June 2009 - PPL-devel - BUGSENG Mailman

Problems with test suite on alpha
by Michael Tautschnig 09 Sep '10

09 Sep '10

Hi PPL developers, I've reported problems with some test cases because of suspected gcc bugs on alpha earlier already. Now it seems that even more test cases are broken, most probably because of the very same bug (I suspect it to be http://gcc.gnu.org/bugzilla/show_bug.cgi?id=8966). My question now is: How to proceed? Shall I specifically disable the failing test cases by surrounding them with an #ifndef __alpha__ or disable the entire test suite on alpha? Thanks, Michael

2 6

Parametric integer linear programming solver
by Sebastian Pop 22 Nov '09

22 Nov '09

Hello, I would like to know whether there exists a parametric integer linear programming solver that can be called from the PPL interface, or how we could use from PPL an existing solver, like http://piplib.org We would need such a tool to improve the precision of the data dependence analysis of Graphite by computing the reaching definitions of array accesses. Thank you, Sebastian Pop -- AMD - GNU Tools

4 10

Re: [PPL-devel] Graphite build fails if PPL configured with --disable-shared
by Roberto Bagnara 08 Jul '09

08 Jul '09

Janis Johnson wrote: > On Mon, 2009-05-11 at 13:07 -0700, Ian Lance Taylor wrote: >> Another Graphite build issue: it appears that I must not use >> --disable-shared when I configure PPL. If I do use --disable-shared, I >> get this: >> >> /home/iant/gnu/ppl-0.10.2-install/lib/libppl_c.a(ppl_c_implementation_common.o): In function `finalize': >> /home/iant/gnu/ppl-0.10.2/interfaces/C/../../src/ppl.hh:1842: undefined reference to `operator delete(void*)' >> >> followed by thousands of similar errors. This is unfortunate, as it >> means that I must manually set LD_LIBRARY_PATH to the directory where >> the PPL library is installed. This also makes it harder for anybody >> else to run the compiler that I build. This needs to be fixed. > > I get around this by setting LDFLAGS for the ppl configure: > > LDFLAGS="-static" \ > ./configure \ > --prefix=$PREFIX \ > --build=powerpc-linux \ > --with-gnu-ld \ > --with-libgmp-prefix=$PREFIX \ > --with-libgmpxx-prefix=$PREFIX \ > --disable-shared I am not sure I understand: we trust that Libtool, which provides us with the --disable-shared option, will do the right thing. And it seems it does here: the static library is built and passes its checks. Perhaps you want something different from what --disable-shared promises, that is, not to build any shared libraries? > I copy libstdc++.a into the directory with the other GCC host > libraries (gmp/mpfr/ppl/cloog/mpc). > > Building these libraries is indeed quite painful. Any suggestion about how to improve the PPL is welcome. This, of course, applies also to the build machinery. All the best, Roberto -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it

6 8

Z-polytopes in PPL
by Michael Classen 30 Jun '09

30 Jun '09

Hello, this might be a trivial question for some, but I just seem to have problems with it: I want to adapt our internal LooPo interface to use PPL Grids instead of Z-Polytopes. Now, Z-Polytopes were typically defined by a domain (or inequality system) and an affine function, which is applied to that domain (also alled Lattice). Now, in PPL, as far as I can see, you can only create Grids out of inequality systems, congruence systems or Grid generators. Is there an easy way (function, constructor) to generate a Grid from an inequality system + affine function, like in the old Polylib? thank you in advance! Greetings, Michael

2 3

Re: [PPL-devel] Compiling ppl-0.10.2 on Solaris Express
by alanpae＠ilkda.com 27 Jun '09

27 Jun '09

> > As requested. > [C_Polyhedron.ii.bz2 attachment] > Some header file different from reg.h is defining the macro CS to 15. > You should discover which one is doing that. > Good luck, > Roberto As I've already stated if you wish you can pursue this on your own merely by downloading either an OpenSolaris, Solaris, or Solaris Express DVD or Live CD and then you can investigate it to your hearts content. I have absolutley zero interest in persuing the matter. alan

1 0

Re: [PPL-devel] Compiling ppl-0.10.2 on Solaris Express
by Roberto Bagnara 27 Jun '09

27 Jun '09

alanpae(a)ilkda.com wrote: > As requested. [C_Polyhedron.ii.bz2 attachment] Some header file different from reg.h is defining the macro CS to 15. You should discover which one is doing that. Good luck, Roberto -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it

1 0

Re: [PPL-devel] Compiling ppl-0.10.2 on Solaris Express
by alanpae＠ilkda.com 25 Jun '09

25 Jun '09

> Here is what I would do in your shoes. I would take the failing compilation > command, i.e., Ok, If I was you I would: > g++ -DHAVE_CONFIG_H -I. -I.. -I.. -I../src -D_SYS_REG_H -I/opt/local/include \ > -g -O2 -frounding-math -W -Wall -MT C_Polyhedron.lo -MD -MP \ > -MF .deps/C_Polyhedron.Tpo -c C_Polyhedron.cc -fPIC -DPIC -o .libs/C_Polyhedron.o > > I would replace -c with -E and erase -g, -O2, ... and anything after > C_Polyhedron.cc, redirecting the output to a file. Then I get: g++ -DHAVE_CONFIG_H -I. -I.. -I.. -I../src -D_SYS_REG_H -I/opt/local/include -frounding-math -W -Wall -MT C_Polyhedron.lo -MD -MP -MF .deps/C_Polyhedron.Tpo -E C_Polyhedron.cc | grep CS static const flags_t CS_PENDING = 1U << 7; return test_any(CS_PENDING); reset(CS_PENDING); set(CS_PENDING); static const flags_t CS_PENDING = 1U << 7; return test_any(CS_PENDING); reset(CS_PENDING); set(CS_PENDING); > Then I would try to identify the header that is defining the macro CS. statit const ... appears in: # 70 "Grid_Status.idefs.hh" and # 85 "Ph_Status.idefs.hh" So then I as you would do the following: ... Thanks, alan > Reading all your files is exactly what I had in mind :-) Yea, it's just that every time I show people my plans for mind control they start to laugh and then they call me stupid and it's really starting to get on my nerves. That's all. :-)

2 1

Compiling ppl-0.10.2 on Solaris Express
by Alan Pae 24 Jun '09

24 Jun '09

This is essentially a repeat of a bug that has already been posted. The original message can be viewed at: http://gcc.gnu.org/ml/gcc-help/2009-04/msg00381.html Instead of using OpenSolaris I'm using Solaris Express which is essentially the same thing but contains all of the proprietary bits that are stripped out to create OpenSolaris. I'm stuck at the exact same spot with the exact same error messages and was just wondering if this was going to be fixed or if there's something I can do to get the compiler to move on down the road. Thanks in advance, alan

2 5

Compiling ppl-0.10.2 on Solaris Express
by Alan Pae 24 Jun '09

24 Jun '09

Should have thought of this sooner. You can grab the OpenSolaris Live CD or Solaris Express at: http://opensolaris.org/os/downloads/ The quick links are on the right hand side of the screen. Both are free for download and after an install you're system would be the same as mine. alan

1 0

[GIT] ppl/w3ppl(master): A few typos in the instructions for the use of git fixed.
by Patricia Hill 22 Jun '09

22 Jun '09

Module: ppl/w3ppl Branch: master Commit: c275e098002ed5518f81d9ffc7ce5102c2504b83 URL: http://www.cs.unipr.it/git/gitweb.cgi?p=ppl/w3ppl.git;a=commit;h=c275e09800… Author: Patricia Hill <p.m.hill(a)comp.leeds.ac.uk> Date: Mon Jun 22 15:32:25 2009 +0100 A few typos in the instructions for the use of git fixed. --- htdocs/Download/Git.raw | 24 ++++++++++++------------ 1 files changed, 12 insertions(+), 12 deletions(-) diff --git a/htdocs/Download/Git.raw b/htdocs/Download/Git.raw index 24f4af4..3ec3b90 100644 --- a/htdocs/Download/Git.raw +++ b/htdocs/Download/Git.raw @@ -55,7 +55,7 @@ In order to clone the repository of the PPL sources, issue the command <BLOCKQUOTE> - <CODE>git clone git://git.cs.unipr.it/ppl/ppl.git</CODE> + <CODE>git clone git://git.cs.unipr.it:/ppl/ppl.git</CODE> </BLOCKQUOTE> This will put your clone into the <CODE>ppl</CODE> directory. @@ -63,7 +63,7 @@ If you want the directory to be named differently you can specify the name as in <BLOCKQUOTE> - <CODE>git clone git://git.cs.unipr.it/ppl/ppl.git my_ppl_repo</CODE> + <CODE>git clone git://git.cs.unipr.it:/ppl/ppl.git my_ppl_repo</CODE> </BLOCKQUOTE> @@ -72,7 +72,7 @@ history. If you only want the latest revision of anything, use the following command instead: <BLOCKQUOTE> - <CODE>git clone --depth 1 git://git.cs.unipr.it/ppl/ppl.git my_ppl_no_history_repo</CODE> + <CODE>git clone --depth 1 git://git.cs.unipr.it:/ppl/ppl.git my_ppl_no_history_repo</CODE> </BLOCKQUOTE> @@ -92,7 +92,7 @@ Another repository you may be interested in is the one containing our web pages. To fully clone it, do, e.g., <BLOCKQUOTE> - <CODE>git clone git://git.cs.unipr.it/ppl/w3ppl.git</CODE> + <CODE>git clone git://git.cs.unipr.it:/ppl/w3ppl.git</CODE> </BLOCKQUOTE> @@ -101,10 +101,10 @@ world via HTTP (possibly with the intermediation of a proxy), you can use the alternative commands <BLOCKQUOTE> - <CODE>git clone http://www.cs.unipr.it/git/ppl/ppl.git</CODE> + <CODE>git clone http://www.cs.unipr.it:/git/ppl/ppl.git</CODE> </BLOCKQUOTE> <BLOCKQUOTE> - <CODE>git clone http://www.cs.unipr.it/git/ppl/w3ppl.git</CODE> + <CODE>git clone http://www.cs.unipr.it:/git/ppl/w3ppl.git</CODE> </BLOCKQUOTE> @@ -157,10 +157,10 @@ solve this problem first. Issue the command <BLOCKQUOTE> - <CODE>git clone ssh://myusername@git.cs.unipr.it/git/ppl/ppl.git ppl</CODE> + <CODE>git clone ssh://myusername@git.cs.unipr.it:/git/ppl/ppl.git ppl</CODE> </BLOCKQUOTE> <BLOCKQUOTE> - <CODE>git clone ssh://myusername@git.cs.unipr.it/git/ppl/w3ppl.git w3ppl</CODE> + <CODE>git clone ssh://myusername@git.cs.unipr.it:/git/ppl/w3ppl.git w3ppl</CODE> </BLOCKQUOTE> to clone the PPL source or the PPL web repositories, respectively. @@ -170,7 +170,7 @@ to clone the PPL source or the PPL web repositories, respectively. After the initial checkout, you can change into one of the project's directories and execute most Git commands omitting -the <CODE>ssh://myusername@git.cs.unipr.it/git/ppl/ppl.git</CODE> argument. +the <CODE>ssh://myusername@git.cs.unipr.it:/git/ppl/ppl.git</CODE> argument. For example, whenever you want to merge with the latest code changes, <CODE>cd</CODE> into your <CODE>ppl</CODE> directory and issue @@ -292,15 +292,15 @@ basic CVS commands and their Git counterparts. <PRE> cvs -d ssh://myusername@git.cs.unipr.it/cvs/ppl checkout ppl | -+---> git clone ssh://myusername@git.cs.unipr.it/git/ppl/ppl.git ++---> git clone ssh://myusername@git.cs.unipr.it:/git/ppl/ppl.git </PRE> <H4>Checkout of a branch vs Cloning of the master branch + Creation of a local branch tracking a remote one</H4> <PRE> -cvs -d ssh://myusername@git.cs.unipr.it/cvs/ppl checkout -r products ppl +cvs -d ssh://myusername@git.cs.unipr.it:/cvs/ppl checkout -r products ppl | -+---> git clone ssh://myusername@git.cs.unipr.it/git/ppl/ppl.git ++---> git clone ssh://myusername@git.cs.unipr.it:/git/ppl/ppl.git git checkout -b products origin/products </PRE>

1 0