Re: [sage-devel] State of sage built with clang/clang++ on OS X - update

2016-10-03 Thread Francois Bissey
You need https://trac.sagemath.org/ticket/12473 for ratpoints. coming in the next beta. François > On 4/10/2016, at 19:07, Ralf Stephan wrote: > > On Sunday, October 2, 2016 at 10:33:38 PM UTC+2, François wrote: > this is the status after about two weeks > > ratpoints compiles for you? On Lin

[sage-devel] Re: State of sage built with clang/clang++ on OS X - update

2016-10-03 Thread Ralf Stephan
On Sunday, October 2, 2016 at 10:33:38 PM UTC+2, François wrote: > > this is the status after about two weeks > ratpoints compiles for you? On Linux with CC=clang CXX=clang++ make (clang-3.7.0) I can't get past ratpoints. -- You received this message because you are subscribed to the Google Gro

[sage-devel] Re: waterfall

2016-10-03 Thread Volker Braun
Yes, I updated over the weekend to the buildbot nine (0.9.0rc4) which is quite different from the old one. On Monday, October 3, 2016 at 9:23:46 PM UTC+2, Jan Groenewald wrote: > > Hi > > http://build.sagemath.org/release/waterfall > > gives me > > No Such Resource > > File not found. > > > Ha

[sage-devel] waterfall

2016-10-03 Thread Jan Groenewald
Hi http://build.sagemath.org/release/waterfall gives me No Such Resource File not found. Has the link changed to http://build.sagemath.org/#/waterfall ? A software update? Regards, Jan -- .~. /V\ Jan Groenewald /( )\www.aims.ac.za ^^-^^ -- You received this message becaus

[sage-devel] Re: SAGE hit tab key for autocomplete

2016-10-03 Thread Harald Schilly
On Monday, October 3, 2016 at 5:00:17 PM UTC+2, Pjl Zaldívar wrote: > > When hitting the key tab to auto-complete the attribute of an object (see > below for an example) I have found that it does not show the window that > should appear. > > for example: > a=501/7 > a.d > You have to evaluate

[sage-devel] SAGE hit tab key for autocomplete

2016-10-03 Thread Pjl Zaldívar
Hi everyone, I am starting to use SM for an introduction course to Maths at University. I am using an online version of the program on my laptop (macos10.11.6, running on both safari/firefox) and I have found the next problem. When hitting the key tab to auto-complete the attribute of an objec

Re: [sage-devel] State of sage built with clang/clang++ on OS X - update

2016-10-03 Thread Jean-Pierre Flori
On Monday, October 3, 2016 at 4:04:20 PM UTC+2, Dima Pasechnik wrote: > > > > On Monday, October 3, 2016 at 11:48:22 AM UTC+1, François wrote: >> >> I haven’t tried SAGE_INSTALL_GCC=no on linux or using clang >> for a full build on linux. Just one packages here and there. >> >> And yes you cann

Re: [sage-devel] State of sage built with clang/clang++ on OS X - update

2016-10-03 Thread Dima Pasechnik
On Monday, October 3, 2016 at 11:48:22 AM UTC+1, François wrote: > > I haven’t tried SAGE_INSTALL_GCC=no on linux or using clang > for a full build on linux. Just one packages here and there. > > And yes you cannot build gfortran without gcc a t this time. > this is a limitation of Sage's gcc

[sage-devel] Re: linbox 64-bit charpoly

2016-10-03 Thread 'Bill Hart' via sage-devel
This is not William's machines causing the delay. It's me. We are currently in the process of moving the sites and redesigning one of them. It's taking time due to multiple hardware failures (not the same one that caused the websites to go down) and tight deadlines for other projects. Hang tight

Re: [sage-devel] State of sage built with clang/clang++ on OS X - update

2016-10-03 Thread Francois Bissey
I haven’t tried SAGE_INSTALL_GCC=no on linux or using clang for a full build on linux. Just one packages here and there. And yes you cannot build gfortran without gcc a t this time. François > On 3/10/2016, at 22:48, Dima Pasechnik wrote: > > And how about Linux with clang(++) and gfortran ins

Re: [sage-devel] State of sage built with clang/clang++ on OS X - update

2016-10-03 Thread Dima Pasechnik
On Monday, October 3, 2016 at 8:56:27 AM UTC, François wrote: > > In the current state of sage starting from scratch with clang/clang++ > via > CC=clang CXX=clang++ make > will trigger the building of sage’s gcc (which is OK on OS X since > you still need gfortran). > The problem is that onc

Re: [sage-devel] linbox 64-bit charpoly

2016-10-03 Thread parisse
The probabilistic early termination does not take much time here, the charpoly stabilizes at about 85% of the primes required to reach the Hadamard bound. Testing with a few random matrices, I often get stabilization at about 80% (+/-10%), in this situation I think it's best to wait a little mo

Re: [sage-devel] State of sage built with clang/clang++ on OS X - update

2016-10-03 Thread Francois Bissey
In the current state of sage starting from scratch with clang/clang++ via CC=clang CXX=clang++ make will trigger the building of sage’s gcc (which is OK on OS X since you still need gfortran). The problem is that once you build sage’s gcc, there is a clause in src/bin/sage-env to use sage’s gcc/g+

Re: [sage-devel] State of sage built with clang/clang++ on OS X - update

2016-10-03 Thread Dima Pasechnik
What exactly should be set to start the Sage build using clang rather than gcc? On Monday, October 3, 2016 at 6:32:45 AM UTC, François wrote: > > > > On 3/10/2016, at 09:33, François Bissey > wrote: > > > >> linbox > > > > No news from upstream at > > https://github.com/linbox-team/linbox/