On 08/27/2014 09:38 PM, Wang Rui wrote:
On 2014/8/28 4:54, John Ferlan wrote:
> I almost didn't want to do this due to the sheer volume, but figured
> at the very least the bulk of these are resource leaks found by the
> much pickier new coverity scanner.
>
> After this there are "only" 70 issues found...
Nice. I did coverity scan yesterday by coincidence. There are 1400+
total errors on my scan environment. I began to analyze 300+ RESOURCE LEAK
errors. You said there were more than 140 on your side. So when you have
finished fixing errors on your side, I'm glad to check if some other
RESOURCE LEAK errors left and fix them
I used "Coverity Static Analysis version 7.5.0 on Linux
3.15.8-200.fc20.x86_64 x86_64" that we have as a shared license of sorts
inside Red Hat.
I generally just take the defaults for the analysis "strength" (eg, the
--aggressiveness-level is the default of low rather than medium or high,
which I know from experience are much more intolerant).
My 141 count covered the gamut of issues and will be reduced to 72 with
these patches applied. Some of the remaining are sourced from the same
problem - some are false positives. It just takes time to go through
and figure that out and what the "best solution" may be.
Perhaps you are running with a different version or a different
aggressiveness especially to have 300+ resource leaks. That's fine -
the more found the better off we are.
John