mdadm: improve the run-ptest

* There are 120+ cases under ${libdir}/mdadm/ptest/tests,
  but the test will break if one test fails as
  below logic in run-ptest.
  ./test &>./test.log

  That's to say, the tests after the failed test
  have no chance to run with the current logic.

  To guarantee all the tests can run even one
  of the tests fails, the option --keep-going
  should be added.

* Refactor the test report to make the report
  more detailed and more common

(From OE-Core rev: 80d17497b719efb2ca9f36b8a730815547e93aa7)

(From OE-Core rev: 1b84064b15022a0834d7f443a8d6bde7f4cfeb6d)

Signed-off-by: Mingli Yu <Mingli.Yu@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This commit is contained in:
Mingli Yu
2018-11-22 22:58:49 -08:00
committed by Richard Purdie
parent f48d96e4df
commit cc4ce7db19

View File

@@ -1,10 +1,19 @@
#!/bin/sh
./test &>./test.log
if [ $? -eq 0 ]
then
echo "PASS: mdadm"
rm test.log
else
echo "FAIL: mdadm"
fi
LOG="$(pwd)/test.log"
# make the test continue to execute even one fail
./test --keep-going 2>&1|tee ${LOG}
# translate the test report
# "tests/18imsm-r10_4d-takeover-r0_2d... succeeded" -> "PASS: tests/18imsm-r10_4d-takeover-r0_2d"
# "tests/19raid6repair... FAILED - see //log for details" -> "FAIL: tests/19raid6repair"
sed -i -e '/succeeded/ s/^/PASS: /' -e '/FAILED/ s/^/FAIL: /' ${LOG}
sed -i -e 's/... FAILED//g' -e 's/... succeeded//g' ${LOG}
passed=`grep PASS: ${LOG}|wc -l`
failed=`grep FAIL: ${LOG}|wc -l`
all=$((passed + failed))
( echo "=== Test Summary ==="
echo "TOTAL: ${all}"
echo "PASSED: ${passed}"
echo "FAILED: ${failed}"
) | tee -a /${LOG}