-
-
Notifications
You must be signed in to change notification settings - Fork 596
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
package polymake 3.0 #20892
Comments
This comment has been minimized.
This comment has been minimized.
Branch: u/vdelecroix/20892 |
Commit: |
New commits:
|
Branch pushed to git repo; I updated commit sha1. New commits:
|
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
comment:6
Installation fails on Mac OS X:
|
comment:7
It finds my Fink installation (though I removed it from PATH for building Sage) -- should this be disabled? |
comment:8
this might be a mismatch between the code emitted by quite new gcc and old assembler, which is unable to compile this into binary. There were a number of such OSX-only reports recently, check the sage-develop... |
comment:9
yeah, try to move Fink totally to some /opt/blah place and try again... |
comment:10
OK. After hiding Fink, I get:
I'll start again with a fresh install of Fink and install some perl there. |
comment:11
I will first try to install polymake from source outside of Sage -- this already fails on my machine; I have asked in the Polymake forum for help. Vincent: I think the right package to use in Sage would be the "minimal" package which picks up cdd, lrs, nauty from the Sage distribution. |
Branch pushed to git repo; I updated commit sha1. New commits:
|
comment:13
Replying to @mkoeppe:
right. done. |
This comment has been minimized.
This comment has been minimized.
Branch pushed to git repo; I updated commit sha1. New commits:
|
comment:16
See #20894 for a first step towards an OS X build |
comment:17
I solved the problem on Mac OS X with the following patch for Polymake, which gets rid of various "-arch" options. The "-arch" options caused the above errors "register %rbp is only available in 64-bit mode" etc.
And here is a quick change to
With these changes, Polymake built successfully. |
comment:18
Can you make a package with this? I tried to test it by creating the right patch directory and changing the spkg-install for applying patches but I tend to always make a mess of that. Anyway, having done that, it did get quite a bit further (including #20894) though I did get some warnings e.g.
But anyway it is making vastly more progress than before, so please bundle this up and with any luck I can help give positive review (I'm assuming it will take a little longer than I want to stay up now to finish!). Great, great work. |
comment:19
#20692 could help with automating the patching. |
comment:20
Replying to @kcrisman:
Yes, I also get a few of those, but the polymake install goes through after that. Details of the 'bundled extensions' can be found in the package build directory (
I don't know what bliss is but I tried to install it -- but it gave me an error. Reported on #20871.
I do have lrslib installed. This needs investigating.
I think singular is standard, but there seems to be no
OK, yes, we don't have SoPlex. This is part of the SCIP Optimization Suite. (#10879) |
Branch pushed to git repo; I updated commit sha1. New commits:
|
Branch pushed to git repo; I updated commit sha1. New commits:
|
Branch pushed to git repo; I updated commit sha1. New commits:
|
comment:94
Branch has now merged the bliss ticket (#20901). |
comment:95
#21175 (Set ARCHFLAGS environment variable - for Perl modules) would make polymake's |
comment:96
Replying to @mkoeppe:
It's now working on Ubuntu (and still on Mac OS X), please test. |
comment:97
Better. Build without any error and
|
comment:98
However, after switching back to the develop branch doing
|
comment:99
Replying to @videlec:
Yes, that's normal. Need to keep #20901 in your branch if you have installed the bliss package from #20901. |
Changed reviewer from Matthias Koeppe, NEEDS MORE REVIEWERS to Matthias Koeppe, Vincent Delecroix |
comment:101
It works well on top of 7.4.beta0. Setting to positive review in order to be easily available for testing. |
Changed reviewer from Matthias Koeppe, Vincent Delecroix to Matthias Koeppe, Vincent Delacroix, Karl-Dieter Crisman |
Changed reviewer from Matthias Koeppe, Vincent Delacroix, Karl-Dieter Crisman to Matthias Koeppe, Vincent Delecroix, Karl-Dieter Crisman |
Changed branch from u/mkoeppe/20892 to |
Changed commit from |
comment:105
For those who really know this stuff, on a related note: http://ask.sagemath.org/question/34457/why-cant-i-compute-the-ehrhart-polynomial-of-a-polytope/?answer=34458#post-id-34458 so hopefully it will be more obvious soon what to do for this type of user. |
comment:106
People report that building Singular with polymake (not this one, a systemwide one) installed leads to a problem. This ought to be investigated. |
Package now compiles out of the box! There will be a next ticket for its interface within Sage.
upstream tarball: https://polymake.org/lib/exe/fetch.php/download/polymake-3.0r2-minimal.tar.bz2
Once installed you should be able to run polymake with
previous attempts (with earlier versions): #13768, #14116
upstream discussion at:
Follow-up tickets:
Depends on #20894
Depends on #20886
Depends on #20901
CC: @dimpase @mkoeppe @fchapoton @stumpc5 @mo271 @tscrim
Component: packages: experimental
Author: Vincent Delecroix, Matthias Koeppe
Branch:
7b096fd
Reviewer: Matthias Koeppe, Vincent Delecroix, Karl-Dieter Crisman
Issue created by migration from https://trac.sagemath.org/ticket/20892
The text was updated successfully, but these errors were encountered: