Hi Richard,

On 3/15/23 15:52, Richard Biener wrote:
> On Wed, Mar 15, 2023 at 3:30 PM Alejandro Colomar via Gcc-patches
> <gcc-patches@gcc.gnu.org> wrote:
>>
>> Ping
> 
> -Wuse-after-free=3 was explicitly added to cover cases with a high
> false-positive rate.  If you want to
> make that the default then instead merge the equality compare case
> back to the =2 case.
> 
> But as I said elsewhere I think that -Wuse-after-free is very much too
> trigger happy, especially
> with value-uses (not accessing released memory but inspecting the old
> pointer value).  Please consider
> looking at https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104075 and
> review the false positives
> reported.
> 
> Also see my very recent patches from today trying to limit
> -Wuse-after-free by not diagnosing
> from late IL.

Hmmm, thanks, didn't know about those.  Please ignore my patch.

Cheers,

Alex


-- 
<http://www.alejandro-colomar.es/>
GPG key fingerprint: A9348594CE31283A826FBDD8D57633D441E25BB5