Re: [sage-release] Sage 6.9.rc0 released

2015-10-03 Thread Clemens Heuberger
Am 2015-10-03 um 11:28 schrieb Jeroen Demeyer: > On 2015-09-28 16:22, Clemens Heuberger wrote: >> >> I recompiled 6.9.beta7 and 6.9.rc0 last night one after the other on the same >> machine with negligible other load. > > What does the word "recompiled" mean in the sentence above? sorry for the

Re: [sage-release] Sage 6.9.rc0 released

2015-10-03 Thread Clemens Heuberger
Am 2015-10-03 um 19:00 schrieb Clemens Heuberger: > rm -rf sage-6.9.beta7 > git clone ... > make -j4; ./sage -i ccache; make I repeated this tonight: 6.9.beta7: real88m39.989s user247m40.494s sys 19m13.816s 6.9.rc0: real181m18.663s user

Re: [sage-release] Sage 6.9.rc0 released

2015-09-29 Thread Jeroen Demeyer
On 2015-09-28 22:05, William Stein wrote: Agreed. Jereon want to do something about this? It's reported upstream to PARI. -- You received this message because you are subscribed to the Google Groups "sage-release" group. To unsubscribe from this group and stop receiving emails from it, send

Re: [sage-release] Sage 6.9.rc0 released

2015-09-28 Thread William Stein
On Monday, September 28, 2015, Volker Braun wrote: > Hmm but /usr/lib/x86_64-linux-gnu/libpari-gmp-tls.so.2.7.2 brings just a > slightly different version of Pari's concat. It may very well be that it > doesn't trigger libkpathsea crashing but thats just by chance; the

Re: [sage-release] Sage 6.9.rc0 released

2015-09-28 Thread Clemens Heuberger
I recompiled 6.9.beta7 and 6.9.rc0 last night one after the other on the same machine with negligible other load. 6.9.beta7: 81 minutes 6.9.rc0:173 minutes (make -j4 with 8 GB RAM, linux 64 bit). I currently do not have a breakdown as "time" would give it (because I did not know that

Re: [sage-release] Sage 6.9.rc0 released

2015-09-27 Thread leif
Volker Braun wrote: > With Sage-6.8 is pretty much the same, maybe ~< 10% less memory usage; > In any case no drastic change. Just about 50+% increase in build time...? ;-) -leif For reference: The figures for 6.9.rc0:

Re: [sage-release] Sage 6.9.rc0 released

2015-09-27 Thread Volker Braun
The wall time isn't comparable, different system load. On Sunday, September 27, 2015 at 3:00:33 PM UTC+2, leif wrote: > > Volker Braun wrote: > > With Sage-6.8 is pretty much the same, maybe ~< 10% less memory usage; > > In any case no drastic change. > > Just about 50+% increase in build

Re: [sage-release] Sage 6.9.rc0 released

2015-09-27 Thread Volker Braun
With Sage-6.8 is pretty much the same, maybe ~< 10% less memory usage; In any case no drastic change. On Saturday, September 26, 2015 at 8:49:43 PM UTC+2, Jeroen Demeyer wrote: > > On

Re: [sage-release] Sage 6.9.rc0 released

2015-09-26 Thread Volker Braun
Though afaik our docs say that you need 2gb per process, so you were quite a bit below that. It could just be slightly different order of execution that blows things up now even though memory consumption didn't increase. On Saturday, September 26, 2015 at 1:03:49 PM UTC+2, Jeroen Demeyer

Re: [sage-release] Sage 6.9.rc0 released

2015-09-26 Thread Clemens Heuberger
ubuntu 14.04, 64 bit, 4 GB RAM: make -j4 ran out of memory while building documentation, last line in logfile: [tutorial ] reading sources... [ 43%] sagetex I then killed 4 processes using 2.7 GB each. On the same machine, there haven't been any problems with make -j4 for any beta or rc

Re: [sage-release] Sage 6.9.rc0 released

2015-09-26 Thread Volker Braun
Graph of CPU + RAM usage of export MAKE='make -j4' make doc -- You received this message because you are subscribed to the Google Groups "sage-release" group. To unsubscribe from this

Re: [sage-release] Sage 6.9.rc0 released

2015-09-26 Thread Jeroen Demeyer
I can confirm that there is indeed a problem with docbuilding. When looking at the patchbot, I saw a similar problem. On 2015-09-26 09:58, Clemens Heuberger wrote: ubuntu 14.04, 64 bit, 4 GB RAM: make -j4 ran out of memory while building documentation, last line in logfile: [tutorial ]

Re: [sage-release] Sage 6.9.rc0 released

2015-09-26 Thread Justin C. Walker
> On Sep 25, 2015, at 03:54, Volker Braun wrote: > > As always, you can get the latest beta version from the "develop" git branch. > Alternatively, the self-contained source tarball is at > http://www.sagemath.org/download-latest.html > > This release now supports

Re: [sage-release] Sage 6.9.rc0 released

2015-09-26 Thread Rob Beezer
On SageMathCloud, built from source, passes long tests ("ptestlong"). https://cloud.sagemath.com/projects/53b77207-8614-4086-a032-432af4b4cdbd/files/sage-dev-images/ -- You received this message because you are subscribed to the Google Groups "sage-release" group. To unsubscribe from this

Re: [sage-release] Sage 6.9.rc0 released

2015-09-26 Thread Jeroen Demeyer
On 2015-09-26 14:48, Volker Braun wrote: Though afaik our docs say that you need 2gb per process, so you were quite a bit below that. It could just be slightly different order of execution that blows things up now even though memory consumption didn't increase. The question is: did it really

[sage-release] Sage 6.9.rc0 released

2015-09-25 Thread Volker Braun
As always, you can get the latest beta version from the "develop" git branch. Alternatively, the self-contained source tarball is at http://www.sagemath.org/download-latest.html This release now supports Xcode 7 but not OSX 10.11. The letter doesn't look encouraging, so I probably won't