1.\"- 2.\" Copyright (c) 2002 Chad David 3.\" All rights reserved. 4.\" 5.\" Redistribution and use in source and binary forms, with or without 6.\" modification, are permitted provided that the following conditions 7.\" are met: 8.\" 1. Redistributions of source code must retain the above copyright 9.\" notice, this list of conditions and the following disclaimer. 10.\" 2. Redistributions in binary form must reproduce the above copyright 11.\" notice, this list of conditions and the following disclaimer in the 12.\" documentation and/or other materials provided with the distribution. 13.\" 14.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 15.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 16.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 17.\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 18.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 19.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 20.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 21.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 22.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 23.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 24.\" SUCH DAMAGE. 25.\" 26.\" $FreeBSD: src/usr.bin/ktrdump/ktrdump.8,v 1.7 2005/03/08 06:58:56 hmp Exp $ 27.\" $DragonFly: src/usr.bin/ktrdump/ktrdump.8,v 1.14 2007/09/14 23:47:53 swildner Exp $ 28.\" 29.Dd June 8, 2007 30.Dt KTRDUMP 8 31.Os 32.Sh NAME 33.Nm ktrdump 34.Nd print kernel ktr trace buffer 35.Sh SYNOPSIS 36.Nm 37.Op Fl acfilnpqrstx 38.Op Fl N Ar execfile 39.Op Fl M Ar corefile 40.Op Fl o Ar outfile 41.Sh DESCRIPTION 42The 43.Nm 44utility is used to dump the contents of the kernel ktr trace buffer. 45.Pp 46The following options are available: 47.Bl -tag -width ".Fl N Ar execfile" 48.It Fl a 49Print most fields. Implies 50.Fl c 51.Fl i 52.Fl r 53.Fl x 54.Fl p 55.Fl r , 56and 57.Fl s . 58Note that 59.Fl f 60is not included. 61.It Fl c 62Print the CPU number that each entry was logged from. 63.It Fl f 64Print the file and line number that each entry was logged from. 65.It Fl i 66Print the ID string field, identifying the facility being logged. 67.It Fl l 68.Nm 69will loop waiting for new data rather than exit. 70.It Fl n 71.Nm 72normally tries to translate the caller fields and (when easily parsed) 73trace arguments into symbols. This option forces hex values to be 74displayed instead. This option will also cause relative timestamps to 75be displayed as TSC timestamps rather than converted to microseconds. 76.It Fl p 77Print the trace data. 78.It Fl q 79Quiet mode; do not print the column header. 80.It Fl r 81Print relative timestamps in microseconds, rather than absolute TSC 82timestamps. 83.It Fl s 84Attempt to merge the KTR logs for all cpus into a single time-sorted 85log. For the numbers to make any sense you probably want to turn 86on the 87.Va debug.ktr.resynchronize 88.Xr sysctl 3 89variable. This sysctl causes the kernel to periodically 90calculate the drift between each CPU's TSC and apply a correction. 91.It Fl x 92Print the call chain leading up to the procedure which issued 93the KTR. 94.It Fl t 95Print the timestamp for each entry. 96.It Fl N Ar execfile 97The kernel image to resolve symbols from. 98The default is the value returned via 99.Xr getbootfile 3 . 100.It Fl M Ar corefile 101The core file or memory image to read from. 102The default is 103.Pa /dev/mem . 104.It Fl o Ar outfile 105The file to write the output to. 106The default is standard output. 107.El 108.Sh OPERATIONAL NOTES 109Users of 110.Nm 111should keep in mind that the act of trace logging will itself affect 112execution overheads. On a 2Ghz cpu a logging call can take anywhere 113from 40ns to 150ns to run. 114.Pp 115The TSC counter is used on cpus equipped with it (which is all modern cpus). 116The TSC counters may not be synchronized on SMP systems and may drift even 117between cores on multi-core cpus. Enabling synchronization between cpus 118via the 119.Va debug.ktr.resynchronize 120sysctl will impose additional system overheads and will generally only be 121accurate to within 100ns or so (and perhaps not even that good). 122Resynchronization only occurs 10 times a second and serious drift will 123cause a great deal of measurement noise when trying to compare events occurring 124on two different cpus. 125.Pp 126Users using the 127.Fl s 128option should be aware that events for each cpu are independently logged 129and one cpu might have more events logged then another, causing earlier 130events to be discarded sooner then other cpus. The beginning portion of 131the sorted output may thus show MP related events for one cpu with no 132corresponding events for other cpus. 133.Pp 134It is possible to somewhat characterize KTR logging overheads by setting 135the 136.Va debug.ktr.testlogcnt 137sysctl and then observing test logging events in the KTR output. Tests 1-3 138log four dummy arguments while tests 4-6 log no arguments. 139.Pp 140It is possible to characterize IPI messaging latencies by setting the 141.Va debug.ktr.testipicnt 142sysctl. A small number between 1 and 1000 is recommended. This will 143cause the system to ping pong a single IPI message between cpu 0 and cpu 1 144asynchronously that number of times, as fast as it can. A 145.Nm 146should be performed almost immediately after setting the sysctl or you 147might miss the logged events. 148.Sh SEE ALSO 149.Xr ktr 4 , 150.Xr ktr 9 151.Sh HISTORY 152The 153.Nm 154utility first appeared in 155.Fx 5.0 . 156.Sh AUTHORS 157.An -nosplit 158The 159.Nm 160utility was originally implemented by 161.An Jake Burkholder Aq jake@FreeBSD.org . 162This manual page was originally written by 163.An Chad David Aq davidc@FreeBSD.org . 164The program and manual page were rewritten pretty much from 165scratch by 166.An Matthew Dillon 167for 168.Dx . 169