brintos

brintos / linux-shallow public Read only

0
0
Text · 3.8 KiB · c972032 Raw
120 lines · plain
1perf-inject(1)2==============3 4NAME5----6perf-inject - Filter to augment the events stream with additional information7 8SYNOPSIS9--------10[verse]11'perf inject <options>'12 13DESCRIPTION14-----------15perf-inject reads a perf-record event stream and repipes it to stdout.  At any16point the processing code can inject other events into the event stream - in17this case build-ids (-b option) are read and injected as needed into the event18stream.19 20Build-ids are just the first user of perf-inject - potentially anything that21needs userspace processing to augment the events stream with additional22information could make use of this facility.23 24OPTIONS25-------26-b::27--build-ids::28	Inject build-ids of DSOs hit by samples into the output stream.29	This means it needs to process all SAMPLE records to find the DSOs.30 31--buildid-all::32	Inject build-ids of all DSOs into the output stream regardless of hits33	and skip SAMPLE processing.34 35--known-build-ids=::36	Override build-ids to inject using these comma-separated pairs of37	build-id and path. Understands file://filename to read these pairs38	from a file, which can be generated with perf buildid-list.39 40-v::41--verbose::42	Be more verbose.43-i::44--input=::45	Input file name. (default: stdin)46-o::47--output=::48	Output file name. (default: stdout)49-s::50--sched-stat::51	Merge sched_stat and sched_switch for getting events where and how long52	tasks slept. sched_switch contains a callchain where a task slept and53	sched_stat contains a timeslice how long a task slept.54 55-k::56--vmlinux=<file>::57        vmlinux pathname58 59--ignore-vmlinux::60	Ignore vmlinux files.61 62--kallsyms=<file>::63	kallsyms pathname64 65--itrace::66	Decode Instruction Tracing data, replacing it with synthesized events.67	Options are:68 69include::itrace.txt[]70 71--strip::72	Use with --itrace to strip out non-synthesized events.73 74-j::75--jit::76	Process jitdump files by injecting the mmap records corresponding to jitted77	functions. This option also generates the ELF images for each jitted function78	found in the jitdumps files captured in the input perf.data file. Use this option79	if you are monitoring environment using JIT runtimes, such as Java, DART or V8.80 81-f::82--force::83	Don't complain, do it.84 85--vm-time-correlation[=OPTIONS]::86	Some architectures may capture AUX area data which contains timestamps87	affected by virtualization. This option will update those timestamps88	in place, to correlate with host timestamps. The in-place update means89	that an output file is not specified, and instead the input file is90	modified.  The options are architecture specific, except that they may91	start with "dry-run" which will cause the file to be processed but92	without updating it. Currently this option is supported only by93	Intel PT, refer linkperf:perf-intel-pt[1]94 95--guest-data=<path>,<pid>[,<time offset>[,<time scale>]]::96	Insert events from a perf.data file recorded in a virtual machine at97	the same time as the input perf.data file was recorded on the host.98	The Process ID (PID) of the QEMU hypervisor process must be provided,99	and the time offset and time scale (multiplier) will likely be needed100	to convert guest time stamps into host time stamps. For example, for101	x86 the TSC Offset and Multiplier could be provided for a virtual machine102	using Linux command line option no-kvmclock.103	Currently only mmap, mmap2, comm, task, context_switch, ksymbol,104	and text_poke events are inserted, as well as build ID information.105	The QEMU option -name debug-threads=on is needed so that thread names106	can be used to determine which thread is running which VCPU. Note107	libvirt seems to use this by default.108	When using perf record in the guest, option --sample-identifier109	should be used, and also --buildid-all and --switch-events may be110	useful.111 112:GMEXAMPLECMD: inject113:GMEXAMPLESUBCMD:114include::guestmount.txt[]115 116SEE ALSO117--------118linkperf:perf-record[1], linkperf:perf-report[1], linkperf:perf-archive[1],119linkperf:perf-intel-pt[1]120