Re: GCC warning options for numerical programs

[Jonathan Wakely <jwakely.gcc@xxxxxxxxx>]

On 23 April 2012 16:26, Jeffrey Walton wrote:
>
> What worries me is missing a detail and later getting pwn'd. Consider
> how many times Dan Rosenberg has dinged the l33t k3rn3l hack3rz for
> CompSci 101 stuff such as not validating parameters or using
> uninitialized variables. GCC will warn for some of these CompSci 101
> transgressions.

How many of them are *not* covered by -Wall -Wextra?

If there are GCC warnings not included by -Wall -Wextra that would
have helped avoid those problems then you might have a point.  If the
warnings are already enabled by one of those options then -Weverything
won't help. Hitting programmers with a cluebat for not using -Wall
would be more effective.


> Additionally, the kernel's random.c and prng.c violate
> GCC's aliasing rules. I believe GCC warns of aliasing and punning
> (correct me here), but the kernel folks are too l33t to use GCC's
> static analysis capabilities [1].

How would ading -Weverything change that?