1perf-report(1) 2============== 3 4NAME 5---- 6perf-report - Read perf.data (created by perf record) and display the profile 7 8SYNOPSIS 9-------- 10[verse] 11'perf report' [-i <file> | --input=file] 12 13DESCRIPTION 14----------- 15This command displays the performance counter profile information recorded 16via perf record. 17 18OPTIONS 19------- 20-i:: 21--input=:: 22 Input file name. (default: perf.data unless stdin is a fifo) 23 24-v:: 25--verbose:: 26 Be more verbose. (show symbol address, etc) 27 28-d:: 29--dsos=:: 30 Only consider symbols in these dsos. CSV that understands 31 file://filename entries. 32-n:: 33--show-nr-samples:: 34 Show the number of samples for each symbol 35 36--showcpuutilization:: 37 Show sample percentage for different cpu modes. 38 39-T:: 40--threads:: 41 Show per-thread event counters 42-c:: 43--comms=:: 44 Only consider symbols in these comms. CSV that understands 45 file://filename entries. 46-S:: 47--symbols=:: 48 Only consider these symbols. CSV that understands 49 file://filename entries. 50 51--symbol-filter=:: 52 Only show symbols that match (partially) with this filter. 53 54-U:: 55--hide-unresolved:: 56 Only display entries resolved to a symbol. 57 58-s:: 59--sort=:: 60 Sort by key(s): pid, comm, dso, symbol, parent. 61 62-p:: 63--parent=<regex>:: 64 regex filter to identify parent, see: '--sort parent' 65 66-x:: 67--exclude-other:: 68 Only display entries with parent-match. 69 70-w:: 71--column-widths=<width[,width...]>:: 72 Force each column width to the provided list, for large terminal 73 readability. 74 75-t:: 76--field-separator=:: 77 78 Use a special separator character and don't pad with spaces, replacing 79 all occurrences of this separator in symbol names (and other output) 80 with a '.' character, that thus it's the only non valid separator. 81 82-D:: 83--dump-raw-trace:: 84 Dump raw trace in ASCII. 85 86-g [type,min[,limit],order]:: 87--call-graph:: 88 Display call chains using type, min percent threshold, optional print 89 limit and order. 90 type can be either: 91 - flat: single column, linear exposure of call chains. 92 - graph: use a graph tree, displaying absolute overhead rates. 93 - fractal: like graph, but displays relative rates. Each branch of 94 the tree is considered as a new profiled object. + 95 96 order can be either: 97 - callee: callee based call graph. 98 - caller: inverted caller based call graph. 99 100 Default: fractal,0.5,callee. 101 102-G:: 103--inverted:: 104 alias for inverted caller based call graph. 105 106--pretty=<key>:: 107 Pretty printing style. key: normal, raw 108 109--stdio:: Use the stdio interface. 110 111--tui:: Use the TUI interface, that is integrated with annotate and allows 112 zooming into DSOs or threads, among other features. Use of --tui 113 requires a tty, if one is not present, as when piping to other 114 commands, the stdio interface is used. 115 116--gtk:: Use the GTK2 interface. 117 118-k:: 119--vmlinux=<file>:: 120 vmlinux pathname 121 122--kallsyms=<file>:: 123 kallsyms pathname 124 125-m:: 126--modules:: 127 Load module symbols. WARNING: This should only be used with -k and 128 a LIVE kernel. 129 130-f:: 131--force:: 132 Don't complain, do it. 133 134--symfs=<directory>:: 135 Look for files with symbols relative to this directory. 136 137-C:: 138--cpu:: Only report samples for the list of CPUs provided. Multiple CPUs can 139 be provided as a comma-separated list with no space: 0,1. Ranges of 140 CPUs are specified with -: 0-2. Default is to report samples on all 141 CPUs. 142 143-M:: 144--disassembler-style=:: Set disassembler style for objdump. 145 146--source:: 147 Interleave source code with assembly code. Enabled by default, 148 disable with --no-source. 149 150--asm-raw:: 151 Show raw instruction encoding of assembly instructions. 152 153--show-total-period:: Show a column with the sum of periods. 154 155-I:: 156--show-info:: 157 Display extended information about the perf.data file. This adds 158 information which may be very large and thus may clutter the display. 159 It currently includes: cpu and numa topology of the host system. 160 161-b:: 162--branch-stack:: 163 Use the addresses of sampled taken branches instead of the instruction 164 address to build the histograms. To generate meaningful output, the 165 perf.data file must have been obtained using perf record -b or 166 perf record --branch-filter xxx where xxx is a branch filter option. 167 perf report is able to auto-detect whether a perf.data file contains 168 branch stacks and it will automatically switch to the branch view mode, 169 unless --no-branch-stack is used. 170 171SEE ALSO 172-------- 173linkperf:perf-stat[1], linkperf:perf-annotate[1] 174