Topics

[PATCH 1/2] tools: Skip backward time entries in xfsslower

Jiri Olsa
 

While using xfsslower on RHEL7 we occasionally get
following screwed up latencies:

# xfsslower
Tracing XFS operations slower than 1 ms
TIME COMM PID T BYTES OFF_KB LAT(ms) FILENAME
13:25:03 git 3385 R 62 4704 18446744073708.55 tmp_pack_bDUbwZ
13:25:03 git 3385 S 0 0 3.05 tmp_idx_Kjb2bW
...

The reason for this is that on RHEL7 it's possible to get backward
timetamp with bpf_ktime_get_ns. This needs to be fixed, but meanwhile
this fix makes sure the latencies with backward times are skipped.

For the rest of the kernels this is just sanity fix with
possibly just single compare instruction overhead.

Signed-off-by: Jiri Olsa <jolsa@...>
---
tools/xfsslower.py | 9 ++++++++-
1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/tools/xfsslower.py b/tools/xfsslower.py
index da70c5705192..b996ce8f143a 100755
--- a/tools/xfsslower.py
+++ b/tools/xfsslower.py
@@ -173,8 +173,15 @@ static int trace_return(struct pt_regs *ctx, int type)

// calculate delta
u64 ts = bpf_ktime_get_ns();
- u64 delta_us = (ts - valp->ts) / 1000;
+ s64 delta_us = ts - valp->ts;
entryinfo.delete(&id);
+
+ // skip entries with backward time
+ if (delta_us < 0)
+ return 0;
+
+ delta_us /= 1000;
+
if (FILTER_US)
return 0;

--
1.8.3.1