tools/perf/Documentation/perf-report.txt

   1 perf-report(1)
   2 ==============
   3
   4 NAME
   5 ----
   6 perf-report - Read perf.data (created by perf record) and display the profile
   7
   8 SYNOPSIS
   9 --------
  10 [verse]
  11 'perf report' [-i <file> | --input=file]
  12
  13 DESCRIPTION
  14 -----------
  15 This command displays the performance counter profile information recorded
  16 via perf record.
  17
  18 OPTIONS
  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 -n::
  29 --show-nr-samples::
  30         Show the number of samples for each symbol
  31
  32 --showcpuutilization::
  33         Show sample percentage for different cpu modes.
  34
  35 -T::
  36 --threads::
  37         Show per-thread event counters.  The input data file should be recorded
  38         with -s option.
  39 -c::
  40 --comms=::
  41         Only consider symbols in these comms. CSV that understands
  42         file://filename entries.  This option will affect the percentage of
  43         the overhead column.  See --percentage for more info.
  44 --pid=::
  45         Only show events for given process ID (comma separated list).
  46
  47 --tid=::
  48         Only show events for given thread ID (comma separated list).
  49 -d::
  50 --dsos=::
  51         Only consider symbols in these dsos. CSV that understands
  52         file://filename entries.  This option will affect the percentage of
  53         the overhead column.  See --percentage for more info.
  54 -S::
  55 --symbols=::
  56         Only consider these symbols. CSV that understands
  57         file://filename entries.  This option will affect the percentage of
  58         the overhead column.  See --percentage for more info.
  59
  60 --symbol-filter=::
  61         Only show symbols that match (partially) with this filter.
  62
  63 -U::
  64 --hide-unresolved::
  65         Only display entries resolved to a symbol.
  66
  67 -s::
  68 --sort=::
  69         Sort histogram entries by given key(s) - multiple keys can be specified
  70         in CSV format.  Following sort keys are available:
  71         pid, comm, dso, symbol, parent, cpu, socket, srcline, weight, local_weight.
  72
  73         Each key has following meaning:
  74
  75         - comm: command (name) of the task which can be read via /proc/<pid>/comm
  76         - pid: command and tid of the task
  77         - dso: name of library or module executed at the time of sample
  78         - symbol: name of function executed at the time of sample
  79         - parent: name of function matched to the parent regex filter. Unmatched
  80         entries are displayed as "[other]".
  81         - cpu: cpu number the task ran at the time of sample
  82         - socket: processor socket number the task ran at the time of sample
  83         - srcline: filename and line number executed at the time of sample.  The
  84         DWARF debugging info must be provided.
  85         - srcfile: file name of the source file of the same. Requires dwarf
  86         information.
  87         - weight: Event specific weight, e.g. memory latency or transaction
  88         abort cost. This is the global weight.
  89         - local_weight: Local weight version of the weight above.
  90         - transaction: Transaction abort flags.
  91         - overhead: Overhead percentage of sample
  92         - overhead_sys: Overhead percentage of sample running in system mode
  93         - overhead_us: Overhead percentage of sample running in user mode
  94         - overhead_guest_sys: Overhead percentage of sample running in system mode
  95         on guest machine
  96         - overhead_guest_us: Overhead percentage of sample running in user mode on
  97         guest machine
  98         - sample: Number of sample
  99         - period: Raw number of event count of sample
 100
 101         By default, comm, dso and symbol keys are used.
 102         (i.e. --sort comm,dso,symbol)
 103
 104         If --branch-stack option is used, following sort keys are also
 105         available:
 106         dso_from, dso_to, symbol_from, symbol_to, mispredict.
 107
 108         - dso_from: name of library or module branched from
 109         - dso_to: name of library or module branched to
 110         - symbol_from: name of function branched from
 111         - symbol_to: name of function branched to
 112         - mispredict: "N" for predicted branch, "Y" for mispredicted branch
 113         - in_tx: branch in TSX transaction
 114         - abort: TSX transaction abort.
 115         - cycles: Cycles in basic block
 116
 117         And default sort keys are changed to comm, dso_from, symbol_from, dso_to
 118         and symbol_to, see '--branch-stack'.
 119
 120 -F::
 121 --fields=::
 122         Specify output field - multiple keys can be specified in CSV format.
 123         Following fields are available:
 124         overhead, overhead_sys, overhead_us, overhead_children, sample and period.
 125         Also it can contain any sort key(s).
 126
 127         By default, every sort keys not specified in -F will be appended
 128         automatically.
 129
 130         If --mem-mode option is used, following sort keys are also available
 131         (incompatible with --branch-stack):
 132         symbol_daddr, dso_daddr, locked, tlb, mem, snoop, dcacheline.
 133
 134         - symbol_daddr: name of data symbol being executed on at the time of sample
 135         - dso_daddr: name of library or module containing the data being executed
 136         on at the time of sample
 137         - locked: whether the bus was locked at the time of sample
 138         - tlb: type of tlb access for the data at the time of sample
 139         - mem: type of memory access for the data at the time of sample
 140         - snoop: type of snoop (if any) for the data at the time of sample
 141         - dcacheline: the cacheline the data address is on at the time of sample
 142
 143         And default sort keys are changed to local_weight, mem, sym, dso,
 144         symbol_daddr, dso_daddr, snoop, tlb, locked, see '--mem-mode'.
 145
 146 -p::
 147 --parent=<regex>::
 148         A regex filter to identify parent. The parent is a caller of this
 149         function and searched through the callchain, thus it requires callchain
 150         information recorded. The pattern is in the exteneded regex format and
 151         defaults to "\^sys_|^do_page_fault", see '--sort parent'.
 152
 153 -x::
 154 --exclude-other::
 155         Only display entries with parent-match.
 156
 157 -w::
 158 --column-widths=<width[,width...]>::
 159         Force each column width to the provided list, for large terminal
 160         readability.  0 means no limit (default behavior).
 161
 162 -t::
 163 --field-separator=::
 164         Use a special separator character and don't pad with spaces, replacing
 165         all occurrences of this separator in symbol names (and other output)
 166         with a '.' character, that thus it's the only non valid separator.
 167
 168 -D::
 169 --dump-raw-trace::
 170         Dump raw trace in ASCII.
 171
 172 -g [type,min[,limit],order[,key][,branch]]::
 173 --call-graph::
 174         Display call chains using type, min percent threshold, optional print
 175         limit and order.
 176         type can be either:
 177         - flat: single column, linear exposure of call chains.
 178         - graph: use a graph tree, displaying absolute overhead rates.
 179         - fractal: like graph, but displays relative rates. Each branch of
 180                  the tree is considered as a new profiled object. +
 181
 182         order can be either:
 183         - callee: callee based call graph.
 184         - caller: inverted caller based call graph.
 185
 186         key can be:
 187         - function: compare on functions
 188         - address: compare on individual code addresses
 189
 190         branch can be:
 191         - branch: include last branch information in callgraph
 192         when available. Usually more convenient to use --branch-history
 193         for this.
 194
 195         Default: fractal,0.5,callee,function.
 196
 197 --children::
 198         Accumulate callchain of children to parent entry so that then can
 199         show up in the output.  The output will have a new "Children" column
 200         and will be sorted on the data.  It requires callchains are recorded.
 201         See the `overhead calculation' section for more details.
 202
 203 --max-stack::
 204         Set the stack depth limit when parsing the callchain, anything
 205         beyond the specified depth will be ignored. This is a trade-off
 206         between information loss and faster processing especially for
 207         workloads that can have a very long callchain stack.
 208
 209         Default: 127
 210
 211 -G::
 212 --inverted::
 213         alias for inverted caller based call graph.
 214
 215 --ignore-callees=<regex>::
 216         Ignore callees of the function(s) matching the given regex.
 217         This has the effect of collecting the callers of each such
 218         function into one place in the call-graph tree.
 219
 220 --pretty=<key>::
 221         Pretty printing style.  key: normal, raw
 222
 223 --stdio:: Use the stdio interface.
 224
 225 --tui:: Use the TUI interface, that is integrated with annotate and allows
 226         zooming into DSOs or threads, among other features. Use of --tui
 227         requires a tty, if one is not present, as when piping to other
 228         commands, the stdio interface is used.
 229
 230 --gtk:: Use the GTK2 interface.
 231
 232 -k::
 233 --vmlinux=<file>::
 234         vmlinux pathname
 235
 236 --kallsyms=<file>::
 237         kallsyms pathname
 238
 239 -m::
 240 --modules::
 241         Load module symbols. WARNING: This should only be used with -k and
 242         a LIVE kernel.
 243
 244 -f::
 245 --force::
 246         Don't complain, do it.
 247
 248 --symfs=<directory>::
 249         Look for files with symbols relative to this directory.
 250
 251 -C::
 252 --cpu:: Only report samples for the list of CPUs provided. Multiple CPUs can
 253         be provided as a comma-separated list with no space: 0,1. Ranges of
 254         CPUs are specified with -: 0-2. Default is to report samples on all
 255         CPUs.
 256
 257 -M::
 258 --disassembler-style=:: Set disassembler style for objdump.
 259
 260 --source::
 261         Interleave source code with assembly code. Enabled by default,
 262         disable with --no-source.
 263
 264 --asm-raw::
 265         Show raw instruction encoding of assembly instructions.
 266
 267 --show-total-period:: Show a column with the sum of periods.
 268
 269 -I::
 270 --show-info::
 271         Display extended information about the perf.data file. This adds
 272         information which may be very large and thus may clutter the display.
 273         It currently includes: cpu and numa topology of the host system.
 274
 275 -b::
 276 --branch-stack::
 277         Use the addresses of sampled taken branches instead of the instruction
 278         address to build the histograms. To generate meaningful output, the
 279         perf.data file must have been obtained using perf record -b or
 280         perf record --branch-filter xxx where xxx is a branch filter option.
 281         perf report is able to auto-detect whether a perf.data file contains
 282         branch stacks and it will automatically switch to the branch view mode,
 283         unless --no-branch-stack is used.
 284
 285 --branch-history::
 286         Add the addresses of sampled taken branches to the callstack.
 287         This allows to examine the path the program took to each sample.
 288         The data collection must have used -b (or -j) and -g.
 289
 290 --objdump=<path>::
 291         Path to objdump binary.
 292
 293 --group::
 294         Show event group information together.
 295
 296 --demangle::
 297         Demangle symbol names to human readable form. It's enabled by default,
 298         disable with --no-demangle.
 299
 300 --demangle-kernel::
 301         Demangle kernel symbol names to human readable form (for C++ kernels).
 302
 303 --mem-mode::
 304         Use the data addresses of samples in addition to instruction addresses
 305         to build the histograms.  To generate meaningful output, the perf.data
 306         file must have been obtained using perf record -d -W and using a
 307         special event -e cpu/mem-loads/ or -e cpu/mem-stores/. See
 308         'perf mem' for simpler access.
 309
 310 --percent-limit::
 311         Do not show entries which have an overhead under that percent.
 312         (Default: 0).
 313
 314 --percentage::
 315         Determine how to display the overhead percentage of filtered entries.
 316         Filters can be applied by --comms, --dsos and/or --symbols options and
 317         Zoom operations on the TUI (thread, dso, etc).
 318
 319         "relative" means it's relative to filtered entries only so that the
 320         sum of shown entries will be always 100%.  "absolute" means it retains
 321         the original value before and after the filter is applied.
 322
 323 --header::
 324         Show header information in the perf.data file.  This includes
 325         various information like hostname, OS and perf version, cpu/mem
 326         info, perf command line, event list and so on.  Currently only
 327         --stdio output supports this feature.
 328
 329 --header-only::
 330         Show only perf.data header (forces --stdio).
 331
 332 --itrace::
 333         Options for decoding instruction tracing data. The options are:
 334
 335 include::itrace.txt[]
 336
 337         To disable decoding entirely, use --no-itrace.
 338
 339 --full-source-path::
 340         Show the full path for source files for srcline output.
 341
 342 --show-ref-call-graph::
 343         When multiple events are sampled, it may not be needed to collect
 344         callgraphs for all of them. The sample sites are usually nearby,
 345         and it's enough to collect the callgraphs on a reference event.
 346         So user can use "call-graph=no" event modifier to disable callgraph
 347         for other events to reduce the overhead.
 348         However, perf report cannot show callgraphs for the event which
 349         disable the callgraph.
 350         This option extends the perf report to show reference callgraphs,
 351         which collected by reference event, in no callgraph event.
 352
 353 include::callchain-overhead-calculation.txt[]
 354
 355 SEE ALSO
 356 --------
 357 linkperf:perf-stat[1], linkperf:perf-annotate[1]