I've spent quite some time debugging it, until I realised that this is
correct behavior, as long as memory access uses index "i", which comes from
here:
data.verlet_list_i = new int[10]{0, 0, 0, 1, 1, 2, 3, 5, 7, 7};
You probably meant using "v_index" when accessing data.particles_fx[].
On Fri, Jun 16, 2017 at 2:05 AM, Dmitry Babokin <[email protected]> wrote:
> I can reproduce and I'm looking at this issue. Though I'm currently on
> vacation, so it may took a little longer that usually.
>
> On Tue, Jun 13, 2017 at 4:28 AM, <[email protected]> wrote:
>
>> Hi,
>>
>> We are writing a molecular dynamics simulation code that will be used at
>> Los Alamos National Laboratory. So far with ISPC we managed to hugely
>> improve performance, but currently I am facing an odd issue while
>> optimizing a function: there seems to be an issue with subtraction. Here's
>> the output of the attached program that reproduces the error. I am adding
>> the change, fx, to the variable p_fx:
>>
>> p_fx [0.002800,0.002800,0.002800,0.037144,0.037144,0.035400,0.259
>> 137,0.021419]
>> -=
>> fx [-0.136454,0.493969,-0.153606,0.042538,0.005820,-0.114653,0.
>> 081846,-0.008912]
>> =
>> p_fx [0.156406,0.156406,0.156406,0.031324,0.031324,0.150053,0.177
>> 291,0.030331]
>>
>> In the first three slots we get the same value, although the subtracted
>> value is different.
>> What do you recommend, how could we solve this issue? Thanks!
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "Intel SPMD Program Compiler Users" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to [email protected].
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>
--
You received this message because you are subscribed to the Google Groups
"Intel SPMD Program Compiler Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
