenient mounting holes itn which to mount the capacitor. A suitable capacitor seems to be ov electrolytic L5-microfarad, 35-velt, axial-lead capaci= tor, available from your locas Radia Shack as part nunber 272-1014 for 49 cents, Although the installation process 18 straightforward, please remenber that you're dealing with a mohy~kilobuck conputer system. Only people that have sig- nificant electronics know-how should attenpt it, The author assuies no respon- sibility for the accldental invocation of the Pp(-1l’s sclfi-destruct nechanisa when you’re through. . Jo dastell the capacitor, aad pain access to the keysuitch mountinn bracket to fernit installation of the pround strap, first null the CP assenbly forvard out cf the systen cabinet by about a foot, Then use a Phillips*-head seretlriver to renove the four screws (one at cach corner) that hold the white hezel tn place fren the rear. Menove the bezel, and you will see five nylon serevs that fasten the purple front panel to the machine; reneve these tith a standard screwriver. "ith the machine pouered dowm disconnect the electrical connectinns to the marel: 2 Nerg connectors at the top left that connect various sinnals, and tun pulleoff tabs at the top right that are the posmr connections for_ the systen yotsr control unit. How you can tate off the front panel hy unscrewing, the three bexanonal costa at the ton, and removiny the three Phillips serevs alone the tottos cine, Ise a lowmvettare soldering tren to install the capacitor; be sure ko install it with the correct polarity, aa sho in the finurest Me-nount the FO bard to the ssic, hut do not tisiten the screws aod posts fully yet; Firce place the white bezel in position momentarily to make sure the switches on the PCG board linc unm correctly. Once they do, fasten the hoard in plnee secure— ly. Me-coaneet the four connectors renovel evelior. Attach the ground strap ag shown in the diacran (Tig, 2); replace the purple plastic panel, and the white bezels; poarer un the systen, scruff your feet across the caract a few tines, and watch the CP blithely ignore your attenpts at panel=locl: homicide. yy. erat Guidolines The solution of static problens in general can be nided by several general tech niques. Tirst, evoid the use of statie~producing, carpets in the machine xroon, Inatall «a hunidifier 4f necessary to keep humidity at a decent level, even on cold dey days. Second, he sure the systen is adequately grounded; install a heavy ground strap if necessary between the systen cabinets and sone rood eirth/buiidiny ground. If all else fails, consider hermetically sealing the ind yout terminal and re-installine the aysten at the deep end of your 8 siimning pcol... 5: a4 - uy ti POCES: lorr} Figue 4. Schematic WES pint D-<5~ 5409701 -O-1) 1823 at Cx pre-d. clled holes Gre und sia Fiqwe 2. Physical Lastallahon PURDUE UNIVERSITY RCHOOL OF MECHANICAL ENGINEERING WEST LAFAYETTE, INDIAMA 47907 March 26, 1976 Professor Melvin Ferentz Physics Department Brooklyn College of CUNY Brooklyn New York 11210 Dear Professor Ferentz: As a "new" user of the UNIX timesharing system, I have found the “UNIX NEWS" very informative and helpful. The UNIX system is slowly gaining recognition (and appreciation) from the numerous POP-11 sites throughout the Purdue campus. I feel that your impressive list of users, as well as the descriptions of their software development efforts will assist a few of us in preaching the merits of UNIX. I would appreciate being added to your subscription list. Please send an invoice to my mailing address below: Professor David C. Anderson Computer Aided Design & Graphics Laboratory Scheol of Mechanical Engineering Purdue University West Lafayette, Indiana 47907 Phone: (317) 49-39385 Our laboratory supports research and instruction in inter- active computer graphics. The activities include graphics systems development (hardware and software), graphic languages, multi-processor graphics, and a wide range of interdisciplinary application areas. We intend to use UNIX as a base timesharing system on which to build a multi-processor, multi-device inter- active graphics environment. The current configuration includes: PDP 11/40 with 56K memory, LA 30, floating point, 2 RKG5 disks, Dectape, 4 DLI1-C, 2 DRII-C, 2 IMLAC PDS-1's (8K and 4K) with disk, 16 channel DAS A/O system with a range of input devices, Tektronix 4014, Calcomp 502 flatbed plotter, Interdata Model 70 with 8K words and drum supporting a 3D display processor, Control Logic (Intel) 8008 micro, Altair 8080, ARDS storage tube and TTY 37. At the present time we are only experimenting with UNIX, anticipating a changeover from DOS within the year. Here is some software that we have developed that may be of interest to others: AC program to read a named DOS linked file from disk or dectape and copy to standard output. - 5-4 ‘Professor Melvin Ferentz March 26, 1976 Page 2 A DRIi-C word-oriented driver resent] 1 t character queues). & piene ee ie An Interdata Model 70 epoescatseabion based on tand totall compatible) with AS. All variations of the branches were consolidated into PDP-11 mneumonics - the assembler handles ail the short/long, register, forward/backward garbage. An extensive FORTRAN graphics Tibrary taking ful of the Tektronix 4014 features. : op BULAaE aun ae A dynamic, structured graphics package for driving an Beane IMLAC PDS-1 from C or FORTRAN through a Sincerely, MH ies David C. Anderson Assistant Professor Mechanical Engineering and Computer Science DCA: mdy University of Technology LOUGHBOROUGH EEICESTERSHIRE LEtr 3TU : Tei: 0509 43171 Telex 34319 Telegrasmu Technology Loughborough DEPARTMENT OF COMPUTER STUDIES Head of Department Professor D. J. EVANS DWT / JMB 12th April; 1976 Prof. Melvin Ferentz, Physics Department, Brooklyn College of CUNY, Brooklyn, : N.Y. 11210, U.S.A. Dear Professor Ferentz, Will you please enrol us as a member of the UNIX users group. I should be extremely grateful for any available literature on UNIX that you could supply me, back issues of newsletters, program write ups, etc.... In particular I should like to contact anyone who has a ‘standard' BASIC or BASIC-PLUS interpreter available, or who has interfaced D.E.C. software onto UNIX (perhaps through an RT-11 or DOS-11 emviator). In addition 1 have heard rumours that Steve Bourne is implementing ALGOL 68C at Bell, do you have any further information? If there is not a U.K. UNIX sub-group in existence then we should be very willing to initiate such a group. For your information our UNIX system will be running on a PDP 11/40 with 60K words of memory, 2 RK@S disc, a dual floppy disc drive (D.E.C. model - anyone with a driver?7), about a dozen teletypes and V.D.U.s, a teletype 40 R.0.P. printer as a lineprinter and various odds and ends. It is used for undergraduate teaching and postgraduate research. We are currently running the 11/40 under RSTS V4 ~- hence the desire for BASIC or BASIC PLUS - and hope to cut over to UNIX completely in October 1976. Software that we hope to implement under UNIX (timescale unknow at present) include the DECnet protocols, a logic patch-board simulator, possibly a LISP interpreter, a computer usage accounting package, UNIQUE - a machine independent JCL compiler for use on networks, a floppy disk driver, graphics packages, etc. Looking forward to hearing from you. Yours sincerely, whe! 3 J D.W. TAYLOR (Mr.) 525: Tevas Student Publications P.O. BOX D AUSTIN, TEXAS 78712 (512) 471-5244 March 23, 1976 Prof. Melvin Ferentz Physics Department Brooklyn College of CUNY Brooklyn, Nev York, 11210 Dear Prof, Ferentz: The Computation Center of The U: ft received a diatribution tape eine te lie doen mean Sixth Edition". That document gives your name as the contact for a UNIX users' group. As the document ves not dated, I have no idea how. oF : it is. If you are no longer the person responsible for the UNTX users! group mailing list, perhe i cere eae hs Uist, perhaps you can redirect this letter to the proper i was able to bring up UNIX vith very little trouble. Now I an vorking oe configuring it for our system. I am having some difficulties there ecause the only documentation I have so fer is an old copy of the UNIX PROGRAMMER'S MANUAL, fifth edition. ; we only the short introduction that I have had to UNIX I am coapieretys a hrallea by it. It is the most unusual, interesting, and nice operati, ystem with which I have worked. In fact. I em so captivated by the system thet I am vw ‘ to UNIX. ery strongly disposed to convert our production operation Naturally, we are hesitant to invest 5 the time, money, and effort th be required to convert our operation to UNIX vithout knoving more Spor aes Before asking the questions I should deser i a ibe our present system confi, bo eee po costing is a PDP-11/45, without Thoating point bavivares were is: K core menory, tvo RK@5 disks, CR11 c : > ard read PAL (eypeset) high speed paper tape reader, PA611 high speed paper Sine unc! level), tvo LPC¢1 Photocomposition mechine interfaces, a DH1L with The University of Texas at Austin » Texas Student Publications Building 3.200 Prof. Melvin Ferentz 3/23/76 - Page 2. 16 lines, DQl1 synchronous interface, and operator console. We also have _on erder an LA36 to go on an edditional DL11 line. “ Qur non-DEC equipment comprises: 64K semiconductor memory system (giving us a total 112X of memory), a dual density (9800/1600) 9-track tape drive, rodel 160CA Versatec printer/plotter, two Photon Pacesetter photocomposition machines, Hendrix OCR-I optical scanner, and sixteen Ontel OP-I programmable intelligent terminals. We ere running under the RSX-11D version 6A operating system. We have designed and written software in MACRO 1] to: drive the PC machine, read and convert TTS coded paper tape, handle OCR output (the OCR is on ea DH11 line end stories ere autonstically spooled to the disk), drive the printer/ pletter, down-line load the terminals, and to implement a "newspaper” editing systen. In FORTRAN we wrote an assembler for the terminal's code, and we are presently developing a newspaper accounting system. This computer system is Jointly supported by the School of Communication and Texes Student Publications, and the Computation Center provides some pro- grascing support, At present, the prinary function of the system is to provide electronic text entry, editing, storage, and typesetting. The system is also used by Journalism students in editing classes. Shortly the system will be used to automate the accounting operations of Texas Student Publications. We would like to expand the text processing capabilities of the system, and here is one area where my interest in UNIX becones keen. We would like to have text formatting capabilities at the level of WROFF, TROFF, EQN, and NEQN. We are also interested in developing or implementing already existing research systens, like: automatic indexing and morgueing, headline writing, text reaiing-level analysis, spelling checker, and others. We also are working to izplezent a link between this system and a CDC 6600 or a DEC 10 system. If you have read this far, I thank you for reading that long-winded exposition about our systen. I hope the foregoing will be helpful to you in formulating answers to my questions by giving a context to these questions. Is there a UNIX users' group? If so, I would like to join. My mailing sddress is: David M. Phillips Computation Center Tne University of Texas Austin, Texas 76712 Can you tell me more about the users' group: How old is it? How large is it? How active is it? Are there members in Texas I can contact? Is UNIX being used in a production environment? In our application reliability 4s a very high priority concern. How does UNIX rate in reliability and stability? We did get a license from Bell Laboratories to use UNIX, but we got a copy of the distribution tape from the Health Sciences Center in Dallas. How can ve Prof. Melvin Ferentz 3/23/76 - Page 3. be sure that this is a current, correct, and complete distribution? In line with that, how can we keep track of new version releases, bug reports end fixes, new softwere developments, and so on? It appears that there have been six editions of UNIX since its inception in i970. Is there an on-going program for support, development, bug fixing and enhancements to the system? How stable is the system? That is how often have new versions come and hoy often are they expected for he future? What is the nature of the changes between editions, or put another way, what kinds of changes in programs are needed to implement new systems? How great are the changes required to sviteh to a new version? Do you know of anyone vho had developed network communication facilities for UNIX? Can I get a list of user-developed programs that are not part of the standard distribution? Is there some sort of formel program for software interchange? Do you know if anyone has made the conversion that I contenplate--from RSX-L1D to UNIX? What is the DEC corporate feeling about UNIX? Is there any chance ‘of UNIX being subsumed by DEC? What would I get out of the UNIX meeting at the DECUS symposium this May? Who do I contact about documentation? How good is "C" as a system implementation lenguage? Well, with thet last flurry of questions I think I have exhausted my store and probably your patience, I really will appreciate whatever help you can give me, even if only to direct me to some other person(s) for the answers. One final thing, I know it is presumptuous of me, particularly in light of how many questions I asked, but I would like to ask one more fevor, and that is for an early reply. I ask this only because we very recently received URIX and in the next month we need to decide whether to purchase a software support contract from DEC, We are sophisticated programmer types, end ordinarily ve maintain our system software ourselves, but we feel the cost of sources for RSX-11D is too great. Thank you. 2 Very truly yours, -f ye pad Rauf Mi. Nn David NM. Phillips =~ DHP/an : From Dennis Richie via Lou Katz (Columbia) ANOTHER BUG FIX C library “7 resetr setexit resel(x) (Will generate the last call a ‘retura® from w setexit() by restoring spy rS The returned value is x on the orisinal call tite returned value is 0. useful for duing back to the main loop” sitter a horrible error routing. From corrected version via DMR of Feb 11,76 x 4 / é 4 4 4 and doing a return. 4 4 / / 4 yi salob1 -setexit edlobl _reset ~setexit? ‘e@lobl csvs cret mov eSrer5 may (sP)ssec mov SPrssP elr ro rts Pc —Teset: dsr rSresv ts mov ACrS) ero cmr (rSdesrs bea if = mov (rS)ers bre ib 7. panic -~ 72-4 lost é, mov SSP rsp mov srSers mow spcr(sp) rts PC tt mov sper2¢r5) jmp cret bss srai spot sspi in a lowlevel UNIVERSITY OF ILLINOIS AT CHICAGO CIRCLE DEPARTMENT OF INFORMATION ENGINEERING COLLEGE OF ENGINEERING BOX 4348, CHICAGO, ILLINOIS 60880 TELEPHONE: 996-3423 April 9, 1976 aS Professor Melvin Ferentz Physics Department Brooklyn College of CUNY Brooklyn, New York 11210 Dear Mel, A gem has popped out of the woodwork. Please add to your mailing list: Mr. Pat Fitzhenry Aviation Research Laboratory Savoy, Illinois 61874 They are covered by the University of Illinois license. They're running a Cal Data with FIS, and have in desperation rewritten UNIX FORTRAN to use FIS and have eliminated fptrap. They've agreed to distribute through me and I've sent them a tape. They claim that FORTRAN programs that used to run 30 sec under DEC FORTRAN, and about 8 minutes under UNIX FORTRAN, are “back at least on the order of 30 sec". Sounds good, if true. I'm enclosing the form I've drawn up for the survey. It's long, but complete. Yours, Mike O'Brien MTO: bb Enc. FIRST ANNUAL UNIX POLL AND SURVEY General Information Name: Organization: Mailing Address: Phone Number: Installation Location: Contact Person(s): Phone Number(s): Applications: Major Application: Major Subprojects: Future Projects: 3. Hardware CPU MEMORY DISK Type: Type: Type: Manufacturer: Size: Manu. : Options (f.p., etc): Manu. : No. Type: Type: Size: Manu. : Manu. : No. type Manu. : No. How old is your system? Terminal Interfaces Magtapes Type: Type: _No.: No.: Manufacturer: Manufacturer: Type: No.: Do you have: 5 Manu: High-speed paper tape? Type: DEC tape? No.: Special Interfaces? (Describe) Manu. : Have you connected other machines to UNIX? If so, how? Software Drivers (other than tty) changed: Driver: Changed: Driver: Changed: Driver: Changed: Drivers Written: Changes to: Scheduler: Swapper: tty: Sys calls added / modified: Calling #: Function: Dependent on other changes? Calling #: Function: Dependent on other changes? Have you made major modifications to any of the distributed Compilers: Other programs: Have you written any major packages of general interest? Do you need any packages or drivers which others may have written? Specify six characters to be used as the installation code Return completed form to: Prof. Melvin Ferentz Physics Dept. Brooklyn College of CUNY Brooklyn, New York 11210 aan tf Mr. Lewis A. Law Director of Technical Services Science Center, Harvard University 1 Oxford Stroet Cambridge, MA 02135 ~ Please have printed the following quantities of manuals for the use of our installation. We are licensed by Western Electric or Bell Telephone for the Sixth Edition of UNIX, Copies of "UNIX PROGRAMNER’S MANUAL" Copics of "DOCUMENTS FOR USE WITH THE UNIX TIME SHARING SYSTEM" Our mailing address is: (Check appropriate lines) ———-— A purchase order is enclosed. ———-— A check for 612.00 per volume is enclosed. Please refund the excess if any. —~-~— Please send an invoice for the correct umount per volume. A check will follow. ——-~ A purchase order will follow. (Additional instructions or comments) signature a ERROR CORRECTION The Line on rade 1 bottom which reads dae 3 /elr/a The complete edited text foliaws} ~~ ( Library 2 ® ¢ a Pipe lf) int fl205 Se OS ;dilobilo creires cerror Pipe = 424 ~PiPat may rore-(se) moy BSP rTs BUS Pire hee ay its cerrer 13 mav P29~(sr) 78K Fix may 4(rS dere mo rOr(p2)+ mOYyV ris(r2) clr ro mov (SP) 49P2 “EK Fin mov (SP )be pS rts PO