Thread (5 messages) 5 messages, 3 authors, 2016-05-04

Re: [PATCH] cyclictest: stop any tracing after hitting a breaktrace threshold

From: Luiz Capitulino <hidden>
Date: 2016-05-03 19:56:49
Also in: lkml

On Tue, 3 May 2016 12:59:53 -0500
Clark Williams [off-list ref] wrote:
John,

This patch is against the devel/v0.98 branch. It turns off tracing in the tracemark() so that we don't lose information about what was going on when we hit the latency:


The current logic of using --tracemark and --notrace works for running
cyclictest with trace-cmd, but even if we are not doing any trace
manipulation in cyclictest, we still need to stop tracing when we hit a
breaktrace threshold (i.e. -b <n>).
Does it solve the problem for you if you revert ba4dd1bf54 and start
cyclictest with:

# cyclictest [...] -bX

Or with:

# cyclictest [...] -bX --tracemark

Also, how do I reproduce your issue? Are you doing tracing by hand?

More comments below.
quoted hunk ↗ jump to hunk
Modify startup logic to hold open file descriptors for the tracemark file
*and* the tracing_on file. When we hit a threshold and call the tracemark()
function, write the marker to the trace buffers and then write a "0\n" to
the tracing_on file to turn off tracing, otherwise we lose the information
immediately prior to the point where we hit the latency.

Signed-off-by: Clark Williams <redacted>
---
 src/cyclictest/cyclictest.c | 32 ++++++++++++++++++++++++++------
 1 file changed, 26 insertions(+), 6 deletions(-)
diff --git a/src/cyclictest/cyclictest.c b/src/cyclictest/cyclictest.c
index 902167010416..00e5f3d59a5b 100644
--- a/src/cyclictest/cyclictest.c
+++ b/src/cyclictest/cyclictest.c
@@ -489,7 +489,12 @@ static void tracemark(char *fmt, ...)
 	va_start(ap, fmt);
 	len = vsnprintf(tracebuf, TRACEBUFSIZ, fmt, ap);
 	va_end(ap);
+
+	/* write the tracemark message */
 	write(tracemark_fd, tracebuf, len);
+
+	/* now stop any trace */
+	write(trace_fd, "0\n", 2);
 }
We do tracing(0) when we hit the latency threshold, so I don't
think this is necessary.

However, have you checked that writing to tracing_on won't break
trace-cmd when it exec()ed cyclictest?
quoted hunk ↗ jump to hunk
 
 
@@ -535,13 +540,28 @@ static void open_tracemark_fd(void)
 {
 	char path[MAX_PATH];
 
-	if (tracemark_fd >= 0)
-		return;
+	/*
+	 * open the tracemark file if it's not already open
+	 */
+	if (tracemark_fd < 0) {
+		sprintf(path, "%s/%s", fileprefix, "trace_marker");
+		tracemark_fd = open(path, O_WRONLY);
+		if (tracemark_fd < 0) {
+			warn("unable to open trace_marker file: %s\n", path);
+			return;
+		}
+	}
 
-	sprintf(path, "%s/%s", fileprefix, "trace_marker");
-	tracemark_fd = open(path, O_WRONLY);
-	if (tracemark_fd < 0)
-		warn("unable to open trace_marker file: %s\n", path);
+	/*
+	 * if we're not tracing and the tracing_on fd is not open,
+	 * open the tracing_on file so that we can stop the trace
+	 * if we hit a breaktrace threshold
+	 */
+	if (notrace && trace_fd < 0) {
+		sprintf(path, "%s/%s", fileprefix, "tracing_on");
+		if ((trace_fd = open(path, O_WRONLY)) < 0)
+			warn("unable to open tracing_on file: %s\n", path);
+	}
 }
 
 static void debugfs_prepare(void)
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help