[DTrace-devel] [PATCH] test: Loosen tolerance for switchrate test

eugene.loh at oracle.com eugene.loh at oracle.com
Wed May 31 05:19:54 UTC 2023


From: Eugene Loh <eugene.loh at oracle.com>

The timing expected for the switchrate test is rather approximate.
We want a "short" switchrate to run in a "short" amount of time and
a "long" one to run "long."  That's about it.

The actual times observed tend to be a little higher than the expected
times.  On some systems (ARM OL9?), the discrepancy is slightly higher.
This test, however, is simply concerned with whether the switchrate
option is working.

Loosen the tolerance for the test.

Signed-off-by: Eugene Loh <eugene.loh at oracle.com>
---
 test/unittest/options/tst.switchrate.sh | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/test/unittest/options/tst.switchrate.sh b/test/unittest/options/tst.switchrate.sh
index 98cb7341..e47d78d9 100755
--- a/test/unittest/options/tst.switchrate.sh
+++ b/test/unittest/options/tst.switchrate.sh
@@ -16,17 +16,17 @@ dtrace=$1
 # very long switchrate.
 
 status=0
-for nexpect in 1 10; do
+for nexpect in 1 16; do
 	# Run the "instantaneous" D script with the prescribed switchrate.
 	# Time it.  Round to the nearest number of seconds with int(t+0.5).
-	# Go ahead and lop off a second just because we expect the script
-	# to take *some* time.
 	nactual=`/usr/bin/time -f "%e" \
 	    $dtrace -xswitchrate=${nexpect}sec -qn 'BEGIN { exit(0) }' \
-	    |& awk 'NF != 0 {print int($1 + 0.5) - 1}'`
+	    |& awk 'NF != 0 {print int($1 + 0.5)}'`
 
 	# Check the actual number of seconds to the expected value.
-	test/utils/check_result.sh $nactual $nexpect 1
+	# Actually, the actual time might be a few seconds longer than expected.
+	# So pad $nexpect.
+	test/utils/check_result.sh $nactual $(($nexpect + 3)) 4
 	status=$(($status + $?))
 done
 
-- 
2.18.4




More information about the DTrace-devel mailing list