On 4/5/16 9:40 AM, Johannes Berg wrote: > On Tue, 2016-04-05 at 09:33 -0400, Jeff Mahoney wrote: >> On 4/5/16 5:43 AM, Johannes Berg wrote: >>> >>> On Mon, 2016-04-04 at 14:15 -0400, Jeff Mahoney wrote: >>>> >>>> This fixes: >>>> >>>> net/mac80211/mesh_hwmp.c:603:26: warning: ‘target_metric’ may be >>>> used >>>> uninitialized in this function >>>> >>>> target_metric is only consumed when reply = true so no bug exists >>>> here, >>>> but gcc doesn't notice that. Initializing to 0 clears the >>>> warning. >>>> >>> What gcc version are you using? Mine doesn't seem to have a problem >>> following the logic here. >> gcc version 4.8.5 (SUSE Linux) > > Ok, mine's significantly newer: > > gcc (Debian 5.3.1-10) 5.3.1 20160224 > > I do wonder though if we really should do this? It might hide errors > later on? AFAICT, the only way errors would be introduced is if someone added a new case in which reply was true *and* needed a different target_metric value (there are 3 cases it's used, 2 of which use 0). Or if there was a new case where the target_metric was used and reply wasn't true -- in which case this code would have been broken anyway. Also, using uninitalized_var would have a higher probability of hiding an error later on. Setting it to 0 seems the safest course. -Jeff -- Jeff Mahoney SUSE Labs
Attachment:
signature.asc
Description: OpenPGP digital signature
