On 07/ 4/10 11:15 AM, Volker Braun wrote:
Does sage compile on anything but linux on itanium systems?
Probably not. But it would be shortsighted to write code that would make future
ports more difficult than necessary.
Several years ago, nobody had attempted to build Sage on 64-bit Solaris, so
someone decided to declare
typedef int int_fast32_t;
typedef long long int_fast64_t;
on just Solaris, which is valid for 32-bit code, but not 64-bit.
So I had to waste time sorting out their mistake.
http://trac.sagemath.org/sage_trac/ticket/9399
With a bit of foresight, they should have seen that this was not accurate.
(Solaris has been a 64-bit operating system for 15 years).
But I
agree that, if the code is itanium linux-specific, then it must be
wrapped into
#if defined(__linux__)&& ( defined(__ia64__)
// itanium linux specific
#endif
About the "readline on itanium" patch, wtf is going on? That patch is
just some random kludgery. Is http://trac.sagemath.org/sage_trac/ticket/488
the corresponding ticket?
http://trac.sagemath.org/sage_trac/ticket/488
is about Readline, but does not mention openSUSE.
The readline patch I was talking about was the one which starts
if [ -f /etc/SuSE-release ]; then
if [ `grep 11.1 /etc/SuSE-release > /dev/null; echo $?` -eq 0 ]; then
echo "OpenSUSE 11.1 detected"
I've no idea if it's related to
http://bugs.python.org/issue1204
or not. I would not have associated them. #488 seems to be about readline, and
not specifically openSUSE.
It doesn't even say what the bug is.
True, which is why I can't tell for 100% sure if
http://trac.sagemath.org/sage_trac/ticket/488
and
http://bugs.python.org/issue1204
are related or not. My gut feeling is they are unrelated.
There
is a sensible discussion on the python bugtracker of something that
sounds very related, see http://bugs.python.org/issue1204. There, you
can find an analysis of what went wrong and a correct fix that has,
moreover, long been incorporated into python. The correct fix is very
much different from what the sage patch mucks around with....
Volker
If we implemented what I proposed the other day on sage-devel, under the title
"Suggestion to make reporting bugs upstream MANDATORY for Positive review"
http://groups.google.co.uk/group/sage-devel/browse_thread/thread/f84c603497d84485/d46bfa7052ea31e7
we might have more hope of being able to remove patches when bugs are fixed
upstream. If we did things the way I suggested,
http://trac.sagemath.org/sage_trac/ticket/488
would have a link to a bug reports on the OpenSUSE bug tracker
http://en.opensuse.org/Submitting_Bug_Reports
and the Python web site
http://bugs.python.org/
It should be noted that
http://trac.sagemath.org/sage_trac/ticket/488
is a very old ticket, and Sage procedures have improved a lot since then. It
would be good if that progress could be maintained by putting even more emphasis
on code quality, and documenting things properly.
Dave
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
