computing exp(1.0) on SPARC.
in [Solaris]
|
From: David Kirkby on 3 Jan 2010 01:17 On 1 Jan, 22:26, Nemo ad Nusquam <inva...(a)invalid.invalid> wrote: > Dave: > > David Kirkby wrote (in part): > > > Building Sage on Solaris is not exactly trivial, but it will build. > > However, it only currently builds as a 32-bit application on Solaris > > 10, and even then, some things do not work right. > > Good work! It would be very nice to have Sage running on more than > Intel boxes. Thank you. It's only recently got it there. The is a lot of code in Sage - the compressed source code is around 260 MB. The university of Washington was donated a T5240 by Sun, but the architecture of that machine has not really been ideal for developing on. I luckily had my own Blade 2000, but Sage took a day or more to build on that. These was until recently a very annoying bug, which meant that Sage would build if Sun Studio was not installed, but would not build if Sun Studio was installed. Finally, after many months, the cause of it has been identified. Of couse, we could have hacked the code to get it to work, or as root you could move Sun Studion to a non-standard place, but these were all hacks. I'm not sure if an HP-UX port will ever happen, but certainly some testing I've done on HP-UX has identified problems that were less apparent on other platforms, but still existed. I believe if there was someone keen on using Sage on HP-UX, a port would be quite easy, as a lot of the GNUisms have gone. (I might try to scronge some HP-UX compiler licences from HP.) Dave
From: David Kirkby on 3 Jan 2010 01:28 On 2 Jan, 07:51, "Uros Nedic" <ur...(a)beotel.rs> wrote: > On OpenSolaris for SPARC it works fine. I have even older version of > SunStudio > and here is the output: > > urosn(a)urosn:~$ uname -a > SunOS urosn 5.11 snv_105 sun4v sparc SUNW,SPARC-Enterprise-T5120 Solaris > urosn(a)urosn:~$ cat exp.c > #include <math.h> > #include <stdio.h> > int main() > { > printf("%.16lf\n",exp(1.0)); > > } > > urosn(a)urosn:~$ cc -lm exp.c > urosn(a)urosn:~$ cc -V > cc: Sun C 5.9 SunOS_sparc Patch 124867-11 2009/04/30 > usage: cc [ options] files. Use 'cc -flags' for details > urosn(a)urosn:~$ ./a.out > 2.7182818284590451 > urosn(a)urosn:~$ > > It was error in SunStudio for Sparc at Solaris 10. In OpenSolaris it has > been fixed. > > Regards, > Uros Nedic Are you saying the problem still exists on Solaris 10? If so, will the fix be backported to Solaris 10? Are you aware of any patch for Solaris 10? GCC has the same issue, but unfortunatley Sage will not yet build with Sun Studio. Does this replace the maths library, or is it just Sun Studio? I know gcc has the exact same issue, though gcc will inline the code if its very obvious exp() will be called with an argument of 1.0, but fails if that is not known in advance. It may need a more complex program than you posted to see this error.
From: Chris Ridd on 3 Jan 2010 02:25 On 2010-01-03 06:17:21 +0000, David Kirkby said: > lot of the GNUisms have gone. (I might try to scronge some HP-UX > compiler licences from HP.) HP's unbundled C and C++ compilers are freely available from HP. Sorry I can't remember which distribution DVD they're on (but the developer's bundle for PA-RISC is called B9007AA) but they're free. They didn't use to be, rather like Sun Studio didn't use to be :-) They get installed into /opt/ansic and /opt/aCC, also /opt/langtools. They're also free for IA64, but I don't have an Itanic box to check details. -- Chris
From: Michael Laajanen on 3 Jan 2010 04:14 Hi, David Kirkby wrote: > On 1 Jan, 22:26, Nemo ad Nusquam <inva...(a)invalid.invalid> wrote: >> Dave: >> >> David Kirkby wrote (in part): >> >>> Building Sage on Solaris is not exactly trivial, but it will build. >>> However, it only currently builds as a 32-bit application on Solaris >>> 10, and even then, some things do not work right. >> Good work! It would be very nice to have Sage running on more than >> Intel boxes. Thank you. > > It's only recently got it there. The is a lot of code in Sage - the > compressed source code is around 260 MB. The university of Washington > was donated a T5240 by Sun, but the architecture of that machine has > not really been ideal for developing on. I luckily had my own Blade > 2000, but Sage took a day or more to build on that. > > These was until recently a very annoying bug, which meant that Sage > would build if Sun Studio was not installed, but would not build if > Sun Studio was installed. Finally, after many months, the cause of it > has been identified. Of couse, we could have hacked the code to get it > to work, or as root you could move Sun Studion to a non-standard > place, but these were all hacks. > Are all these fixes that you do in the main source (4.3)? > I'm not sure if an HP-UX port will ever happen, but certainly some > testing I've done on HP-UX has identified problems that were less > apparent on other platforms, but still existed. I believe if there was > someone keen on using Sage on HP-UX, a port would be quite easy, as a > lot of the GNUisms have gone. (I might try to scronge some HP-UX > compiler licences from HP.) > > Dave What about IRIX? /michael
From: David Kirkby on 3 Jan 2010 11:14
On Jan 3, 9:14 am, Michael Laajanen <michael_laaja...(a)yahoo.com> wrote: > Hi, > > > > David Kirkby wrote: > > On 1 Jan, 22:26, Nemo ad Nusquam <inva...(a)invalid.invalid> wrote: > >> Dave: > > >> David Kirkby wrote (in part): > > >>> Building Sage on Solaris is not exactly trivial, but it will build. > >>> However, it only currently builds as a 32-bit application on Solaris > >>> 10, and even then, some things do not work right. > >> Good work! It would be very nice to have Sage running on more than > >> Intel boxes. Thank you. > > > It's only recently got it there. The is a lot of code in Sage - the > > compressed source code is around 260 MB. The university of Washington > > was donated a T5240 by Sun, but the architecture of that machine has > > not really been ideal for developing on. I luckily had my own Blade > > 2000, but Sage took a day or more to build on that. > > > These was until recently a very annoying bug, which meant that Sage > > would build if Sun Studio was not installed, but would not build if > > Sun Studio was installed. Finally, after many months, the cause of it > > has been identified. Of couse, we could have hacked the code to get it > > to work, or as root you could move Sun Studion to a non-standard > > place, but these were all hacks. > > Are all these fixes that you do in the main source (4.3)?> I'm not sure if an HP-UX port will ever happen, but certainly some > > testing I've done on HP-UX has identified problems that were less > > apparent on other platforms, but still existed. I believe if there was > > someone keen on using Sage on HP-UX, a port would be quite easy, as a > > lot of the GNUisms have gone. (I might try to scronge some HP-UX > > compiler licences from HP.) > > > Dave > > What about IRIX? > > /michael I have an SGI Octane running IRIX, a Dec Alpha running Tru64 and an IBM RS/6000 running AIX. In principle, Sage could be ported to any of them, but I think AIX and HP-UX are the only worthwhile ones now. If you want to try a port to IRIX, you are welcome to. I've started writing the code in a way it will work with IRIX. There are specific tests for IRIX, specific tests for the SGI IRIX compilers. I'm planning for the future, but I suspect IRIX port will never happen. But where I can add code that would make a port less hassle, without too much effort on my part, then it gets added. If you use IRIX, and want to port it, let me know. Dave |