l...@gnu.org (Ludovic Courtès) skribis:
> l...@gnu.org (Ludovic Courtès) skribis:
>
>> Andreas Enge skribis:
>>
>>> On Wed, Oct 07, 2015 at 10:32:43PM +0200, Ludovic Courtès wrote:
Do you think it’s a likely problem? Andreas, could you try again on
your Novena with 2 cores turned off?
l...@gnu.org (Ludovic Courtès) skribis:
> Andreas Enge skribis:
>
>> On Wed, Oct 07, 2015 at 10:32:43PM +0200, Ludovic Courtès wrote:
>>> Do you think it’s a likely problem? Andreas, could you try again on
>>> your Novena with 2 cores turned off?
>>
>> I have trouble believing the explanation of
Andreas Enge skribis:
> On Wed, Oct 07, 2015 at 10:32:43PM +0200, Ludovic Courtès wrote:
>> Do you think it’s a likely problem? Andreas, could you try again on
>> your Novena with 2 cores turned off?
>
> I have trouble believing the explanation of overheating. The gmp test
> suite is carried out
Andreas Enge writes:
> On Wed, Oct 07, 2015 at 10:32:43PM +0200, Ludovic Courtès wrote:
>> Do you think it’s a likely problem? Andreas, could you try again on
>> your Novena with 2 cores turned off?
>
> I have trouble believing the explanation of overheating. The gmp test
> suite is carried out
On Wed, Oct 07, 2015 at 10:32:43PM +0200, Ludovic Courtès wrote:
> Do you think it’s a likely problem? Andreas, could you try again on
> your Novena with 2 cores turned off?
I have trouble believing the explanation of overheating. The gmp test
suite is carried out sequentially (they are not yet u
Mark H Weaver skribis:
> l...@gnu.org (Ludovic Courtès) writes:
>
>> Mark H Weaver skribis:
>>
>>> l...@gnu.org (Ludovic Courtès) writes:
>>>
t...@gmplib.org (Torbjörn Granlund) skribis:
> This is most surely a hardware problem. The most common such problem is
> a bad RAM chip
l...@gnu.org (Ludovic Courtès) writes:
OK, I’ll see if I can reproduce it on another ARMv7 machine and
report back.
Good. Incidentally, we run nightly builds on a bunch of ARMv7 machines.
However, the machine in question is a build machine and it would be
surprising for the RAM issue
Andreas Enge writes:
> On Wed, Oct 07, 2015 at 08:45:09AM -0400, Mark H Weaver wrote:
>> When used as a build slave, the Novena is prone to overheating. The
>> best solution is to add a fan. In the meantime, I've disabled 2 out of
>> the 4 cores on hydra-slave1 using the following commands:
>
>
On Wed, Oct 07, 2015 at 08:45:09AM -0400, Mark H Weaver wrote:
> When used as a build slave, the Novena is prone to overheating. The
> best solution is to add a fan. In the meantime, I've disabled 2 out of
> the 4 cores on hydra-slave1 using the following commands:
Which temperature is overheati
l...@gnu.org (Ludovic Courtès) writes:
> Mark H Weaver skribis:
>
>> l...@gnu.org (Ludovic Courtès) writes:
>>
>>> t...@gmplib.org (Torbjörn Granlund) skribis:
>>>
This is most surely a hardware problem. The most common such problem is
a bad RAM chip.
>>>
>>> OK, I’ll see if I can repr
Mark H Weaver skribis:
> l...@gnu.org (Ludovic Courtès) writes:
>
>> t...@gmplib.org (Torbjörn Granlund) skribis:
>>
>>> This is most surely a hardware problem. The most common such problem is
>>> a bad RAM chip.
>>
>> OK, I’ll see if I can reproduce it on another ARMv7 machine and
>> report bac
[removed t...@gmplib.org and gmp-b...@gmplib.org from the CC list]
l...@gnu.org (Ludovic Courtès) writes:
> t...@gmplib.org (Torbjörn Granlund) skribis:
>
>> This is most surely a hardware problem. The most common such problem is
>> a bad RAM chip.
>
> OK, I’ll see if I can reproduce it on anoth
t...@gmplib.org (Torbjörn Granlund) skribis:
> This is most surely a hardware problem. The most common such problem is
> a bad RAM chip.
OK, I’ll see if I can reproduce it on another ARMv7 machine and
report back.
However, the machine in question is a build machine and it would be
surprising fo
This is most surely a hardware problem. The most common such problem is
a bad RAM chip.
Without special undocumented parameters, GMP's test suite runs the same
exact test vectors for each invocation, and GMP uses just plain old
deterministic algortihms internally.
--
Torbjörn
Please encrypt, ke
14 matches
Mail list logo
