[gmx-users] how can i install sse4.1 and avx_256

2014-10-23 Thread Yaser Hosseini
hi

Reading file md.tpr, VERSION 4.6.5 (single precision)
Using 4 MPI threads
Compiled acceleration: SSE4.1 (Gromacs could use AVX_256 on this machine,
which is better)

how can i use this features for better mdrun speed.

thanke you.
-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.


Re: [gmx-users] how can i install sse4.1 and avx_256

2014-10-23 Thread Milan Melicherčík
Hello,
you need to recompile Gromacs with -DGMX_SIMD=AVX_256 flag. Or not set it and 
cmake should detect and use the best version automatically. 
http://www.gromacs.org/Documentation/Installation_Instructions_4.6
Cheers.

Milan

On Thursday 23 October 2014 13:32:31 Yaser Hosseini wrote:
 hi
 
 Reading file md.tpr, VERSION 4.6.5 (single precision)
 Using 4 MPI threads
 Compiled acceleration: SSE4.1 (Gromacs could use AVX_256 on this machine,
 which is better)
 
 how can i use this features for better mdrun speed.
 
 thanke you.
-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.


Re: [gmx-users] how can i install sse4.1 and avx_256

2014-10-23 Thread Justin Lemkul



On 10/23/14 6:02 AM, Yaser Hosseini wrote:

hi

Reading file md.tpr, VERSION 4.6.5 (single precision)
Using 4 MPI threads
Compiled acceleration: SSE4.1 (Gromacs could use AVX_256 on this machine,
which is better)

how can i use this features for better mdrun speed.



The level of optimization is automatically detected by cmake, but if you're 
compiling for a cluster that may have compute nodes with different 
configurations from the head node, you need to explicitly set 
-DGMX_CPU_ACCELERATION (note this has been changed to -DGMX_SIMD in 5.0) to the 
desired level of optimization.


-Justin

--
==

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalem...@outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==
--
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.