| blob | 59ab1ff9d75f7e3a3d35c5d6df2ac20569fd4d6d |
1 perf-intel-pt(1)
2 ================
4 NAME
5 ----
6 perf-intel-pt - Support for Intel Processor Trace within perf tools
8 SYNOPSIS
9 --------
10 [verse]
11 'perf record' -e intel_pt//
13 DESCRIPTION
14 -----------
16 Intel Processor Trace (Intel PT) is an extension of Intel Architecture that
17 collects information about software execution such as control flow, execution
18 modes and timings and formats it into highly compressed binary packets.
19 Technical details are documented in the Intel 64 and IA-32 Architectures
20 Software Developer Manuals, Chapter 36 Intel Processor Trace.
22 Intel PT is first supported in Intel Core M and 5th generation Intel Core
23 processors that are based on the Intel micro-architecture code name Broadwell.
25 Trace data is collected by 'perf record' and stored within the perf.data file.
26 See below for options to 'perf record'.
28 Trace data must be 'decoded' which involves walking the object code and matching
29 the trace data packets. For example a TNT packet only tells whether a
30 conditional branch was taken or not taken, so to make use of that packet the
31 decoder must know precisely which instruction was being executed.
33 Decoding is done on-the-fly. The decoder outputs samples in the same format as
34 samples output by perf hardware events, for example as though the "instructions"
35 or "branches" events had been recorded. Presently 3 tools support this:
36 'perf script', 'perf report' and 'perf inject'. See below for more information
37 on using those tools.
39 The main distinguishing feature of Intel PT is that the decoder can determine
40 the exact flow of software execution. Intel PT can be used to understand why
41 and how did software get to a certain point, or behave a certain way. The
42 software does not have to be recompiled, so Intel PT works with debug or release
43 builds, however the executed images are needed - which makes use in JIT-compiled
44 environments, or with self-modified code, a challenge. Also symbols need to be
45 provided to make sense of addresses.
47 A limitation of Intel PT is that it produces huge amounts of trace data
48 (hundreds of megabytes per second per core) which takes a long time to decode,
49 for example two or three orders of magnitude longer than it took to collect.
50 Another limitation is the performance impact of tracing, something that will
51 vary depending on the use-case and architecture.
54 Quickstart
55 ----------
57 It is important to start small. That is because it is easy to capture vastly
58 more data than can possibly be processed.
60 The simplest thing to do with Intel PT is userspace profiling of small programs.
61 Data is captured with 'perf record' e.g. to trace 'ls' userspace-only:
63 perf record -e intel_pt//u ls
65 And profiled with 'perf report' e.g.
67 perf report
69 To also trace kernel space presents a problem, namely kernel self-modifying
70 code. A fairly good kernel image is available in /proc/kcore but to get an
71 accurate image a copy of /proc/kcore needs to be made under the same conditions
72 as the data capture. 'perf record' can make a copy of /proc/kcore if the option
73 --kcore is used, but access to /proc/kcore is restricted e.g.
75 sudo perf record -o pt_ls --kcore -e intel_pt// -- ls
77 which will create a directory named 'pt_ls' and put the perf.data file (named
78 simply 'data') and copies of /proc/kcore, /proc/kallsyms and /proc/modules into
79 it. The other tools understand the directory format, so to use 'perf report'
80 becomes:
82 sudo perf report -i pt_ls
84 Because samples are synthesized after-the-fact, the sampling period can be
85 selected for reporting. e.g. sample every microsecond
87 sudo perf report pt_ls --itrace=i1usge
89 See the sections below for more information about the --itrace option.
91 Beware the smaller the period, the more samples that are produced, and the
92 longer it takes to process them.
94 Also note that the coarseness of Intel PT timing information will start to
95 distort the statistical value of the sampling as the sampling period becomes
96 smaller.
98 To represent software control flow, "branches" samples are produced. By default
99 a branch sample is synthesized for every single branch. To get an idea what
100 data is available you can use the 'perf script' tool with all itrace sampling
101 options, which will list all the samples.
103 perf record -e intel_pt//u ls
104 perf script --itrace=iybxwpe
106 An interesting field that is not printed by default is 'flags' which can be
107 displayed as follows:
109 perf script --itrace=iybxwpe -F+flags
111 The flags are "bcrosyiABExghDt" which stand for branch, call, return, conditional,
112 system, asynchronous, interrupt, transaction abort, trace begin, trace end,
113 in transaction, VM-entry, VM-exit, interrupt disabled, and interrupt disable
114 toggle respectively.
116 perf script also supports higher level ways to dump instruction traces:
118 perf script --insn-trace=disasm
120 or to use the xed disassembler, which requires installing the xed tool
121 (see XED below):
123 perf script --insn-trace --xed
125 Dumping all instructions in a long trace can be fairly slow. It is usually better
126 to start with higher level decoding, like
128 perf script --call-trace
130 or
132 perf script --call-ret-trace
134 and then select a time range of interest. The time range can then be examined
135 in detail with
137 perf script --time starttime,stoptime --insn-trace=disasm
139 While examining the trace it's also useful to filter on specific CPUs using
140 the -C option
142 perf script --time starttime,stoptime --insn-trace=disasm -C 1
144 Dump all instructions in time range on CPU 1.
146 Another interesting field that is not printed by default is 'ipc' which can be
147 displayed as follows:
149 perf script --itrace=be -F+ipc
151 There are two ways that instructions-per-cycle (IPC) can be calculated depending
152 on the recording.
154 If the 'cyc' config term (see config terms section below) was used, then IPC
155 and cycle events are calculated using the cycle count from CYC packets, otherwise
156 MTC packets are used - refer to the 'mtc' config term. When MTC is used, however,
157 the values are less accurate because the timing is less accurate.
159 Because Intel PT does not update the cycle count on every branch or instruction,
160 the values will often be zero. When there are values, they will be the number
161 of instructions and number of cycles since the last update, and thus represent
162 the average IPC cycle count since the last IPC for that event type.
163 Note IPC for "branches" events is calculated separately from IPC for "instructions"
164 events.
166 Even with the 'cyc' config term, it is possible to produce IPC information for
167 every change of timestamp, but at the expense of accuracy. That is selected by
168 specifying the itrace 'A' option. Due to the granularity of timestamps, the
169 actual number of cycles increases even though the cycles reported does not.
170 The number of instructions is known, but if IPC is reported, cycles can be too
171 low and so IPC is too high. Note that inaccuracy decreases as the period of
172 sampling increases i.e. if the number of cycles is too low by a small amount,
173 that becomes less significant if the number of cycles is large. It may also be
174 useful to use the 'A' option in conjunction with dlfilter-show-cycles.so to
175 provide higher granularity cycle information.
177 Also note that the IPC instruction count may or may not include the current
178 instruction. If the cycle count is associated with an asynchronous branch
179 (e.g. page fault or interrupt), then the instruction count does not include the
180 current instruction, otherwise it does. That is consistent with whether or not
181 that instruction has retired when the cycle count is updated.
183 Another note, in the case of "branches" events, non-taken branches are not
184 presently sampled, so IPC values for them do not appear e.g. a CYC packet with a
185 TNT packet that starts with a non-taken branch. To see every possible IPC
186 value, "instructions" events can be used e.g. --itrace=i0ns
188 While it is possible to create scripts to analyze the data, an alternative
189 approach is available to export the data to a sqlite or postgresql database.
190 Refer to script export-to-sqlite.py or export-to-postgresql.py for more details,
191 and to script exported-sql-viewer.py for an example of using the database.
193 There is also script intel-pt-events.py which provides an example of how to
194 unpack the raw data for power events and PTWRITE. The script also displays
195 branches, and supports 2 additional modes selected by option:
197 - --insn-trace - instruction trace
198 - --src-trace - source trace
200 The intel-pt-events.py script also has options:
202 - --all-switch-events - display all switch events, not only the last consecutive.
203 - --interleave [<n>] - interleave sample output for the same timestamp so that
204 no more than n samples for a CPU are displayed in a row. 'n' defaults to 4.
205 Note this only affects the order of output, and only when the timestamp is the
206 same.
208 As mentioned above, it is easy to capture too much data. One way to limit the
209 data captured is to use 'snapshot' mode which is explained further below.
210 Refer to 'new snapshot option' and 'Intel PT modes of operation' further below.
212 Another problem that will be experienced is decoder errors. They can be caused
213 by inability to access the executed image, self-modified or JIT-ed code, or the
214 inability to match side-band information (such as context switches and mmaps)
215 which results in the decoder not knowing what code was executed.
217 There is also the problem of perf not being able to copy the data fast enough,
218 resulting in data lost because the buffer was full. See 'Buffer handling' below
219 for more details.
222 perf record
223 -----------
225 new event
226 ~~~~~~~~~
228 The Intel PT kernel driver creates a new PMU for Intel PT. PMU events are
229 selected by providing the PMU name followed by the "config" separated by slashes.
230 An enhancement has been made to allow default "config" e.g. the option
232 -e intel_pt//
234 will use a default config value. Currently that is the same as
236 -e intel_pt/tsc,noretcomp=0/
238 which is the same as
240 -e intel_pt/tsc=1,noretcomp=0/
242 Note there are now new config terms - see section 'config terms' further below.
244 The config terms are listed in /sys/devices/intel_pt/format. They are bit
245 fields within the config member of the struct perf_event_attr which is
246 passed to the kernel by the perf_event_open system call. They correspond to bit
247 fields in the IA32_RTIT_CTL MSR. Here is a list of them and their definitions:
249 $ grep -H . /sys/bus/event_source/devices/intel_pt/format/*
250 /sys/bus/event_source/devices/intel_pt/format/cyc:config:1
251 /sys/bus/event_source/devices/intel_pt/format/cyc_thresh:config:19-22
252 /sys/bus/event_source/devices/intel_pt/format/mtc:config:9
253 /sys/bus/event_source/devices/intel_pt/format/mtc_period:config:14-17
254 /sys/bus/event_source/devices/intel_pt/format/noretcomp:config:11
255 /sys/bus/event_source/devices/intel_pt/format/psb_period:config:24-27
256 /sys/bus/event_source/devices/intel_pt/format/tsc:config:10
258 Note that the default config must be overridden for each term i.e.
260 -e intel_pt/noretcomp=0/
262 is the same as:
264 -e intel_pt/tsc=1,noretcomp=0/
266 So, to disable TSC packets use:
268 -e intel_pt/tsc=0/
270 It is also possible to specify the config value explicitly:
272 -e intel_pt/config=0x400/
274 Note that, as with all events, the event is suffixed with event modifiers:
276 u userspace
277 k kernel
278 h hypervisor
279 G guest
280 H host
281 p precise ip
283 'h', 'G' and 'H' are for virtualization which are not used by Intel PT.
284 'p' is also not relevant to Intel PT. So only options 'u' and 'k' are
285 meaningful for Intel PT.
287 perf_event_attr is displayed if the -vv option is used e.g.
289 ------------------------------------------------------------
290 perf_event_attr:
291 type 6
292 size 112
293 config 0x400
294 { sample_period, sample_freq } 1
295 sample_type IP|TID|TIME|CPU|IDENTIFIER
296 read_format ID
297 disabled 1
298 inherit 1
299 exclude_kernel 1
300 exclude_hv 1
301 enable_on_exec 1
302 sample_id_all 1
303 ------------------------------------------------------------
304 sys_perf_event_open: pid 31104 cpu 0 group_fd -1 flags 0x8
305 sys_perf_event_open: pid 31104 cpu 1 group_fd -1 flags 0x8
306 sys_perf_event_open: pid 31104 cpu 2 group_fd -1 flags 0x8
307 sys_perf_event_open: pid 31104 cpu 3 group_fd -1 flags 0x8
308 ------------------------------------------------------------
311 config terms
312 ~~~~~~~~~~~~
314 The June 2015 version of Intel 64 and IA-32 Architectures Software Developer
315 Manuals, Chapter 36 Intel Processor Trace, defined new Intel PT features.
316 Some of the features are reflect in new config terms. All the config terms are
317 described below.
319 tsc Always supported. Produces TSC timestamp packets to provide
320 timing information. In some cases it is possible to decode
321 without timing information, for example a per-thread context
322 that does not overlap executable memory maps.
324 The default config selects tsc (i.e. tsc=1).
326 noretcomp Always supported. Disables "return compression" so a TIP packet
327 is produced when a function returns. Causes more packets to be
328 produced but might make decoding more reliable.
330 The default config does not select noretcomp (i.e. noretcomp=0).
332 psb_period Allows the frequency of PSB packets to be specified.
334 The PSB packet is a synchronization packet that provides a
335 starting point for decoding or recovery from errors.
337 Support for psb_period is indicated by:
339 /sys/bus/event_source/devices/intel_pt/caps/psb_cyc
341 which contains "1" if the feature is supported and "0"
342 otherwise.
344 Valid values are given by:
346 /sys/bus/event_source/devices/intel_pt/caps/psb_periods
348 which contains a hexadecimal value, the bits of which represent
349 valid values e.g. bit 2 set means value 2 is valid.
351 The psb_period value is converted to the approximate number of
352 trace bytes between PSB packets as:
354 2 ^ (value + 11)
356 e.g. value 3 means 16KiB bytes between PSBs
358 If an invalid value is entered, the error message
359 will give a list of valid values e.g.
361 $ perf record -e intel_pt/psb_period=15/u uname
362 Invalid psb_period for intel_pt. Valid values are: 0-5
364 If MTC packets are selected, the default config selects a value
365 of 3 (i.e. psb_period=3) or the nearest lower value that is
366 supported (0 is always supported). Otherwise the default is 0.
368 If decoding is expected to be reliable and the buffer is large
369 then a large PSB period can be used.
371 Because a TSC packet is produced with PSB, the PSB period can
372 also affect the granularity to timing information in the absence
373 of MTC or CYC.
375 mtc Produces MTC timing packets.
377 MTC packets provide finer grain timestamp information than TSC
378 packets. MTC packets record time using the hardware crystal
379 clock (CTC) which is related to TSC packets using a TMA packet.
381 Support for this feature is indicated by:
383 /sys/bus/event_source/devices/intel_pt/caps/mtc
385 which contains "1" if the feature is supported and
386 "0" otherwise.
388 The frequency of MTC packets can also be specified - see
389 mtc_period below.
391 mtc_period Specifies how frequently MTC packets are produced - see mtc
392 above for how to determine if MTC packets are supported.
394 Valid values are given by:
396 /sys/bus/event_source/devices/intel_pt/caps/mtc_periods
398 which contains a hexadecimal value, the bits of which represent
399 valid values e.g. bit 2 set means value 2 is valid.
401 The mtc_period value is converted to the MTC frequency as:
403 CTC-frequency / (2 ^ value)
405 e.g. value 3 means one eighth of CTC-frequency
407 Where CTC is the hardware crystal clock, the frequency of which
408 can be related to TSC via values provided in cpuid leaf 0x15.
410 If an invalid value is entered, the error message
411 will give a list of valid values e.g.
413 $ perf record -e intel_pt/mtc_period=15/u uname
414 Invalid mtc_period for intel_pt. Valid values are: 0,3,6,9
416 The default value is 3 or the nearest lower value
417 that is supported (0 is always supported).
419 cyc Produces CYC timing packets.
421 CYC packets provide even finer grain timestamp information than
422 MTC and TSC packets. A CYC packet contains the number of CPU
423 cycles since the last CYC packet. Unlike MTC and TSC packets,
424 CYC packets are only sent when another packet is also sent.
426 Support for this feature is indicated by:
428 /sys/bus/event_source/devices/intel_pt/caps/psb_cyc
430 which contains "1" if the feature is supported and
431 "0" otherwise.
433 The number of CYC packets produced can be reduced by specifying
434 a threshold - see cyc_thresh below.
436 cyc_thresh Specifies how frequently CYC packets are produced - see cyc
437 above for how to determine if CYC packets are supported.
439 Valid cyc_thresh values are given by:
441 /sys/bus/event_source/devices/intel_pt/caps/cycle_thresholds
443 which contains a hexadecimal value, the bits of which represent
444 valid values e.g. bit 2 set means value 2 is valid.
446 The cyc_thresh value represents the minimum number of CPU cycles
447 that must have passed before a CYC packet can be sent. The
448 number of CPU cycles is:
450 2 ^ (value - 1)
452 e.g. value 4 means 8 CPU cycles must pass before a CYC packet
453 can be sent. Note a CYC packet is still only sent when another
454 packet is sent, not at, e.g. every 8 CPU cycles.
456 If an invalid value is entered, the error message
457 will give a list of valid values e.g.
459 $ perf record -e intel_pt/cyc,cyc_thresh=15/u uname
460 Invalid cyc_thresh for intel_pt. Valid values are: 0-12
462 CYC packets are not requested by default.
464 pt Specifies pass-through which enables the 'branch' config term.
466 The default config selects 'pt' if it is available, so a user will
467 never need to specify this term.
469 branch Enable branch tracing. Branch tracing is enabled by default so to
470 disable branch tracing use 'branch=0'.
472 The default config selects 'branch' if it is available.
474 ptw Enable PTWRITE packets which are produced when a ptwrite instruction
475 is executed.
477 Support for this feature is indicated by:
479 /sys/bus/event_source/devices/intel_pt/caps/ptwrite
481 which contains "1" if the feature is supported and
482 "0" otherwise.
484 As an alternative, refer to "Emulated PTWRITE" further below.
486 fup_on_ptw Enable a FUP packet to follow the PTWRITE packet. The FUP packet
487 provides the address of the ptwrite instruction. In the absence of
488 fup_on_ptw, the decoder will use the address of the previous branch
489 if branch tracing is enabled, otherwise the address will be zero.
490 Note that fup_on_ptw will work even when branch tracing is disabled.
492 pwr_evt Enable power events. The power events provide information about
493 changes to the CPU C-state.
495 Support for this feature is indicated by:
497 /sys/bus/event_source/devices/intel_pt/caps/power_event_trace
499 which contains "1" if the feature is supported and
500 "0" otherwise.
502 event Enable Event Trace. The events provide information about asynchronous
503 events.
505 Support for this feature is indicated by:
507 /sys/bus/event_source/devices/intel_pt/caps/event_trace
509 which contains "1" if the feature is supported and
510 "0" otherwise.
512 notnt Disable TNT packets. Without TNT packets, it is not possible to walk
513 executable code to reconstruct control flow, however FUP, TIP, TIP.PGE
514 and TIP.PGD packets still indicate asynchronous control flow, and (if
515 return compression is disabled - see noretcomp) return statements.
516 The advantage of eliminating TNT packets is reducing the size of the
517 trace and corresponding tracing overhead.
519 Support for this feature is indicated by:
521 /sys/bus/event_source/devices/intel_pt/caps/tnt_disable
523 which contains "1" if the feature is supported and
524 "0" otherwise.
527 AUX area sampling option
528 ~~~~~~~~~~~~~~~~~~~~~~~~
530 To select Intel PT "sampling" the AUX area sampling option can be used:
532 --aux-sample
534 Optionally it can be followed by the sample size in bytes e.g.
536 --aux-sample=8192
538 In addition, the Intel PT event to sample must be defined e.g.
540 -e intel_pt//u
542 Samples on other events will be created containing Intel PT data e.g. the
543 following will create Intel PT samples on the branch-misses event, note the
544 events must be grouped using {}:
546 perf record --aux-sample -e '{intel_pt//u,branch-misses:u}'
548 An alternative to '--aux-sample' is to add the config term 'aux-sample-size' to
549 events. In this case, the grouping is implied e.g.
551 perf record -e intel_pt//u -e branch-misses/aux-sample-size=8192/u
553 is the same as:
555 perf record -e '{intel_pt//u,branch-misses/aux-sample-size=8192/u}'
557 but allows for also using an address filter e.g.:
559 perf record -e intel_pt//u --filter 'filter * @/bin/ls' -e branch-misses/aux-sample-size=8192/u -- ls
561 It is important to select a sample size that is big enough to contain at least
562 one PSB packet. If not a warning will be displayed:
564 Intel PT sample size (%zu) may be too small for PSB period (%zu)
566 The calculation used for that is: if sample_size <= psb_period + 256 display the
567 warning. When sampling is used, psb_period defaults to 0 (2KiB).
569 The default sample size is 4KiB.
571 The sample size is passed in aux_sample_size in struct perf_event_attr. The
572 sample size is limited by the maximum event size which is 64KiB. It is
573 difficult to know how big the event might be without the trace sample attached,
574 but the tool validates that the sample size is not greater than 60KiB.
577 new snapshot option
578 ~~~~~~~~~~~~~~~~~~~
580 The difference between full trace and snapshot from the kernel's perspective is
581 that in full trace we don't overwrite trace data that the user hasn't collected
582 yet (and indicated that by advancing aux_tail), whereas in snapshot mode we let
583 the trace run and overwrite older data in the buffer so that whenever something
584 interesting happens, we can stop it and grab a snapshot of what was going on
585 around that interesting moment.
587 To select snapshot mode a new option has been added:
589 -S
591 Optionally it can be followed by the snapshot size e.g.
593 -S0x100000
595 The default snapshot size is the auxtrace mmap size. If neither auxtrace mmap size
596 nor snapshot size is specified, then the default is 4MiB for privileged users
597 (or if /proc/sys/kernel/perf_event_paranoid < 0), 128KiB for unprivileged users.
598 If an unprivileged user does not specify mmap pages, the mmap pages will be
599 reduced as described in the 'new auxtrace mmap size option' section below.
601 The snapshot size is displayed if the option -vv is used e.g.
603 Intel PT snapshot size: %zu
606 new auxtrace mmap size option
607 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
609 Intel PT buffer size is specified by an addition to the -m option e.g.
611 -m,16
613 selects a buffer size of 16 pages i.e. 64KiB.
615 Note that the existing functionality of -m is unchanged. The auxtrace mmap size
616 is specified by the optional addition of a comma and the value.
618 The default auxtrace mmap size for Intel PT is 4MiB/page_size for privileged users
619 (or if /proc/sys/kernel/perf_event_paranoid < 0), 128KiB for unprivileged users.
620 If an unprivileged user does not specify mmap pages, the mmap pages will be
621 reduced from the default 512KiB/page_size to 256KiB/page_size, otherwise the
622 user is likely to get an error as they exceed their mlock limit (Max locked
623 memory as shown in /proc/self/limits). Note that perf does not count the first
624 512KiB (actually /proc/sys/kernel/perf_event_mlock_kb minus 1 page) per cpu
625 against the mlock limit so an unprivileged user is allowed 512KiB per cpu plus
626 their mlock limit (which defaults to 64KiB but is not multiplied by the number
627 of cpus).
629 In full-trace mode, powers of two are allowed for buffer size, with a minimum
630 size of 2 pages. In snapshot mode or sampling mode, it is the same but the
631 minimum size is 1 page.
633 The mmap size and auxtrace mmap size are displayed if the -vv option is used e.g.
635 mmap length 528384
636 auxtrace mmap length 4198400
639 Intel PT modes of operation
640 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
642 Intel PT can be used in 3 modes:
643 full-trace mode
644 sample mode
645 snapshot mode
647 Full-trace mode traces continuously e.g.
649 perf record -e intel_pt//u uname
651 Sample mode attaches a Intel PT sample to other events e.g.
653 perf record --aux-sample -e intel_pt//u -e branch-misses:u
655 Snapshot mode captures the available data when a signal is sent or "snapshot"
656 control command is issued. e.g. using a signal
658 perf record -v -e intel_pt//u -S ./loopy 1000000000 &
659 [1] 11435
660 kill -USR2 11435
661 Recording AUX area tracing snapshot
663 Note that the signal sent is SIGUSR2.
664 Note that "Recording AUX area tracing snapshot" is displayed because the -v
665 option is used.
667 The advantage of using "snapshot" control command is that the access is
668 controlled by access to a FIFO e.g.
670 $ mkfifo perf.control
671 $ mkfifo perf.ack
672 $ cat perf.ack &
673 [1] 15235
674 $ sudo ~/bin/perf record --control fifo:perf.control,perf.ack -S -e intel_pt//u -- sleep 60 &
675 [2] 15243
676 $ ps -e | grep perf
677 15244 pts/1 00:00:00 perf
678 $ kill -USR2 15244
679 bash: kill: (15244) - Operation not permitted
680 $ echo snapshot > perf.control
681 ack
683 The 3 Intel PT modes of operation cannot be used together.
686 Buffer handling
687 ~~~~~~~~~~~~~~~
689 There may be buffer limitations (i.e. single ToPa entry) which means that actual
690 buffer sizes are limited to powers of 2 up to 4MiB (MAX_PAGE_ORDER). In order to
691 provide other sizes, and in particular an arbitrarily large size, multiple
692 buffers are logically concatenated. However an interrupt must be used to switch
693 between buffers. That has two potential problems:
694 a) the interrupt may not be handled in time so that the current buffer
695 becomes full and some trace data is lost.
696 b) the interrupts may slow the system and affect the performance
697 results.
699 If trace data is lost, the driver sets 'truncated' in the PERF_RECORD_AUX event
700 which the tools report as an error.
702 In full-trace mode, the driver waits for data to be copied out before allowing
703 the (logical) buffer to wrap-around. If data is not copied out quickly enough,
704 again 'truncated' is set in the PERF_RECORD_AUX event. If the driver has to
705 wait, the intel_pt event gets disabled. Because it is difficult to know when
706 that happens, perf tools always re-enable the intel_pt event after copying out
707 data.
710 Intel PT and build ids
711 ~~~~~~~~~~~~~~~~~~~~~~
713 By default "perf record" post-processes the event stream to find all build ids
714 for executables for all addresses sampled. Deliberately, Intel PT is not
715 decoded for that purpose (it would take too long). Instead the build ids for
716 all executables encountered (due to mmap, comm or task events) are included
717 in the perf.data file.
719 To see buildids included in the perf.data file use the command:
721 perf buildid-list
723 If the perf.data file contains Intel PT data, that is the same as:
725 perf buildid-list --with-hits
728 Snapshot mode and event disabling
729 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
731 In order to make a snapshot, the intel_pt event is disabled using an IOCTL,
732 namely PERF_EVENT_IOC_DISABLE. However doing that can also disable the
733 collection of side-band information. In order to prevent that, a dummy
734 software event has been introduced that permits tracking events (like mmaps) to
735 continue to be recorded while intel_pt is disabled. That is important to ensure
736 there is complete side-band information to allow the decoding of subsequent
737 snapshots.
739 A test has been created for that. To find the test:
741 perf test list
742 ...
743 23: Test using a dummy software event to keep tracking
745 To run the test:
747 perf test 23
748 23: Test using a dummy software event to keep tracking : Ok
751 perf record modes (nothing new here)
752 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
754 perf record essentially operates in one of three modes:
755 per thread
756 per cpu
757 workload only
759 "per thread" mode is selected by -t or by --per-thread (with -p or -u or just a
760 workload).
761 "per cpu" is selected by -C or -a.
762 "workload only" mode is selected by not using the other options but providing a
763 command to run (i.e. the workload).
765 In per-thread mode an exact list of threads is traced. There is no inheritance.
766 Each thread has its own event buffer.
768 In per-cpu mode all processes (or processes from the selected cgroup i.e. -G
769 option, or processes selected with -p or -u) are traced. Each cpu has its own
770 buffer. Inheritance is allowed.
772 In workload-only mode, the workload is traced but with per-cpu buffers.
773 Inheritance is allowed. Note that you can now trace a workload in per-thread
774 mode by using the --per-thread option.
777 Privileged vs non-privileged users
778 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
780 Unless /proc/sys/kernel/perf_event_paranoid is set to -1, unprivileged users
781 have memory limits imposed upon them. That affects what buffer sizes they can
782 have as outlined above.
784 The v4.2 kernel introduced support for a context switch metadata event,
785 PERF_RECORD_SWITCH, which allows unprivileged users to see when their processes
786 are scheduled out and in, just not by whom, which is left for the
787 PERF_RECORD_SWITCH_CPU_WIDE, that is only accessible in system wide context,
788 which in turn requires CAP_PERFMON or CAP_SYS_ADMIN.
790 Please see the 45ac1403f564 ("perf: Add PERF_RECORD_SWITCH to indicate context
791 switches") commit, that introduces these metadata events for further info.
793 When working with kernels < v4.2, the following considerations must be taken,
794 as the sched:sched_switch tracepoints will be used to receive such information:
796 Unless /proc/sys/kernel/perf_event_paranoid is set to -1, unprivileged users are
797 not permitted to use tracepoints which means there is insufficient side-band
798 information to decode Intel PT in per-cpu mode, and potentially workload-only
799 mode too if the workload creates new processes.
801 Note also, that to use tracepoints, read-access to debugfs is required. So if
802 debugfs is not mounted or the user does not have read-access, it will again not
803 be possible to decode Intel PT in per-cpu mode.
806 sched_switch tracepoint
807 ~~~~~~~~~~~~~~~~~~~~~~~
809 The sched_switch tracepoint is used to provide side-band data for Intel PT
810 decoding in kernels where the PERF_RECORD_SWITCH metadata event isn't
811 available.
813 The sched_switch events are automatically added. e.g. the second event shown
814 below:
816 $ perf record -vv -e intel_pt//u uname
817 ------------------------------------------------------------
818 perf_event_attr:
819 type 6
820 size 112
821 config 0x400
822 { sample_period, sample_freq } 1
823 sample_type IP|TID|TIME|CPU|IDENTIFIER
824 read_format ID
825 disabled 1
826 inherit 1
827 exclude_kernel 1
828 exclude_hv 1
829 enable_on_exec 1
830 sample_id_all 1
831 ------------------------------------------------------------
832 sys_perf_event_open: pid 31104 cpu 0 group_fd -1 flags 0x8
833 sys_perf_event_open: pid 31104 cpu 1 group_fd -1 flags 0x8
834 sys_perf_event_open: pid 31104 cpu 2 group_fd -1 flags 0x8
835 sys_perf_event_open: pid 31104 cpu 3 group_fd -1 flags 0x8
836 ------------------------------------------------------------
837 perf_event_attr:
838 type 2
839 size 112
840 config 0x108
841 { sample_period, sample_freq } 1
842 sample_type IP|TID|TIME|CPU|PERIOD|RAW|IDENTIFIER
843 read_format ID
844 inherit 1
845 sample_id_all 1
846 exclude_guest 1
847 ------------------------------------------------------------
848 sys_perf_event_open: pid -1 cpu 0 group_fd -1 flags 0x8
849 sys_perf_event_open: pid -1 cpu 1 group_fd -1 flags 0x8
850 sys_perf_event_open: pid -1 cpu 2 group_fd -1 flags 0x8
851 sys_perf_event_open: pid -1 cpu 3 group_fd -1 flags 0x8
852 ------------------------------------------------------------
853 perf_event_attr:
854 type 1
855 size 112
856 config 0x9
857 { sample_period, sample_freq } 1
858 sample_type IP|TID|TIME|IDENTIFIER
859 read_format ID
860 disabled 1
861 inherit 1
862 exclude_kernel 1
863 exclude_hv 1
864 mmap 1
865 comm 1
866 enable_on_exec 1
867 task 1
868 sample_id_all 1
869 mmap2 1
870 comm_exec 1
871 ------------------------------------------------------------
872 sys_perf_event_open: pid 31104 cpu 0 group_fd -1 flags 0x8
873 sys_perf_event_open: pid 31104 cpu 1 group_fd -1 flags 0x8
874 sys_perf_event_open: pid 31104 cpu 2 group_fd -1 flags 0x8
875 sys_perf_event_open: pid 31104 cpu 3 group_fd -1 flags 0x8
876 mmap size 528384B
877 AUX area mmap length 4194304
878 perf event ring buffer mmapped per cpu
879 Synthesizing auxtrace information
880 Linux
881 [ perf record: Woken up 1 times to write data ]
882 [ perf record: Captured and wrote 0.042 MB perf.data ]
884 Note, the sched_switch event is only added if the user is permitted to use it
885 and only in per-cpu mode.
887 Note also, the sched_switch event is only added if TSC packets are requested.
888 That is because, in the absence of timing information, the sched_switch events
889 cannot be matched against the Intel PT trace.
892 perf script
893 -----------
895 By default, perf script will decode trace data found in the perf.data file.
896 This can be further controlled by new option --itrace.
899 New --itrace option
900 ~~~~~~~~~~~~~~~~~~~
902 Having no option is the same as
904 --itrace
906 which, in turn, is the same as
908 --itrace=cepwxy
910 The letters are:
912 i synthesize "instructions" events
913 y synthesize "cycles" events
914 b synthesize "branches" events
915 x synthesize "transactions" events
916 w synthesize "ptwrite" events
917 p synthesize "power" events (incl. PSB events)
918 c synthesize branches events (calls only)
919 r synthesize branches events (returns only)
920 o synthesize PEBS-via-PT events
921 I synthesize Event Trace events
922 e synthesize tracing error events
923 d create a debug log
924 g synthesize a call chain (use with i or x)
925 G synthesize a call chain on existing event records
926 l synthesize last branch entries (use with i or x)
927 L synthesize last branch entries on existing event records
928 s skip initial number of events
929 q quicker (less detailed) decoding
930 A approximate IPC
931 Z prefer to ignore timestamps (so-called "timeless" decoding)
933 "Instructions" events look like they were recorded by "perf record -e
934 instructions".
936 "Cycles" events look like they were recorded by "perf record -e cycles"
937 (ie., the default). Note that even with CYC packets enabled and no sampling,
938 these are not fully accurate, since CYC packets are not emitted for each
939 instruction, only when some other event (like an indirect branch, or a
940 TNT packet representing multiple branches) happens causes a packet to
941 be emitted. Thus, it is more effective for attributing cycles to functions
942 (and possibly basic blocks) than to individual instructions, although it
943 is not even perfect for functions (although it becomes better if the noretcomp
944 option is active).
946 "Branches" events look like they were recorded by "perf record -e branches". "c"
947 and "r" can be combined to get calls and returns.
949 "Transactions" events correspond to the start or end of transactions. The
950 'flags' field can be used in perf script to determine whether the event is a
951 transaction start, commit or abort.
953 Note that "instructions", "cycles", "branches" and "transactions" events
954 depend on code flow packets which can be disabled by using the config term
955 "branch=0". Refer to the config terms section above.
957 "ptwrite" events record the payload of the ptwrite instruction and whether
958 "fup_on_ptw" was used. "ptwrite" events depend on PTWRITE packets which are
959 recorded only if the "ptw" config term was used. Refer to the config terms
960 section above. perf script "synth" field displays "ptwrite" information like
961 this: "ip: 0 payload: 0x123456789abcdef0" where "ip" is 1 if "fup_on_ptw" was
962 used.
964 "Power" events correspond to power event packets and CBR (core-to-bus ratio)
965 packets. While CBR packets are always recorded when tracing is enabled, power
966 event packets are recorded only if the "pwr_evt" config term was used. Refer to
967 the config terms section above. The power events record information about
968 C-state changes, whereas CBR is indicative of CPU frequency. perf script
969 "event,synth" fields display information like this:
971 cbr: cbr: 22 freq: 2189 MHz (200%)
972 mwait: hints: 0x60 extensions: 0x1
973 pwre: hw: 0 cstate: 2 sub-cstate: 0
974 exstop: ip: 1
975 pwrx: deepest cstate: 2 last cstate: 2 wake reason: 0x4
977 Where:
979 "cbr" includes the frequency and the percentage of maximum non-turbo
980 "mwait" shows mwait hints and extensions
981 "pwre" shows C-state transitions (to a C-state deeper than C0) and
982 whether initiated by hardware
983 "exstop" indicates execution stopped and whether the IP was recorded
984 exactly,
985 "pwrx" indicates return to C0
987 For more details refer to the Intel 64 and IA-32 Architectures Software
988 Developer Manuals.
990 PSB events show when a PSB+ occurred and also the byte-offset in the trace.
991 Emitting a PSB+ can cause a CPU a slight delay. When doing timing analysis
992 of code with Intel PT, it is useful to know if a timing bubble was caused
993 by Intel PT or not.
995 Error events show where the decoder lost the trace. Error events
996 are quite important. Users must know if what they are seeing is a complete
997 picture or not. The "e" option may be followed by flags which affect what errors
998 will or will not be reported. Each flag must be preceded by either '+' or '-'.
999 The flags supported by Intel PT are:
1001 -o Suppress overflow errors
1002 -l Suppress trace data lost errors
1004 For example, for errors but not overflow or data lost errors:
1006 --itrace=e-o-l
1008 The "d" option will cause the creation of a file "intel_pt.log" containing all
1009 decoded packets and instructions. Note that this option slows down the decoder
1010 and that the resulting file may be very large. The "d" option may be followed
1011 by flags which affect what debug messages will or will not be logged. Each flag
1012 must be preceded by either '+' or '-'. The flags support by Intel PT are:
1014 -a Suppress logging of perf events
1015 +a Log all perf events
1016 +e Output only on decoding errors (size configurable)
1017 +o Output to stdout instead of "intel_pt.log"
1019 By default, logged perf events are filtered by any specified time ranges, but
1020 flag +a overrides that. The +e flag can be useful for analyzing errors. By
1021 default, the log size in that case is 16384 bytes, but can be altered by
1022 linkperf:perf-config[1] e.g. perf config itrace.debug-log-buffer-size=30000
1024 In addition, the period of the "instructions" event can be specified. e.g.
1026 --itrace=i10us
1028 sets the period to 10us i.e. one instruction sample is synthesized for each 10
1029 microseconds of trace. Alternatives to "us" are "ms" (milliseconds),
1030 "ns" (nanoseconds), "t" (TSC ticks) or "i" (instructions).
1032 "ms", "us" and "ns" are converted to TSC ticks.
1034 The timing information included with Intel PT does not give the time of every
1035 instruction. Consequently, for the purpose of sampling, the decoder estimates
1036 the time since the last timing packet based on 1 tick per instruction. The time
1037 on the sample is *not* adjusted and reflects the last known value of TSC.
1039 For Intel PT, the default period is 100us.
1041 Setting it to a zero period means "as often as possible".
1043 In the case of Intel PT that is the same as a period of 1 and a unit of
1044 'instructions' (i.e. --itrace=i1i).
1046 Also the call chain size (default 16, max. 1024) for instructions or
1047 transactions events can be specified. e.g.
1049 --itrace=ig32
1050 --itrace=xg32
1052 Also the number of last branch entries (default 64, max. 1024) for instructions or
1053 transactions events can be specified. e.g.
1055 --itrace=il10
1056 --itrace=xl10
1058 Note that last branch entries are cleared for each sample, so there is no overlap
1059 from one sample to the next.
1061 The G and L options are designed in particular for sample mode, and work much
1062 like g and l but add call chain and branch stack to the other selected events
1063 instead of synthesized events. For example, to record branch-misses events for
1064 'ls' and then add a call chain derived from the Intel PT trace:
1066 perf record --aux-sample -e '{intel_pt//u,branch-misses:u}' -- ls
1067 perf report --itrace=Ge
1069 Although in fact G is a default for perf report, so that is the same as just:
1071 perf report
1073 One caveat with the G and L options is that they work poorly with "Large PEBS".
1074 Large PEBS means PEBS records will be accumulated by hardware and the written
1075 into the event buffer in one go. That reduces interrupts, but can give very
1076 late timestamps. Because the Intel PT trace is synchronized by timestamps,
1077 the PEBS events do not match the trace. Currently, Large PEBS is used only in
1078 certain circumstances:
1079 - hardware supports it
1080 - PEBS is used
1081 - event period is specified, instead of frequency
1082 - the sample type is limited to the following flags:
1083 PERF_SAMPLE_IP | PERF_SAMPLE_TID | PERF_SAMPLE_ADDR |
1084 PERF_SAMPLE_ID | PERF_SAMPLE_CPU | PERF_SAMPLE_STREAM_ID |
1085 PERF_SAMPLE_DATA_SRC | PERF_SAMPLE_IDENTIFIER |
1086 PERF_SAMPLE_TRANSACTION | PERF_SAMPLE_PHYS_ADDR |
1087 PERF_SAMPLE_REGS_INTR | PERF_SAMPLE_REGS_USER |
1088 PERF_SAMPLE_PERIOD (and sometimes) | PERF_SAMPLE_TIME
1089 Because Intel PT sample mode uses a different sample type to the list above,
1090 Large PEBS is not used with Intel PT sample mode. To avoid Large PEBS in other
1091 cases, avoid specifying the event period i.e. avoid the 'perf record' -c option,
1092 --count option, or 'period' config term.
1094 To disable trace decoding entirely, use the option --no-itrace.
1096 It is also possible to skip events generated (instructions, branches, transactions)
1097 at the beginning. This is useful to ignore initialization code.
1099 --itrace=i0nss1000000
1101 skips the first million instructions.
1103 The q option changes the way the trace is decoded. The decoding is much faster
1104 but much less detailed. Specifically, with the q option, the decoder does not
1105 decode TNT packets, and does not walk object code, but gets the ip from FUP and
1106 TIP packets. The q option can be used with the b and i options but the period
1107 is not used. The q option decodes more quickly, but is useful only if the
1108 control flow of interest is represented or indicated by FUP, TIP, TIP.PGE, or
1109 TIP.PGD packets (refer below). However the q option could be used to find time
1110 ranges that could then be decoded fully using the --time option.
1112 What will *not* be decoded with the (single) q option:
1114 - direct calls and jmps
1115 - conditional branches
1116 - non-branch instructions
1118 What *will* be decoded with the (single) q option:
1120 - asynchronous branches such as interrupts
1121 - indirect branches
1122 - function return target address *if* the noretcomp config term (refer
1123 config terms section) was used
1124 - start of (control-flow) tracing
1125 - end of (control-flow) tracing, if it is not out of context
1126 - power events, ptwrite, transaction start and abort
1127 - instruction pointer associated with PSB packets
1129 Note the q option does not specify what events will be synthesized e.g. the p
1130 option must be used also to show power events.
1132 Repeating the q option (double-q i.e. qq) results in even faster decoding and even
1133 less detail. The decoder decodes only extended PSB (PSB+) packets, getting the
1134 instruction pointer if there is a FUP packet within PSB+ (i.e. between PSB and
1135 PSBEND). Note PSB packets occur regularly in the trace based on the psb_period
1136 config term (refer config terms section). There will be a FUP packet if the
1137 PSB+ occurs while control flow is being traced.
1139 What will *not* be decoded with the qq option:
1141 - everything except instruction pointer associated with PSB packets
1143 What *will* be decoded with the qq option:
1145 - instruction pointer associated with PSB packets
1147 The Z option is equivalent to having recorded a trace without TSC
1148 (i.e. config term tsc=0). It can be useful to avoid timestamp issues when
1149 decoding a trace of a virtual machine.
1152 dlfilter-show-cycles.so
1153 ~~~~~~~~~~~~~~~~~~~~~~~
1155 Cycles can be displayed using dlfilter-show-cycles.so in which case the itrace A
1156 option can be useful to provide higher granularity cycle information:
1158 perf script --itrace=A --call-trace --dlfilter dlfilter-show-cycles.so
1160 To see a list of dlfilters:
1162 perf script -v --list-dlfilters
1164 See also linkperf:perf-dlfilters[1]
1167 dump option
1168 ~~~~~~~~~~~
1170 perf script has an option (-D) to "dump" the events i.e. display the binary
1171 data.
1173 When -D is used, Intel PT packets are displayed. The packet decoder does not
1174 pay attention to PSB packets, but just decodes the bytes - so the packets seen
1175 by the actual decoder may not be identical in places where the data is corrupt.
1176 One example of that would be when the buffer-switching interrupt has been too
1177 slow, and the buffer has been filled completely. In that case, the last packet
1178 in the buffer might be truncated and immediately followed by a PSB as the trace
1179 continues in the next buffer.
1181 To disable the display of Intel PT packets, combine the -D option with
1182 --no-itrace.
1185 perf report
1186 -----------
1188 By default, perf report will decode trace data found in the perf.data file.
1189 This can be further controlled by new option --itrace exactly the same as
1190 perf script, with the exception that the default is --itrace=igxe.
1193 perf inject
1194 -----------
1196 perf inject also accepts the --itrace option in which case tracing data is
1197 removed and replaced with the synthesized events. e.g.
1199 perf inject --itrace -i perf.data -o perf.data.new
1201 Below is an example of using Intel PT with autofdo. It requires autofdo
1202 (https://github.com/google/autofdo) and gcc version 5. The bubble
1203 sort example is from the AutoFDO tutorial (https://gcc.gnu.org/wiki/AutoFDO/Tutorial)
1204 amended to take the number of elements as a parameter.
1206 $ gcc-5 -O3 sort.c -o sort_optimized
1207 $ ./sort_optimized 30000
1208 Bubble sorting array of 30000 elements
1209 2254 ms
1211 $ cat ~/.perfconfig
1212 [intel-pt]
1213 mispred-all = on
1215 $ perf record -e intel_pt//u ./sort 3000
1216 Bubble sorting array of 3000 elements
1217 58 ms
1218 [ perf record: Woken up 2 times to write data ]
1219 [ perf record: Captured and wrote 3.939 MB perf.data ]
1220 $ perf inject -i perf.data -o inj --itrace=i100usle --strip
1221 $ ./create_gcov --binary=./sort --profile=inj --gcov=sort.gcov -gcov_version=1
1222 $ gcc-5 -O3 -fauto-profile=sort.gcov sort.c -o sort_autofdo
1223 $ ./sort_autofdo 30000
1224 Bubble sorting array of 30000 elements
1225 2155 ms
1227 Note there is currently no advantage to using Intel PT instead of LBR, but
1228 that may change in the future if greater use is made of the data.
1231 PEBS via Intel PT
1232 -----------------
1234 Some hardware has the feature to redirect PEBS records to the Intel PT trace.
1235 Recording is selected by using the aux-output config term e.g.
1237 perf record -c 10000 -e '{intel_pt/branch=0/,cycles/aux-output/ppp}' uname
1239 Originally, software only supported redirecting at most one PEBS event because it
1240 was not able to differentiate one event from another. To overcome that, more recent
1241 kernels and perf tools add support for the PERF_RECORD_AUX_OUTPUT_HW_ID side-band event.
1242 To check for the presence of that event in a PEBS-via-PT trace:
1244 perf script -D --no-itrace | grep PERF_RECORD_AUX_OUTPUT_HW_ID
1246 To display PEBS events from the Intel PT trace, use the itrace 'o' option e.g.
1248 perf script --itrace=oe
1250 XED
1251 ---
1253 include::build-xed.txt[]
1256 Tracing Virtual Machines (kernel only)
1257 --------------------------------------
1259 Currently, kernel tracing is supported with either "timeless" decoding
1260 (i.e. no TSC timestamps) or VM Time Correlation. VM Time Correlation is an extra step
1261 using 'perf inject' and requires unchanging VMX TSC Offset and no VMX TSC Scaling.
1263 Other limitations and caveats
1265 VMX controls may suppress packets needed for decoding resulting in decoding errors
1266 VMX controls may block the perf NMI to the host potentially resulting in lost trace data
1267 Guest kernel self-modifying code (e.g. jump labels or JIT-compiled eBPF) will result in decoding errors
1268 Guest thread information is unknown
1269 Guest VCPU is unknown but may be able to be inferred from the host thread
1270 Callchains are not supported
1272 Example using "timeless" decoding
1274 Start VM
1276 $ sudo virsh start kubuntu20.04
1277 Domain kubuntu20.04 started
1279 Mount the guest file system. Note sshfs needs -o direct_io to enable reading of proc files. root access is needed to read /proc/kcore.
1281 $ mkdir vm0
1282 $ sshfs -o direct_io root@vm0:/ vm0
1284 Copy the guest /proc/kallsyms, /proc/modules and /proc/kcore
1286 $ perf buildid-cache -v --kcore vm0/proc/kcore
1287 kcore added to build-id cache directory /home/user/.debug/[kernel.kcore]/9600f316a53a0f54278885e8d9710538ec5f6a08/2021021807494306
1288 $ KALLSYMS=/home/user/.debug/[kernel.kcore]/9600f316a53a0f54278885e8d9710538ec5f6a08/2021021807494306/kallsyms
1290 Find the VM process
1292 $ ps -eLl | grep 'KVM\|PID'
1293 F S UID PID PPID LWP C PRI NI ADDR SZ WCHAN TTY TIME CMD
1294 3 S 64055 1430 1 1440 1 80 0 - 1921718 - ? 00:02:47 CPU 0/KVM
1295 3 S 64055 1430 1 1441 1 80 0 - 1921718 - ? 00:02:41 CPU 1/KVM
1296 3 S 64055 1430 1 1442 1 80 0 - 1921718 - ? 00:02:38 CPU 2/KVM
1297 3 S 64055 1430 1 1443 2 80 0 - 1921718 - ? 00:03:18 CPU 3/KVM
1299 Start an open-ended perf record, tracing the VM process, do something on the VM, and then ctrl-C to stop.
1300 TSC is not supported and tsc=0 must be specified. That means mtc is useless, so add mtc=0.
1301 However, IPC can still be determined, hence cyc=1 can be added.
1302 Only kernel decoding is supported, so 'k' must be specified.
1303 Intel PT traces both the host and the guest so --guest and --host need to be specified.
1304 Without timestamps, --per-thread must be specified to distinguish threads.
1306 $ sudo perf kvm --guest --host --guestkallsyms $KALLSYMS record --kcore -e intel_pt/tsc=0,mtc=0,cyc=1/k -p 1430 --per-thread
1307 ^C
1308 [ perf record: Woken up 1 times to write data ]
1309 [ perf record: Captured and wrote 5.829 MB ]
1311 perf script can be used to provide an instruction trace
1313 $ perf script --guestkallsyms $KALLSYMS --insn-trace=disasm -F+ipc | grep -C10 vmresume | head -21
1314 CPU 0/KVM 1440 ffffffff82133cdd __vmx_vcpu_run+0x3d ([kernel.kallsyms]) movq 0x48(%rax), %r9
1315 CPU 0/KVM 1440 ffffffff82133ce1 __vmx_vcpu_run+0x41 ([kernel.kallsyms]) movq 0x50(%rax), %r10
1316 CPU 0/KVM 1440 ffffffff82133ce5 __vmx_vcpu_run+0x45 ([kernel.kallsyms]) movq 0x58(%rax), %r11
1317 CPU 0/KVM 1440 ffffffff82133ce9 __vmx_vcpu_run+0x49 ([kernel.kallsyms]) movq 0x60(%rax), %r12
1318 CPU 0/KVM 1440 ffffffff82133ced __vmx_vcpu_run+0x4d ([kernel.kallsyms]) movq 0x68(%rax), %r13
1319 CPU 0/KVM 1440 ffffffff82133cf1 __vmx_vcpu_run+0x51 ([kernel.kallsyms]) movq 0x70(%rax), %r14
1320 CPU 0/KVM 1440 ffffffff82133cf5 __vmx_vcpu_run+0x55 ([kernel.kallsyms]) movq 0x78(%rax), %r15
1321 CPU 0/KVM 1440 ffffffff82133cf9 __vmx_vcpu_run+0x59 ([kernel.kallsyms]) movq (%rax), %rax
1322 CPU 0/KVM 1440 ffffffff82133cfc __vmx_vcpu_run+0x5c ([kernel.kallsyms]) callq 0xffffffff82133c40
1323 CPU 0/KVM 1440 ffffffff82133c40 vmx_vmenter+0x0 ([kernel.kallsyms]) jz 0xffffffff82133c46
1324 CPU 0/KVM 1440 ffffffff82133c42 vmx_vmenter+0x2 ([kernel.kallsyms]) vmresume IPC: 0.11 (50/445)
1325 :1440 1440 ffffffffbb678b06 native_write_msr+0x6 ([guest.kernel.kallsyms]) nopl %eax, (%rax,%rax,1)
1326 :1440 1440 ffffffffbb678b0b native_write_msr+0xb ([guest.kernel.kallsyms]) retq IPC: 0.04 (2/41)
1327 :1440 1440 ffffffffbb666646 lapic_next_deadline+0x26 ([guest.kernel.kallsyms]) data16 nop
1328 :1440 1440 ffffffffbb666648 lapic_next_deadline+0x28 ([guest.kernel.kallsyms]) xor %eax, %eax
1329 :1440 1440 ffffffffbb66664a lapic_next_deadline+0x2a ([guest.kernel.kallsyms]) popq %rbp
1330 :1440 1440 ffffffffbb66664b lapic_next_deadline+0x2b ([guest.kernel.kallsyms]) retq IPC: 0.16 (4/25)
1331 :1440 1440 ffffffffbb74607f clockevents_program_event+0x8f ([guest.kernel.kallsyms]) test %eax, %eax
1332 :1440 1440 ffffffffbb746081 clockevents_program_event+0x91 ([guest.kernel.kallsyms]) jz 0xffffffffbb74603c IPC: 0.06 (2/30)
1333 :1440 1440 ffffffffbb74603c clockevents_program_event+0x4c ([guest.kernel.kallsyms]) popq %rbx
1334 :1440 1440 ffffffffbb74603d clockevents_program_event+0x4d ([guest.kernel.kallsyms]) popq %r12
1336 Example using VM Time Correlation
1338 Start VM
1340 $ sudo virsh start kubuntu20.04
1341 Domain kubuntu20.04 started
1343 Mount the guest file system. Note sshfs needs -o direct_io to enable reading of proc files. root access is needed to read /proc/kcore.
1345 $ mkdir -p vm0
1346 $ sshfs -o direct_io root@vm0:/ vm0
1348 Copy the guest /proc/kallsyms, /proc/modules and /proc/kcore
1350 $ perf buildid-cache -v --kcore vm0/proc/kcore
1351 same kcore found in /home/user/.debug/[kernel.kcore]/cc9c55a98c5e4ec0aeda69302554aabed5cd6491/2021021312450777
1352 $ KALLSYMS=/home/user/.debug/\[kernel.kcore\]/cc9c55a98c5e4ec0aeda69302554aabed5cd6491/2021021312450777/kallsyms
1354 Find the VM process
1356 $ ps -eLl | grep 'KVM\|PID'
1357 F S UID PID PPID LWP C PRI NI ADDR SZ WCHAN TTY TIME CMD
1358 3 S 64055 16998 1 17005 13 80 0 - 1818189 - ? 00:00:16 CPU 0/KVM
1359 3 S 64055 16998 1 17006 4 80 0 - 1818189 - ? 00:00:05 CPU 1/KVM
1360 3 S 64055 16998 1 17007 3 80 0 - 1818189 - ? 00:00:04 CPU 2/KVM
1361 3 S 64055 16998 1 17008 4 80 0 - 1818189 - ? 00:00:05 CPU 3/KVM
1363 Start an open-ended perf record, tracing the VM process, do something on the VM, and then ctrl-C to stop.
1364 IPC can be determined, hence cyc=1 can be added.
1365 Only kernel decoding is supported, so 'k' must be specified.
1366 Intel PT traces both the host and the guest so --guest and --host need to be specified.
1368 $ sudo perf kvm --guest --host --guestkallsyms $KALLSYMS record --kcore -e intel_pt/cyc=1/k -p 16998
1369 ^C[ perf record: Woken up 1 times to write data ]
1370 [ perf record: Captured and wrote 9.041 MB perf.data.kvm ]
1372 Now 'perf inject' can be used to determine the VMX TCS Offset. Note, Intel PT TSC packets are
1373 only 7-bytes, so the TSC Offset might differ from the actual value in the 8th byte. That will
1374 have no effect i.e. the resulting timestamps will be correct anyway.
1376 $ perf inject -i perf.data.kvm --vm-time-correlation=dry-run
1377 ERROR: Unknown TSC Offset for VMCS 0x1bff6a
1378 VMCS: 0x1bff6a TSC Offset 0xffffe42722c64c41
1379 ERROR: Unknown TSC Offset for VMCS 0x1cbc08
1380 VMCS: 0x1cbc08 TSC Offset 0xffffe42722c64c41
1381 ERROR: Unknown TSC Offset for VMCS 0x1c3ce8
1382 VMCS: 0x1c3ce8 TSC Offset 0xffffe42722c64c41
1383 ERROR: Unknown TSC Offset for VMCS 0x1cbce9
1384 VMCS: 0x1cbce9 TSC Offset 0xffffe42722c64c41
1386 Each virtual CPU has a different Virtual Machine Control Structure (VMCS)
1387 shown above with the calculated TSC Offset. For an unchanging TSC Offset
1388 they should all be the same for the same virtual machine.
1390 Now that the TSC Offset is known, it can be provided to 'perf inject'
1392 $ perf inject -i perf.data.kvm --vm-time-correlation="dry-run 0xffffe42722c64c41"
1394 Note the options for 'perf inject' --vm-time-correlation are:
1396 [ dry-run ] [ <TSC Offset> [ : <VMCS> [ , <VMCS> ]... ] ]...
1398 So it is possible to specify different TSC Offsets for different VMCS.
1399 The option "dry-run" will cause the file to be processed but without updating it.
1400 Note it is also possible to get a intel_pt.log file by adding option --itrace=d
1402 There were no errors so, do it for real
1404 $ perf inject -i perf.data.kvm --vm-time-correlation=0xffffe42722c64c41 --force
1406 'perf script' can be used to see if there are any decoder errors
1408 $ perf script -i perf.data.kvm --guestkallsyms $KALLSYMS --itrace=e-o
1410 There were none.
1412 'perf script' can be used to provide an instruction trace showing timestamps
1414 $ perf script -i perf.data.kvm --guestkallsyms $KALLSYMS --insn-trace=disasm -F+ipc | grep -C10 vmresume | head -21
1415 CPU 1/KVM 17006 [001] 11500.262865593: ffffffff82133cdd __vmx_vcpu_run+0x3d ([kernel.kallsyms]) movq 0x48(%rax), %r9
1416 CPU 1/KVM 17006 [001] 11500.262865593: ffffffff82133ce1 __vmx_vcpu_run+0x41 ([kernel.kallsyms]) movq 0x50(%rax), %r10
1417 CPU 1/KVM 17006 [001] 11500.262865593: ffffffff82133ce5 __vmx_vcpu_run+0x45 ([kernel.kallsyms]) movq 0x58(%rax), %r11
1418 CPU 1/KVM 17006 [001] 11500.262865593: ffffffff82133ce9 __vmx_vcpu_run+0x49 ([kernel.kallsyms]) movq 0x60(%rax), %r12
1419 CPU 1/KVM 17006 [001] 11500.262865593: ffffffff82133ced __vmx_vcpu_run+0x4d ([kernel.kallsyms]) movq 0x68(%rax), %r13
1420 CPU 1/KVM 17006 [001] 11500.262865593: ffffffff82133cf1 __vmx_vcpu_run+0x51 ([kernel.kallsyms]) movq 0x70(%rax), %r14
1421 CPU 1/KVM 17006 [001] 11500.262865593: ffffffff82133cf5 __vmx_vcpu_run+0x55 ([kernel.kallsyms]) movq 0x78(%rax), %r15
1422 CPU 1/KVM 17006 [001] 11500.262865593: ffffffff82133cf9 __vmx_vcpu_run+0x59 ([kernel.kallsyms]) movq (%rax), %rax
1423 CPU 1/KVM 17006 [001] 11500.262865593: ffffffff82133cfc __vmx_vcpu_run+0x5c ([kernel.kallsyms]) callq 0xffffffff82133c40
1424 CPU 1/KVM 17006 [001] 11500.262865593: ffffffff82133c40 vmx_vmenter+0x0 ([kernel.kallsyms]) jz 0xffffffff82133c46
1425 CPU 1/KVM 17006 [001] 11500.262866075: ffffffff82133c42 vmx_vmenter+0x2 ([kernel.kallsyms]) vmresume IPC: 0.05 (40/769)
1426 :17006 17006 [001] 11500.262869216: ffffffff82200cb0 asm_sysvec_apic_timer_interrupt+0x0 ([guest.kernel.kallsyms]) clac
1427 :17006 17006 [001] 11500.262869216: ffffffff82200cb3 asm_sysvec_apic_timer_interrupt+0x3 ([guest.kernel.kallsyms]) pushq $0xffffffffffffffff
1428 :17006 17006 [001] 11500.262869216: ffffffff82200cb5 asm_sysvec_apic_timer_interrupt+0x5 ([guest.kernel.kallsyms]) callq 0xffffffff82201160
1429 :17006 17006 [001] 11500.262869216: ffffffff82201160 error_entry+0x0 ([guest.kernel.kallsyms]) cld
1430 :17006 17006 [001] 11500.262869216: ffffffff82201161 error_entry+0x1 ([guest.kernel.kallsyms]) pushq %rsi
1431 :17006 17006 [001] 11500.262869216: ffffffff82201162 error_entry+0x2 ([guest.kernel.kallsyms]) movq 0x8(%rsp), %rsi
1432 :17006 17006 [001] 11500.262869216: ffffffff82201167 error_entry+0x7 ([guest.kernel.kallsyms]) movq %rdi, 0x8(%rsp)
1433 :17006 17006 [001] 11500.262869216: ffffffff8220116c error_entry+0xc ([guest.kernel.kallsyms]) pushq %rdx
1434 :17006 17006 [001] 11500.262869216: ffffffff8220116d error_entry+0xd ([guest.kernel.kallsyms]) pushq %rcx
1435 :17006 17006 [001] 11500.262869216: ffffffff8220116e error_entry+0xe ([guest.kernel.kallsyms]) pushq %rax
1438 Tracing Virtual Machines (including user space)
1439 -----------------------------------------------
1441 It is possible to use perf record to record sideband events within a virtual machine, so that an Intel PT trace on the host can be decoded.
1442 Sideband events from the guest perf.data file can be injected into the host perf.data file using perf inject.
1444 Here is an example of the steps needed:
1446 On the guest machine:
1448 Check that no-kvmclock kernel command line option was used to boot:
1450 Note, this is essential to enable time correlation between host and guest machines.
1452 $ cat /proc/cmdline
1453 BOOT_IMAGE=/boot/vmlinuz-5.10.0-16-amd64 root=UUID=cb49c910-e573-47e0-bce7-79e293df8e1d ro no-kvmclock
1455 There is no BPF support at present so, if possible, disable JIT compiling:
1457 $ echo 0 | sudo tee /proc/sys/net/core/bpf_jit_enable
1458 0
1460 Start perf record to collect sideband events:
1462 $ sudo perf record -o guest-sideband-testing-guest-perf.data --sample-identifier --buildid-all --switch-events --kcore -a -e dummy
1464 On the host machine:
1466 Start perf record to collect Intel PT trace:
1468 Note, the host trace will get very big, very fast, so the steps from starting to stopping the host trace really need to be done so that they happen in the shortest time possible.
1470 $ sudo perf record -o guest-sideband-testing-host-perf.data -m,64M --kcore -a -e intel_pt/cyc/
1472 On the guest machine:
1474 Run a small test case, just 'uname' in this example:
1476 $ uname
1477 Linux
1479 On the host machine:
1481 Stop the Intel PT trace:
1483 ^C
1484 [ perf record: Woken up 1 times to write data ]
1485 [ perf record: Captured and wrote 76.122 MB guest-sideband-testing-host-perf.data ]
1487 On the guest machine:
1489 Stop the Intel PT trace:
1491 ^C
1492 [ perf record: Woken up 1 times to write data ]
1493 [ perf record: Captured and wrote 1.247 MB guest-sideband-testing-guest-perf.data ]
1495 And then copy guest-sideband-testing-guest-perf.data to the host (not shown here).
1497 On the host machine:
1499 With the 2 perf.data recordings, and with their ownership changed to the user.
1501 Identify the TSC Offset:
1503 $ perf inject -i guest-sideband-testing-host-perf.data --vm-time-correlation=dry-run
1504 VMCS: 0x103fc6 TSC Offset 0xfffffa6ae070cb20
1505 VMCS: 0x103ff2 TSC Offset 0xfffffa6ae070cb20
1506 VMCS: 0x10fdaa TSC Offset 0xfffffa6ae070cb20
1507 VMCS: 0x24d57c TSC Offset 0xfffffa6ae070cb20
1509 Correct Intel PT TSC timestamps for the guest machine:
1511 $ perf inject -i guest-sideband-testing-host-perf.data --vm-time-correlation=0xfffffa6ae070cb20 --force
1513 Identify the guest machine PID:
1515 $ perf script -i guest-sideband-testing-host-perf.data --no-itrace --show-task-events | grep KVM
1516 CPU 0/KVM 0 [000] 0.000000: PERF_RECORD_COMM: CPU 0/KVM:13376/13381
1517 CPU 1/KVM 0 [000] 0.000000: PERF_RECORD_COMM: CPU 1/KVM:13376/13382
1518 CPU 2/KVM 0 [000] 0.000000: PERF_RECORD_COMM: CPU 2/KVM:13376/13383
1519 CPU 3/KVM 0 [000] 0.000000: PERF_RECORD_COMM: CPU 3/KVM:13376/13384
1521 Note, the QEMU option -name debug-threads=on is needed so that thread names
1522 can be used to determine which thread is running which VCPU as above. libvirt seems to use this by default.
1524 Create a guestmount, assuming the guest machine is 'vm_to_test':
1526 $ mkdir -p ~/guestmount/13376
1527 $ sshfs -o direct_io vm_to_test:/ ~/guestmount/13376
1529 Inject the guest perf.data file into the host perf.data file:
1531 Note, due to the guestmount option, guest object files and debug files will be copied into the build ID cache from the guest machine, with the notable exception of VDSO.
1532 If needed, VDSO can be copied manually in a fashion similar to that used by the perf-archive script.
1534 $ perf inject -i guest-sideband-testing-host-perf.data -o inj --guestmount ~/guestmount --guest-data=guest-sideband-testing-guest-perf.data,13376,0xfffffa6ae070cb20
1536 Show an excerpt from the result. In this case the CPU and time range have been to chosen to show interaction between guest and host when 'uname' is starting to run on the guest machine:
1538 Notes:
1540 - the CPU displayed, [002] in this case, is always the host CPU
1541 - events happening in the virtual machine start with VM:13376 VCPU:003, which shows the hypervisor PID 13376 and the VCPU number
1542 - only calls and errors are displayed i.e. --itrace=ce
1543 - branches entering and exiting the virtual machine are split, and show as 2 branches to/from "0 [unknown] ([unknown])"
1545 $ perf script -i inj --itrace=ce -F+machine_pid,+vcpu,+addr,+pid,+tid,-period --ns --time 7919.408803365,7919.408804631 -C 2
1546 CPU 3/KVM 13376/13384 [002] 7919.408803365: branches: ffffffffc0f8ebe0 vmx_vcpu_enter_exit+0xc0 ([kernel.kallsyms]) => ffffffffc0f8edc0 __vmx_vcpu_run+0x0 ([kernel.kallsyms])
1547 CPU 3/KVM 13376/13384 [002] 7919.408803365: branches: ffffffffc0f8edd5 __vmx_vcpu_run+0x15 ([kernel.kallsyms]) => ffffffffc0f8eca0 vmx_update_host_rsp+0x0 ([kernel.kallsyms])
1548 CPU 3/KVM 13376/13384 [002] 7919.408803365: branches: ffffffffc0f8ee1b __vmx_vcpu_run+0x5b ([kernel.kallsyms]) => ffffffffc0f8ed60 vmx_vmenter+0x0 ([kernel.kallsyms])
1549 CPU 3/KVM 13376/13384 [002] 7919.408803461: branches: ffffffffc0f8ed62 vmx_vmenter+0x2 ([kernel.kallsyms]) => 0 [unknown] ([unknown])
1550 VM:13376 VCPU:003 uname 3404/3404 [002] 7919.408803461: branches: 0 [unknown] ([unknown]) => 7f851c9b5a5c init_cacheinfo+0x3ac (/usr/lib/x86_64-linux-gnu/libc-2.31.so)
1551 VM:13376 VCPU:003 uname 3404/3404 [002] 7919.408803567: branches: 7f851c9b5a5a init_cacheinfo+0x3aa (/usr/lib/x86_64-linux-gnu/libc-2.31.so) => 0 [unknown] ([unknown])
1552 CPU 3/KVM 13376/13384 [002] 7919.408803567: branches: 0 [unknown] ([unknown]) => ffffffffc0f8ed80 vmx_vmexit+0x0 ([kernel.kallsyms])
1553 CPU 3/KVM 13376/13384 [002] 7919.408803596: branches: ffffffffc0f6619a vmx_vcpu_run+0x26a ([kernel.kallsyms]) => ffffffffb2255c60 x86_virt_spec_ctrl+0x0 ([kernel.kallsyms])
1554 CPU 3/KVM 13376/13384 [002] 7919.408803801: branches: ffffffffc0f66445 vmx_vcpu_run+0x515 ([kernel.kallsyms]) => ffffffffb2290b30 native_write_msr+0x0 ([kernel.kallsyms])
1555 CPU 3/KVM 13376/13384 [002] 7919.408803850: branches: ffffffffc0f661f8 vmx_vcpu_run+0x2c8 ([kernel.kallsyms]) => ffffffffc1092300 kvm_load_host_xsave_state+0x0 ([kernel.kallsyms])
1556 CPU 3/KVM 13376/13384 [002] 7919.408803850: branches: ffffffffc1092327 kvm_load_host_xsave_state+0x27 ([kernel.kallsyms]) => ffffffffc1092220 kvm_load_host_xsave_state.part.0+0x0 ([kernel.kallsyms])
1557 CPU 3/KVM 13376/13384 [002] 7919.408803862: branches: ffffffffc0f662cf vmx_vcpu_run+0x39f ([kernel.kallsyms]) => ffffffffc0f63f90 vmx_recover_nmi_blocking+0x0 ([kernel.kallsyms])
1558 CPU 3/KVM 13376/13384 [002] 7919.408803862: branches: ffffffffc0f662e9 vmx_vcpu_run+0x3b9 ([kernel.kallsyms]) => ffffffffc0f619a0 __vmx_complete_interrupts+0x0 ([kernel.kallsyms])
1559 CPU 3/KVM 13376/13384 [002] 7919.408803872: branches: ffffffffc109cfb2 vcpu_enter_guest+0x752 ([kernel.kallsyms]) => ffffffffc0f5f570 vmx_handle_exit_irqoff+0x0 ([kernel.kallsyms])
1560 CPU 3/KVM 13376/13384 [002] 7919.408803881: branches: ffffffffc109d028 vcpu_enter_guest+0x7c8 ([kernel.kallsyms]) => ffffffffb234f900 __srcu_read_lock+0x0 ([kernel.kallsyms])
1561 CPU 3/KVM 13376/13384 [002] 7919.408803897: branches: ffffffffc109d06f vcpu_enter_guest+0x80f ([kernel.kallsyms]) => ffffffffc0f72e30 vmx_handle_exit+0x0 ([kernel.kallsyms])
1562 CPU 3/KVM 13376/13384 [002] 7919.408803897: branches: ffffffffc0f72e3d vmx_handle_exit+0xd ([kernel.kallsyms]) => ffffffffc0f727c0 __vmx_handle_exit+0x0 ([kernel.kallsyms])
1563 CPU 3/KVM 13376/13384 [002] 7919.408803897: branches: ffffffffc0f72b15 __vmx_handle_exit+0x355 ([kernel.kallsyms]) => ffffffffc0f60ae0 vmx_flush_pml_buffer+0x0 ([kernel.kallsyms])
1564 CPU 3/KVM 13376/13384 [002] 7919.408803903: branches: ffffffffc0f72994 __vmx_handle_exit+0x1d4 ([kernel.kallsyms]) => ffffffffc10b7090 kvm_emulate_cpuid+0x0 ([kernel.kallsyms])
1565 CPU 3/KVM 13376/13384 [002] 7919.408803903: branches: ffffffffc10b70f1 kvm_emulate_cpuid+0x61 ([kernel.kallsyms]) => ffffffffc10b6e10 kvm_cpuid+0x0 ([kernel.kallsyms])
1566 CPU 3/KVM 13376/13384 [002] 7919.408803941: branches: ffffffffc10b7125 kvm_emulate_cpuid+0x95 ([kernel.kallsyms]) => ffffffffc1093110 kvm_skip_emulated_instruction+0x0 ([kernel.kallsyms])
1567 CPU 3/KVM 13376/13384 [002] 7919.408803941: branches: ffffffffc109311f kvm_skip_emulated_instruction+0xf ([kernel.kallsyms]) => ffffffffc0f5e180 vmx_get_rflags+0x0 ([kernel.kallsyms])
1568 CPU 3/KVM 13376/13384 [002] 7919.408803951: branches: ffffffffc109312a kvm_skip_emulated_instruction+0x1a ([kernel.kallsyms]) => ffffffffc0f5fd30 vmx_skip_emulated_instruction+0x0 ([kernel.kallsyms])
1569 CPU 3/KVM 13376/13384 [002] 7919.408803951: branches: ffffffffc0f5fd79 vmx_skip_emulated_instruction+0x49 ([kernel.kallsyms]) => ffffffffc0f5fb50 skip_emulated_instruction+0x0 ([kernel.kallsyms])
1570 CPU 3/KVM 13376/13384 [002] 7919.408803956: branches: ffffffffc0f5fc68 skip_emulated_instruction+0x118 ([kernel.kallsyms]) => ffffffffc0f6a940 vmx_cache_reg+0x0 ([kernel.kallsyms])
1571 CPU 3/KVM 13376/13384 [002] 7919.408803964: branches: ffffffffc0f5fc11 skip_emulated_instruction+0xc1 ([kernel.kallsyms]) => ffffffffc0f5f9e0 vmx_set_interrupt_shadow+0x0 ([kernel.kallsyms])
1572 CPU 3/KVM 13376/13384 [002] 7919.408803980: branches: ffffffffc109f8b1 vcpu_run+0x71 ([kernel.kallsyms]) => ffffffffc10ad2f0 kvm_cpu_has_pending_timer+0x0 ([kernel.kallsyms])
1573 CPU 3/KVM 13376/13384 [002] 7919.408803980: branches: ffffffffc10ad2fb kvm_cpu_has_pending_timer+0xb ([kernel.kallsyms]) => ffffffffc10b0490 apic_has_pending_timer+0x0 ([kernel.kallsyms])
1574 CPU 3/KVM 13376/13384 [002] 7919.408803991: branches: ffffffffc109f899 vcpu_run+0x59 ([kernel.kallsyms]) => ffffffffc109c860 vcpu_enter_guest+0x0 ([kernel.kallsyms])
1575 CPU 3/KVM 13376/13384 [002] 7919.408803993: branches: ffffffffc109cd4c vcpu_enter_guest+0x4ec ([kernel.kallsyms]) => ffffffffc0f69140 vmx_prepare_switch_to_guest+0x0 ([kernel.kallsyms])
1576 CPU 3/KVM 13376/13384 [002] 7919.408803996: branches: ffffffffc109cd7d vcpu_enter_guest+0x51d ([kernel.kallsyms]) => ffffffffb234f930 __srcu_read_unlock+0x0 ([kernel.kallsyms])
1577 CPU 3/KVM 13376/13384 [002] 7919.408803996: branches: ffffffffc109cd9c vcpu_enter_guest+0x53c ([kernel.kallsyms]) => ffffffffc0f609b0 vmx_sync_pir_to_irr+0x0 ([kernel.kallsyms])
1578 CPU 3/KVM 13376/13384 [002] 7919.408803996: branches: ffffffffc0f60a6d vmx_sync_pir_to_irr+0xbd ([kernel.kallsyms]) => ffffffffc10adc20 kvm_lapic_find_highest_irr+0x0 ([kernel.kallsyms])
1579 CPU 3/KVM 13376/13384 [002] 7919.408804010: branches: ffffffffc0f60abd vmx_sync_pir_to_irr+0x10d ([kernel.kallsyms]) => ffffffffc0f60820 vmx_set_rvi+0x0 ([kernel.kallsyms])
1580 CPU 3/KVM 13376/13384 [002] 7919.408804019: branches: ffffffffc109ceca vcpu_enter_guest+0x66a ([kernel.kallsyms]) => ffffffffb2249840 fpregs_assert_state_consistent+0x0 ([kernel.kallsyms])
1581 CPU 3/KVM 13376/13384 [002] 7919.408804021: branches: ffffffffc109cf10 vcpu_enter_guest+0x6b0 ([kernel.kallsyms]) => ffffffffc0f65f30 vmx_vcpu_run+0x0 ([kernel.kallsyms])
1582 CPU 3/KVM 13376/13384 [002] 7919.408804024: branches: ffffffffc0f6603b vmx_vcpu_run+0x10b ([kernel.kallsyms]) => ffffffffb229bed0 __get_current_cr3_fast+0x0 ([kernel.kallsyms])
1583 CPU 3/KVM 13376/13384 [002] 7919.408804024: branches: ffffffffc0f66055 vmx_vcpu_run+0x125 ([kernel.kallsyms]) => ffffffffb2253050 cr4_read_shadow+0x0 ([kernel.kallsyms])
1584 CPU 3/KVM 13376/13384 [002] 7919.408804030: branches: ffffffffc0f6608d vmx_vcpu_run+0x15d ([kernel.kallsyms]) => ffffffffc10921e0 kvm_load_guest_xsave_state+0x0 ([kernel.kallsyms])
1585 CPU 3/KVM 13376/13384 [002] 7919.408804030: branches: ffffffffc1092207 kvm_load_guest_xsave_state+0x27 ([kernel.kallsyms]) => ffffffffc1092110 kvm_load_guest_xsave_state.part.0+0x0 ([kernel.kallsyms])
1586 CPU 3/KVM 13376/13384 [002] 7919.408804032: branches: ffffffffc0f660c6 vmx_vcpu_run+0x196 ([kernel.kallsyms]) => ffffffffb22061a0 perf_guest_get_msrs+0x0 ([kernel.kallsyms])
1587 CPU 3/KVM 13376/13384 [002] 7919.408804032: branches: ffffffffb22061a9 perf_guest_get_msrs+0x9 ([kernel.kallsyms]) => ffffffffb220cda0 intel_guest_get_msrs+0x0 ([kernel.kallsyms])
1588 CPU 3/KVM 13376/13384 [002] 7919.408804039: branches: ffffffffc0f66109 vmx_vcpu_run+0x1d9 ([kernel.kallsyms]) => ffffffffc0f652c0 clear_atomic_switch_msr+0x0 ([kernel.kallsyms])
1589 CPU 3/KVM 13376/13384 [002] 7919.408804040: branches: ffffffffc0f66119 vmx_vcpu_run+0x1e9 ([kernel.kallsyms]) => ffffffffc0f73f60 intel_pmu_lbr_is_enabled+0x0 ([kernel.kallsyms])
1590 CPU 3/KVM 13376/13384 [002] 7919.408804042: branches: ffffffffc0f73f81 intel_pmu_lbr_is_enabled+0x21 ([kernel.kallsyms]) => ffffffffc10b68e0 kvm_find_cpuid_entry+0x0 ([kernel.kallsyms])
1591 CPU 3/KVM 13376/13384 [002] 7919.408804045: branches: ffffffffc0f66454 vmx_vcpu_run+0x524 ([kernel.kallsyms]) => ffffffffc0f61ff0 vmx_update_hv_timer+0x0 ([kernel.kallsyms])
1592 CPU 3/KVM 13376/13384 [002] 7919.408804057: branches: ffffffffc0f66142 vmx_vcpu_run+0x212 ([kernel.kallsyms]) => ffffffffc10af100 kvm_wait_lapic_expire+0x0 ([kernel.kallsyms])
1593 CPU 3/KVM 13376/13384 [002] 7919.408804057: branches: ffffffffc0f66156 vmx_vcpu_run+0x226 ([kernel.kallsyms]) => ffffffffb2255c60 x86_virt_spec_ctrl+0x0 ([kernel.kallsyms])
1594 CPU 3/KVM 13376/13384 [002] 7919.408804057: branches: ffffffffc0f66161 vmx_vcpu_run+0x231 ([kernel.kallsyms]) => ffffffffc0f8eb20 vmx_vcpu_enter_exit+0x0 ([kernel.kallsyms])
1595 CPU 3/KVM 13376/13384 [002] 7919.408804057: branches: ffffffffc0f8eb44 vmx_vcpu_enter_exit+0x24 ([kernel.kallsyms]) => ffffffffb2353e10 rcu_note_context_switch+0x0 ([kernel.kallsyms])
1596 CPU 3/KVM 13376/13384 [002] 7919.408804057: branches: ffffffffb2353e1c rcu_note_context_switch+0xc ([kernel.kallsyms]) => ffffffffb2353db0 rcu_qs+0x0 ([kernel.kallsyms])
1597 CPU 3/KVM 13376/13384 [002] 7919.408804066: branches: ffffffffc0f8ebe0 vmx_vcpu_enter_exit+0xc0 ([kernel.kallsyms]) => ffffffffc0f8edc0 __vmx_vcpu_run+0x0 ([kernel.kallsyms])
1598 CPU 3/KVM 13376/13384 [002] 7919.408804066: branches: ffffffffc0f8edd5 __vmx_vcpu_run+0x15 ([kernel.kallsyms]) => ffffffffc0f8eca0 vmx_update_host_rsp+0x0 ([kernel.kallsyms])
1599 CPU 3/KVM 13376/13384 [002] 7919.408804066: branches: ffffffffc0f8ee1b __vmx_vcpu_run+0x5b ([kernel.kallsyms]) => ffffffffc0f8ed60 vmx_vmenter+0x0 ([kernel.kallsyms])
1600 CPU 3/KVM 13376/13384 [002] 7919.408804162: branches: ffffffffc0f8ed62 vmx_vmenter+0x2 ([kernel.kallsyms]) => 0 [unknown] ([unknown])
1601 VM:13376 VCPU:003 uname 3404/3404 [002] 7919.408804162: branches: 0 [unknown] ([unknown]) => 7f851c9b5a5c init_cacheinfo+0x3ac (/usr/lib/x86_64-linux-gnu/libc-2.31.so)
1602 VM:13376 VCPU:003 uname 3404/3404 [002] 7919.408804273: branches: 7f851cb7c0e4 _dl_init+0x74 (/usr/lib/x86_64-linux-gnu/ld-2.31.so) => 7f851cb7bf50 call_init.part.0+0x0 (/usr/lib/x86_64-linux-gnu/ld-2.31.so)
1603 VM:13376 VCPU:003 uname 3404/3404 [002] 7919.408804526: branches: 55e0c00136f0 _start+0x0 (/usr/bin/uname) => ffffffff83200ac0 asm_exc_page_fault+0x0 ([kernel.kallsyms])
1604 VM:13376 VCPU:003 uname 3404/3404 [002] 7919.408804526: branches: ffffffff83200ac3 asm_exc_page_fault+0x3 ([kernel.kallsyms]) => ffffffff83201290 error_entry+0x0 ([kernel.kallsyms])
1605 VM:13376 VCPU:003 uname 3404/3404 [002] 7919.408804534: branches: ffffffff832012fa error_entry+0x6a ([kernel.kallsyms]) => ffffffff830b59a0 sync_regs+0x0 ([kernel.kallsyms])
1606 VM:13376 VCPU:003 uname 3404/3404 [002] 7919.408804631: branches: ffffffff83200ad9 asm_exc_page_fault+0x19 ([kernel.kallsyms]) => ffffffff830b8210 exc_page_fault+0x0 ([kernel.kallsyms])
1607 VM:13376 VCPU:003 uname 3404/3404 [002] 7919.408804631: branches: ffffffff830b82a4 exc_page_fault+0x94 ([kernel.kallsyms]) => ffffffff830b80e0 __kvm_handle_async_pf+0x0 ([kernel.kallsyms])
1608 VM:13376 VCPU:003 uname 3404/3404 [002] 7919.408804631: branches: ffffffff830b80ed __kvm_handle_async_pf+0xd ([kernel.kallsyms]) => ffffffff830b80c0 kvm_read_and_reset_apf_flags+0x0 ([kernel.kallsyms])
1611 Tracing Virtual Machines - Guest Code
1612 -------------------------------------
1614 A common case for KVM test programs is that the test program acts as the
1615 hypervisor, creating, running and destroying the virtual machine, and
1616 providing the guest object code from its own object code. In this case,
1617 the VM is not running an OS, but only the functions loaded into it by the
1618 hypervisor test program, and conveniently, loaded at the same virtual
1619 addresses. To support that, option "--guest-code" has been added to perf script
1620 and perf kvm report.
1622 Here is an example tracing a test program from the kernel's KVM selftests:
1624 # perf record --kcore -e intel_pt/cyc/ -- tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test
1625 [ perf record: Woken up 1 times to write data ]
1626 [ perf record: Captured and wrote 0.280 MB perf.data ]
1627 # perf script --guest-code --itrace=bep --ns -F-period,+addr,+flags
1628 [SNIP]
1629 tsc_msrs_test 18436 [007] 10897.962087733: branches: call ffffffffc13b2ff5 __vmx_vcpu_run+0x15 (vmlinux) => ffffffffc13b2f50 vmx_update_host_rsp+0x0 (vmlinux)
1630 tsc_msrs_test 18436 [007] 10897.962087733: branches: return ffffffffc13b2f5d vmx_update_host_rsp+0xd (vmlinux) => ffffffffc13b2ffa __vmx_vcpu_run+0x1a (vmlinux)
1631 tsc_msrs_test 18436 [007] 10897.962087733: branches: call ffffffffc13b303b __vmx_vcpu_run+0x5b (vmlinux) => ffffffffc13b2f80 vmx_vmenter+0x0 (vmlinux)
1632 tsc_msrs_test 18436 [007] 10897.962087836: branches: vmentry ffffffffc13b2f82 vmx_vmenter+0x2 (vmlinux) => 0 [unknown] ([unknown])
1633 [guest/18436] 18436 [007] 10897.962087836: branches: vmentry 0 [unknown] ([unknown]) => 402c81 guest_code+0x131 (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test)
1634 [guest/18436] 18436 [007] 10897.962087836: branches: call 402c81 guest_code+0x131 (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test) => 40dba0 ucall+0x0 (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test)
1635 [guest/18436] 18436 [007] 10897.962088248: branches: vmexit 40dba0 ucall+0x0 (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test) => 0 [unknown] ([unknown])
1636 tsc_msrs_test 18436 [007] 10897.962088248: branches: vmexit 0 [unknown] ([unknown]) => ffffffffc13b2fa0 vmx_vmexit+0x0 (vmlinux)
1637 tsc_msrs_test 18436 [007] 10897.962088248: branches: jmp ffffffffc13b2fa0 vmx_vmexit+0x0 (vmlinux) => ffffffffc13b2fd2 vmx_vmexit+0x32 (vmlinux)
1638 tsc_msrs_test 18436 [007] 10897.962088256: branches: return ffffffffc13b2fd2 vmx_vmexit+0x32 (vmlinux) => ffffffffc13b3040 __vmx_vcpu_run+0x60 (vmlinux)
1639 tsc_msrs_test 18436 [007] 10897.962088270: branches: return ffffffffc13b30b6 __vmx_vcpu_run+0xd6 (vmlinux) => ffffffffc13b2f2e vmx_vcpu_enter_exit+0x4e (vmlinux)
1640 [SNIP]
1641 tsc_msrs_test 18436 [007] 10897.962089321: branches: call ffffffffc13b2ff5 __vmx_vcpu_run+0x15 (vmlinux) => ffffffffc13b2f50 vmx_update_host_rsp+0x0 (vmlinux)
1642 tsc_msrs_test 18436 [007] 10897.962089321: branches: return ffffffffc13b2f5d vmx_update_host_rsp+0xd (vmlinux) => ffffffffc13b2ffa __vmx_vcpu_run+0x1a (vmlinux)
1643 tsc_msrs_test 18436 [007] 10897.962089321: branches: call ffffffffc13b303b __vmx_vcpu_run+0x5b (vmlinux) => ffffffffc13b2f80 vmx_vmenter+0x0 (vmlinux)
1644 tsc_msrs_test 18436 [007] 10897.962089424: branches: vmentry ffffffffc13b2f82 vmx_vmenter+0x2 (vmlinux) => 0 [unknown] ([unknown])
1645 [guest/18436] 18436 [007] 10897.962089424: branches: vmentry 0 [unknown] ([unknown]) => 40dba0 ucall+0x0 (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test)
1646 [guest/18436] 18436 [007] 10897.962089701: branches: jmp 40dc1b ucall+0x7b (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test) => 40dc39 ucall+0x99 (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test)
1647 [guest/18436] 18436 [007] 10897.962089701: branches: jcc 40dc3c ucall+0x9c (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test) => 40dc20 ucall+0x80 (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test)
1648 [guest/18436] 18436 [007] 10897.962089701: branches: jcc 40dc3c ucall+0x9c (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test) => 40dc20 ucall+0x80 (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test)
1649 [guest/18436] 18436 [007] 10897.962089701: branches: jcc 40dc37 ucall+0x97 (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test) => 40dc50 ucall+0xb0 (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test)
1650 [guest/18436] 18436 [007] 10897.962089878: branches: vmexit 40dc55 ucall+0xb5 (/home/user/git/work/tools/testing/selftests/kselftest_install/kvm/tsc_msrs_test) => 0 [unknown] ([unknown])
1651 tsc_msrs_test 18436 [007] 10897.962089878: branches: vmexit 0 [unknown] ([unknown]) => ffffffffc13b2fa0 vmx_vmexit+0x0 (vmlinux)
1652 tsc_msrs_test 18436 [007] 10897.962089878: branches: jmp ffffffffc13b2fa0 vmx_vmexit+0x0 (vmlinux) => ffffffffc13b2fd2 vmx_vmexit+0x32 (vmlinux)
1653 tsc_msrs_test 18436 [007] 10897.962089887: branches: return ffffffffc13b2fd2 vmx_vmexit+0x32 (vmlinux) => ffffffffc13b3040 __vmx_vcpu_run+0x60 (vmlinux)
1654 tsc_msrs_test 18436 [007] 10897.962089901: branches: return ffffffffc13b30b6 __vmx_vcpu_run+0xd6 (vmlinux) => ffffffffc13b2f2e vmx_vcpu_enter_exit+0x4e (vmlinux)
1655 [SNIP]
1657 # perf kvm --guest-code --guest --host report -i perf.data --stdio | head -20
1659 # To display the perf.data header info, please use --header/--header-only options.
1660 #
1661 #
1662 # Total Lost Samples: 0
1663 #
1664 # Samples: 12 of event 'instructions'
1665 # Event count (approx.): 2274583
1666 #
1667 # Children Self Command Shared Object Symbol
1668 # ........ ........ ............. .................... ...........................................
1669 #
1670 54.70% 0.00% tsc_msrs_test [kernel.vmlinux] [k] entry_SYSCALL_64_after_hwframe
1671 |
1672 ---entry_SYSCALL_64_after_hwframe
1673 do_syscall_64
1674 |
1675 |--29.44%--syscall_exit_to_user_mode
1676 | exit_to_user_mode_prepare
1677 | task_work_run
1678 | __fput
1681 Event Trace
1682 -----------
1684 Event Trace records information about asynchronous events, for example interrupts,
1685 faults, VM exits and entries. The information is recorded in CFE and EVD packets,
1686 and also the Interrupt Flag is recorded on the MODE.Exec packet. The CFE packet
1687 contains a type field to identify one of the following:
1689 1 INTR interrupt, fault, exception, NMI
1690 2 IRET interrupt return
1691 3 SMI system management interrupt
1692 4 RSM resume from system management mode
1693 5 SIPI startup interprocessor interrupt
1694 6 INIT INIT signal
1695 7 VMENTRY VM-Entry
1696 8 VMEXIT VM-Entry
1697 9 VMEXIT_INTR VM-Exit due to interrupt
1698 10 SHUTDOWN Shutdown
1700 For more details, refer to the Intel 64 and IA-32 Architectures Software
1701 Developer Manuals (version 076 or later).
1703 The capability to do Event Trace is indicated by the
1704 /sys/bus/event_source/devices/intel_pt/caps/event_trace file.
1706 Event trace is selected for recording using the "event" config term. e.g.
1708 perf record -e intel_pt/event/u uname
1710 Event trace events are output using the --itrace I option. e.g.
1712 perf script --itrace=Ie
1714 perf script displays events containing CFE type, vector and event data,
1715 in the form:
1717 evt: hw int (t) cfe: INTR IP: 1 vector: 3 PFA: 0x8877665544332211
1719 The IP flag indicates if the event binds to an IP, which includes any case where
1720 flow control packet generation is enabled, as well as when CFE packet IP bit is
1721 set.
1723 perf script displays events containing changes to the Interrupt Flag in the form:
1725 iflag: t IFLAG: 1->0 via branch
1727 where "via branch" indicates a branch (interrupt or return from interrupt) and
1728 "non branch" indicates an instruction such as CFI, STI or POPF).
1730 In addition, the current state of the interrupt flag is indicated by the presence
1731 or absence of the "D" (interrupt disabled) perf script flag. If the interrupt
1732 flag is changed, then the "t" flag is also included i.e.
1734 no flag, interrupts enabled IF=1
1735 t interrupts become disabled IF=1 -> IF=0
1736 D interrupts are disabled IF=0
1737 Dt interrupts become enabled IF=0 -> IF=1
1739 The intel-pt-events.py script illustrates how to access Event Trace information
1740 using a Python script.
1743 TNT Disable
1744 -----------
1746 TNT packets are disabled using the "notnt" config term. e.g.
1748 perf record -e intel_pt/notnt/u uname
1750 In that case the --itrace q option is forced because walking executable code
1751 to reconstruct the control flow is not possible.
1754 Emulated PTWRITE
1755 ----------------
1757 Later perf tools support a method to emulate the ptwrite instruction, which
1758 can be useful if hardware does not support the ptwrite instruction.
1760 Instead of using the ptwrite instruction, a function is used which produces
1761 a trace that encodes the payload data into TNT packets. Here is an example
1762 of the function:
1764 #include <stdint.h>
1766 void perf_emulate_ptwrite(uint64_t x)
1767 __attribute__((externally_visible, noipa, no_instrument_function, naked));
1769 #define PERF_EMULATE_PTWRITE_8_BITS \
1770 "1: shl %rax\n" \
1771 " jc 1f\n" \
1772 "1: shl %rax\n" \
1773 " jc 1f\n" \
1774 "1: shl %rax\n" \
1775 " jc 1f\n" \
1776 "1: shl %rax\n" \
1777 " jc 1f\n" \
1778 "1: shl %rax\n" \
1779 " jc 1f\n" \
1780 "1: shl %rax\n" \
1781 " jc 1f\n" \
1782 "1: shl %rax\n" \
1783 " jc 1f\n" \
1784 "1: shl %rax\n" \
1785 " jc 1f\n"
1787 /* Undefined instruction */
1788 #define PERF_EMULATE_PTWRITE_UD2 ".byte 0x0f, 0x0b\n"
1790 #define PERF_EMULATE_PTWRITE_MAGIC PERF_EMULATE_PTWRITE_UD2 ".ascii \"perf,ptwrite \"\n"
1792 void perf_emulate_ptwrite(uint64_t x __attribute__ ((__unused__)))
1793 {
1794 /* Assumes SysV ABI : x passed in rdi */
1795 __asm__ volatile (
1796 "jmp 1f\n"
1797 PERF_EMULATE_PTWRITE_MAGIC
1798 "1: mov %rdi, %rax\n"
1799 PERF_EMULATE_PTWRITE_8_BITS
1800 PERF_EMULATE_PTWRITE_8_BITS
1801 PERF_EMULATE_PTWRITE_8_BITS
1802 PERF_EMULATE_PTWRITE_8_BITS
1803 PERF_EMULATE_PTWRITE_8_BITS
1804 PERF_EMULATE_PTWRITE_8_BITS
1805 PERF_EMULATE_PTWRITE_8_BITS
1806 PERF_EMULATE_PTWRITE_8_BITS
1807 "1: ret\n"
1808 );
1809 }
1811 For example, a test program with the function above:
1813 #include <stdio.h>
1814 #include <stdint.h>
1815 #include <stdlib.h>
1817 #include "perf_emulate_ptwrite.h"
1819 int main(int argc, char *argv[])
1820 {
1821 uint64_t x = 0;
1823 if (argc > 1)
1824 x = strtoull(argv[1], NULL, 0);
1825 perf_emulate_ptwrite(x);
1826 return 0;
1827 }
1829 Can be compiled and traced:
1831 $ gcc -Wall -Wextra -O3 -g -o eg_ptw eg_ptw.c
1832 $ perf record -e intel_pt//u ./eg_ptw 0x1234567890abcdef
1833 [ perf record: Woken up 1 times to write data ]
1834 [ perf record: Captured and wrote 0.017 MB perf.data ]
1835 $ perf script --itrace=ew
1836 eg_ptw 19875 [007] 8061.235912: ptwrite: IP: 0 payload: 0x1234567890abcdef 55701249a196 perf_emulate_ptwrite+0x16 (/home/user/eg_ptw)
1837 $
1840 Pipe mode
1841 ---------
1842 Pipe mode is a problem for Intel PT and possibly other auxtrace users.
1843 It's not recommended to use a pipe as data output with Intel PT because
1844 of the following reason.
1846 Essentially the auxtrace buffers do not behave like the regular perf
1847 event buffers. That is because the head and tail are updated by
1848 software, but in the auxtrace case the data is written by hardware.
1849 So the head and tail do not get updated as data is written.
1851 In the Intel PT case, the head and tail are updated only when the trace
1852 is disabled by software, for example:
1853 - full-trace, system wide : when buffer passes watermark
1854 - full-trace, not system-wide : when buffer passes watermark or
1855 context switches
1856 - snapshot mode : as above but also when a snapshot is made
1857 - sample mode : as above but also when a sample is made
1859 That means finished-round ordering doesn't work. An auxtrace buffer
1860 can turn up that has data that extends back in time, possibly to the
1861 very beginning of tracing.
1863 For a perf.data file, that problem is solved by going through the trace
1864 and queuing up the auxtrace buffers in advance.
1866 For pipe mode, the order of events and timestamps can presumably
1867 be messed up.
1870 EXAMPLE
1871 -------
1873 Examples can be found on perf wiki page "Perf tools support for Intel® Processor Trace":
1875 https://perf.wiki.kernel.org/index.php/Perf_tools_support_for_Intel%C2%AE_Processor_Trace
1878 SEE ALSO
1879 --------
1881 linkperf:perf-record[1], linkperf:perf-script[1], linkperf:perf-report[1],
1882 linkperf:perf-inject[1]