32

I am running valgrind in a bash script, and directing the valgrind's output to a file. Like this :

valgrind --leak-check=full --show-reachable=yes --xml=yes --xml-file=unit_tests_valgrind.out.xml ./unit_tests_runner

The RET_VALUE=$? is going to put the return value of the program (unit_tests_runner above), not from valgrind. And the only way to check whether the valgrind found memory problems is to open the log.

Is there a way to check in a script whether there are memory problems? If yes, how?

BЈовић
  • 62,405
  • 41
  • 173
  • 273

1 Answers1

46

You seem to be looking for the --error-exitcode option.

Since it defaults to 0, the return code from Valgrind is the same as that of the process. Set it to a non-zero value instead.

From Valgrind core manual:

--error-exitcode=<number> [default: 0]

Specifies an alternative exit code to return if Valgrind reported any errors in the run. When set to the default value (zero), the return value from Valgrind will always be the return value of the process being simulated. When set to a nonzero value, that value is returned instead, if Valgrind detects any errors. This is useful for using Valgrind as part of an automated test suite, since it makes it easy to detect test cases for which Valgrind has reported errors, just by inspecting return codes.

devnull
  • 118,548
  • 33
  • 236
  • 227
  • Is it me or setting `` to -1 does not work for me, it seems as though it has to be bigger than 0? – chutsu Nov 11 '13 at 14:52
  • 13
    `--leak-check=full` or `yes` is critical for `--error-exitcode` to behave as expected (depending on your use case). Without it, the leak errors will be suppressed and not reported in the exit code (valgrind 3.10 doesn't even report them as suppressed). – jtpereyda Oct 02 '14 at 18:31
  • Thank you so much! I was trying to find a way to have "make check" fail when valgrind failed and this completely solved it! – JDN Oct 04 '15 at 17:28