Poly/ML March 2002

polyml@lists.polyml.org

4 participants
5 discussions

beginner questions
by baxter＠math.umn.edu 29 Mar '02

29 Mar '02

The example works nicely, and the system seems very convenient. I made a small C++ dll now using extern "C" and it's fine too. Thanks very much for the help! John On Fri, 29 Mar 2002, David Matthews wrote: > John, > There's nothing special about a DLL for use with Poly/ML. If you can > build a DLL for use with C or Visual Basic then it will work with > Poly/ML. The Windows programming interface in Poly/ML is largely done > by calling Windows system DLLs. To build a DLL you need some sort of > Windows development environment. I've been using Visual C++ but you > should be able to use anything. You will normally need a .DEF file > which contains a list of the functions you want to export from the DLL. > I made a very simple example some time ago and I'm including it as a ZIP > archive. It was done with Visual C++ and includes an ML file which > calls the DLL. You can write your DLL in C++ but all the exported > functions have to use C calling conventions. You really need to look at > the documentation for your development environment to find out how to > construct a DLL. > > Hope that helps. > David. > > > > >> earlier questions by me

1 0

beginner questions
by baxter＠math.umn.edu 28 Mar '02

28 Mar '02

Just starting using polyml and had a couple of questions about using c functions. I read the documentation on the Poly/ML Interface to the C Programming Language but I'm sure I don't get the whole picture. (I've made little c functions for python & tcl in the past but of course that's different.) 1. The documentation gives an example of making a dll for unix using gcc. Since polyml has a very nice Windows interface it looks as if everything works on windows. I'm on windows 2000 for now and I wonder if I need to use the microsoft vc++ compiler to make dll's for that platform, or if other compilers are ok. I guess in any case I need to figure out what symbols the compiler is using for functions(?) but that shouldn't be hard. 2. There didn't seem to be any mention of special header files that are needed when compiling the c libraries. Are there any? Also do I need to make any special type of dll? 3. Since I have yet to make any c functions, my last question is rather excessive, but what about a c++ library? Can one compile it in such a way that polyml can use it? Thanks for any tips, or pointers to things I should have read more carefully! Cheers, John Baxter

1 0

Re: improve.c problem and patch
by David.Matthews＠prolingua.co.uk 25 Mar '02

25 Mar '02

Hi Bow-Yaw, Thanks for reporting that and sending the fix. I'll incorporate it in the next release. Markus Wenzel reported a problem in the disc garbage collector when compiling with GCC 3.0.2 which sounds exactly like this. There hasn't been any discussion of this on the Poly/ML mailing list because, so far, there hasn't been ANY discussion on the mailing list. As far as ml-lex and ml-yacc are concerned, I don't know whether they will work with Poly/ML or not. If they're written entirely in Standard ML they should work fine but if they use some special features of SML/NJ then there may be a problem. Does anyone else know? Regards, David. On Saturday, March 23, 2002, at 12:42 , Bow-Yaw Wang wrote: > > Hi Mr. Matthews: > > I try to build Isabelle2002 with PolyML 4.1.2. But poly > always crashes at "Bad pointer" when closing the database. > It turns out that gcc-3.0.4 has a warning and generates different > code for improve.c. After I made the following change: > > 739c739 > < while (n--) *pt++ = (word) FixupPointer((word *)*pt); > --- >> while (n--) { *pt = (word) FixupPointer((word *)*pt); pt++; } > > the problem goes away. Since I'm not on the mailing list, I'm not > sure if this is a known issue. Just let you know in case the > problem hasn't been reported. > > Currently, I'm working on a project which needs a theorem prover, > external function interface and yacc. PolyML + Isabelle would be > the ideal environment if tools like yacc were available. I'm thinking > using ml-yacc/ml-lex from SMLNJ and compiling everything in PolyML. > Do you by chance know whether it may work? > > Thanks, > > Bow-Yaw

2 1

polyml executables [was Re: improve.c problem and patch]
by bernard＠laas.fr 25 Mar '02

25 Mar '02

Hello, David wrote in last message: > There hasn't been any discussion of this on the Poly/ML mailing list > because, so far, there hasn't been ANY discussion on the mailing list. Let us start one then: I'm writing applications in Poly/ML that I intend to distribute in binary form. Currently, this requires users to download at least the polyml driver from polyml.org and install it. Not a big deal, but certainly an avoidable burden, specially when applications require several similar prerequisites. I think a nice addition to the polyml tools would be one that packs the polyml driver with a database, in a single executable. The tclpro compiler suite, for instance, has a convenient tool for such purposes, that wraps an interpreter with its libraries, code, and possibly data files, into a single executable. A similar "polywrap" tool would be perfect for my purposes. Any plans for such a tool, or on-the-shelf alternatives ? Other opinions ? Bernard. -- Bernard Berthomieu ^ Bernard.Berthomieu(a)laas.fr LAAS / CNRS . /_\ http://www.laas.fr/~bernard 7, avenue du Colonel Roche / \/ \ phone +33/(0)5 61 33 63 63 31077 Toulouse Cedex / \ \ fax +33/(0)5 61 33 64 11 France / \ \ ____________________________________/ \____\___________________________

2 1

Re: Poly/ML Compilation Performance Problem
by David.Matthews＠prolingua.co.uk 25 Mar '02

25 Mar '02

Rob, I've had a look at this and found the problem. It turned out to be much simpler to fix than I'd feared and applying the diffs below reduces the compilation time from around 45 minutes to around 7 seconds! The problem had to do with the way Poly/ML inlines "small" functions. In order to get reasonably efficient code small functions are inserted inline. If they are tail-recursive they can get converted into something like while-loops. I changed this and improved it quite a bit in recent releases but it seems there was something left from a previous attempt at this which was getting in the way. It was causing recursive calls to be expanded inline when they shouldn't have been, leading to a massive code blow-up. It was actually the fact that you had functions of the form val rec f = fn x => fn y => ..... rather than the equivalent "fun" declaration that meant that the problem only showed up in your code and hadn't appeared before. Poly/ML treats fun declarations specially so that a function of the form fun f a b c = ... gets its arguments passed on the stack. That means that we save the cost of creating a closure except when the function is partially applied. That doesn't apply to the equivalent val rec declaration. Perhaps it could but that would require the front end to look quite closely at the body of the outer function to see whether it was simply another function. Let me know how you get on with this. Regards, David. RCS file: /usr/cvsroot/mlsource/MLCompiler/CodeTree/G_CODE.ML,v retrieving revision 1.22 diff -r1.22 G_CODE.ML 1122,1126c1122 < case pt of < Lambda _ => true < | Declar {value, ...} => isSmall (value, sizeOptVal) < | Newenv [c] => isSmall (c, sizeOptVal) < | _ => size pt < !maxInlineSize --- > size pt < !maxInlineSize On Saturday, March 2, 2002, at 01:01 , Rob Arthan wrote: > David, > > Last month, I ported another tool of ours called ClawZ to Poly/ML. > ClawZ is much smaller than the ML parts of ProofPower. Somewhat to my > surprise it contains some code which provokes a performance problem > in the compiler, which ends up taking about 45 minutes to compile the > last 2,400 or so lines of code on a 450MHz Pentium Xeon with 512Mb of > RAM. This is the only case I've encountered where Poly/ML underperforms > SML/NJ.

1 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

Poly/ML March 2002