1 \documentclass{article}
4 % Copyright (C) 2005 Alan D. Brunelle <Alan.Brunelle@hp.com>
6 % This program is free software; you can redistribute it and/or modify
7 % it under the terms of the GNU General Public License as published by
8 % the Free Software Foundation; either version 2 of the License, or
9 % (at your option) any later version.
11 % This program is distributed in the hope that it will be useful,
12 % but WITHOUT ANY WARRANTY; without even the implied warranty of
13 % MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 % GNU General Public License for more details.
16 % You should have received a copy of the GNU General Public License
17 % along with this program; if not, write to the Free Software
18 % Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
21 \title{blktrace User Guide}
22 \author{blktrace: Jens Axboe (axboe@suse.de)\\
23 User Guide: Alan D. Brunelle (Alan.Brunelle@hp.com)}
28 %---------------------
29 \section{\label{sec:intro}Introduction}
31 blktrace is a block layer IO tracing mechanism which provides detailed
32 information about request queue operations up to user space. There are
33 three major components that are provided:
36 \item[Kernel patch] A patch to the Linux kernel which includes the
37 kernel event logging interfaces, and patches to areas within the block
38 layer to emit event traces.
40 \item[blktrace] A utility which transfers event traces from the kernel
41 into either long-term on-disk storage, or provides direct formatted
42 output (via blkparse).
44 \item[blkparse] A utility which formats events stored in files, or when
45 run in \emph{live} mode directly outputs data collected by blktrace.
48 \subsection{blktrace Download Area}
50 The blktrace and blkparse utilities and associated kernel patch are provided
51 as part of the following git repository:
53 rsync://rsync.kernel.org/pub/scm/linux/kernel/git/axboe/blktrace.git
55 %--------------------------
56 \newpage\section{\label{sec:quick-start}Quick Start Guide}
58 The following sections outline some quick steps towards utilizing
59 blktrace. Some of the specific instructions below may need to be tailored
62 \subsection{\label{sec:get-blktrace}Retrieving blktrace}
64 As noted above, the kernel patch along with the blktrace and blkparse utilities are stored in a git repository. One simple way to get going would be:
67 % git clone rsync://rsync.kernel.org/pub/scm/linux/kernel/git/axboe/blktrace.git bt
69 % git-read-tree -m HEAD && git-checkout-cache -q -f -u -a
72 \subsection{\label{sec:patching}Patching and configuring the Linux kernel}
74 A patch for a \emph{specific Linux kernel} is provided in bt/kernel (where
75 \emph{bt} is the name of the directory from the above git sequence). The
76 detailed actual patching instructions for a Linux kernel is outside the
77 scope of this document, but the following may be used as a sample template.
79 As an example, bt/kernel contains blk-trace-2.6.14-rc1-git-G2, download
80 linux-2.6.13.tar.bz2 and patch-2.6.14-rc1.bz2
83 % tar xjf linux-2.6.13.tar.bz2
84 % mv linux-2.6.13 linux-2.6.14-rc1
86 % bunzip2 -c ../patch-2.6.14-rc1.bz2 | patch -p1
89 At this point you may (optionally) remove linux-2.6.13.tar.bz2 and
92 At this point you should configure the Linux kernel for your specific
93 system -- again, outside the scope of this document -- and then enable
94 \emph{Support for tracing block io actions.} To do this, run
97 % make menuconfig or make xconfig, or edit .config, or ...
100 and navigate through \emph{Device Drivers} and \emph{Block devices}
101 and then down to \emph{Support for tracing block io actions} and hit Y.
103 Install the new kernel (and modules\ldots) and reboot.
105 \subsection{\label{sec:mount}Mounting the RelayFS file system}
107 blktrace utilizes files under the Relay file system, and thus must have
108 the mount point set up -- mounted on the directory /relay. To do this
109 one may do either of the following:
112 \item Manually mount after each boot:
114 % mount -t relayfs relayfs /relay
117 \item Add an entry into /etc/fstab, and have it done automatically at
118 each boot\footnote{Note: after adding the entry to /etc/fstab, you
119 could then mount the directory this time only by doing: \% mount /relay}:
121 relay /relay relayfs default 0 0
125 \subsection{\label{sec:build}Build the tools}
127 To build and install the tools, execute the following sequence (as root):
131 % make && make install
134 \subsection{\label{sec:live-blktrace}blktrace -- live}
136 Now to simply watch what is going on for a specific disk (to stop the
137 trace, hit control-C):
140 % blktrace -d /dev/sda -o - | blkparse -i -
141 8,0 3 1 0.000000000 697 G W 223490 + 8 [kjournald]
142 8,0 3 2 0.000001829 697 P R [kjournald]
143 8,0 3 3 0.000002197 697 Q W 223490 + 8 [kjournald]
144 8,0 3 4 0.000005533 697 M W 223498 + 8 [kjournald]
145 8,0 3 5 0.000008607 697 M W 223506 + 8 [kjournald]
146 8,0 3 6 0.000011569 697 M W 223514 + 8 [kjournald]
147 8,0 3 7 0.000014407 697 M W 223522 + 8 [kjournald]
148 8,0 3 8 0.000017367 697 M W 223530 + 8 [kjournald]
149 8,0 3 9 0.000020161 697 M W 223538 + 8 [kjournald]
150 8,0 3 10 0.000024062 697 D W 223490 + 56 [kjournald]
151 8,0 1 11 0.009507758 0 C W 223490 + 56 [0]
152 8,0 1 12 0.009538995 697 G W 223546 + 8 [kjournald]
153 8,0 1 13 0.009540033 697 P R [kjournald]
154 8,0 1 14 0.009540313 697 Q W 223546 + 8 [kjournald]
155 8,0 1 15 0.009542980 697 D W 223546 + 8 [kjournald]
156 8,0 1 16 0.013542170 0 C W 223546 + 8 [0]
161 Reads Queued: 0, 0KiB Writes Queued: 7, 128KiB
162 Read Dispatches: 0, 0KiB Write Dispatches: 7, 128KiB
163 Reads Completed: 0, 0KiB Writes Completed: 11, 168KiB
164 Read Merges: 0 Write Merges: 25
165 IO unplugs: 0 Timer unplugs: 0
168 Reads Queued: 0, 0KiB Writes Queued: 1, 28KiB
169 Read Dispatches: 0, 0KiB Write Dispatches: 1, 28KiB
170 Reads Completed: 0, 0KiB Writes Completed: 0, 0KiB
171 Read Merges: 0 Write Merges: 6
172 IO unplugs: 0 Timer unplugs: 0
175 Reads Queued: 0, 0KiB Writes Queued: 11, 168KiB
176 Read Dispatches: 0, 0KiB Write Dispatches: 11, 168KiB
177 Reads Completed: 0, 0KiB Writes Completed: 11, 168KiB
178 Read Merges: 0 Write Merges: 31
179 IO unplugs: 0 Timer unplugs: 3
181 Events (8,0): 89 entries, 0 skips
184 \subsection{\label{sec:blktrace-post}blktrace -- post-processing}
186 Another way to run blktrace is to have blktrace save data away for later
187 formatting by blkparse. This would be useful if you want to get
188 measurements while running specific loads.
190 To do this, one would specify the device (or devices) to be watched. Then
191 go run you test cases. Stop the trace, and at your leisure utilize
192 blkparse to see the results.
194 In this example, devices /dev/sdaa, /dev/sdc and /dev/sdo are used in an
195 LVM volume called adb3/vol.
198 % blktrace /dev/sdaa /dev/sdc /dev/sdo &
201 % mkfs -t ext3 /dev/adb3/vol
202 mke2fs 1.35 (28-Feb-2004)
205 Block size=4096 (log=2)
206 Fragment size=4096 (log=2)
207 16793600 inodes, 33555456 blocks
208 1677772 blocks (5.00%) reserved for the super user
210 Maximum filesystem blocks=4294967296
212 32768 blocks per group, 32768 fragments per group
213 16384 inodes per group
214 Superblock backups stored on blocks:
215 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
216 4096000, 7962624, 11239424, 20480000, 23887872
218 Writing inode tables: done
219 Creating journal (8192 blocks): done
220 Writing superblocks and filesystem accounting information: done
222 This filesystem will be automatically checked every 27 mounts or
223 180 days, whichever comes first. Use tune2fs -c or -i to override.
228 Then you could process the events later:
232 % blkparse sdaa sdc sdo > events
234 8,32 1 1 0.000000000 9728 G R 384 + 32 [mkfs.ext3]
235 8,32 1 2 0.000001959 9728 P R [mkfs.ext3]
236 8,32 1 3 0.000002446 9728 Q R 384 + 32 [mkfs.ext3]
237 8,32 1 4 0.000005110 9728 D R 384 + 32 [mkfs.ext3]
238 8,32 3 5 0.000200570 0 C R 384 + 32 [0]
239 8,224 3 1 0.021658989 9728 G R 384 + 32 [mkfs.ext3]
241 65,160 3 163392 41.117070504 0 C W 87469088 + 1376 [0]
242 8,32 3 163374 41.122683668 0 C W 88168160 + 1376 [0]
243 65,160 3 163393 41.129952433 0 C W 87905984 + 1376 [0]
244 65,160 3 163394 41.130049431 0 D W 89129344 + 1376 [swapper]
245 65,160 3 163395 41.130067135 0 D W 89216704 + 1376 [swapper]
246 65,160 3 163396 41.130083785 0 D W 89304096 + 1376 [swapper]
247 65,160 3 163397 41.130099455 0 D W 89391488 + 1376 [swapper]
248 65,160 3 163398 41.130114732 0 D W 89478848 + 1376 [swapper]
249 65,160 3 163399 41.130128885 0 D W 89481536 + 64 [swapper]
250 8,32 3 163375 41.134758196 0 C W 86333152 + 1376 [0]
251 65,160 3 163400 41.142229726 0 C W 89129344 + 1376 [0]
252 65,160 3 163401 41.144952314 0 C W 89481536 + 64 [0]
253 8,32 3 163376 41.147441930 0 C W 88342912 + 1376 [0]
254 65,160 3 163402 41.155869604 0 C W 89478848 + 1376 [0]
255 8,32 3 163377 41.159466082 0 C W 86245760 + 1376 [0]
256 65,160 3 163403 41.166944976 0 C W 89216704 + 1376 [0]
257 65,160 3 163404 41.178968252 0 C W 89304096 + 1376 [0]
258 65,160 3 163405 41.191860173 0 C W 89391488 + 1376 [0]
260 Events (sdo): 0 entries, 0 skips
263 Reads Queued: 0, 0KiB Writes Queued: 9, 5,520KiB
264 Read Dispatches: 0, 0KiB Write Dispatches: 0, 0KiB
265 Reads Completed: 0, 0KiB Writes Completed: 0, 0KiB
266 Read Merges: 0 Write Merges: 336
267 IO unplugs: 0 Timer unplugs: 0
269 Reads Queued: 2,411, 38,576KiB Writes Queued: 769, 425,408KiB
270 Read Dispatches: 2,407, 38,512KiB Write Dispatches: 118, 61,680KiB
271 Reads Completed: 0, 0KiB Writes Completed: 0, 0KiB
272 Read Merges: 0 Write Merges: 25,819
273 IO unplugs: 0 Timer unplugs: 4
275 Reads Queued: 2, 32KiB Writes Queued: 18, 10,528KiB
276 Read Dispatches: 2, 32KiB Write Dispatches: 3, 1,344KiB
277 Reads Completed: 0, 0KiB Writes Completed: 0, 0KiB
278 Read Merges: 0 Write Merges: 640
279 IO unplugs: 0 Timer unplugs: 0
281 Reads Queued: 20,572, 329,152KiB Writes Queued: 594, 279,712KiB
282 Read Dispatches: 20,576, 329,216KiB Write Dispatches: 1,474, 740,720KiB
283 Reads Completed: 22,985, 367,760KiB Writes Completed: 1,390, 721,168KiB
284 Read Merges: 0 Write Merges: 16,888
285 IO unplugs: 0 Timer unplugs: 0
288 Reads Queued: 22,985, 367,760KiB Writes Queued: 1,390, 721,168KiB
289 Read Dispatches: 22,985, 367,760KiB Write Dispatches: 1,595, 803,744KiB
290 Reads Completed: 22,985, 367,760KiB Writes Completed: 1,390, 721,168KiB
291 Read Merges: 0 Write Merges: 43,683
292 IO unplugs: 0 Timer unplugs: 4
296 %----------------------------
297 \newpage\section{\label{sec:blktrace-ug}blktrace User Guide}
299 The \emph{blktrace} utility extracts event traces from the kernel (via
300 the relay file system). Some background details concerning the run-time
301 behaviour of blktrace will help to understand some of the more arcane
302 command line options:
305 \item blktrace receives data from the kernel in buffers passed up
306 through the Relay file system (RelayFS). Each device being traced has
307 a file created in the mounted directory for the RelayFS, which defaults
308 to \emph{/relay} -- this can be overridden with the \emph{-r} command
311 \item blktrace defaults to collecting \emph{all} events that can be
312 traced. To limit the events being captured, you can specify one or
313 more filter masks via the \emph{-a} option.
315 Alternatively, one may specify the entire mask utilizing a hexadecimal
316 value that is version-specific. (Requires understanding of the internal
317 representation of the filter mask.)
319 \item As noted above, the events are passed up via a series of buffers
320 stored into RelayFS files. The size and number of buffers can be
321 specified via the \emph{-b} and \emph{-n} arguments respectively.
323 \item blktrace stores the extracted data into files stored in the
324 \emph{local} directory. The format of the file names is (by default)
325 \emph{device}.blktrace.\emph{cpu}, where \emph{device} is the base
326 device name (e.g, if we are tracing /dev/sda, the base device name would
327 be \emph{sda}); and \emph{cpu} identifies a CPU for the event stream.
329 The \emph{device} portion of the event file name can be changed via
330 the \emph{-o} option.
332 \item blktrace may also be run concurrently with blkparse to produce
333 \emph{live} output -- to do this specify \emph{-o -} for blktrace.
335 \item The default behaviour for blktrace is to run forever until explicitly killed by the user (via a control-C, or \emph{kill} utility invocation). There are two ways to modify this:
338 \item You may utilize the blktrace utility itself to \emph{kill}
339 a running trace -- via the \emph{-k} option.
341 \item You can specify a run-time duration for blktrace via the
342 \emph{-w} option -- then blktrace will run for the specified number
343 of seconds, and then halt.
347 \subsection{\label{sec:blktrace-args}Command line arguments}
348 \begin{tabular}{|l|l|l|}\hline
349 Short & Long & Description \\ \hline\hline
350 -A \emph{hex-mask} & --set-mask=\emph{hex-mask} & Set filter mask to \emph{hex-mask} \\ \hline
351 -a \emph{mask} & --act-mask=\emph{mask} & Add \emph{mask} to current filter (see below for masks) \\ \hline
352 -b \emph{size} & --buffer-size=\emph{size} & Specifies buffer size for event extraction (scaled by $2^{10}$) \\ \hline
353 -d \emph{dev} & --dev=\emph{dev} & Adds \emph{dev} as a device to trace \\ \hline
354 -k & --kill & Kill on-going trace \\ \hline
355 -n \emph{num-sub} & --num-sub=\emph{num-sub} & Specifies number of buffers to use \\ \hline
356 -o \emph{file} & --output=\emph{file} & Prepend \emph{file} to output file name(s) \\ \hline
357 -r \emph{rel-path} & --relay=\emph{rel-path} & Specifies RelayFS mount point \\ \hline
358 -v & --version & Outputs version \\ \hline
359 -w \emph{seconds} & --stopwatch=\emph{seconds} & Sets run time to the number of seconds specified \\ \hline
362 \subsubsection{\label{sec:filter-mask}Filter Masks}
363 The following masks may be passed with the \emph{-a} command line
364 option, multiple filters may be combined via multiple \emph{-a} command
365 line options.\smallskip
367 \begin{tabular}{|l|l|}\hline
368 barrier & \emph{barrier} attribute \\ \hline
369 complete & \emph{completed} by driver \\ \hline
370 fs & \emph{FS} requests \\ \hline
371 issue & \emph{issued} to driver \\ \hline
372 pc & \emph{packet command} events \\ \hline
373 queue & \emph{queue} operations \\ \hline
374 read & \emph{read} traces \\ \hline
375 requeue & \emph{requeue} operations \\ \hline
376 sync & \emph{synchronous} attribute \\ \hline
377 write & \emph{write} traces \\ \hline
380 \subsubsection{\label{sec:request-types}Request types}
381 blktrace disguingishes between two types of block layer requests,
382 file system and scsi commands. The former are dubbed \emph{fs}
383 requests, the latter \emph{pc} requests. File system requests are
384 normal read/write operations, ie any type of read or write from a
385 specific disk location at a given size. These requests typically
386 originate from a user process, but they may also be initiated by
387 the vm flushing dirty data to disk or the file system syncing
388 a super or journal block to disk. \emph{pc} requests are SCSI
389 commands. blktrace sends the command data block as a payload
390 so that blkparse can decode it.
392 %----------------------------
393 \newpage\section{\label{sec:blkparse-ug}blkparse User Guide}
395 The \emph{blkparse} utility will attempt to combine streams of events
396 for various devices on various CPUs, and produce a formatted output of
397 the event information. As with blktrace, some details concerning blkparse
398 will help in understanding the command line options presented below.
401 \item By default, blkparse expects to run in a post-processing mode
402 -- one where the trace events have been saved by a previous run
403 of blktrace, and blkparse is combining event streams and dumping
406 blkparse \emph{may} be run in a \emph{live} manner concurrently with
407 blktrace by specifying \emph{-i -} to blkparse, and combining it with
408 the live option for blktrace. An example would be:
411 % blktrace -d /dev/sda -o - | blkparse -i -
414 \item You can set how many blkparse batches event reads via the
415 \emph{-b} option, the default is to handle events in batches of 512.
417 \item If you have saved event traces in blktrace with different output
418 names (via the \emph{-o} option to blktrace), you must specify the
419 same \emph{input} name via the \emph{-i} option.
421 \item The format of the output data can be controlled via the \emph{-f}
422 or \emph{-F} options -- see section~\ref{sec:blkparse-format} for details.
424 By default, blkparse sends formatted data to standard output. This
425 may be changed via the \emph{-o} option.
429 \newpage\subsection{\label{sec:blkparse-args}Command line arguments}
430 \begin{tabular}{|l|l|l|}\hline
431 Short & Long & Description \\ \hline\hline
432 -b \emph{batch} & --batch={batch} & Standard input read batching \\ \hline
434 -i \emph{file} & --input=\emph{file} & Specifies base name for input files -- default is \emph{device}.blktrace.\emph{cpu}. \\
435 & & As noted above, specifying \emph{-i -} runs in \emph{live} mode with blktrace \\
436 & & (reading data from standard in). \\ \hline
438 -F \emph{typ,fmt} & --format=\emph{typ,fmt} & Sets output format \\
439 -f \emph{fmt} & --format-spec=\emph{fmt} & (See section~\ref{sec:blkparse-format} for details.) \\
441 & & The -f form specifies a format for all events \\
443 & & The -F form allows one to specify a format for a specific \\
444 & & event type. The single-character \emph{typ} field is one of the \\
445 & & action specifiers in section~\ref{sec:act-table} \\ \hline
448 -m & --missing & Print missing entries\\ \hline
450 -n & --hash-by-name & Hash processes by name, not by PID\\ \hline
452 -o \emph{file} & --output=\emph{file} & Output file \\ \hline
454 -q & --quiet & Quite mode \\ \hline
456 -s & --per-program-stats & Displays data sorted by program \\ \hline
458 -t & --track-ios & Display time deltas per IO \\ \hline
460 -w \emph{span} & --stopwatch=\emph{span} & Display traces for the \emph{span} specified -- where span can be: \\
461 & & \emph{end-time} -- Display traces from time 0 through \emph{end-time} (in ns) \\
463 & & \emph{start:end-time} -- Display traces from time \emph{start} \\
464 & & through {end-time} (in ns). \\ \hline
466 -v & --version & Display version \\ \hline
471 \subsection{\label{sec:blkparse-actions}Trace actions}
474 \item[C -- complete] A previously issued request has been completed.
475 The output will detail the sector and size of that request, as well
476 as the success or failure of it.
478 \item[D -- issued] A request that previously resided on the block layer
479 queue or in the io scheduler has been sent to the driver.
481 \item[I -- inserted] A request is being sent to the io scheduler for
482 addition to the internal queue and later service by the driver. The
483 request is fully formed at this time.
485 \item[Q -- queued] This notes intent to queue io at the given location.
486 No real requests exists yet.
488 \item[W -- bounced] The data pages attached to this \emph{bio} are
489 not reachable by the hardware and must be bounced to a lower memory
490 location. This causes a big slowdown in io performance, since the data
491 must be copied to/from kernel buffers. Usually this can be fixed with
492 using better hardware - either a better io controller, or a platform
495 \item[B -- back merge] A previously inserted request exists that ends
496 on the boundary of where this io begins, so the io scheduler can merge
499 \item[F -- front merge] Same as the back merge, except this io ends
500 where a previously inserted requests starts.
502 \item[M -- front or back merge] One of the above.
504 \item[G -- get request] To send any type of request to a block device,
505 a \emph{struct request} container must be allocated first.
507 \item[S -- sleep] No available request structures were available, so
508 the issuer has to wait for one to be freed.
510 \item[P -- plug] When io is queued to a previously empty block device
511 queue, Linux will plug the queue in anticipation of future ios being
512 added before this data is needed.
514 \item[U -- unplug] Some request data already queued in the device,
515 start sending requests to the driver. This may happen automatically
516 if a timeout period has passed (see next entry) or if a number of
517 requests have been added to the queue.
519 \item[T -- unplug due to timer] If nobody requests the io that was queued
520 after plugging the queue, Linux will automatically unplug it after a
521 defined period has passed.
523 \item[X -- split] On raid or device mapper setups, an incoming io may
524 straddle a device or internal zone and needs to be chopped up into
525 smaller pieces for service. This may indicate a performance problem due
526 to a bad setup of that raid/dm device, but may also just be part of
527 normal boundary conditions. dm is notably bad at this and will clone
532 \subsection{\label{sec:blkparse-format}Output Description and Formatting}
534 The output from blkparse can be tailored for specific use - in particular,
535 to ease parsing of output, and/or limit output fields to those the user
536 wants to see. The data for fields which can be output include:
539 \begin{tabular}{|l|l|}\hline
540 Field & Description \\
541 Specifier & \\ \hline\hline
542 \emph{a} & Action, a (small) string (1 or 2 characters) -- see table below for more details \\ \hline
543 \emph{c} & CPU id \\ \hline
544 \emph{C} & Command \\ \hline
545 \emph{d} & RWBS field, a (small) string (1-3 characters) -- see section below for more details \\ \hline
546 \emph{D} & 7-character string containing the major and minor numbers of
547 the event's device \\
548 & (separated by a comma). \\ \hline
549 \emph{e} & Error value \\ \hline
550 \emph{m} & Minor number of event's device. \\ \hline
551 \emph{M} & Major number of event's device. \\ \hline
552 \emph{n} & Number of blocks \\ \hline
553 \emph{N} & Number of bytes \\ \hline
554 \emph{p} & Process ID \\ \hline
555 \emph{P} & Display packet data -- series of hexadecimal values\\ \hline
556 \emph{s} & Sequence numbers \\ \hline
557 \emph{S} & Sector number \\ \hline
558 \emph{t} & Time stamp (nanoseconds) \\ \hline
559 \emph{T} & Time stamp (seconds) \\ \hline
560 \emph{u} & Elapsed value in microseconds (\emph{-t} command line option) \\ \hline
561 \emph{U} & Payload unsigned integer \\ \hline
564 Note that the user can optionally specify field display width, and
565 optionally a left-aligned specifier. These precede field specifiers,
566 with a '\%' character, followed by the optional left-alignment specifer
567 (-) followed by the width (a decimal number) and then the field.
569 Thus, to specify the command in a 12-character field that is left aligned:
576 \subsubsection{\label{sec:act-table}Action Table}
577 The following table shows the various actions which may be output.
579 \begin{tabular}{|l|l|}\hline
580 Act & Description \\ \hline\hline
581 B & IO back merged with request on queue \\ \hline
582 C & IO completion \\ \hline
583 D & IO issued to driver \\ \hline
584 F & IO front merged with request on queue \\ \hline
585 G & Get request \\ \hline
586 I & IO inserted onto request queue \\ \hline
587 P & Plug request \\ \hline
588 Q & IO handled by request queue code \\ \hline
589 S & Sleep request \\ \hline
590 T & Unplug due to timeout \\ \hline
591 U & Unplug request \\ \hline
592 W & IO bounced \\ \hline
596 \subsubsection{\label{sec:act-table}RWBS Description}
597 This is a small string containing at least one character ('R' for read,
598 'W' for write operation), and optionally either a 'B' (for barrier
599 operations) or 'S' (for synchronous operations).
601 \subsubsection{\label{sec:default-output}Default output}
603 The standard \emph{header} (or initial fields displayed) include:
606 "%D %2c %8s %5T.%9t %5p %2a %3d "
612 \item[\%D] Displays the event's device major/minor as: \%3d,\%-3d.
613 \item[\%2c] CPU ID (2-character field).
614 \item[\%8s] Sequence number
615 \item[\%5T.\%9t] 5-charcter field for the seconds portion of the
616 time stamp and a 9-character field for the nanoseconds in the time stamp.
617 \item[\%5p] 5-character field for the process ID.
618 \item[\%2a] 2-character field for one of the actions.
619 \item[\%3d] 3-character field for the RWBS data.
622 Seeing this in action:
625 8,0 3 1 0.000000000 697 G W 223490 + 8 [kjournald]
628 The header is the data in this line up to the 223490 (starting block).
630 The default output for all event types includes this header.
632 \paragraph{Default output per action}
635 \item[C -- complete] If a payload is present, this is presented between
636 parenthesis following the header, followed by the error value.
638 If no payload is present, the sector and number of blocks are presented
639 (with an intervening plus (+) character). If the \emph{-t} option
640 was specified, then the elapsed time is presented. In either case,
641 it is followed by the error value for the completion.
646 \item[W -- bounced] If a payload is present, the number of payload bytes
647 is output, followed by the payload in hexadecimal between parenthesis.
649 If no payload is present, the sector and number of blocks are presented
650 (with an intervening plus (+) character). If the \emph{-t} option was
651 specified, then the elapsed time is presented (in parenthesis). In
652 either case, it is followed by the command associated with the event
653 (surrounded by square brackets).
655 \item[B -- back merge]
656 \item[F -- front merge]
657 \item[G -- get request]
658 \item[M -- front or back merge]
659 \item[S -- sleep] The starting sector and number of blocks is output
660 (with an intervening plus (+) character), followed by the command
661 associated with the event (surrounded by square brackets).
663 \item[P -- plug] The command associated with the event (surrounded by
664 square brackets) is output.
667 \item[T -- unplug due to timer] The command associated with the event
668 (surrounded by square brackets) is output, followed by the number of
669 requests outstanding.
671 \item[X -- split] The original starting sector followed by the new
672 sector (separated by a slash (/) is output, followed by the command
673 associated with the event (surrounded by square brackets).
677 %------------------------------
679 \newpage\section*{\label{sec:blktrace-kg}Appendix: blktrace Kernel Guide}
681 The blktrace facility provides an efficient event transfer mechanism which
682 supplies block IO layer state transition data via the relay
683 filesystem. This section provides some details as to the interfaces
684 blktrace utilizes in the kernel to effect this. It is good background data
685 to help understand some of the outputs and command-line options above.
687 \subsection{blktrace.h Definitions}
688 Files which include $<linux/blktrace.h>$ are supplied with the following
691 \subsubsection{Trace Action Specifiers}
692 \begin{tabular}{|l|l|}\hline
693 BLK\_TA\_QUEUE & (RQ) Command queued to request\_queue. \\
694 & (BIO) Command queued by elevator. \\ \hline
695 BLK\_TA\_BACKMERGE & Back merging elevator operation \\ \hline
696 BLK\_TA\_FRONTMERGE & Front merging elevator operation \\ \hline
697 BLK\_TA\_GETRQ & Free request retrieved. \\ \hline
698 BLK\_TA\_SLEEPRQ & No requests available, device unplugged. \\ \hline
699 BLK\_TA\_REQUEUE & Request requeued. \\ \hline
700 BLK\_TA\_ISSUE & Command set to driver for request\_queue. \\ \hline
701 BLK\_TA\_COMPLETE & Command completed by driver. \\ \hline
702 BLK\_TA\_PLUG & Device is plugged \\ \hline
703 BLK\_TA\_UNPLUG\_IO & Unplug device as IO is made available. \\ \hline
704 BLK\_TA\_UNPLUG\_TIMER & Unplug device after timer expired. \\ \hline
705 BLK\_TA\_INSERT & Insert request into queue. \\ \hline
706 BLK\_TA\_SPLIT & BIO split into 2 or more requests. \\ \hline
707 BLK\_TA\_BOUNCE & BIO was bounced \\ \hline
710 %..........................................
711 \subsection{blktrace.h Routines}
712 Files which include $<linux/blktrace.h>$ are supplied with the following
713 kernel routine invocable interfaces:
716 \item[blk\_add\_trace\_rq(struct request\_queue *q, struct request\_queue
718 Adds a trace event describing the state change of the passed in
719 request\_queue. The \emph{what} parameter describes the change in
720 the request\_queue state, and is one of the request queue action
721 specifiers -- BLK\_TA\_QUEUE, BLK\_TA\_REQUEUE, BLK\_TA\_ISSUE,
722 or BLK\_TA\_COMPLETE.
724 \item[blk\_add\_trace\_bio(struct request\_queue *q, struct bio *bio,
726 Adds a trace event for the BIO passed in. The \emph{what} parameter
727 describes the action being performed on the BIO, and is one of
728 BLK\_TA\_BACKMERGE, BLK\_TA\_FRONTMERGE, or BLK\_TA\_QUEUE.
730 \item[blk\_add\_trace\_generic(struct request\_queue *q, struct bio *bio,
732 Adds a \emph{generic} trace event -- not one of the request queue
733 or BIO traces. The \emph{what} parameter describes the action being
734 performed on the BIO (if bio is non-NULL), and is one of
735 BLK\_TA\_PLUG, BLK\_TA\_GETRQ or BLK\_TA\_SLEEPRQ.
737 \item[blk\_add\_trace\_pdu\_int(struct request\_queue *q, u32 what,
739 Adds a trace with some payload data -- in this case, an unsigned
740 32-bit entity (the \emph{pdu} parameter). The \emph{what} parameter
741 describes the nature of the payload, and is one of
742 BLK\_TA\_UNPLUG\_IO or BLK\_TA\_UNPLUG\_TIMER.