[gmx-developers] Alingment & NULL pointers

[Berk Hess]

Hi,

They should not only be aligned at the start, but also padded at the end 
up to SIMD register width alignment such that SIMD loops can read and 
write without masking at the end.

Cheers,

Berk

On 2015-09-04 12:01, Erik Lindahl wrote:
> Hi,
>
> Yes, aligned allocators should be on our todo-list :-)
>
> Cheers,
>
> ERik
>
>> On 04 Sep 2015, at 11:57, David van der Spoel <spoel at xray.bmc.uu.se> wrote:
>>
>> With the impending requirement of C++11 the use of NULL is deprecated (as I just found out). However we still are dependent on smalloc and friends to give us aligned memory for the inner loops, which means we have to deal with malloc and free for handling pointers.
>>
>> The stackoverflow page below suggests a way of aligning std::vector under the hood, which would have the advantage that we could use std::vector without having to worry about memory management. Is that something we should consider?
>>
>> http://stackoverflow.com/questions/8456236/how-is-a-vectors-data-aligned
>>
>> -- 
>> David van der Spoel, Ph.D., Professor of Biology
>> Dept. of Cell & Molec. Biol., Uppsala University.
>> Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205.
>> spoel at xray.bmc.uu.se    http://folding.bmc.uu.se
>> -- 
>> Gromacs Developers mailing list
>>
>> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_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-developers or send a mail to gmx-developers-request at gromacs.org.