- gsll-devel - mailman3.common-lisp.net

[Gsll-devel] lu-decomp error
by Zach 20 Mar '08

20 Mar '08

It looks like the definition of lu-decomp (linear-algebra/lu.lisp) is incorrect. Instead of (signum :int) it should read (signum :pointer), right? My code (see below) works with this change, but not without it (it complains that the pointer isn't a fixnum). However, signum is really just a return value of the function, depending on how high level this should all be, we may want to write a wrapper that allocates that memory and passes it back as a second value. But then we would have to worry about memory leaks (since the GC won't free the memory, right?) Further, this `function' really just modifies it's arguments. I guess the Lisp way would be to change this into a function that returns the modified structures and mark it as a destructive function... It is hard for me to make Lisp style and C style coexist in my head. How were the GSL definitions generated? How did this :int :pointer confusion slip through? If the point here is to make a system where mathematics can be done interactively, i.e. (invert-matrix #2A((1 0) (3 4))) ==> #2A(...), then a big part of this is going to be making wrapper functions that hide the C like nuts and bolts that make GSL a bit annoying to use in the first place. (defun invert-matrix (mat) (gsl:letm ((mat2 (gsl:matrix-double mat)) (per (gsl:permutation 2)) (inv (gsl:matrix-double 2 2)) (sig (cffi:foreign-alloc :int :count 1)) ) (gsl:lu-decomp mat2 per sig) (gsl:lu-invert mat2 per inv) (cffi:foreign-free sig) (gsl:data inv) )) How can I get rid of the explicit foreign-alloc/free? Is this outside of the scope of GSLL (seems like a border case to me). Thanks, Zach

3 5

[Gsll-devel] How would you like to have your ad on 2 Million Websites ?
by Blog＠common-lisp.net 19 Mar '08

19 Mar '08

How would you like 2 Million Sites linking to your ad ? Weblog or blog population is exploding around the world, resembling the growth of e-mail users in the 1990s. Post your ads where people read them! - What if you could place your ad on all these sites ? Right, that would mean you would have millions of sites linking to your ad. For Full details please read the attached .html file Unsubscribe Please read the attached .txt file

1 0

[Gsll-devel] #'sort-vector-smallest-index and #'sort-vector-largest-index now work on 64-bit platforms; API changed.
by Liam Healy 16 Mar '08

16 Mar '08

The functions #'sort-vector-smallest-index and #'sort-vector-largest-index now work on 64-bit platforms. It was necessary to change the API in order to do this: instead of taking a GSL vector-fixnum as the first argument, they now take either a positive integer, or a CL vector. This is because it appears to be very difficult to find a GSL vector type that matches a C array of size_ts. Thus, the GSLL policy of using only GSL vectors and not raw C vectors has been altered. The returned object is either a newly-created CL fixnum vector, or the CL vector passed in. It will contain the smallest or largest indices. Now the GSLL policy of not requiring an object to be passed as an argument if it is used only for returning values is fulfilled. By the way, I have been sending an email to this list when there is an update to GSLL that changes the input or output of exported functions. I am also listing these updates on the Trac page http://trac.common-lisp.net/gsll. Liam

1 0

[Gsll-devel] Rewrite of data objects (SVN version 34)
by Liam Healy 11 Mar '08

11 Mar '08

I've rewritten how GSLL handles data (vectors, matrices, etc.) so that it is clearer how types are mapped, and the various macros use the same definitions to figure out how to translate from one form to another (e.g. Lisp side to C side). This should make maintenance and adding new vector and matrix types easier. For the most part, this is largely invisible to the library user, except that the GSL object names vector-single, vector-double, matrix-single, matrix-double now have "-float" appended: vector-single-float, vector-double-float, matrix-single-float, matrix-double-float. Because of all the examples/tests using one or more of these, many files were touched. The regression tests all pass as well as they did before, so I think I got everything updated. I wanted to alert you in case your code uses these objects. Liam

1 0

[Gsll-devel] lu-decomp error
by Liam Healy 11 Mar '08

11 Mar '08

Zach, Remove the signum argument from the argument to LU-decomp. That is, it should be defined (defmfun LU-decomp (A p) ...) Then everything will work as desired; signum will be returned from this function. Let me know if it works and I'll check that change in. Note there is no :pointer confusion and defmfun does all the wrapping necessary; look at the macroexpansion and definition of defmfun to see how it works. All of GSLL was written by me, save for the docstrings that came from the GSL documentation, and those were edited by me for style, grammar, and to fit GSLL. I'm sure a few errors like this one crept in. Many I caught with regression tests, but I don't have a test for LU-decomp. If you have a fairly simple example/test with LU-decomp post it and I will include it in the code. Thanks, Liam On 3/10/08, Zach <elzacho(a)gmail.com> wrote: > > It looks like the definition of lu-decomp (linear-algebra/lu.lisp) is > incorrect. Instead of (signum :int) it should read (signum :pointer), > right? My code (see below) works with this change, but not without it (it > complains that the pointer isn't a fixnum). However, signum is really just > a return value of the function, depending on how high level this should all > be, we may want to write a wrapper that allocates that memory and passes it > back as a second value. But then we would have to worry about memory leaks > (since the GC won't free the memory, right?) Further, this `function' > really just modifies it's arguments. I guess the Lisp way would be to > change this into a function that returns the modified structures and mark it > as a destructive function... It is hard for me to make Lisp style and C > style coexist in my head. > > How were the GSL definitions generated? How did this :int :pointer > confusion slip through? If the point here is to make a system where > mathematics can be done interactively, i.e. (invert-matrix #2A((1 0) (3 4))) > ==> #2A(...), then a big part of this is going to be making wrapper > functions that hide the C like nuts and bolts that make GSL a bit annoying > to use in the first place. > > (defun invert-matrix (mat) > (gsl:letm ((mat2 (gsl:matrix-double mat)) > (per (gsl:permutation 2)) > (inv (gsl:matrix-double 2 2)) > (sig (cffi:foreign-alloc :int :count 1)) ) > (gsl:lu-decomp mat2 per sig) > (gsl:lu-invert mat2 per inv) > (cffi:foreign-free sig) > (gsl:data inv) )) > > How can I get rid of the explicit foreign-alloc/free? Is this outside of > the scope of GSLL (seems like a border case to me). > > Thanks, > Zach > > > _______________________________________________ > Gsll-devel mailing list > Gsll-devel(a)common-lisp.net > http://common-lisp.net/cgi-bin/mailman/listinfo/gsll-devel > >

1 0

Re: [Gsll-devel] Bug with image
by Liam Healy 04 Mar '08

04 Mar '08

Jason, Can you do svn update and try again? I think I've fixed this problem, but I'd like you to test it. Thanks. Liam On 3/1/08, Jason Nielsen <jdn(a)math.carleton.ca> wrote: > Done. Ticket #3. > > Thanks, > > Jason > > > On Sat, 1 Mar 2008, Liam Healy wrote: > > > Jason, > > > > Thanks for the report. > > > > I'm not sure what is causing this problem; I think you are doing > > everything you are supposed to. I have to admit I haven't done a > > save-lisp-and-die with a GSLL image. I know that the SBCL developers > > have discussed issues related to save-lisp-and-die in the past; it > > could perhaps be some SBCL state that is not being saved in the image, > > either intentionally or accidentally. > > > > If it's not too much trouble, can you file a bug report on the Trac page > > http://trac.common-lisp.net/gsll/ -> New Ticket > > I will ask the sbcl developers if I don't come up with any other ideas. > > > > Liam > > > > On 3/1/08, Jason Nielsen <jdn(a)math.carleton.ca> wrote: > >> First of all, thank you for making these bindings available. They are > >> great! Much better that the bare-bones cffi wrappers I've been using for > >> the small subset of gsl I've needed. > >> > >> I think there is a small bug with exceptions in the code. I usually make > >> my own image with code that I use regularly (sbcl 1.0.14), i.e. > >> > >> (require 'asdf) > >> (require 'asdf-install) > >> (require 'cffi) > >> (require 'gsll) > >> > >> (sb-ext:save-lisp-and-die "sbcl-with-slime.core") > >> > >> then start up sbcl with "sbcl --core sbcl-with-slime.core". In this mode > >> of operation gsll hangs when an exception is caught, e.g. > >> > >> (jacobian-elliptic-functions 0.61802d0 1.5d0) > >> > >> just sits and the repl prompt is not returned (endless loop or something). > >> > >> If I just fire up sbcl and load gsll normal everything works fine and I > >> get the expected warning that |m|>1. > >> > >> Hope this is a useful enough description of the problem for you to track > >> it down. > >> > >> With kind regards, > >> Jason > >> _______________________________________________ > >> Gsll-devel mailing list > >> Gsll-devel(a)common-lisp.net > >> http://common-lisp.net/cgi-bin/mailman/listinfo/gsll-devel > >> > > >

2 1

Re: [Gsll-devel] GSL version being used
by Liam Healy 04 Mar '08

04 Mar '08

82 failures seems like quite a lot. I think using 1.10 there should be on the order of a dozen. If you can't figure out what's going on, post a sample and I'll see if I can figure it out. Liam On 3/3/08, Felip Alàez Nadal <uu.nix.uu(a)gmail.com> wrote: > Thanks you: > > I wanted to use the most similar configuration to test GSLL. I'm using > gsl-1.10 on a i386 linux pc running Suse linux Enterprise 10. > > First of all, congratulations! GSLL looks really fine. We've been looking > for something like this from long time ago. > > I've ran the regression tests and I get 82 failures and 2 execution errors. > I still couldn't try to solve this, but I'll try to solve some of those > failures and send you a patch. > > Bye! > -- > Felip Alàez Nadal

1 0

[Gsll-devel] MAKE-SYMBOL-FROM-STRINGS does the wrong thing?
by Zach 03 Mar '08

03 Mar '08

Okay, so this is my first post, so I may be completely misguided, After loading asdf loading GSLL, I find that I cannot do any matrix or permutation stuff with it, unless I did it from withing the package itself. After a while I tracked this to MAKE-SYMBOL-FROM-STRINGS (init/utility.lisp) which interns symbols on the fly. The problem is that it creates symbols in whatever package you happen to be in when it is called. This matters because it is used create class name symbols which should be in the package GSLL not where ever I happen to be using GSLL from. For example: (gsl:letm ((perm-1 (gsl:permutation 4 t))) (gsl:set-identity perm-1) (gsl:maref perm-1 2) ) ==> Error: There is no class named GSL-PERMUTATION. (use-package :gsll) (gsl:letm ((perm-1 (gsl:permutation 4 t))) (gsl:set-identity perm-1) (gsl:maref perm-1 2) ) ==> Error: There is no class named GSL-PERMUTATION. So I think we want to change... (defun make-symbol-from-strings (&rest strings) "Construct a symbol by concatenating words with hyphens." (intern (format nil "~{~:@(~a~)~^-~}" (mapcar #'string strings)))) ...to... (defun make-symbol-from-strings (&rest strings) "Construct a symbol by concatenating words with hyphens." (intern (format nil "~{~:@(~a~)~^-~}" (mapcar #'string strings)) :gsll)) With the change the test works (returns 2). What do you think? Zach

2 1

[Gsll-devel] GSL version being used
by Felip Alàez Nadal 02 Mar '08

02 Mar '08

Hello: Which gsl version are you using to develop gsll? Thank you. -- Felip Alàez Nadal

2 1

[Gsll-devel] Undelivered Mail Returned to Sender
by MAILER-DAEMON＠common-lisp.net 01 Mar '08

01 Mar '08

This is the Postfix program at host common-lisp.net. I'm sorry to have to inform you that your message could not be be delivered to one or more recipients. It's attached below. For further assistance, please send mail to <postmaster> If you do so, please include this problem report. You can delete your own text from the attached returned message. The Postfix program <gsll-ticket(a)common-lisp.net>: Command died with status 1: "/usr/lib/mailman/mail/mailman post GSLL-ticket". Command output: post script, list not found: GSLL-ticket

1 0