1 i synthesize instructions events 2 b synthesize branches events (branch misses for Arm SPE) 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 (incl. PSB events for Intel PT) 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 f synthesize first level cache events 13 m synthesize last level cache events 14 M synthesize memory events 15 t synthesize TLB events 16 a synthesize remote access events 17 g synthesize a call chain (use with i or x) 18 G synthesize a call chain on existing event records 19 l synthesize last branch entries (use with i or x) 20 L synthesize last branch entries on existing event records 21 s skip initial number of events 22 q quicker (less detailed) decoding 23 Z prefer to ignore timestamps (so-called "timeless" decoding) 24 25 The default is all events i.e. the same as --itrace=ibxwpe, 26 except for perf script where it is --itrace=ce 27 28 In addition, the period (default 100000, except for perf script where it is 1) 29 for instructions events can be specified in units of: 30 31 i instructions 32 t ticks 33 ms milliseconds 34 us microseconds 35 ns nanoseconds (default) 36 37 Also the call chain size (default 16, max. 1024) for instructions or 38 transactions events can be specified. 39 40 Also the number of last branch entries (default 64, max. 1024) for 41 instructions or transactions events can be specified. 42 43 Similar to options g and l, size may also be specified for options G and L. 44 On x86, note that G and L work poorly when data has been recorded with 45 large PEBS. Refer linkperf:perf-intel-pt[1] man page for details. 46 47 It is also possible to skip events generated (instructions, branches, transactions, 48 ptwrite, power) at the beginning. This is useful to ignore initialization code. 49 50 --itrace=i0nss1000000 51 52 skips the first million instructions. 53 54 The 'e' option may be followed by flags which affect what errors will or 55 will not be reported. Each flag must be preceded by either '+' or '-'. 56 The flags are: 57 o overflow 58 l trace data lost 59 60 If supported, the 'd' option may be followed by flags which affect what 61 debug messages will or will not be logged. Each flag must be preceded 62 by either '+' or '-'. The flags are: 63 a all perf events 64 65 If supported, the 'q' option may be repeated to increase the effect. 66