1 .TH BLKTPARSE 1 "March 6, 2007" "blktrace git\-20070306202522" ""
5 blkparse \- produce formatted output of event streams of block devices
9 .B blkparse [ \fIoptions\fR ]
14 The \fIblkparse\fR utility will attempt to combine streams of events for
15 various devices on various CPUs, and produce a formatted output of the event
16 information. Specifically, it will take the (machine-readable) output of the
17 \fIblktrace\fR utility and convert it to a nicely formatted and human-readable
20 As with \fIblktrace\fR, some details concerning \fIblkparse\fR
21 will help in understanding the command line options presented below.
26 By default, \fIblkparse\fR expects to run in a post-processing mode; one where
27 the trace events have been saved by a previous run of blktrace, and blkparse
28 is combining event streams and dumping formatted data.
30 blkparse may be run in a live manner concurrently with blktrace by specifying
31 \fB\-i \-\fR to blkparse, and combining it with the live option for blktrace.
34 % blktrace \-d /dev/sda \-o \- | blkparse \-i \-
38 You can set how many blkparse batches event reads via the \fB\-b\fR option, the
39 default is to handle events in batches of 512.
43 If you have saved event traces in blktrace with different output names (via
44 the \fB\-o\fR option to blktrace), you must specify the same input name via the
49 The format of the output data can be controlled via the \fB\-f\fR or \fB\-F\fR
50 options \-\- see OUTPUT DESCRIPTION AND FORMATTING for details.
53 By default, blkparse sends formatted data to standard output. This may
54 be changed via the \fB\-o\fR option, or text output can be disabled via the
55 \fB\-O\fR option. A merged binary stream can be produced using the \fB\-d\fR
66 Standard input read batching
73 Specifies base name for input files \-\- default is \fIdevice\fR.blktrace.\fIcpu\fR.
75 As noted above, specifying \fB\-i \-\fR runs in live mode with blktrace
76 (reading data from standard in).
81 \-\-format=\fItyp,fmt\fR
85 \-\-format\-spec=\fIfmt\fR
88 (See OUTPUT DESCRIPTION AND FORMATTING for details.)
90 The \-f form specifies a format for all events
92 The \-F form allows one to specify a format for a specific
93 event type. The single\-character \fItyp\fR field is one of the
94 action specifiers described in ACTION IDENTIFIERS.
101 Print missing entries
108 When \-d is specified, this will stop messages from being output to the
109 file. (Can seriously reduce the size of the resultant file when using
110 the CFQ I/O scheduler.)
117 Hash processes by name, not by PID
122 \-\-output=\fIfile\fR
131 Do \fInot\fR produce text output, used for binary (\fB\-d\fR) only
136 \-\-dump\-binary=\fIfile\fR
150 \-\-per\-program\-stats
152 Displays data sorted by program
159 Display time deltas per IO
164 \-\-stopwatch=\fIspan\fR
166 Display traces for the \fIspan\fR specified \-\- where span can be:
168 \fIend\-time\fR \-\- Display traces from time 0 through \fIend\-time\fR (in ns)
172 \fIstart:end\-time\fR \-\- Display traces from time \fIstart\fR
173 through end\-time (in ns).
180 More verbose marginal on marginal errors
192 The following trace actions are recognised:
196 A previously issued request has been completed. The output will detail the
197 sector and size of that request, as well as the success or failure of it.
201 A request that previously resided on the block layer queue or in the i/o
202 scheduler has been sent to the driver.
206 A request is being sent to the i/o scheduler for addition to the internal queue
207 and later service by the driver. The request is fully formed at this time.
211 This notes intent to queue i/o at the given location. No real requests exists
216 The data pages attached to this \fIbio\fR are not reachable by the hardware
217 and must be bounced to a lower memory location. This causes a big slowdown in
218 i/o performance, since the data must be copied to/from kernel buffers. Usually
219 this can be fixed with using better hardware -- either a better i/o controller,
220 or a platform with an IOMMU.
223 \fBM -- back merge\fR
224 A previously inserted request exists that ends on the boundary of where this i/o
225 begins, so the i/o scheduler can merge them together.
228 \fBF -- front merge\fR
229 Same as the back merge, except this i/o ends where a previously inserted
233 \fBM --front or back merge\fR
237 \fBM -- front or back merge\fR
241 \fBG -- get request\fR
242 To send any type of request to a block device, a \fIstruct request\fR
243 container must be allocated first.
247 No available request structures were available, so the issuer has to wait for
252 When i/o is queued to a previously empty block device queue, Linux will plug the
253 queue in anticipation of future ios being added before this data is needed.
257 Some request data already queued in the device, start sending requests to the
258 driver. This may happen automatically if a timeout period has passed (see next
259 entry) or if a number of requests have been added to the queue.
262 \fBT -- unplug due to timer\fR
263 If nobody requests the i/o that was queued after plugging the queue, Linux will
264 automatically unplug it after a defined period has passed.
268 On raid or device mapper setups, an incoming i/o may straddle a device or
269 internal zone and needs to be chopped up into smaller pieces for service. This
270 may indicate a performance problem due to a bad setup of that raid/dm device,
271 but may also just be part of normal boundary conditions. dm is notably bad at
272 this and will clone lots of i/o.
276 For stacked devices, incoming i/o is remapped to device below it in the i/o
277 stack. The remap action details what exactly is being remapped to what.
282 .SH "OUTPUT DESCRIPTION AND FORMATTING"
284 The output from blkparse can be tailored for specific use -- in particular, to ease
285 parsing of output, and/or limit output fields to those the user wants to see. The
286 data for fields which can be output include:
289 Action, a (small) string (1 or 2 characters) -- see table below for more details
298 RWBS field, a (small) string (1-3 characters) -- see section below for more details
301 7-character string containing the major and minor numbers of
302 the event's device (separated by a comma).
308 Minor number of event's device.
311 Major number of event's device.
323 Display packet data \-\- series of hexadecimal values
332 Time stamp (nanoseconds)
338 Elapsed value in microseconds (\fI\-t\fR command line option)
341 Payload unsigned integer
344 Note that the user can optionally specify field display width, and optionally a
345 left-aligned specifier. These precede field specifiers, with a '%' character,
346 followed by the optional left-alignment specifier (\-) followed by the width (a
347 decimal number) and then the field.
349 Thus, to specify the command in a 12-character field that is left aligned:
354 .SH "ACTION IDENTIFIERS"
356 The following table shows the various actions which may be output:
359 IO was remapped to a different device
371 IO front merged with request on queue
377 IO inserted onto request queue
380 IO back merged with request on queue
386 IO handled by request queue code
392 Unplug due to timeout
401 .SH "RWBS DESCRIPTION"
403 This is a small string containing at least one character ('R' for read, 'W'
404 for write, or 'D' for block discard operation), and optionally either
405 a 'B' (for barrier operations) or 'S' (for synchronous operations).
410 The standard header (or initial fields displayed) include:
412 "%D %2c %8s %5T.%9t %5p %2a %3d"
417 Displays the event's device major/minor as: %3d,%\-3d.
420 CPU ID (2-character field).
426 5-character field for the seconds portion of the time stamp and a 9-character field for the nanoseconds in the time stamp.
429 5-character field for the process ID.
432 2-character field for one of the actions.
435 3-character field for the RWBS data.
437 Seeing this in action:
439 8,0 3 1 0.000000000 697 G W 223490 + 8 [kjournald]
441 The header is the data in this line up to the 223490 (starting block).
442 The default output for all event types includes this header.
446 .SH "DEFAULT OUTPUT PER ACTION"
448 \fBC \-\- complete\fR
450 If a payload is present, this is presented between
451 parenthesis following the header, followed by the error value.
453 If no payload is present, the sector and number of blocks are presented
454 (with an intervening plus (+) character). If the \fB\-t\fR option
455 was specified, then the elapsed time is presented. In either case,
456 it is followed by the error value for the completion.
463 \fBI \-\- inserted\fR
467 If a payload is present, the number of payload bytes
468 is output, followed by the payload in hexadecimal between parenthesis.
470 If no payload is present, the sector and number of blocks are presented
471 (with an intervening plus (+) character). If the \fB\-t\fR option was
472 specified, then the elapsed time is presented (in parenthesis). In
473 either case, it is followed by the command associated with the event
474 (surrounded by square brackets).
477 \fBF \-\- front merge\fR
479 \fBG \-\- get request\fR
481 \fBM \-\- back merge\fR
485 The starting sector and number of blocks is output
486 (with an intervening plus (+) character), followed by the command
487 associated with the event (surrounded by square brackets).
492 The command associated with the event (surrounded by
493 square brackets) is output.
498 \fBT \-\- unplug due to timer\fR
500 The command associated with the event
501 (surrounded by square brackets) is output, followed by the number of
502 requests outstanding.
507 The original starting sector followed by the new
508 sector (separated by a slash (/) is output, followed by the command
509 associated with the event (surrounded by square brackets).
514 Sector and length is output, along with the original
515 device and sector offset.
520 To trace the i/o on the device \fI/dev/hda\fB and parse the output to human
521 readable form, use the following command:
523 % blktrace \-d /dev/sda \-o \- | blkparse \-i \-
525 (see \fIblktrace\fR (8) for more information).
526 This same behaviour can be achieve with the convenience script \fIbtrace\fR.
531 has exactly the same effect as the previous command. See \fIbtrace\fR (8) for
534 To trace the i/o on a device and save the output for later processing with
535 \fIblkparse\fR, use \fIblktrace\fR like this:
537 % blktrace /dev/sda /dev/sdb
539 This will trace i/o on the devices \fI/dev/sda\fR and \fI/dev/sdb\fR and save
540 the recorded information in the files \fIsda\fR and \fIsdb\fR in the current
541 directory, for the two different devices, respectively. This trace
542 information can later be parsed by the \fIblkparse\fR utility:
546 which will output the previously recorded tracing information in human
547 readable form to stdout.
551 \fIblkparse\fR was written by Jens Axboe, Alan D. Brunelle and Nathan Scott. This
552 man page was created from the \fIblktrace\fR documentation by Bas Zoetekouw.
556 Report bugs to <linux\-btrace@vger.kernel.org>
559 Copyright \(co 2006 Jens Axboe, Alan D. Brunelle and Nathan Scott.
561 This is free software. You may redistribute copies of it under the terms of
562 the GNU General Public License <http://www.gnu.org/licenses/gpl.html>.
563 There is NO WARRANTY, to the extent permitted by law.
565 This manual page was created for Debian by Bas Zoetekouw. It was derived from
566 the documentation provided by the authors and it may be used, distributed and
567 modified under the terms of the GNU General Public License, version 2.
569 On Debian systems, the text of the GNU General Public License can be found in
570 /usr/share/common\-licenses/GPL\-2.
573 btrace (8), blktrace (8), verify_blkparse (1), blkrawverify (1), btt (1)