[Halld-offline] SIMD Troubles

[David Lawrence]

Hi Curtis,

     The auto-detection approach has been what was in use previously and 
we seem to have gotten bitten by it a couple of times recently. Our 
thoughts were to switch the default to one that we had the highest 
confidence of working on all systems (no SIMD) and force people who want 
to use it to take some action (make ENABLE_SIMD=yes). Folks should then 
at least get a working system when typing simply "make" for the time being.

     This will be on the agenda for discussion at the Offline Software 
meeting next week where we can discuss it more.

Regards,
-Dave

On 3/2/11 6:14 PM, Curtis A. Meyer wrote:
> Hi Simon -
>
>     these are such a speed up on 64-bit machines that it would be a bit
> of a shame to not have them default on on those boxes. Is it worth
> considering having the make scripts determine 32- or 64-bit OS and
> then do the appropriate thing with the SIMD things?
>
>       Curtis
> On 3/2/11 6:08 PM, Simon Taylor wrote:
>> Hi.
>>
>> Our suspicion is that there is an alignment problem on 32-bit systems
>> with regard to the SIMD instructions;  Dave looked into this some time
>> ago and it is not clear to us how to fix it.
>>
>> I've checked in a change to Makefile.SIMD that changes the default from
>> "SIMD on" to "SIMD off".  To get the SIMD instructions, one would
>> now need to do "make ENABLE_SIMD=yes".
>>
>> Simon
>>
>> Matthew Shepherd wrote:
>>> Hi all,
>>>
>>> It seems that the BMS system doesn't properly understand our SIMD capabilities on the machines here at Indiana.  If we do a default build, then we get a segmentation fault at the first DVector2 operation.  If we build with DISABLE_SIMD=1 then this segfault is avoided.
>>>
>>> This seems to point to two causes:
>>>
>>> (1) there is a bug in the SIMD implementation of DVector2
>>> (2) our machines are not capable of handling current SIMD code
>>>
>>> (1) seems unlikely since other people are using the code.  Assuming it is (2), how do we properly diagnose and fix it?
>>>
>>> -Matt
>>>
>>>
>>>
>>> _______________________________________________
>>> Halld-offline mailing list
>>> Halld-offline at jlab.org
>>> https://mailman.jlab.org/mailman/listinfo/halld-offline
>>>
>> _______________________________________________
>> Halld-offline mailing list
>> Halld-offline at jlab.org
>> https://mailman.jlab.org/mailman/listinfo/halld-offline
>>
>