1
- .TH ustat 8 "2016-11-07 " "USER COMMANDS"
1
+ .TH ustat 8 "2018-10-09 " "USER COMMANDS"
2
2
.SH NAME
3
- ustat, javastat, nodestat, perlstat, phpstat, pythonstat, rubystat \- Activity stats from
3
+ ustat, javastat, nodestat, perlstat, phpstat, pythonstat, rubystat, tclstat \- Activity stats from
4
4
high-level languages.
5
5
.SH SYNOPSIS
6
6
.B javastat [-C] [-S {cload,excp,gc,method,objnew,thread}] [-r MAXROWS] [-d] [interval [count]]
@@ -15,7 +15,9 @@ high-level languages.
15
15
.br
16
16
.B rubystat [-C] [-S {cload,excp,gc,method,objnew,thread}] [-r MAXROWS] [-d] [interval [count]]
17
17
.br
18
- .B ustat [-l {java,perl,python,ruby,node,php}] [-C] [-S {cload,excp,gc,method,objnew,thread}] [-r MAXROWS] [-d] [interval [count]]
18
+ .B tclstat [-C] [-S {cload,excp,gc,method,objnew,thread}] [-r MAXROWS] [-d] [interval [count]]
19
+ .br
20
+ .B ustat [-l {java,node,perl,php,python,ruby,tcl}] [-C] [-S {cload,excp,gc,method,objnew,thread}] [-r MAXROWS] [-d] [interval [count]]
19
21
.SH DESCRIPTION
20
22
This is "top" for high-level language events, such as garbage collections,
21
23
exceptions, thread creations, object allocations, method calls, and more. The
@@ -26,7 +28,7 @@ set of details.
26
28
This uses in-kernel eBPF maps to store per process summaries for efficiency.
27
29
28
30
This tool relies on USDT probes embedded in many high-level languages, such as
29
- Java, Node, Perl, PHP, Python, and Ruby . It requires a runtime instrumented with
31
+ Java, Node, Perl, PHP, Python, Ruby, and Tcl . It requires a runtime instrumented with
30
32
these probes, which in some cases requires building from source with a
31
33
USDT-specific flag, such as "--enable-dtrace" or "--with-dtrace". For Java,
32
34
some probes are not enabled by default, and can be turned on by running the Java
@@ -43,7 +45,7 @@ Since this uses BPF, only the root user can use this tool.
43
45
CONFIG_BPF and bcc.
44
46
.SH OPTIONS
45
47
.TP
46
- \- l {java,node,perl,php,python,ruby}
48
+ \- l {java,node,perl,php,python,ruby,tcl }
47
49
The language to trace. By default, all languages are traced.
48
50
.TP
49
51
\- C
@@ -107,10 +109,10 @@ THR/s
107
109
Count of threads created during interval.
108
110
.SH OVERHEAD
109
111
When using this tool with high-frequency events, such as method calls, a very
110
- significant slow-down can be expected. However, many of the high-level
112
+ significant slow-down can be expected. However, many of the high-level
111
113
languages covered by this tool already have a fairly high per-method invocation
112
- cost, especially when running in interpreted mode. For the lower-frequency
113
- events, such as garbage collections or thread creations, the overhead should
114
+ cost, especially when running in interpreted mode. For the lower-frequency
115
+ events, such as garbage collections or thread creations, the overhead should
114
116
not be significant. Specifically, when probing Java processes and not using the
115
117
"-XX:+ExtendedDTraceProbes" flag, the most expensive probes are not emitted,
116
118
and the overhead should be acceptable.
0 commit comments