* [PATCH application] Allow one to filter on specific trace actions
@ 2008-07-21 18:43 Alan D. Brunelle
0 siblings, 0 replies; only message in thread
From: Alan D. Brunelle @ 2008-07-21 18:43 UTC (permalink / raw)
To: linux-kernel@vger.kernel.org; +Cc: Jens Axboe
[-- Attachment #1: Type: text/plain, Size: 0 bytes --]
[-- Attachment #2: 0002-Allow-one-to-filter-on-specific-trace-actions.patch --]
[-- Type: text/x-diff, Size: 17812 bytes --]
>From 8dbad2d1af70f3dce95e3d38a5f8cb582a02a69e Mon Sep 17 00:00:00 2001
From: Alan D. Brunelle <alan.brunelle@hp.com>
Date: Mon, 21 Jul 2008 14:41:54 -0400
Subject: [PATCH] Allow one to filter on specific trace actions
Added in new '-m' option to blktrace: Can now specify filter at a specific
trace action.
Also: fixed up a missing NOTIFY in the mas_maps table. Updated doc -
still need to do man page.
Signed-off-by: Alan D. Brunelle <alan.brunelle@hp.com>
---
act_mask.c | 43 +++++++++++++++++++++++---
blktrace.c | 30 +++++++++++++++---
blktrace.h | 1 +
blktrace_api.h | 1 +
doc/blktrace.tex | 90 +++++++++++++++++++++++++++++++++++------------------
5 files changed, 124 insertions(+), 41 deletions(-)
diff --git a/act_mask.c b/act_mask.c
index 40a4050..87faa62 100644
--- a/act_mask.c
+++ b/act_mask.c
@@ -1,7 +1,9 @@
#include <strings.h>
#include "blktrace.h"
-#define DECLARE_MASK_MAP(mask) { BLK_TC_##mask, #mask, "BLK_TC_"#mask }
+#define DECLARE_MASK_MAP(mask) { BLK_TC_##mask, #mask, "BLK_TC_"#mask }
+#define DECLARE_TRACE_MAP(mask) { __BLK_TA_##mask, #mask, "__BLK_TA_"#mask }
+
#define COMPARE_MASK_MAP(mmp, str) \
(!strcasecmp((mmp)->short_form, (str)) || \
!strcasecmp((mmp)->long_form, (str)))
@@ -23,22 +25,53 @@ static struct mask_map mask_maps[] = {
DECLARE_MASK_MAP(COMPLETE),
DECLARE_MASK_MAP(FS),
DECLARE_MASK_MAP(PC),
+ DECLARE_MASK_MAP(NOTIFY),
DECLARE_MASK_MAP(AHEAD),
DECLARE_MASK_MAP(META),
};
-int find_mask_map(char *string)
+static struct mask_map trace_maps[] = {
+ DECLARE_TRACE_MAP(QUEUE),
+ DECLARE_TRACE_MAP(BACKMERGE),
+ DECLARE_TRACE_MAP(FRONTMERGE),
+ DECLARE_TRACE_MAP(GETRQ),
+ DECLARE_TRACE_MAP(SLEEPRQ),
+ DECLARE_TRACE_MAP(REQUEUE),
+ DECLARE_TRACE_MAP(ISSUE),
+ DECLARE_TRACE_MAP(COMPLETE),
+ DECLARE_TRACE_MAP(PLUG),
+ DECLARE_TRACE_MAP(UNPLUG_IO),
+ DECLARE_TRACE_MAP(UNPLUG_TIMER),
+ DECLARE_TRACE_MAP(INSERT),
+ DECLARE_TRACE_MAP(SPLIT),
+ DECLARE_TRACE_MAP(BOUNCE),
+ DECLARE_TRACE_MAP(REMAP),
+};
+
+#define MAP_LEN(map) (sizeof(map) / sizeof((map)[0]))
+
+int find_mask(char *string, struct mask_map *maps, unsigned int map_len)
{
unsigned int i;
- for (i = 0; i < sizeof(mask_maps)/sizeof(mask_maps[0]); i++)
- if (COMPARE_MASK_MAP(&mask_maps[i], string))
- return mask_maps[i].mask;
+ for (i = 0; i < map_len; i++)
+ if (COMPARE_MASK_MAP(&maps[i], string))
+ return maps[i].mask;
return -1;
}
+int find_mask_map(char *string)
+{
+ return find_mask(string, mask_maps, MAP_LEN(mask_maps));
+}
+
int valid_act_opt(int x)
{
return (1 <= x) && (x < (1 << BLK_TC_SHIFT));
}
+
+int find_trace_map(char *string)
+{
+ return find_mask(string, trace_maps, MAP_LEN(trace_maps));
+}
diff --git a/blktrace.c b/blktrace.c
index 8892a33..f9131fe 100644
--- a/blktrace.c
+++ b/blktrace.c
@@ -60,7 +60,7 @@ static char blktrace_version[] = "0.99.3";
#define DEBUGFS_TYPE 0x64626720
-#define S_OPTS "d:a:A:r:o:kw:Vb:n:D:lh:p:sI:"
+#define S_OPTS "d:a:A:r:o:kw:Vb:n:D:lh:p:sI:m:"
static struct option l_opts[] = {
{
.name = "dev",
@@ -247,6 +247,7 @@ static char *debugfs_path;
static char *output_name;
static char *output_dir;
static int cat_mask = ~0U;
+static u32 trace_mask = ~0U;
static int kill_running_trace;
static unsigned long buf_size = BUF_SIZE;
static unsigned long buf_nr = BUF_NR;
@@ -398,6 +399,7 @@ static int start_trace(struct device_information *dip)
buts.buf_size = dip->buf_size;
buts.buf_nr = dip->buf_nr;
buts.cat_mask = cat_mask;
+ buts.trace_mask = trace_mask;
if (ioctl(dip->fd, BLKTRACESETUP, &buts) < 0) {
perror("BLKTRACESETUP");
@@ -1756,7 +1758,8 @@ static int net_setup_client(void)
static char usage_str[] = \
"-d <dev> [ -r debugfs path ] [ -o <output> ] [-k ] [ -w time ]\n" \
- "[ -a action ] [ -A action mask ] [ -I <devs file> ] [ -v ]\n\n" \
+ "[ -a action ] [ -A action mask ] [ -I <devs file> ]\n" \
+ "[ -m trace ] [ -M trace mask ][ -v ]\n\n" \
"\t-d Use specified device. May also be given last after options\n" \
"\t-r Path to mounted debugfs, defaults to /sys/kernel/debug\n" \
"\t-o File(s) to send output to\n" \
@@ -1772,6 +1775,8 @@ static char usage_str[] = \
"\t-p Network port to use (default 8462)\n" \
"\t-s Make the network client NOT use sendfile() to transfer data\n" \
"\t-I Add devices found in <devs file>\n" \
+ "\t-m Only capture specified traces. See documentation\n" \
+ "\t-M Give traces mask as a single value. See documentation\n" \
"\t-V Print program version info\n\n";
static void show_usage(char *program)
@@ -1786,6 +1791,7 @@ int main(int argc, char *argv[])
int i, c;
int stop_watch = 0;
int cat_mask_tmp = 0;
+ int trace_mask_tmp = 0;
while ((c = getopt_long(argc, argv, S_OPTS, l_opts, NULL)) >= 0) {
switch (c) {
@@ -1800,7 +1806,7 @@ int main(int argc, char *argv[])
break;
case 'A':
- if ((sscanf(optarg, "%x", &i) != 1) ||
+ if ((sscanf(optarg, "%x", &i) != 1) ||
!valid_act_opt(i)) {
fprintf(stderr,
"Invalid set action mask %s/0x%x\n",
@@ -1810,6 +1816,17 @@ int main(int argc, char *argv[])
cat_mask_tmp = i;
break;
+ case 'm':
+ i = find_trace_map(optarg);
+ if (i < 0) {
+ fprintf(stderr,
+ "Invalid trace action mask %s\n",
+ optarg);
+ return 1;
+ }
+ trace_mask_tmp |= (1 << i);
+ break;
+
case 'd':
if (resize_devices(optarg) != 0)
return 1;
@@ -1820,8 +1837,8 @@ int main(int argc, char *argv[])
FILE *ifp = fopen(optarg, "r");
if (!ifp) {
- fprintf(stderr,
- "Invalid file for devices %s\n",
+ fprintf(stderr,
+ "Invalid file for devices %s\n",
optarg);
return 1;
}
@@ -1920,6 +1937,9 @@ int main(int argc, char *argv[])
if (cat_mask_tmp != 0)
cat_mask = cat_mask_tmp;
+ if (trace_mask_tmp != 0)
+ trace_mask = trace_mask_tmp;
+
if (!debugfs_path)
debugfs_path = default_debugfs_path;
diff --git a/blktrace.h b/blktrace.h
index 816ce61..b26686d 100644
--- a/blktrace.h
+++ b/blktrace.h
@@ -140,6 +140,7 @@ extern void process_fmt(char *, struct per_cpu_info *, struct blk_io_trace *,
unsigned long long, int, unsigned char *);
extern int valid_act_opt(int);
extern int find_mask_map(char *);
+extern int find_trace_map(char *);
extern char *find_process_name(pid_t);
#endif
diff --git a/blktrace_api.h b/blktrace_api.h
index 2b76d76..8490fdc 100644
--- a/blktrace_api.h
+++ b/blktrace_api.h
@@ -115,6 +115,7 @@ struct blk_io_trace_remap {
struct blk_user_trace_setup {
char name[32]; /* output */
__u16 cat_mask; /* input */
+ __u32 trace_mask; /* input */
__u32 buf_size; /* input */
__u32 buf_nr; /* input */
__u64 start_lba;
diff --git a/doc/blktrace.tex b/doc/blktrace.tex
index 1b79c73..bac1495 100644
--- a/doc/blktrace.tex
+++ b/doc/blktrace.tex
@@ -21,7 +21,7 @@
\title{blktrace User Guide}
\author{blktrace: Jens Axboe (jens.axboe@oracle.com)\\
User Guide: Alan D. Brunelle (Alan.Brunelle@hp.com)}
-\date{27 May 2008}
+\date{21 July 2008}
\begin{document}
\maketitle
@@ -82,7 +82,7 @@ As an example, bt/kernel contains blk-trace-2.6.14-rc1-git-G2, download
linux-2.6.13.tar.bz2 and patch-2.6.14-rc1.bz2
\begin{verbatim}
-% tar xjf linux-2.6.13.tar.bz2
+% tar xjf linux-2.6.13.tar.bz2
% mv linux-2.6.13 linux-2.6.14-rc1
% cd linux-2.6.14-rc1
% bunzip2 -c ../patch-2.6.14-rc1.bz2 | patch -p1
@@ -102,7 +102,7 @@ system -- again, outside the scope of this document -- and then enable
and navigate through \emph{Device Drivers} and \emph{Block devices}
and then down to \emph{Support for tracing block io actions} and hit Y.
-Install the new kernel (and modules\ldots) and reboot.
+Install the new kernel (and modules\ldots) and reboot.
\subsection{\label{sec:mount}Mounting the debugfs file system}
@@ -220,14 +220,14 @@ device. An example of that would be \emph{cdrecord} burning.
\subsection{\label{sec:blktrace-post}blktrace -- post-processing}
Another way to run blktrace is to have blktrace save data away for later
-formatting by blkparse. This would be useful if you want to get
+formatting by blkparse. This would be useful if you want to get
measurements while running specific loads.
-To do this, one would specify the device (or devices) to be watched. Then
+To do this, one would specify the device (or devices) to be watched. Then
go run you test cases. Stop the trace, and at your leisure utilize
blkparse to see the results.
-In this example, devices /dev/sdaa, /dev/sdc and /dev/sdo are used in an
+In this example, devices /dev/sdaa, /dev/sdc and /dev/sdo are used in an
LVM volume called adb3/vol.
\begin{verbatim}
@@ -247,11 +247,11 @@ Maximum filesystem blocks=4294967296
1025 block groups
32768 blocks per group, 32768 fragments per group
16384 inodes per group
-Superblock backups stored on blocks:
- 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
+Superblock backups stored on blocks:
+ 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
4096000, 7962624, 11239424, 20480000, 23887872
-Writing inode tables: done
+Writing inode tables: done
Creating journal (8192 blocks): done
Writing superblocks and filesystem accounting information: done
@@ -343,7 +343,7 @@ of the more arcane command line options:
a file created in the mounted directory for the debugfs, which defaults
to \emph{/sys/kernel/debug} -- this can be overridden with the \emph{-r}
command line argument.
-
+
\item blktrace defaults to collecting \emph{all} events that can be
traced. To limit the events being captured, you can specify one or
more filter masks via the \emph{-a} option.
@@ -368,7 +368,7 @@ of the more arcane command line options:
\item blktrace may also be run concurrently with blkparse to produce
\emph{live} output -- to do this specify \emph{-o -} for blktrace.
- \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:
+ \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:
\begin{enumerate}
\item You may utilize the blktrace utility itself to \emph{kill}
@@ -395,6 +395,7 @@ Short & Long & Description \\ \hline\hline
-w \emph{seconds} & --stopwatch=\emph{seconds} & Sets run time to the number of seconds specified \\ \hline
-I \emph{devs file}& --input-devs=\emph{devs file}& Adds devices found in \emph{devs file} to list of devices to trace. \\
& & (One device per line.) \\ \hline
+-m \emph{name} & --act-mask=\emph{name} & Add \emph{name} to current trace filter (see below for masks) \\ \hline
\end{tabular}
\subsubsection{\label{sec:filter-mask}Filter Masks}
@@ -402,17 +403,44 @@ The following category masks may be passed with the \emph{-a} command
line option, multiple filters may be combined via multiple \emph{-a}
command line options.\smallskip
+\begin{tabular}{|l|l|l|}\hline
+\textbf{Name} & \textbf{Description} & \textbf{Hex} \\ \hline\hline
+ahead & \emph{Read ahead} events & 0x0800 \\ \hline
+barrier & \emph{barrier} attribute & 0x0004 \\ \hline
+complete & \emph{completed} by driver & 0x0080 \\ \hline
+fs & \emph{FS} requests & 0x0100 \\ \hline
+issue & \emph{issued} to driver & 0x0040 \\ \hline
+meta & \emph{Metadata} events & 0x1000 \\ \hline
+notify & \emph{Notification} events & 0x0400 \\ \hline
+pc & \emph{packet command} events & 0x0200 \\ \hline
+queue & \emph{queue} operations & 0x0010 \\ \hline
+read & \emph{read} events & 0x0001 \\ \hline
+requeue & \emph{requeue} operations & 0x0020 \\ \hline
+sync & \emph{synchronous} attribute & 0x0008 \\ \hline
+write & \emph{write} events & 0x0002 \\ \hline
+\end{tabular}
+
+The following trace masks may be pased with the \emph{-m} command line
+option, multiple filters may be combined via multiple \emph{-m} command
+line options.\smallskip
+
\begin{tabular}{|l|l|}\hline
-barrier & \emph{barrier} attribute \\ \hline
-complete & \emph{completed} by driver \\ \hline
-fs & \emph{FS} requests \\ \hline
-issue & \emph{issued} to driver \\ \hline
-pc & \emph{packet command} events \\ \hline
-queue & \emph{queue} operations \\ \hline
-read & \emph{read} traces \\ \hline
-requeue & \emph{requeue} operations \\ \hline
-sync & \emph{synchronous} attribute \\ \hline
-write & \emph{write} traces \\ \hline
+\textbf{Name} & \textbf{Description} \\ \hline\hline
+queue & \emph{Queue} traces \\ \hline
+backmerge & \emph{Back merge} traces \\ \hline
+frontmerge & \emph{Merge (front)} traces \\ \hline
+getrq & \emph{Get request} traces \\ \hline
+sleeprq & \emph{Sleep for req} traces \\ \hline
+requeue & \emph{Requeue req} traces \\ \hline
+issue & \emph{Issues} traces \\ \hline
+complete & \emph{Req completion} traces \\ \hline
+plug & \emph{Req queue plug} traces \\ \hline
+unplug\_io & \emph{Unplug req Q} traces \\ \hline
+unplug\_timer & \emph{Unplug req Q timeout} \\ \hline
+insert & \emph{Insertion} traces \\ \hline
+split & \emph{Split req} traces \\ \hline
+bounce & \emph{Req bounce} traces \\ \hline
+remap & \emph{Req remap} traces \\ \hline
\end{tabular}
\subsubsection{\label{sec:request-types}Request types}
@@ -439,7 +467,7 @@ will help in understanding the command line options presented below.
\item By default, blkparse expects to run in a post-processing mode
-- one where the trace events have been saved by a previous run
of blktrace, and blkparse is combining event streams and dumping
- formatted data.
+ formatted data.
blkparse \emph{may} be run in a \emph{live} manner concurrently with
blktrace by specifying \emph{-i -} to blkparse, and combining it with
@@ -476,7 +504,7 @@ Short & Long & Description \\ \hline\hline
& & (reading data from standard in). \\ \hline
-F \emph{typ,fmt} & --format=\emph{typ,fmt} & Sets output format \\
--f \emph{fmt} & --format-spec=\emph{fmt} & (See section~\ref{sec:blkparse-format} for details.) \\
+-f \emph{fmt} & --format-spec=\emph{fmt} & (See section~\ref{sec:blkparse-format} for details.) \\
& & \\
& & The -f form specifies a format for all events \\
& & \\
@@ -500,7 +528,7 @@ Short & Long & Description \\ \hline\hline
-t & --track-ios & Display time deltas per IO \\ \hline
--w \emph{span} & --stopwatch=\emph{span} & Display traces for the \emph{span} specified -- where span can be: \\
+-w \emph{span} & --stopwatch=\emph{span} & Display traces for the \emph{span} specified -- where span can be: \\
& & \emph{end-time} -- Display traces from time 0 through \emph{end-time} (in ns) \\
& & or \\
& & \emph{start:end-time} -- Display traces from time \emph{start} \\
@@ -675,7 +703,7 @@ Seeing this in action:
8,0 3 1 0.000000000 697 G W 223490 + 8 [kjournald]
\end{verbatim}
-The header is the data in this line up to the 223490 (starting block).
+The header is the data in this line up to the 223490 (starting block).
The default output for all event types includes this header.
@@ -683,7 +711,7 @@ The default output for all event types includes this header.
\begin{description}
\item[C -- complete] If a payload is present, this is presented between
- parenthesis following the header, followed by the error value.
+ parenthesis following the header, followed by the error value.
If no payload is present, the sector and number of blocks are presented
(with an intervening plus (+) character). If the \emph{-t} option
@@ -769,24 +797,24 @@ Files which include $<linux/blktrace.h>$ are supplied with the following
kernel routine invocable interfaces:
\begin{description}
- \item[blk\_add\_trace\_rq(struct request\_queue *q, struct request\_queue
+ \item[blk\_add\_trace\_rq(struct request\_queue *q, struct request\_queue
*rq, u32 what)]
Adds a trace event describing the state change of the passed in
request\_queue. The \emph{what} parameter describes the change in
- the request\_queue state, and is one of the request queue action
+ the request\_queue state, and is one of the request queue action
specifiers -- BLK\_TA\_QUEUE, BLK\_TA\_REQUEUE, BLK\_TA\_ISSUE,
or BLK\_TA\_COMPLETE.
- \item[blk\_add\_trace\_bio(struct request\_queue *q, struct bio *bio,
+ \item[blk\_add\_trace\_bio(struct request\_queue *q, struct bio *bio,
u32 what)]
Adds a trace event for the BIO passed in. The \emph{what} parameter
describes the action being performed on the BIO, and is one of
BLK\_TA\_BACKMERGE, BLK\_TA\_FRONTMERGE, or BLK\_TA\_QUEUE.
- \item[blk\_add\_trace\_generic(struct request\_queue *q, struct bio *bio,
+ \item[blk\_add\_trace\_generic(struct request\_queue *q, struct bio *bio,
int rw, u32 what)]
Adds a \emph{generic} trace event -- not one of the request queue
- or BIO traces. The \emph{what} parameter describes the action being
+ or BIO traces. The \emph{what} parameter describes the action being
performed on the BIO (if bio is non-NULL), and is one of
BLK\_TA\_PLUG, BLK\_TA\_GETRQ or BLK\_TA\_SLEEPRQ.
--
1.5.4.3
^ permalink raw reply related [flat|nested] only message in thread
only message in thread, other threads:[~2008-07-21 18:44 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-07-21 18:43 [PATCH application] Allow one to filter on specific trace actions Alan D. Brunelle
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox