| 9 | | This hint amazingly reduces the size of the generated kernel code! Cool. |
| | 9 | This hint amazingly reduces the size of the generated kernel code! Cool! |
| | 10 | |
| | 11 | == How to use "ioc" command equipped with Intel SDK. == |
| | 12 | We need to set the environment variable INTELOCLSDKROOT |
| | 13 | |
| | 14 | {{{ |
| | 15 | export INTELOCLSDKROOT=/usr/lib64/OpenCL/vendors/intel |
| | 16 | }}} |
| | 17 | |
| | 18 | To dump the assembly code: |
| | 19 | {{{ |
| | 20 | ioc -input=kernel_file.cl -asm |
| | 21 | }}} |
| | 22 | |
| | 23 | == Dump IL/ISA with AMD SDK == |
| | 24 | Set the following the environment variable (APP Programming Guide August 2011, section 4.2 (63 page)). |
| | 25 | {{{ |
| | 26 | export GPU_DUMP_DEVICE_KERNEL=3 |
| | 27 | }}} |
| | 28 | |