Archives Discussions

[pwvdendr]

The attached kernel compiles and runs without any problem on CPU. On GPU though, the compiler fails to compile it, but does not return any errors. When running it through the KernelAnalyzer, it seems compilation fails for any platform that I tried (Tahiti, Cayman, Cedar, ...). This is quite serious. Does anyone have any idea what could cause this, or how to work around it? I'm using Catalyst 12.8 and have a HD7950, if it matters.

[drallan · Accepted solution]

Hi pwvdendr,

This version is failing because it allocates about 150,000 bytes of local memory. The maximum per work group is 32768 bytes. (Tahiti will use its full 64K when two work groups run on the same CU). When the local memory is set below the 32K limit, it compiles on Tahiti as well, congratulations.

Certainly, the compiler should report the error, it's pretty easy for it to check the local memory allocation, I would  think.

[pwvdendr]

Aha, well remarked again. Didn't think of memory issues as I thought that was only checked on runtime. Thanks a bunch!