[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [ESPResSo-users] Configuration file for building Espresso with LB
From: |
Axel Arnold |
Subject: |
Re: [ESPResSo-users] Configuration file for building Espresso with LB |
Date: |
Wed, 17 Dec 2014 16:46:28 +0100 |
Hi,
Just to clarify: You need to remove all features ending in _GPU, because those
require a CUDA capable device. In addition, remove all below the Debugging
comment, including ADDITIONAL_CHECKS, which will speed up things drastically,
and reduce output to a human level.
Best,
Axel
JP Dr. Axel Arnold
ICP, Universität Stuttgart
Allmandring 3
70569 Stuttgart
Email: address@hidden
Phone: +49 711 685 67609
> On 17 Dec 2014, at 16:02, Owen Hickey <address@hidden> wrote:
>
> Hello Edith,
>
> The message indicates that no GPU capable of running CUDA code was
> detected. The easiest way to solve this is to run configure with the
> option --without-cuda. The LB is much faster on a GPU but if you don't
> have one available the CPU version is an alternative.
>
> Good luck,
>
> Owen
>
>> On Wed, Dec 17, 2014 at 1:48 PM, Edith Euan <address@hidden> wrote:
>> Dear Axel,
>>
>> Thanks for your answer, a this is the output for the sample
>> "espresso_logo.tcl"
>>
>> 0: node_grid 1 1 1, pos 0 0 0, node_neighbors
>> =======================================================
>> = espresso_logo.tcl =
>> =======================================================
>>
>> cup: bottom_rad 2.38732
>> handle rad=1.41421 circ=6.66431 beta -0.229935
>> saucer: r_out=4.77465, r_in=2.38732, rings=3
>> Parts: total 194, cup 124, saucer 70 (valency -1.77143)
>> #particles realloc 224
>> CUDA error: no CUDA-capable device is detected
>>
>> Plus, we have a lot text printed on the screen that, indeed, we are not able
>> to understand. Please, find attached the file that contains part of the
>> text.
>>
>> Would you be so kind and give us some suggestions to improve our
>> configuration file for a better installation?
>>
>> Thank you so much in advance.
>>
>> Best,
>> Edith
>>
>>
>>
>>
>> On Tue, Dec 16, 2014 at 10:50 PM, Axel Arnold <address@hidden>
>> wrote:
>>>
>>>
>>> On 16.12.2014, at 17:34, Edith Euan <address@hidden> wrote:
>>>
>>> Unfortunately, with this configuration file, we are able to compile
>>> Espresso without problem, but it doesn't run correctly any test code.
>>>
>>>
>>> Could you be a bit more specific? What means “doesn’t run correctly”?
>>> Which "test code”? The problem is unlikely with the myconfig, since our CI
>>> also uses such a maximal set, and it works.
>>>
>>> Note that you also enabled all debug output and expensive debug checks,
>>> so you will be flooded with debug output that you will not be able to
>>> interpret, but makes your simulation terribly slow.
>>>
>>> Axel
>>>
>>> ------------------------------------------------
>>> JP Dr. Axel Arnold
>>> ICP, Universität Stuttgart
>>> Allmandring 3
>>> 70569 Stuttgart, Germany
>>> Email: address@hidden
>>> Phone: +49 711 685 67609
>>
>>
>> --
>> Dr. Edith Cristina Euán-Díaz
>> ---------------------------------------------
>> address@hidden
>> ---------------------------------------------
>> Postdoctoral Fellow
>> Condensed Matter Theory Group
>> University of Antwerp
>> Groenenborgerlaan 171
>> BE-2020 Antwerp, Belgium
>> ---------------------------------------------
>> T: +32 3 265 35 42 3527
>>