[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CABob6irSm=qfVGZMBk_GrWNT8C8bXohwf4SE2zn6QkTNu2Og5Q@mail.gmail.com>
Date: Fri, 13 Apr 2012 06:57:39 +0200
From: Lukas Odzioba <lukas.odzioba@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: OpenCL runtime errors
2012/4/11 magnum <john.magnum@...hmail.com>:
> Most current OpenCL formats
> (I think all except mine) use this to determine the device's maximum
> local worksize:
>
> clGetDeviceInfo(devices[gpu_id], CL_DEVICE_MAX_WORK_GROUP_SIZE,
> sizeof(max_group_size), &max_group_size, NULL);
>
> ...but this figure does not help you much. It just shows the maximum
> supported worksize for any (lean) kernel on this device. I use this instead:
>
> clGetKernelWorkGroupInfo(crypt_kernel, devices[gpu_id],
> CL_KERNEL_WORK_GROUP_SIZE, sizeof(max_group_size), &max_group_size, NULL);
>
> This one tells us the maximum local worksize for *this very kernel* on
> this device. That is, the OpenCL implementation uses the resource
> requirements of the kernel (register usage etc.) to determine the max
> usable local worksize.
Thank you for proposing it. I'll change it in other formats, and send
you patch today.
BTW I am trying to clean find_best_workgroup() because its code is
almost exactly the same in every format.
Lukas
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.
