Hello all,
I wrote:
> I'm having an unexpected linking error with code that has previously
> built without problem against earlier GSL versions.
This turns out to be a g++/autotools issue rather than GSL -- sorry for
the noise.
FWIW, the workaround I have found is to add
CXXFLAGS="$CXXFLAGS -
I'm having an unexpected linking error with code that has previously
built without problem against earlier GSL versions.
I'm using Kubuntu 11.10 with GSL 1.15, newly installed. The errors
during build are as follows:
/bin/b
On 05/28/2010 08:16 PM, Peter Johansson wrote:
> You could do
>
> AC_CHECK_HEADER([gsl/gsl_version.h],
>[foo="#include "],
>[foo="#define INLINE_FUN extern inline"])
> AC_SUBST(foo)
> AC_CONFIG_FILES([config_public.h])
>
> in configure.ac
>
> and have a config_pub
On 05/28/2010 05:38 PM, Peter Johansson wrote:
> It make sense as long as HAVE_GSL_INLINE is defined/undefined
> accurately. If the header file is only used within your package there
> should be no problem. If you install the header file it is a bit
> trickier.
Exactly my situation ... :-)
> Then
On 05/28/2010 04:50 AM, Peter Johansson wrote:
> This is a pefect use case for autoconf. Rather than checking which GSL
> version your client is using, you can implement a compilation test in
> your configure checking if 'gsl/gsl_inline.h' exists. If it exists
> define HAVE_GSL_INLINE and modify yo
On 05/27/2010 07:42 PM, Rhys Ulerich wrote:
> As of this particular commit
> http://git.savannah.gnu.org/cgit/gsl.git/commit/?id=ffb8708d2cf636801de0b9f1f03c65a5ab0bcb13
> there's a GSL_MAJOR_VERSION and GSL_MINOR_VERSION macro defined within
> gsl_version.h. I know this is in 1.14. Offhand I can
Hello all,
A small problem I came up against while working on an extension I'm
building to GSL.
Later versions of GSL (1.13+ I think?) have the gsl_inline.h header,
which defines various preprocessor macros with respect to inline
functions -- in particular, INLINE_FUN. This allows code to be def
On 05/07/2010 09:22 PM, Thomas Weber wrote:
> mysql.h (and all of MySQL) is licensed under GPL2-only.
> GSL (at least the latest version) is licensed under GPL3-or-later.
Depends on what is being linked to -- isn't libmysql distributed under
LGPL? That might(*) make things easier.
Also, Sun (a
Hello all,
A couple of days ago I installed a 64-bit Linux distro instead of a
32-bit one, and discovered an interesting bug in some C++ code I'd written.
The bug comes down to this little bit of code:
unsigned long int M_ul = floor(log2(ULONG_MAX-1))-1;
unsigned long int M_rng = floor(l
Marco Tului wrote:
> I have read the GSL manual, and in my Makefile I wrote:
>
> :::
> # RANDOM_NUMBER_GENERATOR
> random_generator.o: random_generator.hpp random_generator.cpp
> g++ -c random_generator.cpp -lgsl -lgslcblas -lm
>
Marco Tului wrote:
> By the way, if I write instead by terminal simply "g++ -c
> random_generator.cpp" I don't receive errors: it seems to be fine. How's
> possible?
The point is that if you use g++ -c then you are compiling the code to
object form -- not linking it together to form an executable.
jan dithmer wrote:
> i have problems to install gsl. I want tu use a software
> that requires gsl to be installed before, but unfortunately I am
> not able to install it properly on linux (ubuntu). I downloaded and unzipped
> gsl-1.12 from your webpage, but if i try to configure the software prog
Will Farr wrote:
> There is a gsl_heapsort function (see section 11.1 of the manual)
> which allows the caller to pass an arbitrary comparison function.
I'll give that a go. Thanks for the suggestion! :-)
___
Help-gsl mailing list
Help-gsl@gnu.org
htt
Hello,
A few queries about the sorting functions in GSL.
I have a few different arrays which I need to sort, depending on what
they are, into either ascending or descending order.
Later on in the program, I will need to perform index-sorts on the same
arrays -- again, respectively in either asce
Hello,
A few queries about the sorting functions in GSL.
I have a few different arrays which I need to sort, depending on what
they are, into either ascending or descending order.
Later on in the program, I will need to perform index-sorts on the same
arrays -- again, respectively in either asce
Fantastic, thank you to all who replied!
John D Lamb wrote:
>> Secondly, do I need to reference -lgsl -lgslcblas when compiling my
>> library, or can I leave this in the hands of the person compiling a
>> program that makes use of it? I have created an appropriate .pc file
>> for pkg-config.
>
>
I'm building a small simulation library that makes use of some GSL
functions. However, I'm a novice with autotools and I'm having some
trouble working out how to check that GSL is installed and its header
files are in place.
What lines do I need to add to configure.ac to make sure that GSL is
pre
Jordi Gutierrez Hermoso wrote:
> I don't understand one thing. Why are you compiling, instead of using
> apt-get?
The Ubuntu apt repositories only contain 1.8, I wanted to use 1.9. I'm
not aware of other repositories for GSL ... ?
In any case having started I wanted to understand why what I was
Patrick Alken wrote:
> Put /usr/local/lib in /etc/ld.so.conf and run "ldconfig". Then type
> "ldconfig -v | grep gsl" and make sure its got an entry for libgsl -
> if it does, gcc ... -lgsl -lgslcblas should work fine from any shell.
Thank you very much---exactly what I was looking for!
A big tha
John Pye wrote:
> That works, but it's very naughty. You should only install to /usr
> indirectly, eg by creating an packaged installation such as an RPM or
> DEB file.
Yes, true. There's a nice explanation of why it's a bad idea here:
http://hivelogic.com/narrative/articles/using_usr_local
> A
John Pye wrote:
> Did you try typing
>
> export LD_LIBRARY_PATH=/usr/local/lib:/usr/lib
>
> then running again?
Damn! I knew there was something I'd forgotten.
I tried an alternative route, also with success: when preparing for the
install, typing,
./configure --prefix=/usr
make
sudo make ins
I've installed gsl 1.9 from source on Ubuntu 7.04 and am having trouble
linking programs with the library.
My Makefile includes the use of C flags -I/usr/local/include and
-L/usr/local/lib along with the library calls -lgsl, -lgslcblas and -lm.
However, when I attempt to run the program I still g
Joseph Wakeling wrote:
> Of course, in many cases it's fine to do this in a serial program and I
> often do. I was wondering about the scenario outlined in the discussion
> below, of using one random number generator (or an incremented sequence)
> to generate seeds for the
ould it be worth
doing this in a _serial_ program, generating different seeds for the
individual runs? Or is it better to stick with one rng for a serial
program?
Thanks in advance for any suggestions,
-- Joe
> Martin Jansche wrote:
>> On 11/12/06, Joseph Wakeling <[EMAIL PROTECT
Rodney Sparapani wrote:
> Here's my 2 cents. Independent parallel streams is a tricky problem.
> I'm not an expert in this area so rather than my advice,
> take a look at JE Gentle's Random Number Generation and Monte Carlo
> Methods, 2nd Ed. He mentions several possible solutions and a free
> so
Whoops, that earlier email was meant to go to the list as well. Let's
get it back on track.
Martin Jansche wrote:
> On 11/12/06, Joseph Wakeling <[EMAIL PROTECTED]> wrote:
>
>> Can you advise on appropriately seeding the rng's for different threads,
>> so a
Hello all,
I'm curious about the random number generators and how they can be
effectively used in different situations. I'm not particularly
experienced with large-scale or parallel programming, so bear with me if
what follows seems stupid.
My "typical" mode of use for the rng would be to create
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I'm writing some code right now in which a parameter is meant to
fluctuate as a random walk. I'm using ran2 as the rng.
Suppose I write a simple piece of code of the form,
{
int i; double walk, increment;
/* initialise variables */
for
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
John D Lamb wrote:
> There is nothing to stop you using two random number generators, though
> you probably only need one. If you use two random number generators with
> the same seed and call them equally often then the results will be
> strongly corr
Hello all,
As I mentioned in earlier emails I've been playing around a bit with the
random number generators in the GSL.
I'm wondering if someone can advise on appropriate ways to generate
uniformly-distributed ints in the range [0,n-1]. My traditional way has
always been to take a function
Apologies to Martin who will be getting this mail twice. The first time
I replied, I accidentally sent it to him only...
Martin Jansche wrote:
Note that I deliberately *don't* want to just compile to object form the
whole GSL.
Why not?
Well, we could consider that to be merely part of my
Hello all,
I recently downloaded the GSL but have been having trouble with the
#include options in these files.
Header include commands for GSL invariably take the form of, say,
#include
but the header file concerned is usually not actually located in the
gsl/ directory, so the compiler fa
32 matches
Mail list logo
