Archives Discussions

[cguenther]

Hi there. I have a really strange problem and i hope you can help me.

I have a shrinked kernel to my problem:

__kernel

void softwareRender()

{

          __private float zeroDotFive;

          zeroDotFive = 0.5f;

  printf( (__constant char*) "%f \n", zeroDotFive);

}

The output of the printf is for every time "0.0000000". Passing data from shared GL VBOs as float is no problem. But when i want to multiply them with 0.5, everything gets 0.00000. The CodeXL 0.93 also verifies that the value of zeroDotFive is zero.

I am using Ubuntu 12.04 with kernel version 3.2.0-31-generic. I use the AMD APP SDK 2.7 and the AMD Catalyst™ 12.8 Proprietary Linux x86 Display Driver. I tried also the 12.9 beta one, and the fglrx driver of the partner Ubuntu repo and both with the same results.

The same Program and Kernel compiled with mingw runs in Windows 7 x64 with Catalyst 12.8 without this problem. Nvidia GPUs have no problem to execute this Kernel in 64bit linux and Windows.

[MicahVillmow · Accepted solution]

Thanks for reporting this, we have fixed this internally as of yesterday, so it will take a month or so to propagate into the catalyst beta's.

[cguenther]

Thanks for the quick response and the internal development information. This helps me a lot, that i have not to try to fix this problem myself.

[cguenther]

I am using currently the 12.10 driver version and it seems to show the same effect. Should i use the 12.11 beta for it?

[cguenther]

Using the AMD catalyst 13.1 has finially fixed all regarding problems.

Thanks a lot.

Christian Günther

[himanshu_gautam]

Thanks for the confirmation. Nice to hear that your issue was resolved.