[CIG-SEISMO] CPML and GPU

Zhang, Chang-Hua chhzhang at phys.ksu.edu
Wed Apr 29 07:36:47 PDT 2015


Hi Dimitri, 

Sorry that I am slow for merging my cpml implementation to the devel version. I don't want to change the devel to fit my version. One of big issues is that I only focus on the forward problem. So my changed "devel" generates a lot bugs in my tests.  

Is it better that I send my version to one of you guys to merge to devel? I do think it is a good piece of work. 

Best Regards, 
Chang-hua   


________________________________________
From: Dimitri Komatitsch [komatitsch at lma.cnrs-mrs.fr]
Sent: Tuesday, April 28, 2015 4:31 PM
To: cig-seismo at geodynamics.org; xin wang
Cc: Zhang, Chang-Hua
Subject: Re: [CIG-SEISMO] CPML and GPU

Dear Xin,

Thanks for your message. That is already done by ZhangChang-Hua (but not
committed yet I think?), thus let me cc him.

Chang-Hua, could you please commit it when you have time?

Thanks,
Dimitri.

On 04/28/2015 04:12 PM, xin wang wrote:
> Dear SPECFEM3D developer,
>
> I realized GPU mode cannot use when CPML are added in the model. It is
> really a time consuming to run the model only CPUs are used. I am
> wondering about this issue. i do not know if it is possible to run them
> together and if there any plan to add this function in new release
> version? Thanks in advance.
>
> Cheers,
> xin
>
> _______________________________________________
> CIG-SEISMO mailing list
> CIG-SEISMO at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-seismo
>

--
Dimitri Komatitsch
CNRS Research Director (DR CNRS), Laboratory of Mechanics and Acoustics,
UPR 7051, Marseille, France    http://komatitsch.free.fr


More information about the CIG-SEISMO mailing list