[PATCH] Fix maybe-used-uninitialised warning.

Christopher Faylor cgf-use-the-mailinglist-please@cygwin.com
Sat Jan 9 16:40:00 GMT 2010

On Sat, Jan 09, 2010 at 11:31:03AM +0000, Dave Korn wrote:
>    Hi,
>  Here are two small fixes shown up by more sensitive warnings in gcc-4.5.0.
>In hookapi.cc, it notices that the loop might not run even once; in
>fhandler_tty, it appears to miss that the loops can never exit.  That probably
>needs fixing upstream (but it may be some odd artifact of C++ language rules,
>since it only happens there, not in plain C; something to do with exceptional
>exits, maybe), but until then it seemed harmless to add a trivial return zero;
>it'll only add a handful of bytes to the dll.  (I tested attribute noreturn
>and it didn't help.)
>	* hookapi.cc (hook_or_detect_cygwin): Initialise i earlier to avoid
>	warning.
>  OK?

I'd prefer i be initialized to zero.

>	* fhandler_tty.cc (process_input): Add redundant final return to
>	silence (bogus?) warning.

These are ok.  As usual, I hate that we have to make these pointless
accommodations.  But I've been hating that for many years so it's
nothing newo.


More information about the Cygwin-patches mailing list