[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 2 Jun 2015 21:48:25 +0200
From: Agnieszka Bielec <bielecagnieszka8@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: PHC: Parallel in OpenCL
2015-06-02 21:37 GMT+02:00 Solar Designer <solar@...nwall.com>:
> On Tue, Jun 02, 2015 at 08:36:36PM +0200, Agnieszka Bielec wrote:
>> the speed decreases after change this loop:
>>
>> for (int i = 0; i < 16; i++) {
>> t1 = k[i] + w[i] + h + Sigma1(e) + Ch(e, f, g);
> [...]
>> a = t1 + t2;
>> }
>>
>> to
>>
>> for (int i = 0; i < 10; i++) {
>> t1 = k[i] + w[i] + h + Sigma1(e) + Ch(e, f, g);
> [...]
>> a = t1 + t2;
>> }
>>
>> for (int i = 10; i < 15; i++) {
>> t1 = k[i] + h + Sigma1(e) + Ch(e, f, g);
> [...]
>> a = t1 + t2;
>> }
>>
>> t1 = k[15] + w[15] + h + Sigma1(e) + Ch(e, f, g);
> [...]
>> a = t1 + t2;
>
> What were you trying to accomplish with this experiment?
I must to change first loop to the second one if I want to remove +
w[i] instructions if I know that they are equal to 0
>> in the same branch parallel_opt as I said you on IRC.
>> this can't be fault of split kernels
>
> What does this have to do with split kernels?
I think that nothing but
2015-06-02 5:37 GMT+02:00 Lukas Odzioba <lukas.odzioba@...il.com>:
> I just compared code on two branches and I don't think that what you
> did it is the proper way of doing split kernel...
maybe I should reorganize my previous mail
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
