On 08/21/2013 11:03 PM, Peter Colberg wrote: > Right now it defines an array of size 1, which is specific to the > scalarwave test. The backend should forward the sizes of constant > and local memory parameters to the LLVM pass, as these are not > in the OpenCL kernel code but queried from the OpenCL buffer sizes.
OK. Got it. So you'd like to invoke the "local arg to a global buffer" pass just before the actual code generation is done when you know the dynamically set sizes. Should be OK to call LLVM directly there, I think. > As a consequence, the kernel cache needs to account for the sizes > of these parameters. A cached kernel must not be reused if a > buffer passed as a constant/local argument has a larger size > than the buffer passed with a prior kernel invocation. True. The kernel compiler cache has other known "false hit" bugs too. At least the options in the clBuildProgram are not included in the "cache key". It needs some rework. -- --Pekka ------------------------------------------------------------------------------ Introducing Performance Central, a new site from SourceForge and AppDynamics. Performance Central is your source for news, insights, analysis and resources for efficient Application Performance Management. Visit us today! http://pubads.g.doubleclick.net/gampad/clk?id=48897511&iu=/4140/ostg.clktrk _______________________________________________ pocl-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/pocl-devel
