| b.liu | e958203 | 2025-04-17 19:18:16 +0800 | [diff] [blame] | 1 | i synthesize instructions events |
| 2 | b synthesize branches events |
| 3 | c synthesize branches events (calls only) |
| 4 | r synthesize branches events (returns only) |
| 5 | x synthesize transactions events |
| 6 | w synthesize ptwrite events |
| 7 | p synthesize power events |
| 8 | o synthesize other events recorded due to the use |
| 9 | of aux-output (refer to perf record) |
| 10 | e synthesize error events |
| 11 | d create a debug log |
| 12 | g synthesize a call chain (use with i or x) |
| 13 | l synthesize last branch entries (use with i or x) |
| 14 | s skip initial number of events |
| 15 | |
| 16 | The default is all events i.e. the same as --itrace=ibxwpe, |
| 17 | except for perf script where it is --itrace=ce |
| 18 | |
| 19 | In addition, the period (default 100000, except for perf script where it is 1) |
| 20 | for instructions events can be specified in units of: |
| 21 | |
| 22 | i instructions |
| 23 | t ticks |
| 24 | ms milliseconds |
| 25 | us microseconds |
| 26 | ns nanoseconds (default) |
| 27 | |
| 28 | Also the call chain size (default 16, max. 1024) for instructions or |
| 29 | transactions events can be specified. |
| 30 | |
| 31 | Also the number of last branch entries (default 64, max. 1024) for |
| 32 | instructions or transactions events can be specified. |
| 33 | |
| 34 | It is also possible to skip events generated (instructions, branches, transactions, |
| 35 | ptwrite, power) at the beginning. This is useful to ignore initialization code. |
| 36 | |
| 37 | --itrace=i0nss1000000 |
| 38 | |
| 39 | skips the first million instructions. |