| 499 | Compiling codes with '-g' option to tells the compiler to generate full debugging information in the object file. |
| 500 | {{{#!bash |
| 501 | [fuji@cypress1 ~]$ icc -g -o mytest mytest.c |
| 502 | }}} |
| 503 | |
| 504 | ==== Run and Collect Information ==== |
| 505 | Start an interactive job, |
| 506 | {{{#!bash |
| 507 | [fuji@cypress1 ~]$ idev |
| 508 | }}} |
| 509 | To collect information, run the code, for example, |
| 510 | {{{#!bash |
| 511 | [fuji@cypress1 ~]$ amplxe-cl -collect hotspot ./mytest |
| 512 | }}} |
| 513 | This will create a holder like '''r000hs'''. |
| 514 | |
| 515 | '''-collect ''' options |
| 516 | |
| 517 | ||= concurrency =|| Concurrency analysis || |
| 518 | ||= hotspots =|| Hotspots analysis || |
| 519 | ||= lightweight-hotspots =|| Lightweight Hotspots analysis || |
| 520 | ||= locksandwaits =|| Locks and Waits analysis || |
| 521 | |
| 522 | To show results, for example, |
| 523 | {{{#!bash |
| 524 | [fuji@cypress1 ~]$ amplxe-cl -report hotspot -r r000hs |
| 525 | }}} |
| 526 | |
| 527 | '''-report ''' options |
| 528 | |
| 529 | ||= summary =|| Display data for the overall performance of the target. || |
| 530 | ||= hotspots =|| Display functions with the highest CPU time. || |
| 531 | ||= wait-time =|| Display Wait time. || |
| 532 | ||= perf =|| Display performance data for each module of the target. || |
| 533 | ||= perf-detail =|| Display performance data for each function of the target. || |
| 534 | ||= callstacks =|| Display CPU or Wait time for call stacks. || |
| 535 | ||= top-down =|| Display a call tree for your target application and provide CPU and Wait time for each function. || |
| 536 | ||= gprof-cc =|| Display CPU or wait time in the gprof-like format. || |
| 537 | |