Changes between Version 39 and Version 40 of OpenCL_Memo
- Timestamp:
- Dec 12, 2011 7:43:51 PM (13 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
OpenCL_Memo
v39 v40 3 3 In the section 6.7.2 (187 page) in the OpenCL Specification Version: 1.1 (Revision: 33), "__attribute__((vec_type_hint(<typen>))" is described. This hint controls the autovectorizer in the compiler for OpenCL C. I tested this feature by dumping the assembly code for a kernel targeted for AVX instructions with Intel SDK (version 1.5). 4 4 5 || || lines || performance ||6 || no hint || 1567 || 23.1 sec ||7 || with the hint || 333 || 43.0 sec ||5 || || lines || 6 || no hint || 1567 || 7 || with the hint || 333 || 8 8 9 This hint amazingly reduces the size of the generated kernel code. But... OK, a shorter code is not always a faster code. Still working... 9 This hint amazingly reduces the size of the generated kernel code. 10 Still working... 10 11 11 12 == How to use "ioc" command equipped with Intel SDK. ==