sts.
4. The header files used in the build are then copied into the output
directory.
5. The built library files (mpir.dll, mpir.exp, mpir.lib and mpir.pdb
for a DLL, mpir.lib and mpir.pdb for a static library) are then
copied into the output directory.
ACKNOWLEDGEMENTS
On Fri, Mar 28, 2014 at 2:25 PM, Brian Gladman wrote:
> I'll ask if we have volunteers who are willing to try 2010 builds. If we
> don't get any, we should remove them.
I will double check but I think I still have VS2010 Professional on
one of my machines that I could test.
Jeff.
--
You recei
On Fri, Mar 28, 2014 at 1:54 PM, Brian Gladman wrote:
> Visual C/C++ EXpress 2013 only - since this is free and has important
> improvements over earlier versions in terms of C/C++ language support,
> there is little reason to stay with earlier versions.
I'm ok with that, it makes my life easier
I can probably help you with that. I have a box with nothing on it so
I will install Express. Do you want me to try multiple versions of
Express or just the latest 2013?
Jeff.
On Fri, Mar 28, 2014 at 1:21 PM, Brian Gladman wrote:
> As we are approaching the next release of MPIR, I now have ful
On Fri, Nov 16, 2012 at 1:24 PM, Bill Hart wrote:
> What happens if you build a small program using a function from math.h and
> compile it with gcc with the -lm option? Perhaps gcc doesn't use lib math.
No complaints, it compiles without errors/warnings, it works fine as expected.
> I'll try t
On Thu, Nov 8, 2012 at 6:09 PM, Bill Hart wrote:
> After a very long alpha/beta release cycle, MPIR 2.6.0 has finally
> been released!
Hi Bill, I just noticed something with MPIR's configure script that
seems to be behaving differently than GMP 5.x. On the Linux system
there is both GCC and ICC
On Tue, Oct 23, 2012 at 5:47 PM, Bill Hart wrote:
> Thanks Jeff. I didn't have the Nehalem values yet, so I've just put
> those into my local copy. Will commit later.
Don't forget that is for ICC so I'm not sure if GCC values will be
different or if that matters.
Jeff.
--
You received this mes
On Tue, Oct 23, 2012 at 3:25 PM, Bill Hart wrote:
> Is there any chance you could send me the output of make tune on
> whichever machines it actually works on (from memory you may have to
> go into the tune directory to run it). I'll also need the output of
> config.guess in each case.
Linux 2.6
On Tue, Oct 23, 2012 at 3:25 PM, Bill Hart wrote:
> Is there any chance you could send me the output of make tune on
> whichever machines it actually works on (from memory you may have to
> go into the tune directory to run it). I'll also need the output of
> config.guess in each case.
Linux 2.6
On Tue, Oct 23, 2012 at 1:25 PM, Bill Hart wrote:
> OK, alpha6 is now up. I didn't touch the choke test, as I think it
> shouldn't be causing the problem. But now configure should run the
> right gcc test.
Alpha 6 is now configuring properly with ICC. Here are some test
results on various Linux
On Tue, Oct 23, 2012 at 11:58 AM, Bill Hart wrote:
> * (hopefully) fixes an issue with the Intel C Compiler noted by Jeff Gilchrist
It still blows up. Looking at config.log I see:
cclist=gcc icc cc
configure:4996: icc -c conftest.c >&5
conftest.c(2): error: identifier "chok
On Tue, Oct 23, 2012 at 9:54 AM, Bill Hart wrote:
> I'm confused. ICC defines _GNUC_?
Apparently ICC tries to be "GCC compatible"
> Is there anything ICC defines that gcc does not?
I can try __INTEL_COMPILER
http://stackoverflow.com/questions/5736790/detect-icc-vs-gcc-at-compile-time
Yes, that
On Tue, Oct 23, 2012 at 9:00 AM, Bill Hart wrote:
> Could you try compiling it with your icc with the flags configure is
> using (see your configure trace below) and tell us what happens when
> you run it?
First it wouldn't compile because it didn't know what
"__builtin_malloc" was, so I changed
On Sun, Oct 21, 2012 at 1:15 PM, Bill Hart wrote:
> I have just uploaded alpha4 to our website http://mpir.org/
>
> This will hopefully be the last alpha. It merely adds a patch to speed
> up the FFT slightly in certain instances.
I found some issues with ICC.
If I use mpir 2.5.1 with just a si
On Mon, May 21, 2012 at 1:14 PM, Bill Hart wrote:
> At last there is a serious alternative for users of Git on Windows:
> https://github.com/blog/1127-github-for-windows
Have you looked at Git Extensions? I find it a very useful GUI for Windows:
http://code.google.com/p/gitextensions/
Jeff.
--
On Sun, Jan 15, 2012 at 4:35 AM, GeekDude wrote:
> I have compiled the lib_mpir_core2 version and looking in the Release
> folder, I only see object files.
>
> So now what do I do since the documentation does not state further on
> what I should do to use MPIR.
As the documentation states, "the
On a positive note, I have been doing lots of computations lately
related to Wilson primes. Robert Gerbicz has written some code
(https://sites.google.com/site/robertgerbicz/wilson) that uses
MPIR/GMP and performs lots of pow, mod, mul operations. The MPIR
version is about twice as fast as the GM
On Wed, Sep 28, 2011 at 3:46 PM, Jason wrote:
> I found a performance loss of mpir 2.4.0 compared to gmp 4.3.2. when running
> gmp bench 0.2 gcd examples on a Core 2 Duo T6400, 2Ghz, 4GB Ram Notebook with
> Ubuntu 11.04, 2.6.38-11-generic-pae i686, gcc 4.5.2.
I'm not sure how GCD code compares i
On Wed, Jul 13, 2011 at 9:49 AM, Cactus wrote:
> I now have the Python build configuration working for Visual Studio 2010.
> It scans the assembler directory tree to identify the possible builds,
> allows the user to select one and then constructs the project files needed
> for a Visual Studio 201
Here is report by a GMP-ECM user that uses MPIR about a strange effect
when trying to compile on mingw64 without giving it a CPU hint. Not
sure if this is a mingw thing or something that can be fixed in the
config script.
The original post:
http://www.mersenneforum.org/showpost.php?p=264649&postc
On Sun, Jun 19, 2011 at 5:58 PM, Jason wrote:
> cheers , all testing is welcome at any time , out of curiosity did you use the
> GUI build , the MSBUILD script or the CL script ?
That time I did the GUI build as I was pressed for time to use it for
something else. I should try the other two as
On Tue, Jun 14, 2011 at 1:48 AM, Jason wrote:
> MPIR 2.4.0 was released on 14th June 2011.
I know this is a little late but 2.4.0 tests fine in VS2010 on Windows
7 Ultimate 64bit for both the 32bit and 64bit builds. I was away at a
conference in Italy and just catching up on things now.
Jeff.
On Wed, Apr 13, 2011 at 6:26 AM, elect wrote:
> I would need to use mpir with one program I am writing on VS10 in
> 64bit..
>
> Based on what I found, I should need mpir.h, mpir.lib and mpir.pdb
You probably just need mpir.h and mpir.lib.
> Well, the problem is that i cant find the last two of
On Tue, Apr 5, 2011 at 12:14 PM, Jason wrote:
> * Bug fixed where compilers without GAS inline assembler on Unix like systems
> had a bug on AMD chips
It compiles and make check works fine on my AMD system now that had problems.
Jeff.
--
You received this message because you are subscribed to
On Thu, Mar 17, 2011 at 8:32 PM, Jason wrote:
> Yep , it was just a hunch that was wrong , could you try putting some
> printf("here\n");fflush(0);
> between he tests in t-constants.c , so we can work out which one is the
> problem.
It stops inside this section:
#ifdef PP_INVERTED
{
mp_li
On Thu, Mar 17, 2011 at 7:23 PM, Jason wrote:
> t-constants appears to check that %l %ll do what they are meant to , but
> perhaps icc does do it ?
>
> Thanks , bit of weird one , can you try the mpn and mpz tests , just cd into
> tests/mpn and make check , to see if we any other more familiar fa
We just got a new 24 core AMD 6174 system.
k102-unknown-linux-gnu
Using ICC v11.0 I'm getting a failure in make check after just doing
./configure without any other parameters.
PASS: t-bswap
/bin/sh: line 1: 13906 Floating point exception"$tst" > t-constants.log-t 2>&1
FAIL: t-constants
PASS: t
On Thu, Mar 3, 2011 at 5:04 AM, Jason wrote:
> OK , but how about tuning and timings?
> Does anyone here have any experience in this ? , or any recommendations
I don't think I have ever actually tested things in virtual machines
such as tuning/timings. Since you still have your 32bit "native"
i
On Thu, Mar 3, 2011 at 11:23 AM, Jason wrote:
> if you do
> gcc -v
> what is the exact output , perhaps when gcc was built some different option
> was
> used , we have quite a few gcc-3.4's on various machines and they all work
> fine
> , although some have broken c++
Reading specs from /usr/l
On Thu, Mar 3, 2011 at 11:02 AM, Jason wrote:
> As we don't seem to have any regressions from mpir-2.2 , I think we can
> release 2.3 this evening.
I'm running the test script now that I know it exists (thanks). Is
there a known issue with fat builds with GCC 3.4.x on K8
(k8-unknown-linux-gnu)?
On Thu, Mar 3, 2011 at 4:30 AM, Jason wrote:
> it fails because sizeof(long)=8 still
>
> Did this work before on previous MPIR's ?
> You could try CFLAGS=-m32 +whatever
No idea if this failed before since ICC is a fairly new to that system
so I had never tried it before. Those systems seem to b
These all passed with C++ enabled:
- 64bit Core2 MSVC 2010 build on Windows 7 Ultimate SP1 64bit
- 32bit P4 MSVC 2010 build on Windows 7 Ultimate SP1 64bit
- 64bit k8-unknown-linux-gnu GCC version 3.4.6
- 32bit (ABI=32) k8-unknown-linux-gnu GCC version 3.4.6
- 64bit penryn-unknown-linux-gnu ICC
On Fri, Nov 19, 2010 at 3:24 PM, Jason wrote:
> In regard to Mingw* , well , what can I say , for mingw32 I have to install it
> twice to get it to work ,apparently once just isn't good enough! I'm not
> installing the latest though , but then again I didn't have this problem a
> year ago. For mi
Finished testing with a Core2 based Windows 7 Ultimate 64bit machine
and Visual Studio 2010. Testing the core2 64bit project with C++
project as well.
Compile happened fine. But when running the python test script using
python 2.6.2 I get:
Traceback (most recent call last):
File "D:\Data\Jeff_
On Fri, Nov 19, 2010 at 1:47 PM, Jason wrote:
> Cheers , My best guess is that this is an autotool problem. I just tried using
> the latest autotools , but it is a lot stricter than our current version , and
> I would have to do some work on our configure.in and makefile.am's before I
> can
> tr
On Fri, Nov 19, 2010 at 11:56 AM, Jason wrote:
> To get just a static build you have to --disable-shared as well and for just a
> shared build --disable-static as linux defaults to building both. Note for
> cygwin/mingw* just putting --disable-static will fail as it only builds static
> by defaul
On Fri, Nov 19, 2010 at 3:57 AM, Jason wrote:
> Thanks , I have no idea what it is , the only thing I can suggest for the
> moment is to try a build for static only , and then in a clean dir a shared
> build only. I'm not sure what it will tell us !!!
Using just --enable-static works fine compil
On Thu, Nov 18, 2010 at 5:29 PM, Jason wrote:
> I'm assuming case 3 is on the same machine as the failed case ?
Yes, they are the same machine.
> Did the previous mpir v2.1 fail with the same options ?
I just checked an v2.1.3 also fails with the same options enabled
> Can you try it without
On Thu, Nov 18, 2010 at 3:31 PM, Bill Hart wrote:
> $ diff cpuid.old cpuid.c
> 124c124,125
> < break;
> ---
>> if (model == 9) { CPUIS(k102);break;}
>> break;
>
> Otherwise tests pass with only the following warnings when doing make check:
I just tried from a new tar bal
On Thu, Nov 18, 2010 at 3:20 PM, Bill Hart wrote:
> Just checking that you did make distclean before the test that failed.
Oh crap, maybe not. Let me start from scratch again.
Jeff.
--
You received this message because you are subscribed to the Google Groups
"mpir-devel" group.
To post to th
Some rc2 tests. Here are the ones that passed:
1. Intel Xeon X5550 running Linux 64bit
./config.guess = nehalem-unknown-linux-gnu
ICC v11.0 with --enable-cxx = All tests passed
ICC v11.0 with --enable-cxx --enable-assert --enable-fat = All tests passed
GCC v4.1.2 with --enable-cxx = All tests pas
On Wed, May 26, 2010 at 3:00 AM, Minh Nguyen wrote:
> MPIR 2.1.0-rc1 was released on 25th May 2010.
Mac OSX test results:
- OSX Leopard - gcc version 4.0.1 (Apple Inc. build 5493) - 64bit Core2
All tests passed
- OSX Leopard - gcc version 4.0.1 (Apple Inc. build 5493) - 32bit Core2
All tests p
On Wed, May 26, 2010 at 3:00 AM, Minh Nguyen wrote:
> MPIR 2.1.0-rc1 was released on 25th May 2010.
Here are some testing results with: configure -enable-cxx; make; make check.
- Linux 64bit - Core2 - icc (ICC) 11.0 20090318
All tests passed
- Linux 64bit - Core2 - gcc version 4.1.2 20080704
On Fri, Apr 23, 2010 at 12:17 PM, Bill Hart wrote:
> I've just issued release candidate 4. This fixes a minor issue with
> versioning for Windows, discovered by Brian Gladman, and fixes a major
> segfault with exact division reported by Yves Gallot.
RC4 all tests ok with VS2008 for both 32bit and
On Thu, Jan 28, 2010 at 12:51 PM, Bill Hart wrote:
> Oh thanks Jeff, that's an error in the matrix. I'll rectify it now. I
> did see your reports, thanks.
Same thing with AMD K8 on Linux, my Opteron reports are K8.
Jeff.
--
You received this message because you are subscribed to the Google Gro
On Thu, Jan 28, 2010 at 12:37 PM, Bill Hart wrote:
> We are missing numerous test reports this time around as Jason Moxham
> is on holidays from development and can't test for us. If anyone can
> fill any of the gaps in our testing matrix on our website
> http://www.mpir.org/ it would be most app
On Sat, Jan 16, 2010 at 9:37 PM, Bill Hart wrote:
> I've put up rc7. There's no need to test this if you already tested
> rc6. The only differences should be in paperwork and the fact that the
> reported warnings about unrecognised options should be gone (I patched
> the yasm configure so it would
On Thu, Jan 14, 2010 at 2:50 PM, Bill Hart wrote:
> You have to do make distclean before going from a 64 bit build to a 32
> bit build.
>
> Let me know if this doesn't fix it for you.
That was my problem, working fine now.
Jeff.
--
You received this message because you are subscribed to the Goo
On Thu, Jan 14, 2010 at 2:40 PM, Bill Hart wrote:
>> If we are on a 64bit OS and want to build a 32bit mpir library, how
>> would we do this use: ./configure ABI=32 ?
>
> Yes, I believe this is the correct method. This will force it to check
> only 32 bit compiler flags.
I just wanted to double
On Thu, Jan 14, 2010 at 11:16 AM, Bill Hart wrote:
> I have fixed the issue with t-invert being missing from the tarball,
> and so rc6 is now available on our site for testing.
The following systems have passed ./configure --enable-cxx; make; make check
- Linux 64bit Intel(R) Xeon(R) CPU E5440 @
On Wed, Jan 13, 2010 at 7:58 PM, Bill Hart wrote:
> Did you build with --enable-cxx?
I realized after I hit SEND that 0 tests was probably the correct
value. I did not use that switch so I'm ok then.
Jeff.
--
You received this message because you are subscribed to the Google Groups
"mpir-deve
On Wed, Jan 13, 2010 at 2:43 PM, Bill Hart wrote:
> believe the @MKDIR_P@ issue is also gone, so make install should also
> work now.
I'm getting 0 tests passed for the cxx tests on all my Linux systems:
make[4]: Leaving directory `/work/jeffg/mpir-1.3.0/tests/misc'
make[3]: Leaving directory `
On Sat, Jan 9, 2010 at 6:35 AM, Cactus wrote:
> Now that GMP 5 has been released, we will have to make a decision on
> whether to make changes to MPIR or ive up GMP compatibility. An
> outline of the features in GMP 5 is given here:
I didn't even notice that it has been released. Has anyone run
On Wed, Dec 9, 2009 at 2:20 PM, Bill Hart wrote:
> If it does cause problems for anyone the way it is now implemented,
> we'll increase the number of Miller-Rabin tests performed for
> mpz_next_prime. Now is the time to speak up if it is going to be an
> issue for you!
It was set to 1 before but
On Wed, Dec 9, 2009 at 1:54 PM, Bill Hart wrote:
> All that remains now in order to produce a new release candidate for
> testing is the following:
What about the new prime() functions, has that been settled yet? I
haven't seen Jason make any comments on that.
Jeff.
--
You received this mess
On Mon, Dec 7, 2009 at 8:03 AM, Børge Strand-Bergesen
wrote:
> I just installed Visual C++ Express 2008 and yasm according to the
> instructions in readme.txt. When I try to open mpir.sln in VC++Ex, a
> dialog box says "The projects sonsists entirely of configurations that
> require support for p
On Fri, Dec 4, 2009 at 2:34 PM, Bill Hart wrote:
> No, I meant, was trial factoring done before testing for base 2
> pseudoprimality. I guess not. I guess that you have a list of *all*
> base 2 pseudoprimes up to 2^64 and worked with that, i.e. base 2
> pseudoprime test followed by BPSW is uncond
On Fri, Dec 4, 2009 at 1:56 PM, Bill Hart wrote:
> I'm not sure I understand exactly what check you did up to 2^64. Did
> you take all the (composite) base 2 strong pseudoprimes and then test
> that BPSW eliminates them all? If you consider that to be a reliable
> result, I could use this in FLIN
On Fri, Dec 4, 2009 at 12:41 PM, Bill Hart wrote:
> Do we still have an optimised version of Nicely's code? I recall that
> for the benchmark we were uninterested in all the optimisations, as
> they weren't profiling MPIR itself, but Nicely's code. However I did
> not keep track of whether we jus
On Fri, Dec 4, 2009 at 12:29 PM, Bill Hart wrote:
> Whoah, it just does trial division up to 1000 then uses a *single*
> *random* base for a strong pseudoprime test!
>
> I have a definite problem with this. What is the probability of
> returning a composite? Assuming the factors are bigger than 10
On Thu, Dec 3, 2009 at 5:18 PM, Borge wrote:
> I undestand that MPIR is supposed to work for Windows. Is Visual C#
> Express 2008 an option? If it is, could you please point me towards an
> installation procedure?
You will need Visual C++ Express 2008 to build a .lib or .dll of MPIR.
If you loo
On Wed, Dec 2, 2009 at 7:57 AM, Cactus wrote:
> In order to reduce the complexity of the Windows build projects, I am
> now inclined to split the mpir Windows build solution into two
> separate build solutions, the first for 32-bit builds and the second
> for 64-bit builds. I would be interested
On Tue, Dec 1, 2009 at 12:07 PM, Bill Hart wrote:
> mpz_nextprime seems to report lots of composites as primes on 32 bit
> machines, including some even numbers! Is this the same issue as we
> had on 64 bit machines?
Is that the old built-in function or the new one Jason developed?
Jeff.
--
Yo
On Sun, Nov 29, 2009 at 8:24 AM, Jeff Gilchrist
wrote:
> If I choose Build Solution again, those 7 projects build properly and
> everything compiles. Are those being compiled before the test-lib.lib
> file is created perhaps?
As long as I build the add-test-lib project first manually
On Sat, Nov 28, 2009 at 4:21 PM, Cactus wrote:
> I have tested this faairly extensively and I believe it is all OK but
> it is a big configuration change so I would be grateful if anyone who
> is able to build MPIR and the MPIR tests on a Windows system could
> check out and test the revised Windo
On Sat, Nov 28, 2009 at 4:21 PM, Cactus wrote:
> I have tested this faairly extensively and I believe it is all OK but
> it is a big configuration change so I would be grateful if anyone who
> is able to build MPIR and the MPIR tests on a Windows system could
> check out and test the revised Wind
On Mon, Nov 16, 2009 at 11:12 AM, Bill Hart wrote:
> Delivery of MPIR 1.3.0 is still at least two weeks out as we still
> haven't had anyone write a new dc_divappr_q_n.c file. At this stage we
> are hoping Jason Moxham might find the time in the next two weeks. We
> will need to go through an e
On Mon, Oct 19, 2009 at 4:47 AM, Bill Hart wrote:
> I've just put up MPIR 1.3 release candidate 1. We are still waiting on some
> final testing from Jeff Gilchrist, which will probably happen tomorrow. If
> everything is fine, the release candidate will become the final release.
On Sat, Oct 17, 2009 at 6:15 AM, Bill Hart wrote:
> I have put MPIR 1.3 (alpha 3) up on our website:
>
> This fixes some further assembly issues Jeff Gilchrist discovered
> during testing.
Third alpha's a charm. ;-) So the Linux opteron system works fine
now, compiles and ch
On Thu, Oct 15, 2009 at 7:20 PM, Bill Hart wrote:
> I've uploaded MPIR 1.3 alpha 2 at http://www.mpir.org/
>
> This fixes some assembly issues discovered by Jeff Gilchrist during testing.
So the good news is that I just tested 1.3alpha2 on Windows7 Pro 64bit
with VS2008 64bit
On Thu, Oct 15, 2009 at 1:13 AM, Bill Hart wrote:
> I've just put up MPIR 1.3 (alpha) for testing.
I should be able to do the Windows 64bit tests tonight.
Tests that have passed:
- Core2, Vista 32bit, cygwin gcc 3.4.4: ./configure; make; make check
- Core2, Vista 32bit, VS2008 Express SP1: b
On Thu, Oct 15, 2009 at 1:13 AM, Bill Hart wrote:
> I've just put up MPIR 1.3 (alpha) for testing.
Maybe I missed something obvious but how do I enable multi-threaded
support? I don't see anything in the docs on what you use to enable
the new MT code. If you just use ./configure, make, make c
On Thu, Oct 15, 2009 at 1:13 AM, Bill Hart wrote:
> We especially would appreciate owners of Apple machines and anything
> slightly exotic, to test MPIR.
I'm in the process of testing Windows builds, and on a MacBook Pro,
./configure passes but I get an error during make:
../mpn/m4-ccas --m4=m
On Wed, Oct 7, 2009 at 6:04 PM, William Stein wrote:
> VirtualBox is vastly better, so after this change I'll be able to
> configure these virtual machines with more cores, RAM, and managing
> them will be much easier.
That is good to know. I have heard about VirtualBox but haven't had
the cha
On Fri, Oct 2, 2009 at 4:04 PM, Cactus wrote:
> The changes to the Windows builds have however been significant so I
> would appreciate assistance from Jeff Gilchrist and any other Windows
> users in the testing of the SVN trunk version. The OpenMP stuff
> needs building an
On Sun, Sep 13, 2009 at 10:25 PM, jason wrote:
> Hi , did you mean that results should be repeatable across different
> arch's ie 64/32bit ? or just repeatable on a set arch?
A set architecture would be fine, so people could say, using arch X,
with MPIR vY.Z then using this function with this s
On Thu, Sep 3, 2009 at 10:51 PM, Jason Moxham wrote:
> Or getting rid of the mpz_next_prime_thing all-together , cant say I like it
> much. What do people use it for ?
I actually use it and find it very useful. If I'm doing some kind of
testing or other calculations on increasing primes it is v
On Sun, May 17, 2009 at 3:42 PM, Jason
Martin wrote:
> Are there any OpenCL compilers available yet?
It looks like AMD is releasing an SDK to run OpenCL code on x86
processors, so you can debug things without a compliant GPU available.
Might be interesting to some people here:
http://arstechnic
On Sun, Jul 26, 2009 at 4:00 AM, Chris Saunders wrote:
> I believe I did the full install of VS2008. I ran the setup again in
> "Maintenance Mode" and all of the check boxes are checked. Service Pack 1
> has also been installed. My operating system is Vista Ultimate 64-bit. Any
> other ideas.
On Mon, Jul 6, 2009 at 10:36 PM, Peter_APIIT wrote:
> What type of library i build static or DLL ?
That is up to you, the easiest is probably build the static version.
> Which file to include in linker additional dependencies ?
If you are building a 32bit library (using the lib_mpir_p4 or
simi
On Thu, Jul 2, 2009 at 12:40 PM, Cactus wrote:
> MSVC has had OpenMP support for some time now.
>
> Visual Studio 2008 supports OpenMP 2.0; Visual Studio 2010 supports
> OpenMP 3.0
Nice! I learned two new things today, built-in support for OpenMP in
both gcc and MSVC. If only that was availabl
On Thu, Jul 2, 2009 at 10:49 AM, Bill Hart wrote:
> In particular gcc 4.2.4 and following have OpenMP support built in.
> That means all the compilers which are derivatives of gcc will fairly
> shortly have OpenMP support if they don't already. Certainly the Intel
> compiler supports OpenMP and h
On Thu, Jul 2, 2009 at 10:22 AM, Bill Hart wrote:
> That is an interesting question. pthreads is certainly available on
> much older compilers. But the issue is, it is very hard to code.
I guess it is the only thing I know, so maybe I should move to OpenMP. ;-)
> I'd be interested in supportin
On Wed, Jul 1, 2009 at 11:48 AM, Bill Hart wrote:
> AN FAQ **
The FAQ is a great idea and you are off to a good start.
> ** FORTNIGHTLY MPIR DIGEST
I think this is a really good idea, makes it easy for people not doing
active development to
On Sun, May 17, 2009 at 3:42 PM, Jason Martin
wrote:
> Are there any OpenCL compilers available yet?
Looks like NVIDIA is releasing a driver pack with an OpenCL compiler:
http://www.techreport.com/discussions.x/17088
Jeff.
--~--~-~--~~~---~--~~
You received thi
On Thu, Jun 11, 2009 at 1:00 PM, gerrob wrote:
> I've uploaded a new code (wagstaff2.c) in google groups to test
> Wagstaff numbers by Anton Vrba's conjecture. It's faster than
> wagstaff.c from Jeff Gilchrist because I'm using bitshift+bitmask
> instead of the
On Thu, Jun 11, 2009 at 11:50 AM, Bill Hart wrote:
> Oh wait, is this on Windows? If so, then drat!
>
> However Jason Moxham is soon going to be helping out with optimisation
> on Windows which should pick the speed up. When the code gets
> converted to the Windows calling conventions, sometimes
On Thu, Jun 11, 2009 at 11:40 AM, Bill Hart wrote:
> These timings look suspiciously close, but I don't think they should
> be. Perhaps we are wrong and the time complexity of this benchmark
> doesn't depend much on MPIR at all, but other things? Or is it
> possible we've picked up a system wide
On Sat, Jun 6, 2009 at 8:57 PM, Bill Hart wrote:
> The MPIR team is proud to release MPIR 1.2, which is available on our
> website:
>
> http://www.mpir.org/
> The new features of MPIR 1.2 include:
>
> * Further improvements to assembly support for K8/K10/Core2/Pentium 4
> including improvements t
On Sat, Jun 6, 2009 at 5:41 PM, Cactus wrote:
> It may be OK since it seems that this stuff is not essential - it is
> pretty close to building in Windows x64 now that I have changed some
> library subroutine names.
I forgot about the assembler. If you can get it working in Windows
x64 that wou
On Sat, Jun 6, 2009 at 3:45 PM, Bill Hart wrote:
> Thanks, both your code and the code from that site sound good. We
> should write to the author and see whether they would be happy to
> contribute it to the project.
Should I be sending my code to Brian to include in his new benchmark?
Do you w
On Sat, Jun 6, 2009 at 10:11 AM, Jason Moxham wrote:
> Jeff Gilchrist has a real world benchmark , ie something thats run for
> purposes other than generating benchmarks. I would like to see a few of these
> in our benchmark suite.
I would be happy to donate my code for that purpose.
Linux Testing
==
Opteron K8 gcc 3 with no python = make check ok
Core2 Penryn gcc 4 = make check ok
Core2 Penryn gcc 4 (with 32bit pathCC setup) = make check ok
Cygwin32 Testing
=
Core2 Conroe = make check ok
PentiumD = make check ok
I will do the MSVC testing once I get the
On Wed, May 27, 2009 at 10:25 AM, Cactus wrote:
> We are in the final stages of preparing for the MPIR 1.2 release and
> we need a volunteer to help in producing FFT tuning values for the
> Core2 processor.
I can do this if someone hasn't already. I was away at a conference
for the past week a
On Thu, May 21, 2009 at 2:40 AM, Cactus wrote:
> So the only way to easily build native 64-bit GMP systems on WIndows
> is to use the Visual Studio build system.
>
> I don't build with VC++ Express but Jeff Gilchrist does and he tells
> me that it works provided the lates
On Sun, May 17, 2009 at 3:42 PM, Jason Martin
wrote:
> Are there any OpenCL compilers available yet?
Since when has something a trivial as a lack of compiler stopped the
amazingly talented MPIR crew. ;-) I guess I should have said, it
would be nice for MPIR to support OpenCL when it becomes m
On Sun, May 17, 2009 at 3:05 PM, Bill Hart wrote:
> We intend to discuss:
>
> * CUDA
> * Assembly language for NVIDIA cards
Everything seems to be CUDA and NVIDIA these days. Is there no love
for the also powerful ATI cards? It would be really nice if MPIR
supports OpenCL so that any GPGPU ca
Looks like Google Code has decided to adopt mercurial along side
subversion and opted not to use git.
News article:
http://arstechnica.com/open-source/news/2009/04/google-code-adds-mercurial-version-control-system.ars
Google's detailed analysis of Mercurial vs Git:
http://code.google.com/p/suppo
On Fri, Apr 24, 2009 at 2:04 PM, Bill Hart wrote:
> As no further build issues have been encountered, even on the new test
> platforms, MPIR 1.1.1 has been released. It is available on the main
> project website:
Not that this is a big deal, but in a lot of the config. files such as
mpir-1.1.1/
On Thu, Apr 23, 2009 at 12:08 PM, Bill Hart wrote:
> This fixes a build issue on very recent apple machines reported by
> Marshall Hampton.
So I had temporary access to a brand new MacBook Pro and tried RC4, it
detected it as penryn-apple-darwin9.6.2 and configured, make, and make
check all tes
1 - 100 of 205 matches
Mail list logo
