Search:

Type: Posts; User: guillona

Search: Search took 0.01 seconds.

  1. Replies
    3
    Views
    615

    Hi Piyush, The best way to do this is to break...

    Hi Piyush,

    The best way to do this is to break apart your single kernel into multiple kernels so that you can measure each component separately. You might also consider the use of profiling tools...
  2. Replies
    3
    Views
    666

    Hi, So here is something to think of. You can...

    Hi,

    So here is something to think of. You can slice up your execution of data many ways, but ultimately the total amount of computation will remain the same, however memory accesses and load...
  3. Replies
    3
    Views
    692

    Hello From An OpenCL Committee Member!

    Hello OpenCL forum!

    My name is AJ Guillon, and I recently joined Khronos and the OpenCL standards committee. I wanted to announce to the forum that I will try to actively monitor discussions...
  4. Replies
    0
    Views
    944

    OpenCL YouTube Training Videos

    Hi everyone,

    I have uploaded a couple YouTube training videos on OpenCL. I plan to upload another video by the weekend that covers GPU architecture.

    http://youtu.be/8D6yhpiQVVI...
  5. This is one of my major issues with OpenCL. I am...

    This is one of my major issues with OpenCL. I am currently working on a library and header files that will deal with this issue for you, by breaking OpenCL apart into discrete units of functionality...
  6. Replies
    32
    Views
    16,179

    Sticky: I just want to let you know that I have posted...

    I just want to let you know that I have posted another article regarding SPIR, and how it might fit with the standard overall. I have a few more articles to write on OpenCL 2.0 before I am finished...
  7. Replies
    4
    Views
    5,227

    Hi. I have written some feedback for SPIR on my...

    Hi. I have written some feedback for SPIR on my blog. I have to admit that I do not have extensive experience working on a compiler, but I have addressed concerns that I have regarding...
  8. Replies
    32
    Views
    16,179

    Sticky: Hi everyone, I decided to provide detailed...

    Hi everyone, I decided to provide detailed feedback in the form of blog articles. You can read them here: https://blog.ajguillon.com/

    The first relevant blog is:...
  9. Replies
    32
    Views
    16,179

    Sticky: Hi everyone. Just a quick note that I have...

    Hi everyone.

    Just a quick note that I have read the OpenCL 2.0 specification and have made extensive notes to provide feedback. I am currently writing it up, and it will take me a week or so to...
  10. Re: OpenCL Spec: What Happens on Event/Command Failure

    Well, I don't know about your programs, but I cannot assume mine are so well-behaved.

    Consider a situation in which there are two completely independent event graphs, one is operating on memory A,...
  11. Re: OpenCL Spec: What Happens on Event/Command Failure

    Hi.

    The issue is from OpenCL 1.2 Section 5.9. The execution status can be an error code (must be negative), and the standard specifies:


    Later it specifies:



    Given these two pieces of...
  12. OpenCL Spec: What Happens on Event/Command Failure

    Hi,

    I am having a hard time understanding the specification right now. What is supposed to happen if an enqueued command fails? What function call reports the error? What happens if I use...
  13. Replies
    3
    Views
    2,414

    Re: Interpretation of CL_DEVICE_VENDOR_ID

    Consider independent processes that are attempting to load-balance in a multi-device configuration. There is no order of devices guaranteed so far as I know, so it is possible that multiple...
  14. Replies
    0
    Views
    1,569

    OpenCL Spec: release/retain errors

    The specification lists a few unusual possible errors (e.g. CL_OUT_OF_RESOURCES, CL_OUT_OF_HOST_MEMORY) for release/retain operations. I think that the specification should guarantee that these...
  15. Replies
    3
    Views
    2,414

    Interpretation of CL_DEVICE_VENDOR_ID

    I'd like to request clarification on the spec. I have posted to AMD's Stream SDK forums as well here: http://forums.amd.com/devforum/messagev ... did=145374

    What is the intended purpose of...
  16. Replies
    2
    Views
    3,782

    Global Barriers?

    Currently I'm writing an algorithm where I need a single (very quick) global barrier, and then processing can resume in parallel as it was... so basically I have a large amount of parallel work, then...
  17. Re: Possible Specification Ambiguity - Host Side Vector Types

    Thank you for your response.

    This is a rather dangerous omission on the part of the spec, and hopefully it can be corrected within the specification itself. The spec should say something about...
  18. Possible Specification Ambiguity - Host Side Vector Types

    Device-side vector types can be accessed like struct types. e.g. float2 T.x, T.y.

    However, the spec is silent (or I don't see where it is mentioned) on how this maps to the host-side types e.g....
  19. Replies
    4
    Views
    5,278

    Lack of double Support in C Types

    I have been reading the released OpenCL specification. I have noticed that double is a reserved type, but not supported by the draft specification. In particular, in section 6.1.4 it is listed as...
Results 1 to 19 of 19