blob: d3740c8f399bac1a03b75cc8484a40d21080fb60 [file] [log] [blame]
Adrian Hunter60b88d82015-07-17 19:33:44 +03001 i synthesize instructions events
Tan Xiaojun9f74d772020-05-30 20:24:41 +08002 b synthesize branches events (branch misses for Arm SPE)
Adrian Hunter60b88d82015-07-17 19:33:44 +03003 c synthesize branches events (calls only)
4 r synthesize branches events (returns only)
5 x synthesize transactions events
Adrian Hunter3bdafdf2017-05-26 11:17:24 +03006 w synthesize ptwrite events
Adrian Hunter70d110d2017-05-26 11:17:25 +03007 p synthesize power events
Adrian Hunter181ebb52019-08-06 11:46:03 +03008 o synthesize other events recorded due to the use
9 of aux-output (refer to perf record)
Adrian Hunter60b88d82015-07-17 19:33:44 +030010 e synthesize error events
11 d create a debug log
Tan Xiaojun9f74d772020-05-30 20:24:41 +080012 f synthesize first level cache events
13 m synthesize last level cache events
14 t synthesize TLB events
15 a synthesize remote access events
Adrian Hunter60b88d82015-07-17 19:33:44 +030016 g synthesize a call chain (use with i or x)
Adrian Hunter1c5c25b2020-04-01 13:16:05 +030017 G synthesize a call chain on existing event records
Adrian Hunter601897b2015-09-25 16:15:39 +030018 l synthesize last branch entries (use with i or x)
Adrian Hunterec90e422020-04-29 18:07:46 +030019 L synthesize last branch entries on existing event records
Andi Kleend1706b32016-03-28 10:45:38 -070020 s skip initial number of events
Adrian Hunter51971532020-07-10 18:11:02 +030021 q quicker (less detailed) decoding
Adrian Hunter60b88d82015-07-17 19:33:44 +030022
Andi Kleen4eb06812018-09-20 11:05:37 -070023 The default is all events i.e. the same as --itrace=ibxwpe,
24 except for perf script where it is --itrace=ce
Adrian Hunter60b88d82015-07-17 19:33:44 +030025
Andi Kleen4eb06812018-09-20 11:05:37 -070026 In addition, the period (default 100000, except for perf script where it is 1)
27 for instructions events can be specified in units of:
Adrian Hunter60b88d82015-07-17 19:33:44 +030028
29 i instructions
30 t ticks
31 ms milliseconds
32 us microseconds
33 ns nanoseconds (default)
34
35 Also the call chain size (default 16, max. 1024) for instructions or
36 transactions events can be specified.
Adrian Hunter601897b2015-09-25 16:15:39 +030037
38 Also the number of last branch entries (default 64, max. 1024) for
39 instructions or transactions events can be specified.
Andi Kleend1706b32016-03-28 10:45:38 -070040
Adrian Hunter43358d92020-04-29 18:07:50 +030041 Similar to options g and l, size may also be specified for options G and L.
42 On x86, note that G and L work poorly when data has been recorded with
43 large PEBS. Refer linkperf:perf-intel-pt[1] man page for details.
44
Adrian Hunter3bdafdf2017-05-26 11:17:24 +030045 It is also possible to skip events generated (instructions, branches, transactions,
Adrian Hunter70d110d2017-05-26 11:17:25 +030046 ptwrite, power) at the beginning. This is useful to ignore initialization code.
Andi Kleend1706b32016-03-28 10:45:38 -070047
48 --itrace=i0nss1000000
49
50 skips the first million instructions.
Adrian Huntercb971432020-07-10 18:10:57 +030051
52 The 'e' option may be followed by flags which affect what errors will or
53 will not be reported. Each flag must be preceded by either '+' or '-'.
54 The flags are:
55 o overflow
56 l trace data lost
Adrian Hunter935aac22020-07-10 18:10:59 +030057
58 If supported, the 'd' option may be followed by flags which affect what
59 debug messages will or will not be logged. Each flag must be preceded
60 by either '+' or '-'. The flags are:
61 a all perf events
Adrian Hunter51971532020-07-10 18:11:02 +030062
63 If supported, the 'q' option may be repeated to increase the effect.