free up -c for something else

examples/funcgraph_example.txt

@@ -1620,10 +1620,10 @@ by completing a 19 second read (time between commands), then has a series of
 The function times printed are inclusive of their children.
 
 
-The -c option will print on-CPU times only, excluding sleeping or blocking
+The -C option will print on-CPU times only, excluding sleeping or blocking
 events from the function duration times. Eg:
 
-# ./funcgraph -ctp 25285 vfs_read
+# ./funcgraph -Ctp 25285 vfs_read
 Tracing "vfs_read" for PID 25285... Ctrl-C to end.
 7338520.591816 |   0)               |              finish_task_switch() {
 7338520.591820 |   0)               |                xen_evtchn_do_upcall() {
@@ -2155,9 +2155,9 @@ performance investigation.
 Use -h to print the USAGE message:
 
 # ./funcgraph -h
-USAGE: funcgraph [-acDhHPtT] [-m maxdepth] [-p PID] [-d secs] funcstring
+USAGE: funcgraph [-aCDhHPtT] [-m maxdepth] [-p PID] [-d secs] funcstring
                  -a              # all info (same as -HPt)
-                 -c              # measure on-CPU time only
+                 -C              # measure on-CPU time only
                  -d seconds      # trace duration, and use buffers
                  -D              # do not show function duration
                  -h              # this usage message

examples/funcslower_example.txt

@@ -57,7 +57,7 @@ earlier output, this tells me that the latency in this function is due to time
 spent blocked off-CPU, and not on-CPU. This makes sense: this function is
 ultimately being blocked on disk I/O.
 
-Were the function duration times to be similar with and without -c, that would
+Were the function duration times to be similar with and without -C, that would
 tell us that the high latency is due to time spent on-CPU executing code.
 
 
@@ -94,9 +94,9 @@ switch on the given CPU.
 Use -h to print the USAGE message:
 
 # ./funcslower -h
-USAGE: funcslower [-achHPt] [-p PID] [-d secs] funcstring latency_us
+USAGE: funcslower [-aChHPt] [-p PID] [-d secs] funcstring latency_us
                  -a              # all info (same as -HPt)
-                 -c              # measure on-CPU time only
+                 -C              # measure on-CPU time only
                  -d seconds      # trace duration, and use buffers
                  -h              # this usage message
                  -H              # include column headers

kernel/funcgraph

@@ -9,7 +9,7 @@
 # using other, lower overhead tools. This is a proof of concept using Linux
 # ftrace capabilities on older kernels.
 #
-# USAGE: funcgraph [-acDhHPtT] [-m maxdepth] [-p PID] [-d secs] funcstring
+# USAGE: funcgraph [-aCDhHPtT] [-m maxdepth] [-p PID] [-d secs] funcstring
 #
 # Run "funcgraph -h" for full usage.
 #
@@ -75,9 +75,9 @@ trap ':' INT QUIT TERM PIPE HUP	# sends execution to end tracing section
 
 function usage {
 	cat <<-END >&2
-	USAGE: funcgraph [-acDhHPtT] [-m maxdepth] [-p PID] [-d secs] funcstring
+	USAGE: funcgraph [-aCDhHPtT] [-m maxdepth] [-p PID] [-d secs] funcstring
 	                 -a              # all info (same as -HPt)
-	                 -c              # measure on-CPU time only
+	                 -C              # measure on-CPU time only
 	                 -d seconds      # trace duration, and use buffers
 	                 -D              # do not show function duration
 	                 -h              # this usage message
@@ -140,11 +140,11 @@ function edie {
 }
 
 ### process options
-while getopts acd:DhHm:p:PtT opt
+while getopts aCd:DhHm:p:PtT opt
 do
 	case $opt in
 	a)	opt_headers=1; opt_proc=1; opt_time=1 ;;
-	c)	opt_cpu=1; ;;
+	C)	opt_cpu=1; ;;
 	d)	opt_duration=1; duration=$OPTARG ;;
 	D)	opt_nodur=1; ;;
 	m)	opt_max=1; max=$OPTARG ;;
@@ -202,7 +202,7 @@ if ! echo function_graph > current_tracer; then
 fi
 if (( opt_cpu )); then
 	if ! echo nosleep-time > trace_options; then
-		edie "ERROR: setting -c (nosleep-time). Exiting."
+		edie "ERROR: setting -C (nosleep-time). Exiting."
 	fi
 fi
 # the following must be done after setting current_tracer

kernel/funcslower

@@ -7,7 +7,7 @@
 # and filter them based on a latency threshold. This is a proof of concept using
 # Linux ftrace capabilities on older kernels.
 #
-# USAGE: funcslower [-achHPt] [-p PID] [-d secs] funcstring latency_us
+# USAGE: funcslower [-aChHPt] [-p PID] [-d secs] funcstring latency_us
 #
 # Run "funcslower -h" for full usage.
 #
@@ -59,9 +59,9 @@ trap ':' INT QUIT TERM PIPE HUP	# sends execution to end tracing section
 
 function usage {
 	cat <<-END >&2
-	USAGE: funcslower [-achHPt] [-p PID] [-d secs] funcstring latency_us
+	USAGE: funcslower [-aChHPt] [-p PID] [-d secs] funcstring latency_us
 	                 -a              # all info (same as -HPt)
-	                 -c              # measure on-CPU time only
+	                 -C              # measure on-CPU time only
 	                 -d seconds      # trace duration, and use buffers
 	                 -h              # this usage message
 	                 -H              # include column headers
@@ -116,11 +116,11 @@ function edie {
 }
 
 ### process options
-while getopts acd:hHp:Pt opt
+while getopts aCd:hHp:Pt opt
 do
 	case $opt in
 	a)	opt_headers=1; opt_proc=1; opt_time=1 ;;
-	c)	opt_cpu=1; ;;
+	C)	opt_cpu=1; ;;
 	d)	opt_duration=1; duration=$OPTARG ;;
 	p)	opt_pid=1; pid=$OPTARG ;;
 	H)	opt_headers=1; ;;
@@ -190,7 +190,7 @@ if ! echo function_graph > current_tracer; then
 fi
 if (( opt_cpu )); then
 	if ! echo nosleep-time > trace_options; then
-		edie "ERROR: setting -c (nosleep-time). Exiting."
+		edie "ERROR: setting -C (nosleep-time). Exiting."
 	fi
 fi
 # the following must be done after setting current_tracer

man/man8/funcgraph.8

@@ -3,7 +3,7 @@
 funcgraph \- trace kernel function graph, showing child function calls and times. Uses Linux ftrace.
 .SH SYNOPSIS
 .B funcgraph
-[\-acDhHPtT] [\-m maxdepth] [\-p PID] [\-d secs] funcstring
+[\-aCDhHPtT] [\-m maxdepth] [\-p PID] [\-d secs] funcstring
 .SH DESCRIPTION
 This is an exploratory tool that shows the graph of child function calls
 for a given kernel function. This can cost moderate overhead to execute, and
@@ -38,7 +38,7 @@ kernels.
 \-a
 All info. Same as \-HPt. (But no -T, which isn't available in older kernels.)
 .TP
-\-c
+\-C
 Function durations measure on-CPU time only (exclude sleep time).
 .TP
 \-d seconds
@@ -120,7 +120,7 @@ TASK/PID
 .TP
 DURATION
 Elapsed time during the function call, inclusive of children. This is also
-inclusive of sleep time, unless -c is used. The time is either displayed on
+inclusive of sleep time, unless -C is used. The time is either displayed on
 the return of a function ("}"), or for a leaf function (no children), on the
 same line.
 

man/man8/funcslower.8

@@ -3,7 +3,7 @@
 funcslower \- trace kernel functions slower than a threshold (microseconds). Uses Linux ftrace.
 .SH SYNOPSIS
 .B funcslower
-[\-achHPt] [\-p PID] [\-d secs] funcstring latency_us
+[\-aChHPt] [\-p PID] [\-d secs] funcstring latency_us
 .SH DESCRIPTION
 This uses the Linux ftrace function graph profiler to time kernel functions
 and filter them based on a latency threshold. Latency outliers can be studied
@@ -28,7 +28,7 @@ recent kernels. And awk.
 \-a
 All info. Same as \-HPt.
 .TP
-\-c
+\-C
 Function durations measure on-CPU time only (exclude sleep time).
 .TP
 \-d seconds
@@ -95,7 +95,7 @@ TASK/PID
 .TP
 DURATION
 Elapsed time during the function call, inclusive of children. This is also
-inclusive of sleep time, unless -c is used.
+inclusive of sleep time, unless -C is used.
 .TP
 FUNCTION CALLS
 Kernel function returns.