Forums - Who maintains OpenCL.so for Adreno 650 on Ubuntu 18.04?

2 posts / 0 new
or Register
Last post
Who maintains OpenCL.so for Adreno 650 on Ubuntu 18.04?
Matt
Join Date: 21 Sep 23
Posts: 1
Posted: Fri, 2023-11-17 07:03
Top

I am trying to write an application that uses clFFT. When I use clFFT on an NVIDIA card, it runs correctly. When I use clFFT on the Adreno 650, I get incorrect results. I've done some debugging and found that the OpenCL barrier() function doesn't seem to be working correctly on the Adreno 650. Who would I talk to about possibly fixing this, possibly under a paid contract?

  • Up0
  • Down0
Re: Who maintains OpenCL.so for Adreno 650 on Ubuntu 18.04?
nelson7896mari
Join Date: 14 May 24
Posts: 1
Posted: Tue, 2024-05-14 02:52
Top

 

For addressing OpenCL-related issues, especially regarding the clFFT library and its compatibility with specific hardware like the Adreno 650 GPU, you would typically reach out to the relevant support channels provided by the GPU manufacturer or the developer of the OpenCL implementation.
 
Here are some steps you can take to address this issue:
 
Contact Qualcomm:
Since the Adreno 650 GPU is developed by Qualcomm, you can start by reaching out to Qualcomm's support channels. They may have dedicated developer support or forums where you can report issues related to OpenCL functionality on their GPUs. They might also offer paid support options or consulting services for resolving such issues.
 
OpenCL Community Forums:
Engage with the OpenCL community forums or mailing lists. Developers and experts in the OpenCL community may have insights or workarounds for issues related to specific GPU implementations. They can also provide guidance on how to approach GPU vendors for support or bug fixes.
 
Vendor-Specific Forums:
Check if Qualcomm or the manufacturer of your device (if different from Qualcomm) has developer forums or support portals specifically for their GPU products. These forums often have sections dedicated to OpenCL development and troubleshooting.
 
Consider Consulting Services:
If the issue is critical and time-sensitive, you might consider hiring OpenCL experts or consultants who have experience working with GPU-specific optimizations and debugging. They can analyze your code, identify potential issues, and work towards resolving them, possibly under a paid contract or consulting arrangement.
 
Report Bugs:
If you believe the issue is a bug in the OpenCL implementation for Adreno 650, make sure to report it to Qualcomm or the relevant GPU manufacturer through their official bug reporting channels. Provide detailed information, including code samples, debug logs, and steps to reproduce the issue.
 
Explore Alternative Approaches:
While waiting for a resolution or workaround, consider exploring alternative approaches or libraries for your application's FFT computations. There are other FFT libraries available that may offer better compatibility with a wide range of GPU hardware.
 
By combining these strategies, you can increase Sedgwick the chances of resolving the barrier() function issue in clFFT on the Adreno 650 GPU or finding suitable workarounds to ensure correct functionality in your application.
  • Up0
  • Down0
or Register

Opinions expressed in the content posted here are the personal opinions of the original authors, and do not necessarily reflect those of Qualcomm Incorporated or its subsidiaries (“Qualcomm”). The content is provided for informational purposes only and is not meant to be an endorsement or representation by Qualcomm or any other party. This site may also provide links or references to non-Qualcomm sites and resources. Qualcomm makes no representations, warranties, or other commitments whatsoever about any non-Qualcomm sites or third-party resources that may be referenced, accessible from, or linked to this site.