Compiling PPL with OS X Lion
Hi all!
I am considering upgrading to OS X 10.7 (Lion). Will I be able to compile the PPL? Thanks in advance,
Fred
On 12/18/11 11:11, Fred Mesnard wrote:
Hi all!
I am considering upgrading to OS X 10.7 (Lion). Will I be able to compile the PPL?
Hi Fred!
It is difficult to say: it depends on the version of the PPL you want to compile and on the tools you have installed on your machine. Many people use the latest version of the PPL under OS X 10.7 so my guess is that there are no big obstacles in compiling it for that platform. In any case, we are here to help. Cheers,
Roberto
ppl has been ported to fink on 10.7, so you should be able to build it too.
http://pdb.finkproject.org/pdb/package.php/ppl9
Fang
On 12/18/11 11:11, Fred Mesnard wrote:
Hi all!
I am considering upgrading to OS X 10.7 (Lion). Will I be able to compile the PPL?
Hi Fred!
It is difficult to say: it depends on the version of the PPL you want to compile and on the tools you have installed on your machine. Many people use the latest version of the PPL under OS X 10.7 so my guess is that there are no big obstacles in compiling it for that platform. In any case, we are here to help. Cheers,
Roberto
On Sun, Dec 18, 2011 at 02:11:45PM +0400, Fred Mesnard wrote:
Hi all!
I am considering upgrading to OS X 10.7 (Lion). Will I be able to compile the PPL? Thanks in advance,
Fred, One caveat about building ppl under Lion with the system llvm-gcc is that -frounding-math support doesn't exist in the llvm-based compilers yet. So either you have to pass --disable-fpmath to configure with llvm-gcc/clang or build with gcc-4.2 which has a functional -frounding-math option. Note that Xcode 4.2 and later no longer provide gcc-4.2 so that option is problematic. Jack
Fred
PPL-devel mailing list PPL-devel@cs.unipr.it http://www.cs.unipr.it/mailman/listinfo/ppl-devel
On 12/19/11 15:27, Jack Howarth wrote:
On Sun, Dec 18, 2011 at 02:11:45PM +0400, Fred Mesnard wrote:
Hi all!
I am considering upgrading to OS X 10.7 (Lion). Will I be able to compile the PPL? Thanks in advance,
Fred, One caveat about building ppl under Lion with the system llvm-gcc is that -frounding-math support doesn't exist in the llvm-based compilers yet. So either you have to pass --disable-fpmath to configure with llvm-gcc/clang or build with gcc-4.2 which has a functional -frounding-math option. Note that Xcode 4.2 and later no longer provide gcc-4.2 so that option is problematic.
Right. In the Git version of the PPL (to be released as PPL 0.12), clang is automatically recognized so that explicitly passing --disable-fpmath will no longer be required.
The point is, though: does Fred need to use any of the PPL abstractions based on machine floating-point numbers? Cheers,
Roberto
On Mon, Dec 19, 2011 at 03:38:34PM +0100, Roberto Bagnara wrote:
On 12/19/11 15:27, Jack Howarth wrote:
On Sun, Dec 18, 2011 at 02:11:45PM +0400, Fred Mesnard wrote:
Hi all!
I am considering upgrading to OS X 10.7 (Lion). Will I be able to compile the PPL? Thanks in advance,
Fred, One caveat about building ppl under Lion with the system llvm-gcc is that -frounding-math support doesn't exist in the llvm-based compilers yet. So either you have to pass --disable-fpmath to configure with llvm-gcc/clang or build with gcc-4.2 which has a functional -frounding-math option. Note that Xcode 4.2 and later no longer provide gcc-4.2 so that option is problematic.
Right. In the Git version of the PPL (to be released as PPL 0.12), clang is automatically recognized so that explicitly passing --disable-fpmath will no longer be required.
The point is, though: does Fred need to use any of the PPL abstractions based on machine floating-point numbers?
Roberto, Well, in fink the only real use for ppl is for FSF gcc's graphite support. I have been told that graphite doesn't use any the floating-point abstractions in PPL and we see no graphite gcc testsuite regressions from using a PPL built with --disable-fpmath. Jack ps FYI, it appears that FSF gcc may switch graphite from PPL to ISL in gcc 4.8 or at least 4.9. This wasn't completed in time for gcc 4.7.
http://gcc.gnu.org/ml/gcc-patches/2011-08/msg01070.html
Cheers,
Roberto
-- Prof. Roberto Bagnara Applied Formal Methods Laboratory Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it _______________________________________________ PPL-devel mailing list PPL-devel@cs.unipr.it http://www.cs.unipr.it/mailman/listinfo/ppl-devel
Le 19 déc. 2011 à 19:02, Jack Howarth a écrit :
On Mon, Dec 19, 2011 at 03:38:34PM +0100, Roberto Bagnara wrote:
On 12/19/11 15:27, Jack Howarth wrote:
On Sun, Dec 18, 2011 at 02:11:45PM +0400, Fred Mesnard wrote:
Hi all!
I am considering upgrading to OS X 10.7 (Lion). Will I be able to compile the PPL? Thanks in advance,
Fred, One caveat about building ppl under Lion with the system llvm-gcc is that -frounding-math support doesn't exist in the llvm-based compilers yet. So either you have to pass --disable-fpmath to configure with llvm-gcc/clang or build with gcc-4.2 which has a functional -frounding-math option. Note that Xcode 4.2 and later no longer provide gcc-4.2 so that option is problematic.
Right. In the Git version of the PPL (to be released as PPL 0.12), clang is automatically recognized so that explicitly passing --disable-fpmath will no longer be required.
The point is, though: does Fred need to use any of the PPL abstractions based on machine floating-point numbers?
Roberto, Well, in fink the only real use for ppl is for FSF gcc's graphite support. I have been told that graphite doesn't use any the floating-point abstractions in PPL and we see no graphite gcc testsuite regressions from using a PPL built with --disable-fpmath.
No I don't need FP numbers. So I understand that either the GIT version of the PPL or the fink port should compile easily under Lion.
Fred
Jackps FYI, it appears that FSF gcc may switch graphite from PPL to ISL in gcc 4.8 or at least 4.9. This wasn't completed in time for gcc 4.7.
http://gcc.gnu.org/ml/gcc-patches/2011-08/msg01070.html
Cheers,
Roberto
-- Prof. Roberto Bagnara Applied Formal Methods Laboratory Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it _______________________________________________ PPL-devel mailing list PPL-devel@cs.unipr.it http://www.cs.unipr.it/mailman/listinfo/ppl-devel
PPL-devel mailing list PPL-devel@cs.unipr.it http://www.cs.unipr.it/mailman/listinfo/ppl-devel
participants (4)
-
David Fang -
Fred Mesnard -
Jack Howarth -
Roberto Bagnara