Markus Elfring

I find it interesting that you [added input parameter validation](https://github.com/wangbin579/tcpcopy/commit/ff8dd0549feb43f9eb7e5500c6143df57182cfa0#L2R54). How do you think about to use a symbol like "[SIGUSR2](http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/signal.h.html#top)" instead of the value "1"? Another implementation detail remains....

The CERT Secure Coding Standards [mention a different information for a strictly conforming program](https://www.securecoding.cert.org/confluence/display/seccode/SIG31-C.+Do+not+access+shared+objects+in+signal+handlers). It might be that attributes like "volatile static" are already provided by the data type definition...

I suggest to avoid [ignorance of return values](https://cwe.mitre.org/data/definitions/252.html "Checking more return values") a bit more. Would you like to detect every error situation as early as possible?

I prefer a more complete error detection and exception handling.

Thanks for [your small source code improvement](https://github.com/glouw/tinn/commit/72dd9ed70a5bcb697b1261037592c391a71fa82c "Checking return values from two functions"). There are still [more function calls to reconsider](https://github.com/glouw/tinn/blob/2617d72ec523d4ffd89f0e6e6d6d80056326db8b/Tinn.c#L155 "Update candidate: xtsave()").

Functions like fclose() and fprintf() will need also further software development attention, won't they?

Your assumptions are questionable in this case. There are additional failure possibilities to consider. How do you think about to improve [static source code analysis](https://en.wikipedia.org/wiki/List_of_tools_for_static_code_analysis#C.2C_C.2B.2B "Tools for source code analysis")...

Would you like to extend the development methodology with any software tools?

>It kind of feels like it's just an automated scan of the code base. I performed a small source code analysis to find a few update candidates. * How do...

>… if there was a change that clearly improved the performance or correctness in measurable ways, … * Would anybody dare to measure effects of questionable extra checks in the...