From ta02@academia.swt.edu Tue Feb 8 13:43:25 1994 Received: from nyssa.swt.edu by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA15220; Tue, 8 Feb 1994 12:43:58 MST Received: from [147.26.101.110] (grasshopper.cs.swt.edu) by academia.swt.edu (PMDF V4.2-15 #3941) id <01H8N88BNWGG936LZL@academia.swt.edu>; Tue, 8 Feb 1994 13:43:25 CST Date: Tue, 08 Feb 1994 13:43:25 -0600 (CST) Date-Warning: Date header was inserted by academia.swt.edu From: ta02@academia.swt.edu (Tod Amon) Subject: SR solution to Stable Marriage Problem To: info-sr@cs.arizona.edu Message-Id: <01H8N88BVXSY936LZL@academia.swt.edu> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7BIT I teach a class at Southwest Texas State University on Concurrent Programming. If anyone has an SR solution to the stable-marriage problem (exercise 2.20 in the Andrews text) I'd love to see it. I have a message-passing solution. I'm more interested in seeing a simple solution that a student who has not read past chapter 2 could potentially come up with. Much Thanks. -Tod Amon (ta02@academia.swt.edu) p.s. yeah, you guessed it, I assigned this problem as homework, but am having trouble seeing the 'obvious' simple solution, :=). Prompt reply appreciated! From @exeter.ac.uk:steve@dcs.exeter.ac.uk Wed Feb 9 13:55:58 1994 Received: from sun2.nsfnet-relay.ac.uk by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA10127; Wed, 9 Feb 1994 07:13:56 MST Via: uk.ac.exeter; Wed, 9 Feb 1994 13:56:26 +0000 Received: from atlas.exeter.ac.uk by exub.exeter.ac.uk with SMTP (PP) id <11237-0@exub.exeter.ac.uk>; Wed, 9 Feb 1994 13:56:06 +0000 Received: from sirius.dcs.exeter.ac.uk by atlas.dcs.exeter.ac.uk; Wed, 9 Feb 94 13:55:59 GMT From: steve@atlas.ex.ac.uk Date: Wed, 9 Feb 94 13:55:58 GMT Message-Id: <24719.9402091355@sirius.dcs.exeter.ac.uk> To: ta02@academia.swt.edu Cc: info-sr@cs.arizona.edu In-Reply-To: Tod Amon's message of Tue, 08 Feb 1994 13:43:25 -0600 (CST) <01H8N88BVXSY936LZL@academia.swt.edu> Subject: SR solution to Stable Marriage Problem Below is a solution to the stable marriage problem. It also uses message passing, but then this problem is particularly suitable for solving using message passing. The first part of the program reads in the data and records the preferences -- an example of a data file together with its output is given after the program. (Note: this solution leaves blocked processes after solving the problem.) If you get an alternative solution I'd quite like to see it. Steve ------------------------------------------------------------------------------- Parallel Systems Research Laboratory | Stephen Turner Email: steve@dcs.exeter.ac.uk | Department of Computer Science Fax: +44 392 264067 | University of Exeter Tel: +44 392 264048 | Prince of Wales Road Telex: 42894 EXUNIV G | Exeter EX4 4PT England ------------------------------------------------------------------------------- # Stable Marriage problem resource StableMarriage() const strlen := 20 var fname: string[strlen] getarg(1, fname) var f: file := open(fname, READ) # open file of names and preferences var N: int # number of Men and Women read(f, N) var name[1:N]: string[strlen] fa i := 1 to N -> read(f, name[i]) af # read names of Men and Women var M: int := N/2 # number of each sex var prefer[1:M, 1:M]: int # Men's preferences var rank[M+1:N, 1:M]: int # Women's rankings var p: int fa i := 1 to N -> writes(name[i], " prefers: ") fa j := 1 to M -> read(f, p) # read preferences of Men and Women if i <= M -> prefer[i,j] := p # record Man's preference [] else -> rank[i,p] := j # or Woman's ranking fi writes(j, ", ", name[p], "; ") af write() af write() op propose[M+1:N](int) # indexed by Woman number op answer [1:M](bool) # indexed by Man number # life history of Man process Man (i := 1 to M) var accepted: bool := false var choice: int var c: int := 0 do true -> nap(int(random()*1000)) # propose to Women in order of preference do not accepted -> choice := prefer[i,++c] send propose[choice](i) receive answer[i](accepted) od # wait to see if jilted receive answer[i](accepted) od end # life history of Woman process Woman (i := M+1 to N) var best, suitor: int receive propose[i](best) # receive first proposal send answer[best](true) # accept first suitor write(name[best], "accepted by", name[i]) do true -> receive propose[i](suitor) # receive new proposal if rank[i, suitor] < rank[i, best] -> # new suitor better than previous best send answer[best](false) # previous best jilted write(name[best], "jilted by", name[i]) best := suitor send answer[suitor](true) # accept new suitor write(name[suitor], "accepted by", name[i]) [] else -> send answer[suitor](false) # reject new suitor write(name[suitor], "rejected by", name[i]) fi od end end StableMarriage ------------------------------------------------------------------------------- 10 Andrew Brian Charles David Edward Angela Brenda Carol Diana Elizabeth 6 8 10 9 7 9 10 7 6 8 6 8 9 7 10 8 10 7 9 6 6 8 7 10 9 2 5 3 1 4 3 4 2 5 1 5 1 3 4 2 5 1 4 2 3 1 2 3 5 4 ------------------------------------------------------------------------------- 51 % a.out mdata10 Andrew prefers: 1, Angela; 2, Carol; 3, Elizabeth; 4, Diana; 5, Brenda; Brian prefers: 1, Diana; 2, Elizabeth; 3, Brenda; 4, Angela; 5, Carol; Charles prefers: 1, Angela; 2, Carol; 3, Diana; 4, Brenda; 5, Elizabeth; David prefers: 1, Carol; 2, Elizabeth; 3, Brenda; 4, Diana; 5, Angela; Edward prefers: 1, Angela; 2, Carol; 3, Brenda; 4, Elizabeth; 5, Diana; Angela prefers: 1, Brian; 2, Edward; 3, Charles; 4, Andrew; 5, David; Brenda prefers: 1, Charles; 2, David; 3, Brian; 4, Edward; 5, Andrew; Carol prefers: 1, Edward; 2, Andrew; 3, Charles; 4, David; 5, Brian; Diana prefers: 1, Edward; 2, Andrew; 3, David; 4, Brian; 5, Charles; Elizabeth prefers: 1, Andrew; 2, Brian; 3, Charles; 4, Edward; 5, David; Andrew accepted by Angela Brian accepted by Diana Andrew jilted by Angela Charles accepted by Angela David accepted by Carol Charles jilted by Angela Edward accepted by Angela David jilted by Carol Andrew accepted by Carol Charles rejected by Carol Charles rejected by Diana Charles accepted by Brenda David accepted by Elizabeth From pheras@tornasol.uc3m.es Mon Mar 21 15:44:59 1994 Message-Id: <199403211448.AA14142@optima.CS.Arizona.EDU> Received: from tornasol.uc3m.es by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA14142; Mon, 21 Mar 1994 07:48:11 MST Received: by tornasol.uc3m.es (16.7/15.6) id AA20246; Mon, 21 Mar 94 15:44:59 GMT From: Pedro de las Heras Subject: Linux- SR 2.2 To: info-sr@cs.arizona.edu Date: Mon, 21 Mar 94 15:44:59 GMT Mailer: Elm [revision: 66.33] Hello. The last release of SR (2.2), supports Linux plattforms. Some time ago I compiled it with gcc-2.4.5 and it went ok. Yesterday I installed Linux 0.99.15 (Version 1.0 is current Linux release), and I had problems when I recompiled SR-2.2 Due to changes in the new Linux's libgcc, some SR files which depend on this stuff didn't compile. (rts/io.c and others). SOLUTION: The problem is within the FILE structure. It has changed to become Posix compliant in Linux 0.99.15 and later. Simply defining 4 macros in stdio.h we haveagain SR compiled in this version: ----------------------------- #define _pptr _IO_write_ptr #define _gptr _IO_read_ptr #define _egptr _IO_read_end #define _pbase _IO_write_base ------------------------------ Hope this helps someone. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Pedro de las Heras Quiros. C.S. Engineer e-mail: pheras@tornasol.uc3m.es Universidad Carlos III. C/ Butarque 15. E-28911, Leganes SPAIN +-+-+-+-----------------=======--------=-=-_+_=-=+_+__++_-+_=-__++_-+_+-_+_+__ From GK15004@academia.swt.edu Mon Apr 4 16:37:53 1994 Received: from tegan.swt.edu by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA19949; Mon, 4 Apr 1994 14:38:26 MST Received: from academia.swt.edu by academia.swt.edu (PMDF V4.2-15 #3941) id <01HAS8C31O5S99ELNJ@academia.swt.edu>; Mon, 4 Apr 1994 16:37:54 CST Date: Mon, 04 Apr 1994 16:37:53 -0600 (CST) From: GK15004@academia.swt.edu Subject: SR and X windows. Help! To: info-sr@cs.arizona.edu Message-Id: <01HAS8C33TBM99ELNJ@academia.swt.edu> Organization: Southwest Texas State University X-Vms-To: IN%"info-sr@cs.arizona.edu" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Hi, I hope I have the right address, I am looking for the forum on the SR programming language. I am a rather new user of SR and I need some help with a couple of things. I am working on an SR project and I would like to give it a graphical interface using X windows. The way I see it, I have two options. Either use the SRWin subsystem to call the X functions or link C and SR code and handle the X interface with C. (I am quite experienced with X through C) The problem with SRWin is that I cannot it to work. I can compile the SRWin examples that came with SR, and I can run them but I only get a black window. Nothing is drawn on it. Any suggestions are welcomed. However, my prefered solution would be to link C and SR code as I have a substancial X toolkit written in C. I need to know what protocol I have to use in order to call C functions from SR and SR functions from C. The machine I am using is a UNIX workstation based on the DEC Alpha chip and is running DEC OSF/1 V1.3 Any sugestions on either of the above would be most apprechiated. -- George Kazangas From russh@galileo.tracor.com Tue Apr 5 11:15:23 1994 Received: from galileo.tracor.com by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA19624; Tue, 5 Apr 1994 09:15:45 MST Received: by galileo.tracor.com (4.1/SMI-4.1) id AA14353; Tue, 5 Apr 94 11:15:23 CDT Date: Tue, 5 Apr 94 11:15:23 CDT From: russh@galileo.tracor.com (Russell Hill) Message-Id: <9404051615.AA14353@galileo.tracor.com> To: GK15004@academia.swt.edu Subject: SR questions Cc: info-sr@cs.arizona.edu Hey George: I'm in your Concurrent Programming class at SWT. I had the same results when trying to use the Alphas. I use SRWin on Linux at home and a Silicon Graphics at work and it works great. I think getting SRWin to work on the Alphas is out of your control. Its up to the Sys Admin. The Alphas may run Unix(OSF/Motif), but Digital always has to change their stuff and make it proprietary, (they screwed with Motif to make it look like DECWindows.) I could flame DEC all day... In response to your other question SR can call C functions. It's in the SR manual under "External Operations," check the index but I think it's link page 112. You will need to declare a prototype at the begining of the resource using the keyword "external" and then you can call it just like any other procedure. SR scans through the C libs looking for the procedure or if it is routine you wrote just include the ".o" filename as an argument to "srl". If you have any more questions you can email me directly and I give it shot. -Russell __________________ ---- | | |-\_ ---- | | | |_\ \o/ ---- | |_| | | | ^^oo^^^^^^^^^oo^^^ o^^o^ _/ \_ ----------------------------------------------------------- Russell Hill Russell_Hill@tracor.com Tracor Applied Sciences, Inc. 6500 Tracor Lane MD 1-7 512-929-4082 office AUSTIN TX 78725-2050 512-929-2241 fax From azhao@cc.gatech.edu Tue Apr 5 21:34:52 1994 Received: from burdell.cc.gatech.edu by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA22937; Tue, 5 Apr 1994 18:34:55 MST Received: from oakmont.cc.gatech.edu (azhao@oakmont.cc.gatech.edu [130.207.119.218]) by burdell.cc.gatech.edu (8.6.4/8.6.4) with ESMTP id VAA10613; Tue, 5 Apr 1994 21:34:53 -0400 Received: from localhost (azhao@localhost) by oakmont.cc.gatech.edu (8.6.4/8.6.4) id VAA24428; Tue, 5 Apr 1994 21:34:52 -0400 From: azhao@cc.gatech.edu (Q. Alex Zhao) Message-Id: <199404060134.VAA24428@oakmont.cc.gatech.edu> Subject: Re: SR and X windows. Help! To: GK15004@academia.swt.edu Date: Tue, 5 Apr 1994 21:34:52 -0400 (EDT) Cc: info-sr@cs.arizona.edu In-Reply-To: <01HAS8C33TBM99ELNJ@academia.swt.edu> from "GK15004@academia.swt.edu" at Apr 4, 94 04:37:53 pm X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 1737 According to GK15004@academia.swt.edu in a previous message: ] ... ] The problem with SRWin is that I cannot it to work. I can compile the ] SRWin examples that came with SR, and I can run them but I only get a ] black window. Nothing is drawn on it. Any suggestions are welcomed. I just worked out a patch to the SRWin lib in the 2.2.1 distribution. It fixes most of the problems on DEC Alpha's. However "WinAddPixel()" doesn't work -- this was on a DEC Alpha OSF/1 1.3 and displaying on its own 24-bit display. "WinAddPixel()" is implemented by directly calling Xlib function "XAddPixel()". I tried to use "WinGetPixel()" then "WinPutPixel()" to go around it and that worked ok. This patch has been tested on SGI's too (8-bit and 24-bit monitors). If you are interested, please let me know and I can send you the patch. Other user-visible changes include: winCursor type changed to "pointer to any" winPixel type changed to "pointer to any" (to use 64-bit integers on the Alpha). This is not an official patch though... ] However, my prefered solution would be to link C and SR code ] as I have a substancial X toolkit written in C. I need to know what ] protocol I have to use in order to call C functions from SR and SR ] functions from C. As pointed out by Russell Hill, this could be done by first declaring the C functions as "external" in the SR code, then link the SR code with C code. Examples: see "gsrwin.sr" and "srwin.[ch]" in the "srwin" directory. Cheers, = Q. Alex Zhao ~{0"UT~} ! Graphics, Visualization & Usability Center Email: azhao@cc.gatech.edu ! College of Computing FAX: 404-853-9378 ! Georgia Institute of Technology Lab: 404-894-9633 ! Atlanta, Georgia 30332-0280 From @jeeves.engr.mun.ca:kwong@engr.mun.ca Thu Apr 7 02:15:00 1994 Received: from jeeves.engr.mun.ca by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA27763; Wed, 6 Apr 1994 21:51:05 MST Received: from nano.engr.mun.ca by jeeves.engr.mun.ca id <59426>; Thu, 7 Apr 1994 02:21:39 -0230 Date: Thu, 7 Apr 1994 02:15:00 -0230 From: Kai Wong Subject: Anyone have the source for concurrent eight queens problem! To: info-sr@cs.arizona.edu Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I got stuck with my eight queens assignment, wonder if anyone on the net have the source. thanks! --KAI-- From Jacob.Levy@eng.sun.com Wed Apr 6 22:41:16 1994 Received: from Sun.COM by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA00546; Wed, 6 Apr 1994 22:39:13 MST Received: from Eng.Sun.COM (zigzag.Eng.Sun.COM) by Sun.COM (sun-barr.Sun.COM) id AA20713; Wed, 6 Apr 94 22:39:01 PDT Received: from burgess.Eng.Sun.COM by Eng.Sun.COM (4.1/SMI-4.1) id AA17963; Wed, 6 Apr 94 22:38:02 PDT Received: by burgess.Eng.Sun.COM (5.0/SMI-SVR4) id AA10802; Wed, 6 Apr 1994 22:41:16 +0800 Date: Wed, 6 Apr 1994 22:41:16 +0800 From: Jacob.Levy@eng.sun.com (Jacob Levy) Message-Id: <9404070541.AA10802@burgess.Eng.Sun.COM> To: Kai Wong Cc: info-sr@cs.arizona.edu In-Reply-To: Subject: Anyone have the source for concurrent eight queens problem! Reply-To: jyl@toss.eng.sun.com Content-Length: 175 Homework alert! :-) --JYL Kai Wong writes: > > I got stuck with my eight queens assignment, wonder if anyone on > the net have the source. thanks! > > --KAI-- > From olsson@cs.ucdavis.edu Thu Apr 14 20:52:02 1994 Received: from ivy.cs.ucdavis.edu by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA07260; Thu, 14 Apr 1994 20:52:12 MST Received: by ivy.cs.ucdavis.edu (5.65/UCD.CS.2.4) id AA06648; Thu, 14 Apr 1994 20:52:02 -0700 Date: Thu, 14 Apr 1994 20:52:02 -0700 From: olsson@cs.ucdavis.edu (Ron Olsson) Message-Id: <9404150352.AA06648@ivy.cs.ucdavis.edu> To: info-sr@cs.arizona.edu Subject: SRgetopt -- a global for handling command line options Attached is a very handy global for handling command line options. (It is also available via anonymous ftp from cs.arizona.edu in the sr subdirectory.) Enjoy! # OVERVIEW: # # SRgetopt provides a general means for an SR program to parse command # line arguments in accordance with the standard Unix conventions; # it is analogous to, and based on, getopt(3) for C programs. # (The following documentation is based on the man page for getopt(3).) # DESCRIPTION: # # SRgetopt is an SR global that provides one procedure, getopt, # and some variables that control behavior of or return additional # information from getopt. # # getopt interprets command arguments in accordance with the standard # Unix conventions: option arguments of a command are introduced by "-" # followed by a key character, and a non-option argument terminates # the processing of options. getopt's option interpretation is controlled # by its parameter optstring, which specifies what characters designate # legal options and which of them require associated values. # # getopt returns the next, moving left to right, option letter # in the command line arguments that matches a letter in optstring. # optstring must contain the option letters the command using getopt # will recognize. For example, getopt("ab") specifies that the command # line should contain no options, only "-a", only "-b", or both "-a" and # "-b" in either order. (The command line can also contain non-option # arguments after any option arguments.) Multiple options per argument # are allowed, e.g., "-ab" for the last case above. # # If a letter in optstring is followed by a colon, the option is expected # to have an argument. The argument may or may not be separated by # whitespace from the option letter. For example, getopt("w:") allows # either "-w 80" or "-w80". The variable optarg is set to the option # argument, e.g., "80" in either of the previous examples. Predefined # conversion functions such as int, char, etc. can then be applied to # optarg. (Use optarg[1] instead of char(optarg) if the first blank or # non-blank character is wanted.) The constant optMAXLEN defines the # length of the longest string that getopt can read from the command line # (extra characters are truncated silently) or will return in optarg. # # getopt places in the variable optind the index of the next command # line argument to be processed; optind is automatically initialized # to 1 before the first call to getopt. # # When all options have been processed (that is, up to the first # non-option argument), getopt returns optEOF. getopt recognizes the # command line argument "--" (i.e., two dashes) to delimit the end of # the options; getopt returns optEOF and skips "--". Subsequent, # non-option arguments can be retrieved using the predefined # function getarg, beginning in with argument number optind. # DIAGNOSTICS: # # getopt prints an error message on stderr and returns a question mark # ('?') when it encounters an option letter in a command line argument # that is not included in optstring. Setting the variable opterr to # false disables this error message. # NOTES: # # The following notes describe SRgetopt's behavior in a few interesting # or special cases; these behaviors are consistent with getopt(3)'s behaviors. # # -- A '-' by itself is treated as a non-option argument. # -- If optstring is "a:" and the command line arguments are "-a -x", # then "-x" is treated as the argument associated with the "-a". # -- Duplicate command line options are allowed; it is up to user to # deal with them as appropriate. # -- A command line option like "-b-" is considered as the two options # "b" and "-" (so "-" should appear in option string); this differs # from "-b --". # -- Sun and DEC getopt(3)'s differ w.r.t. how "---" is handled. # Sun treats "---" (or anything starting with "--") the same as "--" # DEC treats "---" as two separate "-" options # (so "-" should appear in option string). # SRgetopt follows the DEC convention. # -- An option `letter' can be a letter, number, or most special character. # Like getopt(3), SRgetopt disallows a colon as an option letter. # EXAMPLE: # # The following code fragment shows how to use SRgetopt to # process a command that takes the options 'a', 'f', and 'w' # where 'f' is followed by a file name and 'w' is followed by an integer. /**** resource main() import SRgetopt var ch: char # command line arguments var aflg := 0 var filename: string[optMAXLEN] := "out" var width := 80 do (ch := getopt("abf:w:")) != optEOF -> if ch = 'a' -> aflg++ [] ch = 'f' -> filename := optarg [] ch = 'w' -> width := int(optarg) [] else -> stop(1) fi od write("-a", aflg) write("-f", filename) write("-w", width) fa k := optind to numargs() -> var xx: string[40] getarg(k,xx) write("normal argument", k, "is", xx) af end ****/ # SEE ALSO: # # getopt(3) # WARNING: # # Changing the value of the variable optind may lead to unexpected behavior. global SRgetopt var optind := 1 var opterr := true # maximum length of an argument # (it is a const, not a var, so not redefinable by user, # because optarg declaration must be here in spec.) const optMAXLEN := 256 var optarg: string[optMAXLEN] # optEOF is a var so user can redefine it if desired. # (also, as a const, it uncovered a bug in 2.2.1 # which requires a patch to fold.c (94-04-06)). var optEOF := char(EOF) op getopt(optstring: string[*]) returns char body SRgetopt procedure write_error(msg: string[*]; ch: char) var cmd: string[optMAXLEN] getarg(0, cmd) writes(stderr, cmd, ": ", msg, " -- ", ch, "\n") end const argc := numargs() var optpos := 2 proc getopt(optstring) returns ch optarg := "" if optind < 1 | optind > argc -> ch := optEOF return fi var thisarg: string[optMAXLEN] getarg(optind, thisarg) var len := length(thisarg) # handle special cases if len <= 1 | thisarg[1] != '-' -> # e.g., "", "a", "abc", or just "-" ch := optEOF return [] thisarg = "--" -> # end of non-option args ch := optEOF optind++ return fi # get next "letter" from option argument ch := thisarg[optpos] # find this option in optstring var found := false var pos: int fa x := 1 to length(optstring) -> if optstring[x] = ch -> found := true pos := x exit fi af if not found | ch = ':' -> if opterr -> write_error("illegal option", ch) fi ch := '?' [] else -> # handle colon, if present if pos < length(optstring) & optstring[pos+1] = ':' -> if optpos != len -> # take rest of current arg as optarg optarg := thisarg[optpos+1:*] optpos := len # to force advance to next arg below [] else -> # take next arg as optarg optind++ if optind <= argc -> getarg(optind,optarg) [] else -> if opterr -> write_error("option requires an argument", ch) fi ch := '?' fi fi fi fi # advance to next option argument, # which might be in thisarg or next arg optpos++ if optpos > len -> optind++ optpos := 2 fi end end From pheras@tornasol.uc3m.es Fri Apr 22 03:29:54 1994 Message-Id: <199404221029.AA13968@optima.CS.Arizona.EDU> Received: from tornasol.uc3m.es by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA13968; Fri, 22 Apr 1994 03:29:54 MST Received: by tornasol.uc3m.es (16.7/15.6) id AA17083; Fri, 22 Apr 94 12:25:47 +0100 From: Pedro de las Heras Subject: Exceptions To: info-sr@cs.arizona.edu Date: Fri, 22 Apr 94 12:25:47 BST Mailer: Elm [revision: 66.33] Hello. Is there any implementation available of exceptions in SR as the one described in the article on Trans. on ACM Prog. Lang. and Syst. in Jan. 1988 ? I'm currently using SR 2.1 Thanks in advance --------------------------------------------------------------------------- Pedro de las Heras Quiros. C.S. Engineer e-mail: pheras@tornasol.uc3m.es Universidad Carlos III. Tlf: +34-1-624 94 97 C/ Butarque 15. E-28911, Leganes SPAIN --------------------------------------------------------------------------- From pheras@tornasol.uc3m.es Fri Apr 22 03:40:52 1994 Message-Id: <199404221040.AA14714@optima.CS.Arizona.EDU> Received: from tornasol.uc3m.es by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA14714; Fri, 22 Apr 1994 03:40:52 MST Received: by tornasol.uc3m.es (16.7/15.6) id AA17109; Fri, 22 Apr 94 12:36:04 +0100 From: Pedro de las Heras Subject: source repository To: info-sr@cs.arizona.edu Date: Fri, 22 Apr 94 12:36:04 BST Mailer: Elm [revision: 66.33] Hello. Is there any repository of source code? Is there any faq or some place where the messages in the list are stored? Regards --------------------------------------------------------------------------- Pedro de las Heras Quiros. C.S. Engineer e-mail: pheras@tornasol.uc3m.es Universidad Carlos III. Tlf: +34-1-624 94 97 C/ Butarque 15. E-28911, Leganes SPAIN --------------------------------------------------------------------------- From pheras@tornasol.uc3m.es Fri Apr 22 03:41:34 1994 Message-Id: <199404221041.AA14763@optima.CS.Arizona.EDU> Received: from tornasol.uc3m.es by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA14763; Fri, 22 Apr 1994 03:41:34 MST Received: by tornasol.uc3m.es (16.7/15.6) id AA17122; Fri, 22 Apr 94 12:37:33 +0100 From: Pedro de las Heras Subject: SR as a teaching tool To: info-sr@cs.arizona.edu Date: Fri, 22 Apr 94 12:37:32 BST Mailer: Elm [revision: 66.33] Hello. I'd like to hear about experiences in the use of SR as a teaching tool at the university. We are planning to use it in a course on distributed systems. Is there any report on this theme? Regards --------------------------------------------------------------------------- Pedro de las Heras Quiros. C.S. Engineer e-mail: pheras@tornasol.uc3m.es Universidad Carlos III. Tlf: +34-1-624 94 97 C/ Butarque 15. E-28911, Leganes SPAIN --------------------------------------------------------------------------- From olsson@cs.ucdavis.edu Fri Apr 22 03:54:19 1994 Received: from ivy.cs.ucdavis.edu by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA15528; Fri, 22 Apr 1994 03:54:49 MST Received: by ivy.cs.ucdavis.edu (5.65/UCD.CS.2.4) id AA16828; Fri, 22 Apr 1994 03:54:19 -0700 Date: Fri, 22 Apr 1994 03:54:19 -0700 From: olsson@cs.ucdavis.edu (Ron Olsson) Message-Id: <9404221054.AA16828@ivy.cs.ucdavis.edu> To: pheras@tornasol.uc3m.es Cc: info-sr@cs.arizona.edu In-Reply-To: <199404221029.AA13968@optima.CS.Arizona.EDU> (message from Pedro de las Heras on Fri, 22 Apr 94 12:25:47 BST) Subject: Exceptions From: Pedro de las Heras Date: Fri, 22 Apr 94 12:25:47 BST Is there any implementation available of exceptions in SR as the one described in the article on Trans. on ACM Prog. Lang. and Syst. in Jan. 1988 ? Those features were never implemented. You might be interested in the paper by Daniel Huang and myself cited in the SR book's bibliography, although those features were never implemented either. You might also be interested in FT-SR, developed by Vic Thomas and Rick Schlichting (rick@cs.arizona.edu); I'm not fully aware of the status of that work, but I believe at least parts of that were implemented. From olsson@cs.ucdavis.edu Fri Apr 22 03:58:53 1994 Received: from ivy.cs.ucdavis.edu by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA15602; Fri, 22 Apr 1994 03:59:05 MST Received: by ivy.cs.ucdavis.edu (5.65/UCD.CS.2.4) id AA16863; Fri, 22 Apr 1994 03:58:53 -0700 Date: Fri, 22 Apr 1994 03:58:53 -0700 From: olsson@cs.ucdavis.edu (Ron Olsson) Message-Id: <9404221058.AA16863@ivy.cs.ucdavis.edu> To: pheras@tornasol.uc3m.es Cc: info-sr@cs.arizona.edu In-Reply-To: <199404221041.AA14763@optima.CS.Arizona.EDU> (message from Pedro de las Heras on Fri, 22 Apr 94 12:37:32 BST) Subject: SR as a teaching tool The only such report of which I'm aware is the one by Hartley cited in the bibliography of the SR book. (You might contact him directly, shartley@mcs.drexel.edu, if you don't have access to those proceedings.) From olsson@cs.ucdavis.edu Fri Apr 22 04:02:58 1994 Received: from ivy.cs.ucdavis.edu by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA16001; Fri, 22 Apr 1994 04:03:15 MST Received: by ivy.cs.ucdavis.edu (5.65/UCD.CS.2.4) id AA17058; Fri, 22 Apr 1994 04:02:58 -0700 Date: Fri, 22 Apr 1994 04:02:58 -0700 From: olsson@cs.ucdavis.edu (Ron Olsson) Message-Id: <9404221102.AA17058@ivy.cs.ucdavis.edu> To: pheras@tornasol.uc3m.es Cc: info-sr@cs.arizona.edu In-Reply-To: <199404221040.AA14714@optima.CS.Arizona.EDU> (message from Pedro de las Heras on Fri, 22 Apr 94 12:36:04 BST) Subject: source repository From: Pedro de las Heras Date: Fri, 22 Apr 94 12:36:04 BST Is there any repository of source code? Not of which I'm aware. If you're looking for specific code, post to info-sr. Is there any faq or some place where the messages in the list are stored? No faq. Old mail to info-sr is available by ftp from cs.arizona.edu in the sr subdirectory. From thomas@orion.ssdc.honeywell.com Tue Apr 26 11:39:38 1994 Received: from orion.ssdc.honeywell.com by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA14140; Tue, 26 Apr 1994 09:39:58 MST Received: from zephyr.ssdc.honeywell.com by orion.ssdc.honeywell.com; Tue, 26 Apr 94 11:39:49 CDT From: "Vic Thomas 951-7470" Message-Id: <9404261639.AA02921@zephyr.ssdc.honeywell.com> Received: by zephyr.ssdc.honeywell.com; Tue, 26 Apr 94 11:39:40 CDT Subject: Re: Exceptions To: pheras@tornasol.uc3m.es Date: Tue, 26 Apr 1994 11:39:38 -0500 (CDT) Cc: info-sr@cs.arizona.edu, rick@cs.arizona.edu In-Reply-To: <9404221054.AA16828@ivy.cs.ucdavis.edu> from "Ron Olsson" at Apr 22, 94 03:54:19 am X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Length: 1442 Ron Olsson writes > > From: Pedro de las Heras > Date: Fri, 22 Apr 94 12:25:47 BST > > Is there any implementation available of exceptions in SR as the > one described in the article on Trans. on ACM Prog. Lang. and > Syst. in Jan. 1988 ? > > Those features were never implemented. You might be interested in the > paper by Daniel Huang and myself cited in the SR book's bibliography, > although those features were never implemented either. You might also > be interested in FT-SR, developed by Vic Thomas and Rick Schlichting > (rick@cs.arizona.edu); I'm not fully aware of the status of that work, > but I believe at least parts of that were implemented. > FT-SR, which is SR with extensions for writing fault-tolerant programs, has a limited exception handling facility that is used to signal loss of resources due to machine crashes. It is therefore not a full fledged exception mechanism that can be used to signal arbitrary exceptions. FT-SR also has an asynchrounous failure notification/handling mechanism. A paper describing FT-SR is due to appear in the IEEE Transactions on Computers in August. If you want, I will be happy to mail you a postscript version of the paper. For a more complete description of FT-SR you might want to look at my dissertation, a copy of which can be ftp'd from cs.arizona.edu. Look for TR93-23.ps (or TR93-23.ps.Z) in directory reports/1993. < Vic From djesau@barrow.uwaterloo.ca Thu Jun 23 16:06:37 1994 Received: from barrow.uwaterloo.ca by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA20748; Thu, 23 Jun 1994 13:06:57 MST Received: by barrow.uwaterloo.ca id <106128>; Thu, 23 Jun 1994 16:06:39 -0400 Subject: Help finding SR applications From: Darren Esau To: info-sr@cs.arizona.edu Date: Thu, 23 Jun 1994 16:06:37 -0400 X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 337 Message-Id: <94Jun23.160639edt.106128@barrow.uwaterloo.ca> Hi. I'm a student doing some work on some software for detecting data races in SR programs, and I was wondering if anyone might be able to point me to some existing SR applications that I could test it with, preferably making substantial use of shared variables. Thanks, Darren Esau University of Waterloo djesau@barrow.uwaterloo.ca From scip3161@leonis.nus.sg Sat Jun 25 09:17:14 1994 Received: from leonis.nus.sg by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA28329; Fri, 24 Jun 1994 18:17:27 MST Received: from localhost (scip3161@localhost) by leonis.nus.sg (8.6.4/8.6.4/CNS-2.0) id JAA11472 for info-sr@cs.arizona.edu; Sat, 25 Jun 1994 09:17:14 +0800 From: E Z A M Message-Id: <199406250117.JAA11472@leonis.nus.sg> Subject: about the nap() function... To: info-sr@cs.arizona.edu (arizona info_sr) Date: Sat, 25 Jun 1994 09:17:14 +0800 (SST) X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Length: 578 Hello, I would like to check that will the nap() function put a process in a waiting state, i.e. not using any cpu resources? I would like to have the process still seeking for cpu services for a specific time period. Is there better way than to use for loops? Or even the for loops, if empty, will be optimize to no operation or suspension? thanks. -- ........................................................ Lai_Hong Cheong (E Z A M) scip3161@leonis.nus.sg Learn from the atom and particles about the way of life. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ From greg Sat Jun 25 08:49:11 1994 Received: from paloverde.CS.Arizona.EDU by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA09947; Sat, 25 Jun 1994 08:49:12 MST Date: Sat, 25 Jun 1994 08:49:11 MST From: "Greg Andrews" Message-Id: <199406251549.AA00846@paloverde.cs.arizona.edu> Received: by paloverde.cs.arizona.edu; Sat, 25 Jun 1994 08:49:11 MST To: info-sr@cs.arizona.edu, scip3161@leonis.nus.sg Subject: Re: about the nap() function... A call to nap() suspends (blocks) the executing process (unless the argument is zero). Thus, you will need to use a loop to busy wait. However, I don't know why you would want to continue to "seek cpu services" while waiting. Loops are "interrupted" every so often to reenter the SR runtime. The default is 10,000 iterations. This can be altered using the -L option of srl, the SR linker. (See the man page, or execute "srl -l" to see all runtime limits.) Greg Andrews From shartley@mcs.drexel.edu Fri Jul 8 13:52:24 1994 Received: from mcs.drexel.edu (king.mcs.drexel.edu) by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA01352; Fri, 8 Jul 1994 10:51:12 MST Received: by mcs.drexel.edu (4.1/SMI-4.0) id AA26030; Fri, 8 Jul 94 13:52:24 EDT Date: Fri, 8 Jul 94 13:52:24 EDT From: shartley@mcs.drexel.edu (Stephen J. Hartley) Message-Id: <9407081752.AA26030@mcs.drexel.edu> To: info-sr@cs.arizona.edu Subject: Re: SR as a teaching tool >The only such report of which I'm aware is the one by Hartley cited in >the bibliography of the SR book. (You might contact him directly, >shartley@mcs.drexel.edu, if you don't have access to those proceedings.) "Experience with the Language SR in an Undergraduate Operating Systems Course," ACM SIGCSE Bulletin, Vol. 24, No. 1, March 1992. I can send uuencoded PostScript of the above to anyone interested. Also, a related article is the following. "An Operating Systems Laboratory Based on the SR (Synchronizing Resources) Programming Language," Computer Science Education, Vol. 3, No. 3, 1992, pp. 251-276. Finally, I am under contract to Oxford University Press to expand the above article into a book. Look for it in about a year. Steve Hartley shartley@mcs.drexel.edu From shartley@king.mcs.drexel.edu Fri Jul 8 14:07:17 1994 Received: from mcs.drexel.edu (king.mcs.drexel.edu) by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA02286; Fri, 8 Jul 1994 11:06:04 MST Received: by mcs.drexel.edu (4.1/SMI-4.0) id AA26488; Fri, 8 Jul 94 14:07:17 EDT Date: Fri, 8 Jul 94 14:07:17 EDT From: shartley@king.mcs.drexel.edu (Stephen J. Hartley) Message-Id: <9407081807.AA26488@mcs.drexel.edu> To: info-sr@cs.arizona.edu Subject: Re: SR and X windows. Help! >I am a rather new user of SR and I need some help with a couple >of things. I am working on an SR project and I would like to give >it a graphical interface using X windows. The way I see it, I have >two options. Either use the SRWin subsystem to call the X functions >or link C and SR code and handle the X interface with C. (I am quite >experienced with X through C) There is a third choice, described in the article: "Animating Operating Systems Algorithms with XTANGO," ACM SIGCSE Bulletin, Vol. 26, No. 1, March 1994. This article shows (with an example) how to use SR with XTANGO. I can send uuencoded PostScript to anybody who is interested. The article shows how to insert additional write statements into a program so that the algorithm in the program is "animated" when the program output is piped into an XTANGO program called animator. Contrast this with SRWin, where you put procedure calls into your SR program. I am working on blending the use of XTANGO's animator into an SR program, similar to the way SRWin works, so that procedures can be called to drive the animation. The advantage over SRWin is that I think animator is easier to use, particularly for students. I am hoping the SR people will include this in a future release of the SR system. From shartley@mcs.drexel.edu Mon Jul 11 19:16:20 1994 Received: from mcs.drexel.edu (king.mcs.drexel.edu) by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA24587; Mon, 11 Jul 1994 16:15:11 MST Received: by mcs.drexel.edu (4.1/SMI-4.0) id AA00356; Mon, 11 Jul 94 19:16:20 EDT Date: Mon, 11 Jul 94 19:16:20 EDT From: shartley@mcs.drexel.edu (Stephen J. Hartley) Message-Id: <9407112316.AA00356@mcs.drexel.edu> To: info-sr@cs.arizona.edu Subject: citation help If someone out there has the last 7 years of ACM SIGOPS OSR, could you supply me with the complete citation for the article: "A Corrent and Unrestrictive Implementation of General Semaphores" by Phil Kearns, Operating Systems Review, 1988 or 1989 or so I need the year, volume number, and issue number. Thanks, Steve Hartley shartley@mcs.drexel.edu From greg Wed Jul 13 09:25:08 1994 Received: from paloverde.CS.Arizona.EDU by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA07064; Wed, 13 Jul 1994 09:25:09 MST Date: Wed, 13 Jul 1994 09:25:08 MST From: "Greg Andrews" Message-Id: <199407131625.AA16641@paloverde.cs.arizona.edu> Received: by paloverde.cs.arizona.edu; Wed, 13 Jul 1994 09:25:08 MST To: info-sr@cs.arizona.edu, shartley@king.mcs.drexel.edu Subject: Re: citation help Date: Mon, 11 Jul 94 19:16:20 EDT From: shartley@king.mcs.drexel.edu (Stephen J. Hartley) Subject: citation help If someone out there has the last 7 years of ACM SIGOPS OSR, could you supply me with the complete citation for the article: "A Correct and Unrestrictive Implementation of General Semaphores" by Phil Kearns, Operating Systems Review, 1988 or 1989 or so I need the year, volume number, and issue number. Thanks, Steve Hartley shartley@mcs.drexel.edu I have a complete set of OSR for the past 20 years. The citation for the Kearns article is Volume 22, Number 4, October 1988, pp. 46-48. If anyone else has a similar question, holler. Greg Andrews From pheras@tornasol.uc3m.es Fri Jul 29 04:22:26 1994 Message-Id: <199407291122.AA02297@optima.cs.arizona.edu> Received: from tornasol.uc3m.es by optima.cs.arizona.edu (5.65c/15) via SMTP id AA02297; Fri, 29 Jul 1994 04:22:26 MST Received: by tornasol.uc3m.es (16.7/15.6) id AA24729; Fri, 29 Jul 94 13:21:54 +0100 From: Pedro de las Heras Subject: SR on NetBSD To: info-sr@cs.arizona.edu Date: Fri, 29 Jul 94 13:21:54 BST Mailer: Elm [revision: 66.33] We have compiled SR-2.2 in NetBSD boxes. If anyone is interested on it, let us know. -- Pedro de las Heras Quiros Internet: pheras@tornasol.uc3m.es // snail-mail: pheras@bizcoyo.uc3m.es // Universidad Carlos III Voice: +34-1-624 94 97 // C/ Butarque 15. E-28911 Fax: +34-1-624 94 30 // Leganes SPAIN WWW: http://www.uc3m.es/~pheras/ From pingu!edgar@tmpuhf.saar.de Mon Aug 1 14:34:39 1994 Received: from shug-internet.saar.de by optima.cs.arizona.edu (5.65c/15) via SMTP id AA24891; Mon, 1 Aug 1994 14:34:39 MST Received: from TMPuhf.Saar.DE (tmpuhf.saar.de [192.109.53.3]) by shug-internet.saar.de (8.6.8.1/8.5) with SMTP id XAA11182; Mon, 1 Aug 1994 23:34:23 +0200 Received: from pingu by TMPuhf.Saar.DE with uucp (Smail3.1.28.1 #1) id m0qV5sG-00021LC; Mon, 1 Aug 94 23:30 WET DST Received: by pingu.saar.de (Smail3.1.28.1 #4) id m0qV4w2-000025C; Mon, 1 Aug 94 23:30 MET DST Message-Id: Date: Mon, 1 Aug 94 23:30 MET DST From: edgar@pingu.saar.de (Edgar Greuter) To: info-sr@cs.arizona.edu Subject: patch for rts/scan.c to allow scansets in scanf Hello! The following patch - applied to rts/scan.c in the SR source tree - enables the use of scansets in scanf-format specifiers like in ANSI-C. A typical usage would be: scanf(inputFile, "%[^\n]\n", line) to read a string up to the next linefeed character (not necessarily separated by whitespace chars). I know that read() can do the same thing as the example above, but it's not as general and i didn't want to mix scanf and read. And BTW the %[-format is mentioned in the book (The SR Programming Language) at page 305 :-). - Edgar --- rts/scan.c~ Tue Sep 28 00:58:32 1993 +++ rts/scan.c Mon Aug 1 23:16:31 1994 @@ -5,6 +5,7 @@ #include "rts.h" static int scanToken (), scanInteger (), scanReal (), scanPointer (); +static int scanTokenFromSet (); #define BACKCH(c,f,S,s) (S ? ((s>DATA (S)) ? (s)-- : 0) : (char*)ungetc (c, f)) #define NEXTCH(l,f,S,s) (S ? ((*s=='\0') ? EOF : * (Char*)(s)++) : INCH (l, f)) @@ -204,6 +205,140 @@ } break; /*---- end of integer conversion */ + case '[': + { + char charset[256]; + int in_set = 1; + + c = *fmt++; + if (c == '^') { + in_set = 0; + c = *fmt++; + } + if (c == '\0') + ABORT_IO (fp, locn, "premature termination of format string"); + memset(charset, !in_set, 256); + if (fmt[1] == '-') { + if (fmt[2] < c) { + if (fmt[2] == '\0') + ABORT_IO (fp, locn, "premature termination of +format string"); + } else { + for (ch = c; ch <= fmt[2]; ch++) + charset[ch] = in_set; + } + c = fmt[3]; + fmt += 3; + } else { + if (c == ']') { + charset[c] = in_set; + c = *fmt++; + } else if (c == '-') { + charset[c] = in_set; + c = *fmt++; + } + } + while (1) { + if (c == '\0') + ABORT_IO (fp, locn, "premature termination of format string"); + if (c == ']') break; + if (fmt[1] == '-') { + if (fmt[2] < c) { + if (fmt[2] == '\0') + ABORT_IO (fp, locn, "premature termination of +format string"); + } else { + for (ch = c; ch <= fmt[2]; ch++) + charset[ch] = in_set; + } + c = fmt[3]; + fmt += 3; + } else { + charset[c] = in_set; + c = *fmt++; + } + } + /* + * handle the string (s) format type here. + * if this field is suppressed then do not consume an arg. + * have to special case the format %1s to allow assignment + * to a char. + */ + if (fldWidth == 1 && *argt == 'c' && !suppress) { + /* + * do this only if format is '%1s', arg is char AND + * the field is NOT suppressed. + */ + ch = NEXTCH (locn, fp, sarg, inString); + + if (ch == EOF) + IO_RETURN (fp, nread ? nread : EOF); + + if (!charset[ch]) + IO_RETURN (fp, nread ? nread : EOF); + + cp = va_arg (ap, Char *); + *cp = ch; + ++nread; + ++argt; + + } else if (*argt == 'a' && !suppress) { + + /* + * scan into array of characters + */ + ++argt; + a = (Array *) va_arg (ap, Array *); + l = a->ub - a->lb + 1; + + /* + * Scan no more than max (length of string, field width) + */ + i = (l < fldWidth) ? l : fldWidth; + if ((n = scanTokenFromSet (locn, fp, sarg, &inString, + buf, i, charset)) == EOF) + IO_RETURN (fp, nread ? nread : EOF); + if (n == 0) + IO_RETURN (fp, nread); + + ++nread; + strncpy (DATA (a), buf, n); + /* + * blank out the remainder of the input array + */ + while (n < l) + DATA (a) [n++] = ' '; + + } else { + n = scanTokenFromSet (locn, fp, sarg, &inString, buf, + fldWidth, charset); + if (n == EOF) + IO_RETURN (fp, nread ? nread : EOF); + if (n == 0) + IO_RETURN (fp, nread); + + if (!suppress) { + if (*argt != 's') + ABORT_IO (fp, locn, + "arguments do not match format string"); + + sp = (String *) va_arg (ap, String *); + ++argt; + + /* + * set length limit (l) to lesser of + * size of String or or the number of chars read. + */ + l = MAXLENGTH (sp); + l = (l < n) ? l : n; + + strncpy (DATA (sp), buf, l); + sp->length = l; + ++nread; + } + } + break; /*---- end of string format conversion */ + } case 's': /* * handle the string (s) format type here. @@ -528,6 +663,42 @@ n = 0; while (ch != EOF && !isspace (ch) && n < len) { + buf[n++] = ch; + ch = NEXTCH (locn, fp, S, *s); + } + + BACKCH (ch, fp, S, *s); + + buf[n] = '\0'; + return n; +} + +/* + * Scan a token from either the file or string into the buffer. + * Token is any string of chars from the charset supplied. + * Scans up to len chars into buf. + * + * This is always called between a BEGIN_IO and an END_IO so we + * do not do it here. + */ +static int +scanTokenFromSet (locn, fp, S, s, buf, len, charset) +char *locn; +FILE *fp; +String *S; +char **s, *buf; +int len; +char *charset; +{ + int n, ch; + + ch = NEXTCH (locn, fp, S, *s); + + if (ch == EOF) + return EOF; + + n = 0; + while (ch != EOF && charset[ch] && n < len) { buf[n++] = ch; ch = NEXTCH (locn, fp, S, *s); } From benson@cs.ucdavis.edu Tue Aug 9 03:26:56 1994 Received: from toadflax.cs.ucdavis.edu by optima.cs.arizona.edu (5.65c/15) via SMTP id AA04597; Tue, 9 Aug 1994 03:27:02 MST Received: from elysium.cs.ucdavis.edu by toadflax.cs.ucdavis.edu (4.1/UCD.CS.2.5) id AA17189; Tue, 9 Aug 94 03:26:59 PDT Received: from localhost.cs.ucdavis.edu by elysium.cs.ucdavis.edu (5.65/UCD.CS.2.5) id AA26773; Tue, 9 Aug 1994 03:26:58 -0700 Message-Id: <9408091026.AA26773@elysium.cs.ucdavis.edu> To: info-sr@cs.arizona.edu Subject: SR and pthreads Date: Tue, 09 Aug 94 03:26:56 -0700 From: benson@cs.ucdavis.edu X-Mts: smtp Here are two SR implementation questions: 1. Has anyone rewritten the the SR RTS using pthreads? 2. Has anyone ported SR to Mach? (by this I mean a rewrite of the SR RTS to use cthreads or kernel threads, not a port to a BSD single server) I am interested in both these tasks and I don't want to duplicate any work. Even if someone hasn't completed either project I would be interested in hearing about any efforts towards these ends. In addition, if anyone has put some thought to these projects or simply has some ideas I would appreciate any comments. Thanks Greg Benson benson@cs.ucdavis.edu From mmunoz@inf.utfsm.cl Tue Aug 9 09:23:07 1994 Received: from inti.inf.utfsm.cl by optima.cs.arizona.edu (5.65c/15) via SMTP id AA13135; Tue, 9 Aug 1994 06:21:52 MST Received: by inti.inf.utfsm.cl id AA08837 (5.67b8/IDA-1.5 for info-sr@cs.arizona.edu); Tue, 9 Aug 1994 09:23:07 -0400 Date: Tue, 9 Aug 1994 09:23:07 -0400 From: Marcelo Munoz Jofre Message-Id: <199408091323.AA08837@inti.inf.utfsm.cl> To: info-sr@cs.arizona.edu unsubcribe me From gangrene!gene@netcom.com Wed Aug 10 05:47:57 1994 Received: from netcomsv.netcom.com (uucp7.netcom.com) by optima.cs.arizona.edu (5.65c/15) via SMTP id AA04245; Wed, 10 Aug 1994 06:18:18 MST Received: from gangrene.UUCP by netcomsv.netcom.com with UUCP (8.6.4/SMI-4.1) id GAA05913; Wed, 10 Aug 1994 06:18:29 -0700 Received: by gangrene. (NX5.67c/NX3.0M) id AA08515; Wed, 10 Aug 94 05:47:57 -0700 Date: Wed, 10 Aug 94 05:47:57 -0700 From: gene michael stover Message-Id: <9408101247.AA08515@gangrene.> Received: by NeXT.Mailer (1.87.1) Received: by NeXT Mailer (1.87.1) To: info-sr@cs.arizona.edu Subject: SR and pthreads Reply-To: gangrene!CyberTiggyr.com!gene@netcom.com Dr Olssen lectured at California State University in Hayward almost two years ago, and I asked him about SR on Mach in terms of cthreads. He said this had been done. I was surprised when he said that SR in terms of "simulated" threads (setjmp/longjmp, I guess) was more efficient. I don't know if the Mach implementation is available or where you might get it. Cheers. gene ----------------------------------------------------------- gene m. stover "Making the world safe from democracy." Software Engineer CyberTiggyr Solutions Internet: gene@CyberTiggyr.com NeXT Mail spoken here For PGP public key, finger gene@netcom.com Begin forwarded message: Errors-To: cs.arizona.edu!info-sr-errors@netcomsv.netcom.com To: info-sr@cs.arizona.edu Subject: SR and pthreads Date: Tue, 09 Aug 94 03:26:56 -0700 From: cs.ucdavis.edu!benson@netcomsv.netcom.com X-Mts: smtp Errors-To: cs.arizona.edu!info-sr-errors@netcomsv.netcom.com Here are two SR implementation questions: 1. Has anyone rewritten the the SR RTS using pthreads? 2. Has anyone ported SR to Mach? (by this I mean a rewrite of the SR RTS to use cthreads or kernel threads, not a port to a BSD single server) I am interested in both these tasks and I don't want to duplicate any work. Even if someone hasn't completed either project I would be interested in hearing about any efforts towards these ends. In addition, if anyone has put some thought to these projects or simply has some ideas I would appreciate any comments. Thanks Greg Benson benson@cs.ucdavis.edu From thomas@p01.htc.honeywell.com Wed Aug 10 09:30:42 1994 Received: from smoke.htc.honeywell.com by optima.cs.arizona.edu (5.65c/15) via SMTP id AA08401; Wed, 10 Aug 1994 07:29:49 MST From: "Vicraj Thomas 951-7470" Message-Id: <9408101430.AA01937@smoke.htc.honeywell.com> Received: by smoke.htc.honeywell.com; Wed, 10 Aug 94 09:30:43 CDT Subject: Re: SR and pthreads To: gangrene!CyberTiggyr.com!gene@netcom.com Date: Wed, 10 Aug 1994 09:30:42 -0500 (CDT) Cc: info-sr@cs.arizona.edu In-Reply-To: <9408101247.AA08515@gangrene.> from "gene michael stover" at Aug 10, 94 05:47:57 am X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Length: 1873 gene michael stover writes > > Dr Olssen lectured at California State University in Hayward > almost two years ago, and I asked him about SR on Mach in terms > of cthreads. He said this had been done. I was surprised when he > said that SR in terms of "simulated" threads (setjmp/longjmp, I > guess) was more efficient. > A couple of years ago I ported SR to run on the x-kernel vers. 3.1, which had kernel level threads. My observations were similar: the Unix version that used simultated threads performed better than the native-threads version in most cases. There were two reasons for this behaviour. 1) In the simulated threads version a thread was preempted only when the rts deemed it safe. Hence there was no need for concurrency control mechanisms such as locks on data structures inside the rts. To make matters worse, lock operations in v3.1 of the x-kernel were extremely expensive because they involved kernel calls. 2) While it was possible to associate thread specific data with x-kernel threads, retrieving this information was expensive, again because it involved a kernel call. This thread specific information was necessary to determine which SR thread corresponded to an executing kernel thread. The simulated version does this by having the scheduler set the sr_curr_proc variable. The x-kernel version performed much better than the Unix version in programs the involved inter-vm communication. This was no surprise and could be attributed to the superior communication support provided by the x-kernel. The port itself was relatively straight-forward and didn't involve much change to the SR rts. < Vic ----- Vicraj Thomas thomas@htc.honeywell.com Senior Research Scientist +1 612 951 7470 Honeywell Technology Center, Minneapolis, MN 55418 +1 612 951 7438 (fax) From scip3161@leonis.nus.sg Mon Aug 22 12:06:03 1994 Received: from leonis.nus.sg by optima.cs.arizona.edu (5.65c/15) via SMTP id AA29911; Sun, 21 Aug 1994 21:06:07 MST Received: (from scip3161@localhost) by leonis.nus.sg (8.6.9/8.6.9/CNS-3.5) id MAA15618 for info-sr@cs.arizona.edu; Mon, 22 Aug 1994 12:06:04 +0800 From: E Z A M Message-Id: <199408220406.MAA15618@leonis.nus.sg> Subject: priority setting To: info-sr@cs.arizona.edu (arizona info_sr) Date: Mon, 22 Aug 1994 12:06:03 +0800 (SST) X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Length: 558 There is a process in my program that does something like allocating proccessors to tasks generated by some other parts of the program. Is there a way to make this allocation be done immediately? Can I set a priority, or anything that will cause the process to be run by the CPU immediately whenever the situation arrises? Thanks. -- ........................................................ Lai_Hong Cheong (E Z A M) scip3161@leonis.nus.sg Learn from the atom and particles about the way of life. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ From gmt@owl Mon Aug 22 09:49:56 1994 Received: from owl.CS.Arizona.EDU by optima.cs.arizona.edu (5.65c/15) via SMTP id AA13108; Mon, 22 Aug 1994 09:49:58 MST Received: (from gmt@localhost) by owl.CS.Arizona.EDU (8.6.9/8.6.6) id JAA02918; Mon, 22 Aug 1994 09:49:56 -0700 Date: Mon, 22 Aug 1994 09:49:56 -0700 From: Gregg Townsend Message-Id: <199408221649.JAA02918@owl.CS.Arizona.EDU> To: info-sr@cs.arizona.edu, scip3161@leonis.nus.sg Subject: Re: priority setting Content-Length: 530 Every process inherits the priority of its creator; the initial resource of the main resource is created with a process of 0. A process can change its own priority using setpriority(n). If you want to run a process at a higher priority than yourself, one way is this: raise your own priority create the new process (via send() or whatever) lower your priority back Gregg Townsend / Computer Science Dept / Univ of Arizona / Tucson, AZ 85721 +1 602 621 4325 gmt@cs.arizona.edu 110 57 16 W / 32 13 45 N / +758m From Per-Arne.Wiberg@cca.hh.se Tue Aug 30 16:50:01 1994 Received: from hh.se (garfield.hh.se) by optima.cs.arizona.edu (5.65c/15) via SMTP id AA08272; Tue, 30 Aug 1994 07:49:36 MST Received: from hamlet.hh.se by hh.se (5.65+bind 1.7+ida 1.4.2/HH-4-NS); Tue, 30 Aug 94 16:50:01 +0200 (MET) Date: Tue, 30 Aug 94 16:50:01 +0200 From: Per-Arne Wiberg Message-Id: <9408301450.AA08864@hh.se> To: info-sr@cs.arizona.edu Subject: sr exercises Hi I am running a cource in concurrent programming using SR (exelent language). I wonder if I can get hold of solutions to the exercises in the SR languge book? Best regards Pelle Wiberg From scip3161@leonis.nus.sg Thu Sep 1 14:15:06 1994 Received: from leonis.nus.sg by optima.cs.arizona.edu (5.65c/15) via SMTP id AA11098; Wed, 31 Aug 1994 23:15:28 MST Received: (from scip3161@localhost) by leonis.nus.sg (8.6.9/8.6.9/CNS-3.5) id OAA02023 for info-sr@cs.arizona.edu; Thu, 1 Sep 1994 14:15:12 +0800 From: E Z A M Message-Id: <199409010615.OAA02023@leonis.nus.sg> Subject: event marker To: info-sr@cs.arizona.edu (arizona info_sr) Date: Thu, 1 Sep 1994 14:15:06 +0800 (SST) X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Length: 556 I have implemented a parallel algorithm using sr and it is running on a Sun elc. All along I am using write statements to output event markers onto a file. I would like to know are there any better methods? Or should it be significantly much better if I were to send different types of event markers to different files? thanks -- ........................................................ Lai_Hong Cheong (E Z A M) scip3161@leonis.nus.sg Learn from the atom and particles about the way of life. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ From olsson@cs.ucdavis.edu Thu Sep 1 04:30:58 1994 Received: from toadflax.cs.ucdavis.edu by optima.cs.arizona.edu (5.65c/15) via SMTP id AA27614; Thu, 1 Sep 1994 04:31:00 MST Received: from ivy.cs.ucdavis.edu by toadflax.cs.ucdavis.edu (4.1/UCD.CS.2.6) id AA19331; Thu, 1 Sep 94 04:30:58 PDT Received: by ivy.cs.ucdavis.edu (5.65/UCD.CS.2.6) id AA02616; Thu, 1 Sep 1994 04:30:58 -0700 Date: Thu, 1 Sep 1994 04:30:58 -0700 From: olsson@cs.ucdavis.edu (Ron Olsson) Message-Id: <9409011130.AA02616@ivy.cs.ucdavis.edu> To: Per-Arne.Wiberg@cca.hh.se Cc: info-sr@cs.arizona.edu In-Reply-To: <9408301450.AA08864@hh.se> (message from Per-Arne Wiberg on Tue, 30 Aug 94 16:50:01 +0200) Subject: sr exercises Date: Tue, 30 Aug 94 16:50:01 +0200 From: Per-Arne Wiberg Errors-To: info-sr-errors@cs.arizona.edu I wonder if I can get hold of solutions to the exercises in the SR languge book? Sorry, there is no solutions manual. From owner-info-sr Fri Sep 2 10:30:30 1994 Received: from owl.CS.Arizona.EDU by optima.cs.arizona.edu (5.65c/15) via SMTP id AA06864; Fri, 2 Sep 1994 10:30:32 MST Received: (from gmt@localhost) by owl.CS.Arizona.EDU (8.6.9/8.6.6) id KAA03526 for info-sr; Fri, 2 Sep 1994 10:30:30 -0700 Date: Fri, 2 Sep 1994 10:30:30 -0700 From: Gregg Townsend Message-Id: <199409021730.KAA03526@owl.CS.Arizona.EDU> To: info-sr@owl Subject: Re: event marker Content-Length: 658 From: E Z A M I have implemented a parallel algorithm using sr and it is running on a Sun elc. All along I am using write statements to output event markers onto a file. I would like to know are there any better methods? Or should it be significantly much better if I were to send different types of event markers to different files? Your approach sounds reasonable to me. I can't see any advantage to splitting things among different files. Gregg Townsend / Computer Science Dept / Univ of Arizona / Tucson, AZ 85721 +1 602 621 4325 gmt@cs.arizona.edu 110 57 16 W / 32 13 45 N / +758m From owner-info-sr Sun Sep 4 09:43:03 1994 Received: from owl.CS.Arizona.EDU by optima.cs.arizona.edu (5.65c/15) via SMTP id AA25355; Sun, 4 Sep 1994 09:43:12 MST Received: from paloverde.cs.arizona.edu (paloverde.CS.Arizona.EDU [192.12.69.16]) by owl.CS.Arizona.EDU (8.6.9/8.6.6) with SMTP id JAA04720; Sun, 4 Sep 1994 09:43:04 -0700 Date: Sun, 4 Sep 1994 09:43:03 MST From: "Greg Andrews" Message-Id: <199409041643.AA26127@paloverde.cs.arizona.edu> Received: by paloverde.cs.arizona.edu; Sun, 4 Sep 1994 09:43:03 MST To: gmt@owl, info-sr@owl Subject: Re: event marker I'd add that if what you really want is internal event tracing, then see the man page for srtrace. -- Greg Andrews From warren@darien.cray.com Tue Sep 6 09:14:27 1994 Received: from timbuk.cray.com by optima.cs.arizona.edu (5.65c/15) via SMTP id AA21331; Tue, 6 Sep 1994 06:14:36 MST Received: from darien.cray.com (warren@darien.cray.com [128.162.41.1]) by timbuk.cray.com (8.6.9/8.6.9L) with SMTP id IAA02008 for ; Tue, 6 Sep 1994 08:14:30 -0500 Received: by darien.cray.com (4.1/CRI-5.13) id AA17874; Tue, 6 Sep 94 09:14:27 EDT Date: Tue, 6 Sep 94 09:14:27 EDT From: warren@darien.cray.com (Warren Carranza) Message-Id: <9409061314.AA17874@darien.cray.com> To: info-sr@cs.arizona.edu Subject: group removal Please remove me from the sr group. thank-you From grabow@venus.baylor.edu Tue Sep 6 13:19:46 1994 Received: from baylor.edu (buvax2.baylor.edu) by optima.cs.arizona.edu (5.65c/15) via SMTP id AA16773; Tue, 6 Sep 1994 11:34:30 MST Received: from venus (venus.baylor.edu) by baylor.edu (PMDF V4.3-7 #5363) id <01HGSL2WW47K987VSZ@baylor.edu>; Tue, 6 Sep 1994 13:33:46 CDT Received: by venus (4.1/SMI-4.1) id AA02800; Tue, 6 Sep 94 13:19:46 CDT Date: Tue, 06 Sep 1994 13:19:46 -0500 (CDT) From: grabow@venus.baylor.edu (Paul Grabow) Subject: group removal To: info-sr@cs.arizona.edu Message-Id: <9409061819.AA02800@venus> Content-Transfer-Encoding: 7BIT Please remove me from the sr group. Thanks. From jerijian@hurricane.seas.ucla.edu Tue Sep 6 15:31:30 1994 Received: from hurricane.seas.ucla.edu by optima.cs.arizona.edu (5.65c/15) via SMTP id AA03902; Tue, 6 Sep 1994 15:31:34 MST Received: by hurricane.seas.ucla.edu (5.65/1.37(UCLA esa 1.005)) id AA07076; Tue, 6 Sep 94 15:31:30 -0700 From: jerijian@hurricane.seas.ucla.edu Message-Id: <9409062231.AA07076@hurricane.seas.ucla.edu> Subject: UNSUBSCRIBE To: info-sr@cs.arizona.edu Date: Tue, 6 Sep 1994 15:31:30 -0800 (PDT) Cc: info-sr-request@cs.arizona.edu X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Length: 12 UNSUBSCRIBE From cl@GOEDEL.UNI-MUENSTER.DE Thu Sep 8 09:24:54 1994 Received: from math.uni-muenster.de (GOEDEL.UNI-MUENSTER.DE) by optima.cs.arizona.edu (5.65c/15) via SMTP id AA26407; Thu, 8 Sep 1994 00:25:13 MST Received: from zenon.uni-muenster.de by math.uni-muenster.de (4.1/SMI-4.1/smah-bp-1.37) id AA03763; Thu, 8 Sep 94 09:24:54 +0200 Organization: Westfaelische Wilhelms-Universitaet, Muenster, Germany Department of Mathematics Date: Thu, 8 Sep 94 09:24:54 +0200 From: Achim Clausing Message-Id: <9409080724.AA03763@math.uni-muenster.de> To: info-sr@cs.arizona.edu Subject: UNSUBSCRIBE UNSUBSCRIBE From CR05427@academia.swt.edu Thu Sep 8 14:51:33 1994 Received: from nyssa.swt.edu by optima.cs.arizona.edu (5.65c/15) via SMTP id AA07104; Thu, 8 Sep 1994 12:51:37 MST Received: from academia.swt.edu by academia.swt.edu (PMDF V4.3-7 #6249) id <01HGVGD9JX0WCA041D@academia.swt.edu>; Thu, 8 Sep 1994 14:51:33 CDT Date: Thu, 08 Sep 1994 14:51:33 -0500 (CDT) From: Rufus Subject: unsubscribe To: info-sr@cs.arizona.edu Message-Id: <01HGVGD9JYWYCA041D@academia.swt.edu> Organization: Southwest Texas State University X-Vms-To: IN%"info-sr@cs.arizona.edu" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT unsubscribe From cliff Thu Sep 8 13:25:05 1994 Received: from cheltenham.CS.Arizona.EDU by optima.cs.arizona.edu (5.65c/15) via SMTP id AA09924; Thu, 8 Sep 1994 13:25:06 MST Date: Thu, 8 Sep 1994 13:25:05 MST From: "Cliff Hathaway" Message-Id: <199409082025.AA26668@cheltenham.cs.arizona.edu> Received: by cheltenham.cs.arizona.edu; Thu, 8 Sep 1994 13:25:05 MST To: info-sr Subject: please use "info-sr-request@cs.arizona.edu" for administrative matters Requests to subscribe or unsubscribe, and change of address messages should be sent to info-sr-request@cs.arizona.edu, rather than info-sr@cs.arizona.edu. Thanks. Cliff Hathaway Dept. of Computer Science (602)621-4291 University of Arizona cliff@cs.arizona.edu (internet) Tucson, Ariz. 85721 {cmcl2,noao,uunet}!arizona!cliff (uucp) From pheras@bizcoyo.uc3m.es Fri Sep 9 13:56:28 1994 Received: from bizcoyo.uc3m.es by optima.cs.arizona.edu (5.65c/15) via SMTP id AA16077; Fri, 9 Sep 1994 09:21:18 MST Message-Id: From: pheras@bizcoyo.uc3m.es (Pedro de las Heras) Subject: SR on WWW To: info-sr@cs.arizona.edu Date: Fri, 9 Sep 1994 13:56:28 +0000 () X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 653 Hello. What about organizing the mail-list with hypermail in the SR pages on the WWW? Hypermail provides access to mail archives, being it possible to search for a topic, date, person,... It presents the information in a tree where answers to previous mails appear nested,... Info on hypermail can be found in http://www.eit.com -- Pedro de las Heras Quiros Internet: pheras@bizcoyo.uc3m.es // snail-mail: pheras@tornasol.uc3m.es // Universidad Carlos III Voice: +34-1-624 94 97 // C/ Butarque 15. E-28911 Fax: +34-1-624 94 30 // Leganes SPAIN WWW: http://www.uc3m.es/~pheras/ From pheras@bizcoyo.uc3m.es Fri Sep 9 13:57:53 1994 Received: from bizcoyo.uc3m.es by optima.cs.arizona.edu (5.65c/15) via SMTP id AA16112; Fri, 9 Sep 1994 09:22:42 MST Message-Id: From: pheras@bizcoyo.uc3m.es (Pedro de las Heras) Subject: XTANGO To: info-sr@cs.arizona.edu Date: Fri, 9 Sep 1994 13:57:53 +0000 () X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 469 Hello. Some time ago there were an announcement of XTango, and the development of a library written-accessible from SR. Any news about this topic? -- Pedro de las Heras Quiros Internet: pheras@bizcoyo.uc3m.es // snail-mail: pheras@tornasol.uc3m.es // Universidad Carlos III Voice: +34-1-624 94 97 // C/ Butarque 15. E-28911 Fax: +34-1-624 94 30 // Leganes SPAIN WWW: http://www.uc3m.es/~pheras/ From shartley@king.mcs.drexel.edu Fri Sep 9 13:21:28 1994 Received: from mcs.drexel.edu (king.mcs.drexel.edu) by optima.cs.arizona.edu (5.65c/15) via SMTP id AA20388; Fri, 9 Sep 1994 10:20:31 MST Received: by mcs.drexel.edu (4.1/SMI-4.0) id AA11209; Fri, 9 Sep 94 13:21:28 EDT Date: Fri, 9 Sep 94 13:21:28 EDT From: shartley@king.mcs.drexel.edu (Stephen J. Hartley) Message-Id: <9409091721.AA11209@mcs.drexel.edu> To: pheras@bizcoyo.uc3m.es Subject: Re: XTANGO Cc: info-sr@cs.arizona.edu I have sent the files to the SR people at Arizona for inclusion in a future release. >From: pheras@bizcoyo.uc3m.es (Pedro de las Heras) >Subject: XTANGO >Date: Fri, 9 Sep 1994 13:57:53 +0000 () > >Some time ago there were an announcement of XTango, and the development >of a library written-accessible from SR. > >Any news about this topic? From owner-info-sr Fri Sep 9 10:59:40 1994 Received: from owl.CS.Arizona.EDU by optima.cs.arizona.edu (5.65c/15) via SMTP id AA23097; Fri, 9 Sep 1994 10:59:41 MST Received: (from gmt@localhost) by owl.CS.Arizona.EDU (8.6.9/8.6.6) id KAA10820 for info-sr; Fri, 9 Sep 1994 10:59:40 -0700 Date: Fri, 9 Sep 1994 10:59:40 -0700 From: Gregg Townsend Message-Id: <199409091759.KAA10820@owl.CS.Arizona.EDU> To: info-sr@owl Subject: Re: XTANGO Content-Length: 273 The XTANGO interface will be included in the next version of SR. I expect that to be ready sometime in October. Gregg Townsend / Computer Science Dept / Univ of Arizona / Tucson, AZ 85721 +1 602 621 4325 gmt@cs.arizona.edu 110 57 16 W / 32 13 45 N / +758m From owner-info-sr Fri Sep 9 11:04:09 1994 Received: from owl.CS.Arizona.EDU by optima.cs.arizona.edu (5.65c/15) via SMTP id AA23727; Fri, 9 Sep 1994 11:04:11 MST Received: (from gmt@localhost) by owl.CS.Arizona.EDU (8.6.9/8.6.6) id LAA10833 for info-sr; Fri, 9 Sep 1994 11:04:09 -0700 Date: Fri, 9 Sep 1994 11:04:09 -0700 From: Gregg Townsend Message-Id: <199409091804.LAA10833@owl.CS.Arizona.EDU> To: info-sr@owl Subject: Re: SR on WWW Content-Length: 341 Hypermail looks interesting, and we'll keep it in mind; but we have a lot of other things going on right now, so I don't think we'll be doing anything with it in the near future. Gregg Townsend / Computer Science Dept / Univ of Arizona / Tucson, AZ 85721 +1 602 621 4325 gmt@cs.arizona.edu 110 57 16 W / 32 13 45 N / +758m From gwpatter@king.mcs.drexel.edu Sat Sep 24 09:56:07 1994 Received: from mcs.drexel.edu (king.mcs.drexel.edu) by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA12336; Sat, 24 Sep 1994 06:55:08 MST Received: from queen.mcs.drexel.edu by mcs.drexel.edu (4.1/SMI-4.0) id AA06392; Sat, 24 Sep 94 09:56:07 EDT Date: Sat, 24 Sep 94 09:56:07 EDT From: gwpatter@king.mcs.drexel.edu (William R. Patterson) Message-Id: <9409241356.AA06392@mcs.drexel.edu> To: info-sr@cs.arizona.edu Subject: Linux I would be interested in hearing from anyone who has successfully implemented SR on Linux. My installation is proceeding more slowly than I would like. I have been able to successfully install SR on a Sun account, but when the same methods are applied to Linux problems result: make cc -g -c util.c -o util.o In file included from util.c:4: /usr/include/sys/types.h:4: linux/types.h: No such file or directory ... make: *** [util.o] Error 1 and it ends there. There is paobably something obvious here to a Linux pro that I am missing, but even after applying the patches in linux.patch, I am still unable to come up with an answer to this. I would be glad to provide further information by email or generally to the broadcast address if others could benefit. Thanks in advance for any help! --Bill Patterson gwpatter@mcs.drexel.edu From hennus@sky.ow.org Sun Sep 25 03:00:35 1994 Received: from eleet.mimuw.edu.pl by optima.cs.arizona.edu (5.65c/15) via SMTP id AA21757; Sun, 25 Sep 1994 21:54:14 MST Received: by eleet.mimuw.edu.pl (8.6.9/8.6.9) with ESMTP id FAA01312; Mon, 26 Sep 1994 05:52:10 +0100 Received: by mail.OW.org (8.6.9/8.6.9) with UUCP id FAA11155; Mon, 26 Sep 1994 05:43:15 +0100 Received: from sky.ow.org by sun4ow.ow.org (8.6.9/8.6.9) with UUCP id DAA20133; Mon, 26 Sep 1994 03:13:23 GMT Received: by sky.ow.org (Linux Smail3.1.28.1 #2) id m0qoitH-000vniC; Sun, 25 Sep 94 03:00 MET Message-Id: From: hennus@sky.ow.org (Hennus Bergman) Subject: Re: Linux To: gwpatter@king.mcs.drexel.edu (William R. Patterson) Date: Sun, 25 Sep 1994 03:00:35 +0100 (MET) Cc: info-sr@cs.arizona.edu In-Reply-To: <9409241356.AA06392@mcs.drexel.edu> from "William R. Patterson" at Sep 24, 94 09:56:07 am X-Mailer: ELM [version 2.4 PL22] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1621 > I would be interested in hearing from anyone who has successfully > implemented SR on Linux. My installation is proceeding more > slowly than I would like. I have been able to successfully install > SR on a Sun account, but when the same methods are applied to > Linux problems result: > make > cc -g -c util.c -o util.o > In file included from util.c:4: > /usr/include/sys/types.h:4: linux/types.h: No such file or directory > ... > make: *** [util.o] Error 1 > and it ends there. There is paobably something obvious here to a Linux pro > that I am missing, but even after applying the patches in linux.patch, > I am still unable to come up with an answer to this. Looks like the compiler can't find the kernel include files. /usr/include/sys/types.h includes the types.h file from the Linux kernel sources, which are normally located in /usr/src/linux. /usr/include/linux usually is a symlink to the /usr/src/linux/include/linux directory where the kernel include file types.h lives. So, you need to un-tar the linux kernel sources in /usr/src, or create the symlink /usr/include/linux to point to your kernel include files. You don't really need to have the entire kernel sources online, just the /usr/src/linux/include part should be sufficient. BTW, there should also be a /usr/include/asm symlink to /usr/src/linux/include/asm. If you want smaller binaries, remove the -g (debugging) option. > --Bill Patterson > gwpatter@mcs.drexel.edu Hennus -- Hennus Bergman [my Linux box, UUCP-connected (slow)] PGP fingerprint: 76 99 FD 31 91 E1 96 1C 90 BB 22 80 62 F6 BD 63 From chris@chaos.aerojet.com Mon Sep 26 09:45:27 1994 Message-Id: <199409261643.AA09880@optima.cs.arizona.edu> Received: from chaos.aerojet.com by optima.cs.arizona.edu (5.65c/15) via SMTP id AA09880; Mon, 26 Sep 1994 09:43:25 MST Subject: Documentation on SR RTS To: info-sr@cs.arizona.edu Date: Mon, 26 Sep 1994 09:45:27 -0700 (PDT) From: Christopher Arndt-Kohlway X-Mailer: ELM [version 2.4 PL21] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 799 Hello - I am trying to develop a thorough understanding of the SR runtime system for a project I am working on. Does anyone have or know of the existence of any detailed documentation on the RTS. I'm looking for something a bit more detailed than that in the appendix of the book "The SR Programming Language". Any help would be greatly appreciated. Thank you. - Chris -- =============================================================================== Christopher Arndt-Kohlway E-Mail: chris@aerojet.com Aerojet, Sacramento Plant Phone : (916) 355-5634 Sacramento, California (US) Fax : (916) 355-6635 =============================================================================== From greg Mon Sep 26 09:58:04 1994 Received: from paloverde.CS.Arizona.EDU by optima.cs.arizona.edu (5.65c/15) via SMTP id AA11268; Mon, 26 Sep 1994 09:58:08 MST Date: Mon, 26 Sep 1994 09:58:04 MST From: "Greg Andrews" Message-Id: <199409261658.AA26083@paloverde.cs.arizona.edu> Received: by paloverde.cs.arizona.edu; Mon, 26 Sep 1994 09:58:04 MST To: chris@chaos.aerojet.com Subject: Re: Documentation on SR RTS Cc: info-sr@cs.arizona.edu Subject: Documentation on SR RTS To: info-sr@cs.arizona.edu Date: Mon, 26 Sep 1994 09:45:27 -0700 (PDT) From: Christopher Arndt-Kohlway Hello - I am trying to develop a thorough understanding of the SR runtime system for a project I am working on. Does anyone have or know of the existence of any detailed documentation on the RTS. I'm looking for something a bit more detailed than that in the appendix of the book "The SR Programming Language". Any help would be greatly appreciated. Thank you. - Chris -- Christopher Arndt-Kohlway E-Mail: chris@aerojet.com Aerojet, Sacramento Plant Phone : (916) 355-5634 Sacramento, California (US) Fax : (916) 355-6635 Have you looked at The SR Run-Time System Interface, which is part of the standard distribution? It contains quite a few details, although not to the level of a "walkthrough". The files are runtime.ms (in "doc") and runtime.ps (in "ps"). -- Greg Andrews From renganag@next2.msci.memst.edu Mon Sep 26 21:39:55 1994 Received: from next2 (next2.msci.memst.edu) by optima.cs.arizona.edu (5.65c/15) via SMTP id AA22727; Mon, 26 Sep 1994 19:44:11 MST Received: by next2 id AA11220 (5.65c/IDA-1.4.4 for info-sr@cs.arizona.edu); Mon, 26 Sep 1994 21:44:08 -0500 Date: Mon, 26 Sep 1994 21:39:55 -0500 (CDT) From: Gopal Renganathan Subject: Problems Compiling SR To: info-sr@cs.arizona.edu Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hello: I have implemented SR in a SUN machine in the /usr/local/src/sr/bin directory and included that path in my .profile. When I try to compile even simple SR programs it gives me the message sr: no resource bodies; linking supressed. Is there something I am doing wrong here or am I missing some files. Please help me. ================================================================================ Gopal Renganathan PH # : (off) : (901)-325-6014 Dept of Computer Sciences (res) : (901)-454-7760 Memphis State University Memphis, TN - 38111 ================================================================================ From shartley@king.mcs.drexel.edu Tue Sep 27 11:34:42 1994 Received: from mcs.drexel.edu (king.mcs.drexel.edu) by optima.cs.arizona.edu (5.65c/15) via SMTP id AA01356; Tue, 27 Sep 1994 08:33:45 MST Received: by mcs.drexel.edu (4.1/SMI-4.0) id AA20910; Tue, 27 Sep 94 11:34:42 EDT Date: Tue, 27 Sep 94 11:34:42 EDT From: shartley@king.mcs.drexel.edu (Stephen J. Hartley) Message-Id: <9409271534.AA20910@mcs.drexel.edu> To: info-sr@cs.arizona.edu Subject: new OS book using SR In a few months, Oxford University Press will publish my book "Operating Systems Programming: The SR Programming Language" Steve Hartley Drexel University shartley@mcs.drexel.edu Here is the Preface: SR is a language for concurrent programming. This book describes the SR language, presents some examples of SR programs in the context of an operating systems or concurrent programming course, and provides some programming as- signments in the form of open Student Laboratories. The SR language can be used by instructors of concurrent programming or operating systems courses to give students experience in writing concurrent programs that use multiple processes, semaphores, monitors, message passing, remote procedure calls, and the rendezvous. The language can also be used for parallel computing in a shared-memory multiprocessor or a distributed memory cluster environment. The intended audience is undergraduate and graduate students enrolled in concurrent programming and operating systems classes. Prerequisites for stu- dents are knowledge of a high-level programming language like C or Pascal and operating systems concepts. See the Instructor's Manual for more detailed prerequisites. This book is designed to be used during an operating systems course in conjunction with one of the standard texts to provide concurrent programming experience. The book can also be used in a full semester follow- on course to operating systems to provide more extensive programming experi- ence. Since most concepts and terms are defined, the book can be used as the sole text in an introductory concurrent programming course that precedes the operating systems course. The required computing environment is a UNIX platform such as a Sun 3, Sun 4, Sequent Symmetry, DECstation, SGI Iris, NeXT, HP RISC, or PC compatible running Linux. The SR examples and programming assignments in this book have been used suc- cessfully in undergraduate concurrent programming and operating systems courses at Drexel University, Philadelphia, Pennsylvania, and Trinity Univer- sity, San Antonio, Texas. The book arose out of a collection of handouts and class notes distributed to students during 1991-94 and is a greatly expanded version of [some earlier journal papers]. This book has several important features. * The material is keyed to four standard operating systems texts: Deitel, Silberschatz and Galvin, Stallings, and Tanenbaum. The relevant sections of these texts for the major concepts, such as semaphores and monitors, is indi- cated. * Each of the numerous example programs in this book includes output from one or two sample runs to show how the program works. * Algorithm animation using the animator interpreter from the XTANGO sys- tem is described and used in several of the examples. * Numerous programming assignments in the form of open Student Labora- tories are given. The example programs in this book will be made available by anonymous ftp at site mcs.drexel.edu. Contact the author for further details. Here is the TOC: Operating Systems Programming Hardware Basics Hardware and Software Interrupts Hardware Protection CPU Scheduling The SR Programming Language Sequential Features Additional Sequential Examples and Features Lab: Sequential SR Multiple Resources Lab: Resources and Capabilities Additional Features Animating Programs with XTANGO Lab: XTANGO Animation Concurrent Programming Multiple Processes in One Resource Multiple Processes in Multiple Resources Simulated versus Real Concurrency Debugging Techniques Race Conditions and Process Synchronization Lab: Race Conditions Architecture and Shared Memory Systems The Mutual Exclusion Problem Semaphores SR Semaphores Classical Operating Systems Problems Semaphore Solutions The Producers and Consumers The Sleeping Barber The Dining Philosophers The Readers and Writers Binary Semaphores Lab: Semaphores Assignment: Fix Race Condition Assignment: Multiple Producers and Consumers Assignment: Another Classical Problem Assignment: Multiple Sleeping Barbers Assignment: Dining Philosophers Assignment: Fair Readers and Writers Assignment: Baboons Crossing a Canyon Assignment: Fraternity Party Assignment: Jurassic Park Animating Operating Systems Algorithms Lab: Classical Problem Animation Monitors Monitors and SR The Producers and Consumers The Dining Philosophers The Readers and Writers Implementing Monitors with Semaphores Lab: Monitors Assignment: Fair Baboons Assignment: Sleeping Barber Assignment: Fair Dining Philosophers Assignment: Fraternity Party Assignment: Bakery Message Passing and the Rendezvous Message Passing Message Passing in SR Concurrent Programs and Message Passing Distributed Mutual Exclusion Miscellaneous Examples Lab: Message Passing Assignment: Merge Sort Assignment: Pipeline Sieve of Eratosthenes Assignment: Compare/Exchange Sort Assignment: Speedup Assignment: Algorithm Animation Rendezvous Classical Problems Using Rendezvous Nested in Statements A Lock Resource The Distributed Dining Philosophers RPC and Client-Server Programming Summary of Operations and Invocations Lab: Rendezvous Assignment: Fair Baboons Assignment: Sleeping Barber Assignment: Fair Readers and Writers Assignment: Fair Dining Philosophers Assignment: Fraternity Party Assignment: Bakery Assignment: Banker's Algorithm More Animation with XTANGO Lab: Algorithm Animation Parallel Computing Coarse-Grained Concurrency in SR Patterns of Communication Data Parallelism Shared Memory Examples Message Passing Examples Integrating Animator into SR Lab: Parallel Computing Assignment: Speedup Assignment: Race Condition Assignment: No Race Condition Instructor's Manual Sample Lab Solutions Semaphores Monitors Message Passing Rendezvous Parallel Computing From @exeter.ac.uk:steve@dcs.exeter.ac.uk Mon Oct 3 18:47:37 1994 Received: from sun2.nsfnet-relay.ac.uk by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA27488; Tue, 4 Oct 1994 03:00:50 MST Via: uk.ac.exeter; Mon, 3 Oct 1994 18:50:01 +0100 Received: from atlas.exeter.ac.uk by exub.exeter.ac.uk with SMTP (PP) id <10967-0@exub.exeter.ac.uk>; Mon, 3 Oct 1994 18:50:27 +0100 Received: from sirius.dcs.exeter.ac.uk by atlas.dcs.exeter.ac.uk; Mon, 3 Oct 94 18:50:23 BST From: steve@atlas.ex.ac.uk Date: Mon, 3 Oct 94 18:47:37 +0100 Message-Id: <28199.9410031747@sirius.dcs.exeter.ac.uk> To: gmt@cs.arizona.edu Cc: info-sr@cs.arizona.edu Subject: SR on SGI under IRIX 5.2 We have been using SR on our SGIs for over two years. However, we have recently upgraded our operating system to IRIX 5.2 When I came to rebuild SR, I found I could not link the context switching code: cd csw; make CC="cc" CFLAGS="-g " cc -g -c cstest.c cc -g -E asm.c | sed -e '/^ *$/d' -e '/^#/'d >asm.s as -o asm.o asm.s . ../paths.sh; cc -g -o cstest \ cstest.o asm.o $LIBM $LIBC ld: Object file format error in: asm.o: non-shared Elf object "asm.o" cannot be linked shared. *** Error code 1 (bu21) As a temporary fix, we can setenv SGI_IRIX4 1 and use the old IRIX 4 libraries, but I wondered if anyone had successfully build SR under IRIX 5.2 Steve ------------------------------------------------------------------------------- Parallel Systems Research Laboratory | Stephen Turner Email: steve@dcs.exeter.ac.uk | Department of Computer Science WWW: http://www.dcs.exeter.ac.uk | University of Exeter Fax: +44 392 264067 | Prince of Wales Road Tel: +44 392 264048 | Exeter EX4 4PT England ------------------------------------------------------------------------------- From MOORE_KEITH@Tandem.COM Tue Oct 4 06:19:00 1994 Received: from comm.Tandem.COM (comm.cpd.tandem.com) by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA05827; Tue, 4 Oct 1994 06:21:40 MST Received: by comm.Tandem.COM (4.12/4.5) id AA29497; 4 Oct 94 06:21:14 -0700 Date: 4 Oct 94 06:19:00 -0700 From: MOORE_KEITH@tandem.com Message-Id: <199410040621.AA29497@comm.Tandem.COM> To: steve@atlas.ex.ac.uk Cc: info-sr@cs.arizona.edu Subject: RE:SR on SGI under IRIX 5.2 Steve, I'm having the same problem, but I'm new to Unix, so I wasn't so sure what it was. I tried re-installing everything -- compilers, linkers, and the SR source. I think I got through the sr install the other day, but I haven't yet done anything to test it yet. I'm still at the very bottom of an extremely steep learning curve. Please let me know what you hear. Thanks in advance. - Keith ------------ ORIGINAL ATTACHMENT -------- SENT 10-04-94 FROM SMTPGATE (steve@atlas.ex.ac.uk) We have been using SR on our SGIs for over two years. However, we have recently upgraded our operating system to IRIX 5.2 When I came to rebuild SR, I found I could not link the context switching code: cd csw; make CC="cc" CFLAGS="-g " cc -g -c cstest.c cc -g -E asm.c | sed -e '/^ *$/d' -e '/^#/'d >asm.s as -o asm.o asm.s . ../paths.sh; cc -g -o cstest \ cstest.o asm.o $LIBM $LIBC ld: Object file format error in: asm.o: non-shared Elf object "asm.o" cannot be linked shared. *** Error code 1 (bu21) As a temporary fix, we can setenv SGI_IRIX4 1 and use the old IRIX 4 libraries, but I wondered if anyone had successfully build SR under IRIX 5.2 Steve ------------------------------------------------------------------------------- Parallel Systems Research Laboratory | Stephen Turner Email: steve@dcs.exeter.ac.uk | Department of Computer Science WWW: http://www.dcs.exeter.ac.uk | University of Exeter Fax: +44 392 264067 | Prince of Wales Road Tel: +44 392 264048 | Exeter EX4 4PT England ------------------------------------------------------------------------------- ------------ TEXT ATTACHMENT -------- SENT 10-04-94 FROM MOORE_KEITH @DETROIT _________________________________________________________________________ / \ | Keith Moore (M. Keith Moore) | | Advisory Analyst Guardian: moore_keith | | Tandem Computers, Inc. Internet: moore_keith@tandem.com | | 19500 Victor Pkwy. #400 | | Livonia, Michigan 48152-1061 U.S.A. Phone: (313) 953-2422 or 8800 | | | | Opinions expressed here are my own. | \_________________________________________________________________________/ From owner-info-sr Tue Oct 4 07:41:43 1994 Received: from owl.CS.Arizona.EDU by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA09254; Tue, 4 Oct 1994 07:41:46 MST Received: (from gmt@localhost) by owl.CS.Arizona.EDU (8.6.9/8.6.6) id HAA15891; Tue, 4 Oct 1994 07:41:43 -0700 Date: Tue, 4 Oct 1994 07:41:43 -0700 From: Gregg Townsend Message-Id: <199410041441.HAA15891@owl.CS.Arizona.EDU> To: steve@atlas.ex.ac.uk Cc: info-sr@owl Subject: Re: SR on SGI under IRIX 5.2 Content-Length: 1244 From steve@dcs.exeter.ac.uk Tue Oct 4 03:01 MST 1994 as -o asm.o asm.s ... asm.o: non-shared Elf object "asm.o" cannot be linked shared. I don't know how to explain that particular error. Is it possible somehow that /bin/as didn't get updated with 5.2? The mips.s file *does* require a fix for Irix 5.2, though; I've included that patch below. SR version 2.3 will be released this month if you'd rather just wait for that. Gregg Townsend / Computer Science Dept / Univ of Arizona / Tucson, AZ 85721 +1 602 621 4325 gmt@cs.arizona.edu 110 57 16 W / 32 13 45 N / +758m ----------------------------------------------------------------------------- diff -c old/csw/mips.s new/csw/mips.s *** old/csw/mips.s Thu May 14 15:12:10 1992 --- new/csw/mips.s Mon Sep 12 17:58:22 1994 *************** *** 64,70 **** move $6,$18 move $7,$19 la $31,under /* detect underflow if called code returns */ ! j $20 /* jump to entry point */ .end sr_build_context --- 64,71 ---- move $6,$18 move $7,$19 la $31,under /* detect underflow if called code returns */ ! move $25,$20 /* MIPS ABI requires $25 for indirect call */ ! j $25 /* jump to entry point */ .end sr_build_context From owner-info-sr Mon Oct 10 08:46:22 1994 Received: from owl.CS.Arizona.EDU by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA04695; Mon, 10 Oct 1994 08:46:25 MST Received: (from gmt@localhost) by owl.CS.Arizona.EDU (8.6.9/8.6.6) id IAA06231 for info-sr; Mon, 10 Oct 1994 08:46:22 -0700 Date: Mon, 10 Oct 1994 08:46:22 -0700 From: Gregg Townsend Message-Id: <199410101546.IAA06231@owl.CS.Arizona.EDU> To: info-sr@owl Subject: further Linux help Content-Length: 1053 From gwpatter@king.mcs.drexel.edu Sat Oct 8 19:33:04 1994 Date: Sat, 8 Oct 94 22:33:59 EDT From: gwpatter@king.mcs.drexel.edu (William R. Patterson) Subject: further linux help Some very nice and helpful people have responded to my previous message requesting help in porting SR to Linux. I have set up one symbolic link that appears to have been helpful, and I have doctored up the util.c file to provide additional references. But... alas, while the effort has gone farther than it did before, the result is still that I am getting numerous undefined conditions. Does anyone have a clear and concise set of symlinks or whatever other things are needed for a successful port to Linux? So far the portation has been piecemeal, and I am hoping some good soul can provide me with a silver bullet or a quick fix, or, if not, some comiseration. Is there a particular directory onto which SR should be placed? I have been trying to place it on /home/sr, is this OK? Any and all ideas are welcome, and I thank you. --Bill gwpatter@mcs.drexel.edu From tage@cs.uit.no Tue Oct 11 05:32:01 1994 Received: from hpserv0.cs.UiT.No by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA16486; Tue, 11 Oct 1994 05:32:01 MST Received: from dslab2.cs.uit.no by hpserv0.cs.uit.no (1.37.109.11/Task/HJ-5) id AA052018718; Tue, 11 Oct 1994 13:31:58 +0100 Received: from localhost by dslab2.cs.uit.no (1.37.109.9/HJ/Task-5) id AA0466434237; Tue, 11 Oct 1994 13:31:57 +0100 Message-Id: <199410111231.AA0466434237@dslab2.cs.uit.no> To: info-sr@cs.arizona.edu Subject: Passing rec (struct) with String (back) from C to SR Date: Tue, 11 Oct 94 13:31:57 MET From: Tage Stabell-Kulo I'm trying to pass a rec from an external routine in C back into SR. In SR I have type to_c = rec( x : int; c : char) type from_c = rec( x : int; s: string[10]; y : int; z : int) external pasing_test(i :to_c) returns x : ptr from_c and in C I write: struct from_sr { int x; char c; } struct to_sr { int x; String s; int y; int z; } But in /usr/local/lib/sr/sr.h, the char* is commented out from the struct String, and the Interfaces/main.c represent the struct from_c as typedef struct { Int x; String c[3]; Int y; Int z; } rs2; How do I declare the struct from_sr, and how do I fill it in so that I can access the string within SR ? //// Tage Stabell-Kuloe | e-mail: tage@cs.uit.no //// /// Department of Computer Science/IMR | Finger: tage@tage.cs.uit.no /// // 9037 University of Tromsoe, Norway | Phone : +47-776-44032 // / "'oe' is '\o' in TeX" | Fax : +47-776-44580 / From owner-info-sr Tue Oct 11 11:43:24 1994 Received: from owl.CS.Arizona.EDU by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA10257; Tue, 11 Oct 1994 11:43:28 MST Received: (from gmt@localhost) by owl.CS.Arizona.EDU (8.6.9/8.6.6) id LAA07704; Tue, 11 Oct 1994 11:43:24 -0700 Date: Tue, 11 Oct 1994 11:43:24 -0700 From: Gregg Townsend Message-Id: <199410111843.LAA07704@owl.CS.Arizona.EDU> To: tage@cs.uit.no Cc: info-sr@owl Subject: Re: Passing rec (struct) with String (back) from C to SR Content-Length: 931 [re: question about passing a particular struct back from C to SR] The declaration String c[3] generated by SR allocates space for a string header followed by space to hold the string data. The [3] varies depending on the declared maxlength of the string. An equivalent declaration in C would be typedef struct { int x; String cheader; /* header = c[0] */ unsigned char cdata[2*sizeof(String)]; /* data = c[1],c[2] */ int y; int z; } rs2; To set up the string for return into SR, put the string data in the "cdata" field above. Set cheader.length to the number of characters (which should be between 0 and 10 inclusive per the declaration in SR). Also, set cheader.size to 10 + sizeof(String) + 1 to set the maxlength. Gregg Townsend / Computer Science Dept / Univ of Arizona / Tucson, AZ 85721 +1 602 621 4325 gmt@cs.arizona.edu 110 57 16 W / 32 13 45 N / +758m From smartt@ariel.msci.memst.edu Thu Oct 13 17:14:30 1994 Received: from ariel (ariel.msci.memst.edu) by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA25268; Thu, 13 Oct 1994 15:14:32 MST Received: by ariel id AA13871 (5.65c/IDA-1.4.4 for info-sr@cs.arizona.edu); Thu, 13 Oct 1994 17:14:30 -0500 Date: Thu, 13 Oct 1994 17:14:30 -0500 From: Todd Smart Message-Id: <199410132214.AA13871@ariel> Received: by NeXT.Mailer (1.95) Received: by NeXT Mailer (1.95) To: info-sr@cs.arizona.edu Subject: SR on AIX??? Cc: ordmanc@next1 SR developers/users: I am trying to find out if there has been any effort made to implement SR on IBM's AIX environment. If anyone has information on this matter please route it to: smartt@next1.msci.memst.edu Thanks for your time... Todd Smart (The University of Memphis) From gmt Mon Oct 17 13:54:20 1994 Date: Mon, 17 Oct 1994 13:54:20 MST From: "Gregg Townsend" Message-Id: <199410172054.AA17809@optima.cs.arizona.edu> Received: by optima.cs.arizona.edu (5.65c/15) id AA17809; Mon, 17 Oct 1994 13:54:20 MST To: info-sr Subject: Announcing version 2.3 of SR [ This is a special mailing that includes people who've shown a recent interest in SR. If you're not on the Info-SR mailing list, you won't receive any further mail from this list. ] Version 2.3 of the SR programming language is now available, superseding version 2.2.1 of last December. Version 2.3 can be obtained: -- by anonymous FTP from the /sr directory on ftp.cs.arizona.edu -- by electronic mail: for details, send a "help" message to ftpmail@cs.arizona.edu -- on tape, cartridge or diskette: write to the address below for details SR (Synchronizing Resources) is a language for writing concurrent programs and is documented in "The SR Programming Language: Concurrency in Practice", by Gregory R. Andrews and Ronald A. Olsson (Benjamin/Cummings, 1993, ISBN 0-8053-0088-0). An overview of version 1 of the language and implementation appeared in the January, 1988, issue of ACM TOPLAS (10,1, 51-86). SR runs on Sun, DEC, HP, IBM, SGI, Linux, and other Unix platforms. Version 2.3 improves array performance, adds support for the Intel Paragon and the IBM RS/6000, and includes other improvements. Details are given in the Release Notes, which are available as a separate PostScript file in the FTP area. For more information, send a message to sr-project@cs.arizona.edu. Gregg Townsend / Computer Science Dept / Univ of Arizona / Tucson, AZ 85721 +1 602 621 4325 gmt@cs.arizona.edu 110 57 16 W / 32 13 45 N / +758m From marco@nextiac.iac.rm.cnr.it Tue Oct 18 10:54:53 1994 Received: from vaxiac.iac.rm.cnr.it by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA10770; Tue, 18 Oct 1994 06:13:51 MST Received: from nextiac.iac.rm.cnr.it by vaxiac.iac.rm.cnr.it with PMDF#10339; Tue, 18 Oct 1994 11:59 GMT Received: by nextiac.iac.rm.cnr.it (NX5.67c/NX3.0S) id AA00867; Tue, 18 Oct 94 10:54:53 +0200 Received: by NeXT.Mailer (1.87.1.RR) Received: by NeXT Mailer (1.87.1.RR) Date: Tue, 18 Oct 94 10:54:53 +0200 From: marco@nextiac.iac.rm.cnr.it Subject: Re: Announcing version 2.3 of SR To: Info-SR@cs.arizona.edu Message-Id: <9410180854.AA00867@nextiac.iac.rm.cnr.it> I subscribe this mail-list bests marco From gwpatter@king.mcs.drexel.edu Mon Oct 31 10:20:50 1994 Received: from mcs.drexel.edu (king.mcs.drexel.edu) by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA07682; Mon, 31 Oct 1994 08:19:57 MST Received: from queen.mcs.drexel.edu by mcs.drexel.edu (4.1/SMI-4.0) id AA13410; Mon, 31 Oct 94 10:20:50 EST Date: Mon, 31 Oct 94 10:20:50 EST From: gwpatter@king.mcs.drexel.edu (William R. Patterson) Message-Id: <9410311520.AA13410@mcs.drexel.edu> To: info-sr@cs.arizona.edu Subject: Thanks\ Thanks to all who helped me. I now have an implementation of SR under Linux on both a Compaq notebook computer and on a desktop. Your help is much appreciated. --Bill From gwpatter@king.mcs.drexel.edu Mon Nov 14 10:16:22 1994 Received: from mcs.drexel.edu (king.mcs.drexel.edu) by optima.cs.arizona.edu (5.65c/15) via SMTP id AA21088; Mon, 14 Nov 1994 08:15:29 MST Received: from queen.mcs.drexel.edu by mcs.drexel.edu (4.1/SMI-4.0) id AA14305; Mon, 14 Nov 94 10:16:22 EST Date: Mon, 14 Nov 94 10:16:22 EST From: gwpatter@king.mcs.drexel.edu (William R. Patterson) Message-Id: <9411141516.AA14305@mcs.drexel.edu> To: info-sr@cs.arizona.edu Subject: clock Does anyone know a way to dissociate the age() and nap() functions from the real-time clock to enable a simulation clock that is not closely coupled to real time? Doing this would extend the utility of SR as a simulation language, allowing it to examine processes that take a long time and doing the examination in a short time, and allowing it to do computationally intensive analysis of processes that take a short time. It would free the computation from the constraint of real time. --Bill gwpatter@mcs.drexel.edu From @exeter.ac.uk:steve@dcs.exeter.ac.uk Mon Nov 14 19:04:27 1994 Received: from sun2.nsfnet-relay.ac.uk by optima.cs.arizona.edu (5.65c/15) via SMTP id AA19130; Mon, 14 Nov 1994 14:50:21 MST Via: uk.ac.exeter; Mon, 14 Nov 1994 19:04:41 +0000 Received: from atlas.exeter.ac.uk by exub.exeter.ac.uk with SMTP (PP) id <14646-0@exub.exeter.ac.uk>; Mon, 14 Nov 1994 19:04:30 +0000 Received: from sirius.dcs.exeter.ac.uk by atlas.dcs.exeter.ac.uk; Mon, 14 Nov 94 19:04:26 GMT From: steve@atlas.ex.ac.uk Date: Mon, 14 Nov 94 19:04:27 GMT Message-Id: <4150.9411141904@sirius.dcs.exeter.ac.uk> To: gwpatter@king.mcs.drexel.edu Cc: info-sr@cs.arizona.edu In-Reply-To: William R. Patterson's message of Mon, 14 Nov 94 10:16:22 EST <9411141516.AA14305@mcs.drexel.edu> Subject: clock >>>>> On Mon, 14 Nov 94 10:16:22 EST, gwpatter@edu.drexel.mcs.king (William R. Patterson) said: > Does anyone know a way to dissociate the age() and > nap() functions from the real-time clock to enable > a simulation clock that is not closely coupled to > real time? Doing this would extend the utility of SR > as a simulation language, allowing it to examine > processes that take a long time and doing the examination > in a short time, and allowing it to do computationally > intensive analysis of processes that take a short time. > It would free the computation from the constraint of > real time. What you seem to be asking for is a language for parallel discrete event simulation. This leads to the principle of causality (the future cannot affect the past). If a process' time is dissociated from real-time, one process could get ahead of another and then there is the possibility of it receiving a message in its simulated past. There are a number of approaches from parallel discrete event simulation that can be used to solve this problem such as the conservative approach (null messages) or optimistic (time warp). One of my students implemented the conservative approach in SR two years ago. Another way of preserving causality is to have a central scheduler. Chapter 18 of the SR book shows how you can do discrete event simulation in SR in this way. The scheduler provides time() and delay() operations which are similar to age() and nap() but use simulated rather than real-time. Steve ------------------------------------------------------------------------------- Parallel Systems Research Laboratory | Stephen Turner Email: steve@dcs.exeter.ac.uk | Department of Computer Science WWW: http://www.dcs.exeter.ac.uk | University of Exeter Fax: +44 392 264067 | Prince of Wales Road Tel: +44 392 264048 | Exeter EX4 4PT England ------------------------------------------------------------------------------- From dkarumur@eecs.uic.edu Tue Nov 15 10:33:27 1994 Received: from mail.eecs.uic.edu by optima.cs.arizona.edu (5.65c/15) via SMTP id AA13841; Tue, 15 Nov 1994 09:32:03 MST Received: from bert.eecs.uic.edu (bert.eecs.uic.edu [128.248.166.23]) by mail.eecs.uic.edu (8.6.9/8.6.9) with ESMTP id KAA01646 for ; Tue, 15 Nov 1994 10:33:06 -0600 Received: (dkarumur@localhost) by bert.eecs.uic.edu (8.6.9/8.6.9) id KAA15860; Tue, 15 Nov 1994 10:33:27 -0600 Date: Tue, 15 Nov 1994 10:33:27 -0600 (CST) From: Dinkar Karumuri To: info-sr@cs.arizona.edu Subject: UNSUBSCRIBE Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Please unsubscribe me --Dinkar ||||| (o o) /-----------------------------OOOo-(_)-oOOO---------------------------------\ | Dinkar Karumuri Phone: Res: (312)693-8120 | | 5461, N.East River Road, #909 Off: (708)-531-1310 x 358 | | Chicago. IL-60656 | | | | E-Mail: dinkar@eecs.uic.edu | \------------------------------ooo0---0ooo----------------------------------/ From chris@chaos.aerojet.com Thu Nov 17 15:43:55 1994 Received: from chaos.aerojet.com by optima.cs.arizona.edu (5.65c/15) via SMTP id AA17089; Thu, 17 Nov 1994 15:45:06 MST Received: (from chris@localhost) by chaos.aerojet.com (8.6.9/8.6.9) id PAA26415 for info-sr@cs.arizona.edu; Thu, 17 Nov 1994 15:43:55 -0800 Message-Id: <199411172343.PAA26415@chaos.aerojet.com> Subject: Questions about srtrace To: info-sr@cs.arizona.edu Date: Thu, 17 Nov 1994 15:43:55 -0800 (PST) From: Christopher Arndt-Kohlway X-Mailer: ELM [version 2.4 PL21] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1363 I have a few questions about SR trace. I am trying to profile SR by placing some hooks in the SR RTS and I have come across some srtrace behavior that I can't explain. Using the program remote.sr in examples/other, I set up a test which should output beginning and end statitics for each process. My problem is that I do not see the ENDBODY trace for the main resource. Looking at the intermediate code, I see the code to write the trace as follows: L2:; sr_stop(0,0); end: if (sr_profile_flag) sr_profile("main end",0); if (sr_trc_flag) sr_trace("ENDBODY",SRC0+23,0); sr_finished_init(); } (with my sr_profile call inserted). When I look at the trace output this trace is missing. The BODY trace and others do appear. Also, I do not get traces from VM's running on remote systems. Is there something special which needs to be done to enable remote VM's to trace? I would appreciate any help you can give. TIA :^) Chris -- =============================================================================== Christopher Arndt-Kohlway E-Mail: chris@aerojet.com Aerojet, Sacramento Plant Phone : (916) 355-5634 Sacramento, California (US) Fax : (916) 355-6635 =============================================================================== From gmt@owl Fri Nov 18 09:19:45 1994 Received: from owl.CS.Arizona.EDU by optima.cs.arizona.edu (5.65c/15) via SMTP id AA07108; Fri, 18 Nov 1994 09:19:47 MST Received: (from gmt@localhost) by owl.CS.Arizona.EDU (8.6.9/8.6.6) id JAA01585; Fri, 18 Nov 1994 09:19:45 -0700 Date: Fri, 18 Nov 1994 09:19:45 -0700 From: Gregg Townsend Message-Id: <199411181619.JAA01585@owl.CS.Arizona.EDU> To: chris@chaos.aerojet.com, info-sr@cs.arizona.edu Subject: Re: Questions about srtrace Content-Length: 790 From: Christopher Arndt-Kohlway Using the program remote.sr in examples/other, I set up a test which should output beginning and end statitics for each process. My problem is that I do not see the ENDBODY trace for the main resource. The main resource executes an explicit `stop' statement, so it never reaches the end of the body. Also, I do not get traces from VM's running on remote systems. Is there something special which needs to be done to enable remote VM's to trace? Remote VMs are traced if the trace is written to stdout or stderr, but not otherwise. Gregg Townsend / Computer Science Dept / Univ of Arizona / Tucson, AZ 85721 +1 602 621 4325 gmt@cs.arizona.edu 110 57 16 W / 32 13 45 N / +758m From canedo@taec.enet.dec.com Tue Nov 29 08:36:59 1994 Received: from inet-gw-2.pa.dec.com by optima.cs.arizona.edu (5.65c/15) via SMTP id AA10017; Tue, 29 Nov 1994 08:36:59 MST Received: from vbormc.vbo.dec.com by inet-gw-2.pa.dec.com (5.65/10Aug94) id AA15767; Tue, 29 Nov 94 07:31:49 -0800 Received: by vbormc.vbo.dec.com; id AA00672; Tue, 29 Nov 94 16:15:52 +0100 From: canedo@taec.enet.dec.com Message-Id: <9411291515.AA00672@vbormc.vbo.dec.com> Received: from taec.enet; by vbormc.enet; Tue, 29 Nov 94 16:15:52 MET Date: Tue, 29 Nov 94 16:15:52 MET To: info-sr@cs.arizona.edu Apparently-To: info-sr@cs.arizona.edu Subject: Graphical interface support with sr Hello everyone, I'm new in sr programming and I wish to have some info about the following: I plan to model a network system using sr concurrency (each node is a single "thread" of control). This network will be viewed with an X11 interface written with tcl (and tk extension) language. I need also to monitor an external process which gives the real state of the network (or at least some nodes of it). Is it possible to program the "kernel" (network model) in sr and use tcl-tk interface (making tk calls in sr program for example) ? I saw that sr has 2 grahical extensions but I need widgets (menus, dialogs...) Sorry if this question is too obvious but I have little time to choose a language to implement my system (the interface is partially written). Thanks in advance, Best regards Joseph Canedo From owner-info-sr Tue Nov 29 13:29:09 1994 Received: from owl.CS.Arizona.EDU by optima.cs.arizona.edu (5.65c/15) via SMTP id AA26730; Tue, 29 Nov 1994 13:29:12 MST Received: (from gmt@localhost) by owl.CS.Arizona.EDU (8.6.9/8.6.6) id NAA02280; Tue, 29 Nov 1994 13:29:09 -0700 Date: Tue, 29 Nov 1994 13:29:09 -0700 From: Gregg Townsend Message-Id: <199411292029.NAA02280@owl.CS.Arizona.EDU> To: canedo@taec.enet.dec.com, info-sr@owl Subject: Re: Graphical interface support with sr Content-Length: 681 An SR main program can call functions written in C, so you could write an interface to the tcl/tk library. I'd think that would fairly straightforward, though it would take some time. However, if SR calls a C function that blocks (waiting for input, for instance) the whole SR program comes to a halt -- not just the particular process. Also, it's not possible for the C program to call back into SR, which would make callbacks tricky. Either of these problems might be enough to make a tcl/tk interface impossible. Gregg Townsend / Computer Science Dept / Univ of Arizona / Tucson, AZ 85721 +1 602 621 4325 gmt@cs.arizona.edu 110 57 16 W / 32 13 45 N / +758m From canedo@taec.enet.dec.com Thu Dec 1 09:40:42 1994 Received: from inet-gw-2.pa.dec.com by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA23032; Thu, 1 Dec 1994 09:40:42 MST Received: from vbormc.vbo.dec.com by inet-gw-2.pa.dec.com (5.65/10Aug94) id AA11376; Thu, 1 Dec 94 08:34:35 -0800 Received: by vbormc.vbo.dec.com; id AA26575; Thu, 1 Dec 94 17:18:20 +0100 From: canedo@taec.enet.dec.com Message-Id: <9412011618.AA26575@vbormc.vbo.dec.com> Received: from taec.enet; by vbormc.enet; Thu, 1 Dec 94 17:18:21 MET Date: Thu, 1 Dec 94 17:18:21 MET To: info-sr@cs.arizona.edu Apparently-To: info-sr@cs.arizona.edu Subject: Little problem with a sr program Hello, I wrote a little program in order to learn sr language. It is composed of 1 ressource in which there is 3 processes: one of them writes "toto" in the screen every 4 seconds. The 2 others wait an integer in standard input and print it when is entered by user. When I launch the program the string "toto" is written as expected but when I enter the first integer the program "blocks" (does not print "toto" anymore). I enter the second integer and the program writes "toto" again and so on. This is the script: resource essai() process readline var x: int do true -> writes("First source\n") read(x) write("result 1 is ",x) if x = 0 -> exit fi od end process readline2 var x: int do true -> writes("Second source\n") read(x) write("result 2 is ",x) if x = 0 -> exit fi od end process print do true -> writes("toto\n") nap(400) od end end I'm sure it is a trivial problem in concurrency (conflict in sharing the same resource or something like that) but I'm very new in this domain. Can some one explain to me this phenomemon please ? Thanks in advance Best regards Joseph Canedo From greg Thu Dec 1 09:53:48 1994 Received: from paloverde.CS.Arizona.EDU by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA24191; Thu, 1 Dec 1994 09:53:50 MST Date: Thu, 1 Dec 1994 09:53:48 MST From: "Greg Andrews" Message-Id: <199412011653.AA01080@paloverde.cs.arizona.edu> Received: by paloverde.cs.arizona.edu; Thu, 1 Dec 1994 09:53:48 MST To: canedo@taec.enet.dec.com, info-sr@cs.arizona.edu Subject: Re: Little problem with a sr program Input has to be ended with a newline (carriage return) character. I believe this is what happened to you. Greg Andrews From magnus@cs.chalmers.se Fri Dec 2 12:42:27 1994 Received: from muppet101.cs.chalmers.se by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA01667; Fri, 2 Dec 1994 04:46:52 MST Received: (from magnus@localhost) by muppet101.cs.chalmers.se (8.6.9/8.6.9) id MAA02759; Fri, 2 Dec 1994 12:42:27 +0100 Date: Fri, 2 Dec 1994 12:42:27 +0100 From: Magnus Carlsson Message-Id: <199412021142.MAA02759@muppet101.cs.chalmers.se> To: canedo@taec.enet.dec.com Cc: info-sr@cs.arizona.edu Subject: Re: Little problem with a sr program In-Reply-To: <9412011618.AA26575@vbormc.vbo.dec.com> References: <9412011618.AA26575@vbormc.vbo.dec.com> You have two processes that contend for input, which is not a good idea. You have no idea which characters from input will end up in readline1, and which of them that readline2 will get. Therefore, SR-programs usually do not try to read from the same file simultaneously in many processes. Of course, the whole program shouldn't block if you do it, so I gather that's a bug in the runtime system. One fix could be to protect the reading by a semaphore. See the enclosed program. If anybody thinks I'm wrong here, please correct me! Regards /Magnus --- resource essai() sem m := 1 # New process readline var x: int do true -> P(m) # New writes("First source\n") read(x) V(m) # New write("result 1 is ",x) if x = 0 -> exit fi od end process readline2 var x: int do true -> P(m) # New writes("Second source\n") read(x) V(m) # New write("result 2 is ",x) if x = 0 -> exit fi od end process print do true -> writes("toto\n") nap(400) od end end From canedo@taec.enet.dec.com Mon Dec 5 08:15:48 1994 Received: from inet-gw-2.pa.dec.com by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA04415; Mon, 5 Dec 1994 08:15:48 MST Received: from vbormc.vbo.dec.com by inet-gw-2.pa.dec.com (5.65/10Aug94) id AA08592; Mon, 5 Dec 94 07:09:58 -0800 Received: by vbormc.vbo.dec.com; id AA16897; Mon, 5 Dec 94 15:52:28 +0100 From: canedo@taec.enet.dec.com Message-Id: <9412051452.AA16897@vbormc.vbo.dec.com> Received: from taec.enet; by vbormc.enet; Mon, 5 Dec 94 15:52:29 MET Date: Mon, 5 Dec 94 15:52:29 MET To: info-sr@cs.arizona.edu Apparently-To: info-sr@cs.arizona.edu Subject: More detailed documentation on sr Hello every one, I wish to learn to use sr language but unfortunately I have only one article (TR 92-09 "SR : A language for Parallel and Distributed Programming"). I know there is a book on sr language programming but I think it will be difficult to me to obtain from France. Is there a more detailed documentation on sr (than the article TR 92-09 : "SR : A language for Parallel and Distributed Programming") available under electronic form and FTPable ? If not, does some one know how to order te book from France (and its price) ? Thanks in advance Best regards Joseph Canedo From owner-info-sr Mon Dec 5 08:59:41 1994 Received: from owl.CS.Arizona.EDU by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA07385; Mon, 5 Dec 1994 08:59:44 MST Received: (from gmt@localhost) by owl.CS.Arizona.EDU (8.6.9/8.6.6) id IAA04408 for info-sr; Mon, 5 Dec 1994 08:59:41 -0700 Date: Mon, 5 Dec 1994 08:59:41 -0700 From: Gregg Townsend Message-Id: <199412051559.IAA04408@owl.CS.Arizona.EDU> To: info-sr@owl Subject: Re: More detailed documentation on sr Content-Length: 801 From: canedo@taec.enet.dec.com Is there a more detailed documentation on sr (than the article TR 92-09 : "SR : A language for Parallel and Distributed Programming") available under electronic form and FTPable ? Sorry, no. If not, does some one know how to order te book from France (and its price) ? The real experts on ordering things from the US are probably to be found in technical and university bookstores in France. They can probably get a copy from the stock of a European distributor. Alternatively, you could call the publisher at +1 415 594 4400. [In the US, call 1 800 950 2665.] Gregg Townsend / Computer Science Dept / Univ of Arizona / Tucson, AZ 85721 +1 602 621 4325 gmt@cs.arizona.edu 110 57 16 W / 32 13 45 N / +758m From jmwagner@ucdavis.edu Mon Dec 5 08:51:07 1994 Received: from ucdavis.ucdavis.edu by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA14931; Mon, 5 Dec 1994 10:35:25 MST Received: from ike.ucdavis.edu by ucdavis.ucdavis.edu (8.6.9/UCD2.50) id IAA11389; Mon, 5 Dec 1994 08:50:24 -0800 Received: from james.ucdavis.edu.ucdavis.edu by ike.ucdavis.edu (4.1/UCD2.04) id AA22307; Mon, 5 Dec 94 08:51:47 PST Received: by james.ucdavis.edu.ucdavis.edu (931110.SGI/client-1.5) id AA16572; Mon, 5 Dec 94 08:51:07 -0800 From: "John Wagner" Message-Id: <9412050851.ZM16570@james.ucdavis.edu> Date: Mon, 5 Dec 1994 08:51:07 -0800 In-Reply-To: canedo@taec.enet.dec.com "More detailed documentation on sr" (Dec 5, 3:52pm) References: <9412051452.AA16897@vbormc.vbo.dec.com> X-Mailer: Z-Mail (3.1.0 22feb94 MediaMail) To: canedo@taec.enet.dec.com, info-sr@cs.arizona.edu Subject: Re: More detailed documentation on sr Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 Joseph, Try The SR Programming Language -- Concurrency in Practice By Gregory R. Andrews and Ronald A. Olsson. The Benjamin/Cummings Publishing Co, Inc. 1993. ISBN 0-8053-0088-0 -- John Wagner Institute of Theoretical Dynamics Hydrology Modeling Lab (LAWR) Graduate Group in Applied Mathematics University of California, Davis Disclaimer: NOBODY at UCD agrees with ANYTHING I say... +----------------------+-------------------------------------------------+ | jmwagner@ucdavis.edu | I spent four years prostrate to the higher mind | | jmwagner@ucdavis | Got my paper and I was free -Indigo Girls | +----------------------+-------------------------------------------------+ From shartley@king.mcs.drexel.edu Wed Dec 7 07:27:43 1994 Received: from mcs.drexel.edu (king.mcs.drexel.edu) by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA13409; Wed, 7 Dec 1994 05:26:52 MST Received: by mcs.drexel.edu (4.1/SMI-4.0) id AA01791; Wed, 7 Dec 94 07:27:43 EST Date: Wed, 7 Dec 94 07:27:43 EST From: shartley@king.mcs.drexel.edu (Stephen J. Hartley) Message-Id: <9412071227.AA01791@mcs.drexel.edu> To: info-sr@cs.arizona.edu Subject: teaching operating systems In late January or early February, Oxford Univ. Press will publish my book "Operating Systems Programming: The SR Programming Language". The ISBN is 0-19-5095790. Contact them at (800) 451-7556 for a copy. Suggested list price is $25.00. The example SR programs used in the book are avaliable by anonymous ftp from site mcs.drexel.edu in directory pub/shartley; retrieve file book_SR_examples.tar.gz or book_SR_examples.tar.Z (remember to use binary mode) to get a tar archive compressed with the GNU compression program gzip or the UNIX utility compress, respectively. Steve Hartley Drexel University shartley@mcs.drexel.edu Here is the Preface: SR is a language for concurrent programming. This book describes the SR language, presents some examples of SR programs in the context of an operating systems or concurrent programming course, and provides some programming as- signments in the form of open Student Laboratories. The SR language can be used by instructors of concurrent programming or operating systems courses to give students experience in writing concurrent programs that use multiple processes, semaphores, monitors, message passing, remote procedure calls, and the rendezvous. The language can also be used for parallel computing in a shared-memory multiprocessor or a distributed memory cluster environment. The intended audience is undergraduate and graduate students enrolled in concurrent programming and operating systems classes. Prerequisites for stu- dents are knowledge of a high-level programming language like C or Pascal and operating systems concepts. See the Instructor's Manual for more detailed prerequisites. This book is designed to be used during an operating systems course in conjunction with one of the standard texts to provide concurrent programming experience. The book can also be used in a full semester follow- on course to operating systems to provide more extensive programming experi- ence. Since most concepts and terms are defined, the book can be used as the sole text in an introductory concurrent programming course that precedes the operating systems course. The required computing environment is a UNIX platform such as a Sun 3, Sun 4, Sequent Symmetry, DECstation, SGI Iris, NeXT, HP RISC, or PC compatible running Linux. The SR examples and programming assignments in this book have been used suc- cessfully in undergraduate concurrent programming and operating systems courses at Drexel University, Philadelphia, Pennsylvania, and Trinity Univer- sity, San Antonio, Texas. The book arose out of a collection of handouts and class notes distributed to students during 1991-94 and is a greatly expanded version of [some earlier journal papers]. This book has several important features. * The material is keyed to four standard operating systems texts: Deitel, Silberschatz and Galvin, Stallings, and Tanenbaum. The relevant sections of these texts for the major concepts, such as semaphores and monitors, are indi- cated. * Each of the numerous example programs in this book includes output from one or two sample runs to show how the program works. * Algorithm animation using the animator interpreter from the XTANGO sys- tem is described and used in several of the examples. * Numerous programming assignments in the form of open Student Labora- tories are given. The example programs in this book will be made available by anonymous ftp at site mcs.drexel.edu. Contact the author for further details. Here is the TOC: Operating Systems Programming Hardware Basics Hardware and Software Interrupts Hardware Protection CPU Scheduling The SR Programming Language Sequential Features Additional Sequential Examples and Features Lab: Sequential SR Multiple Resources Lab: Resources and Capabilities Additional Features Animating Programs with XTANGO Lab: XTANGO Animation Concurrent Programming Multiple Processes in One Resource Multiple Processes in Multiple Resources Simulated versus Real Concurrency Debugging Techniques Race Conditions and Process Synchronization Lab: Race Conditions Architecture and Shared Memory Systems The Mutual Exclusion Problem Semaphores SR Semaphores Classical Operating Systems Problems Semaphore Solutions The Producers and Consumers The Sleeping Barber The Dining Philosophers The Readers and Writers Binary Semaphores Lab: Semaphores Assignment: Fix Race Condition Assignment: Multiple Producers and Consumers Assignment: Another Classical Problem Assignment: Multiple Sleeping Barbers Assignment: Dining Philosophers Assignment: Fair Readers and Writers Assignment: Baboons Crossing a Canyon Assignment: Fraternity Party Assignment: Jurassic Park Animating Operating Systems Algorithms Lab: Classical Problem Animation Monitors Monitors and SR The Producers and Consumers The Dining Philosophers The Readers and Writers Implementing Monitors with Semaphores Lab: Monitors Assignment: Fair Baboons Assignment: Sleeping Barber Assignment: Fair Dining Philosophers Assignment: Fraternity Party Assignment: Bakery Message Passing and the Rendezvous Message Passing Message Passing in SR Concurrent Programs and Message Passing Distributed Mutual Exclusion Miscellaneous Examples Lab: Message Passing Assignment: Merge Sort Assignment: Pipeline Sieve of Eratosthenes Assignment: Compare/Exchange Sort Assignment: Speedup Assignment: Algorithm Animation Rendezvous Classical Problems Using Rendezvous Nested in Statements A Lock Resource The Distributed Dining Philosophers RPC and Client-Server Programming Summary of Operations and Invocations Lab: Rendezvous Assignment: Fair Baboons Assignment: Sleeping Barber Assignment: Fair Readers and Writers Assignment: Fair Dining Philosophers Assignment: Fraternity Party Assignment: Bakery Assignment: Banker's Algorithm More Animation with XTANGO Lab: Algorithm Animation Parallel Computing Coarse-Grained Concurrency in SR Patterns of Communication Data Parallelism Shared Memory Examples Message Passing Examples Integrating Animator into SR Lab: Parallel Computing Assignment: Speedup Assignment: Race Condition Assignment: No Race Condition Instructor's Manual Sample Lab Solutions Semaphores Monitors Message Passing Rendezvous Parallel Computing From scip3161@cobra.nus.sg Thu Dec 8 09:07:51 1994 Received: from cobra.nus.sg by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA01625; Wed, 7 Dec 1994 18:08:17 MST Received: (from scip3161@localhost) by cobra.nus.sg (8.6.9/8.6.9/CNS-3.5) id JAA22621 for info-sr@cs.arizona.edu; Thu, 8 Dec 1994 09:07:52 +0800 From: E Z A M Message-Id: <199412080107.JAA22621@cobra.nus.sg> Subject: any shared memory machine with SR? To: info-sr@cs.arizona.edu (arizona info_sr) Date: Thu, 8 Dec 1994 09:07:51 +0800 (SST) X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Length: 558 Hello, I am looking for a place where I can test a program running in parallel environment. There are many problems in trying to understand or visualise the behaviour of concurrent algorithms. I hope that if possible, I would like to run the program in parallel environment. -- ........................................................ Lai_Hong Cheong (E Z A M) scip3161@leonis.nus.sg, cheongla@iscs.nus.sg http://www.iscs.nus.sg/~cheongla Learn from the atom and particles about the way of life. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ From canedo@taec.enet.dec.com Tue Dec 13 03:45:57 1994 Received: from inet-gw-2.pa.dec.com by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA08008; Tue, 13 Dec 1994 03:45:57 MST Received: from vbormc.vbo.dec.com by inet-gw-2.pa.dec.com (5.65/10Aug94) id AA06266; Tue, 13 Dec 94 02:42:12 -0800 Received: by vbormc.vbo.dec.com; id AA22327; Tue, 13 Dec 94 11:24:35 +0100 From: canedo@taec.enet.dec.com Message-Id: <9412131024.AA22327@vbormc.vbo.dec.com> Received: from taec.enet; by vbormc.enet; Tue, 13 Dec 94 11:24:36 MET Date: Tue, 13 Dec 94 11:24:36 MET To: info-sr@cs.arizona.edu Apparently-To: info-sr@cs.arizona.edu Subject: Passing a SR proc in argument to an external C function Hello everyone, Is it possible to pass a SR proc in argument to an external C function ? I want to make a callback (GUI) using SR code instead of C code... If that is not possible what do you think about this solution : 1 process written in SR + 1 process written in C (Motif) or tcl (tk) which communicate via UNIX pipes. Does this solution garantee synchronization between interface and state of objects in SR process ? Any help is greatly appreciated Best regards Joseph Canedo From owner-info-sr Tue Dec 13 07:40:32 1994 Received: from owl.CS.Arizona.EDU by optima.CS.Arizona.EDU (5.65c/15) via SMTP id AA14469; Tue, 13 Dec 1994 07:40:35 MST Received: (from gmt@localhost) by owl.CS.Arizona.EDU (8.6.9/8.6.6) id HAA11977; Tue, 13 Dec 1994 07:40:32 -0700 Date: Tue, 13 Dec 1994 07:40:32 -0700 From: Gregg Townsend Message-Id: <199412131440.HAA11977@owl.CS.Arizona.EDU> To: canedo@taec.enet.dec.com Cc: info-sr@owl Subject: Re: Passing a SR proc in argument to an external C function Content-Length: 898 From: canedo@taec.enet.dec.com Is it possible to pass a SR proc in argument to an external C function ? I want to make a callback (GUI) using SR code instead of C code... No. SR procs use a different calling convention than C. If that is not possible what do you think about this solution : 1 process written in SR + 1 process written in C (Motif) or tcl (tk) which communicate via UNIX pipes. That should be feasible. Just be sure to explicitly flush the output from both sides so that it doesn't stall in a buffer. Does this solution garantee synchronization between interface and state of objects in SR process ? It should