On Sun, Oct 25, 2015 at 5:10 PM, David Joyner wrote:
> On Sun, Oct 25, 2015 at 3:46 PM, Volker Braun wrote:
>> Are those the new binaries that I built with SAGE_FAT_BINARY=yes, or the
>> previous ones? The new md5sum is:
>>
>> $ md5 *
>> MD5 (sage-6.10.beta1-x86_64-Darwin-OSX_10.11-app.dmg) =
>>
On Sun, Oct 25, 2015 at 3:46 PM, Volker Braun wrote:
> Are those the new binaries that I built with SAGE_FAT_BINARY=yes, or the
> previous ones? The new md5sum is:
>
> $ md5 *
> MD5 (sage-6.10.beta1-x86_64-Darwin-OSX_10.11-app.dmg) =
> 879b8f796ffacc84a372987539a3fc4d
> MD5 (sage-6.10.beta1-x86_64
Are those the new binaries that I built with SAGE_FAT_BINARY=yes, or the
previous ones? The new md5sum is:
$ md5 *
MD5 (sage-6.10.beta1-x86_64-Darwin-OSX_10.11-app.dmg) =
879b8f796ffacc84a372987539a3fc4d
MD5 (sage-6.10.beta1-x86_64-Darwin-OSX_10.11.dmg) =
f5eba161c01c37aa0422519dd4f028d6
On
On Sun, Oct 25, 2015 at 4:35 AM, Christian Nassau
wrote:
> On 25.10.2015 01:03, Volker Braun wrote:
>>
>> I've used that ticket to build relocatable OSX 10.11 binaries and uploaded
>> them to http://files.sagemath.org/osx/intel/index.html. They work except you
>> can't compile stuff on top of thes
On Sun, Oct 25, 2015 at 9:54 AM, Volker Braun wrote:
> On Sunday, October 25, 2015 at 5:49:45 PM UTC+1, William wrote:
>>
>> At least openssl very recently changed their license to be
>> GPL-compatible, so it is legal to include with Sage... finally.
>
>
> No they didn't; They announced their inte
On Sunday, October 25, 2015 at 5:49:45 PM UTC+1, William wrote:
>
> At least openssl very recently changed their license to be
> GPL-compatible, so it is legal to include with Sage... finally.
>
No they didn't; They announced their intention but nothing has happened
yet. Considering that they'l
On Sun, Oct 25, 2015 at 3:11 AM, Volker Braun wrote:
> On Sunday, October 25, 2015 at 10:13:39 AM UTC+1, François wrote:
>>
>> That explain why I had to change the configuration for git.
>> No openssl headers, I have heard they are going to switch to libressl,
>
>
> Not sure about that, they proba
The speedups you are reporting are really impressive. 10-12x on 16 cores is
quite difficult to achieve for many things!
Bill.
On Wednesday, 21 October 2015 23:38:34 UTC+2, Victor Shoup wrote:
>
> Just in case anyone is interested...
>
> I released a new version of NTL that offers a new "thread b
On Sunday, October 25, 2015 at 10:13:39 AM UTC+1, François wrote:
>
> That explain why I had to change the configuration for git.
> No openssl headers, I have heard they are going to switch to libressl,
>
Not sure about that, they probably want everybody to use their own
proprietary crypto fram
That explain why I had to change the configuration for git.
No openssl headers, I have heard they are going to switch to libressl,
but even then, no headers? for something like ssl?
François
> On 25/10/2015, at 21:48, Volker Braun wrote:
>
> Ah yes forgot to set that environment variable during
Ah yes forgot to set that environment variable during build, so it requires
a recent Core i-series processor.
On an unrelated note, Apple dropped another giant t***d on us by not
including the openssl headers any more. Only the dylib is in OSX 10.11, but
without headers we can't compile agains
On 25.10.2015 01:03, Volker Braun wrote:
I've used that ticket to build relocatable OSX 10.11 binaries and
uploaded them to http://files.sagemath.org/osx/intel/index.html. They
work except you can't compile stuff on top of these binaries since the
library install names are still wrong. In any c
12 matches
Mail list logo