Fix ' warning: delete called on non-final 'PeerLogicValidation' that has virtual functions but non-virtual destructor'#2363
Conversation
…has virtual functions but non-virtual destructor'
|
The failures on Jenkins do not seem to be related to my code change. |
|
Thanks but this one is going to conflict with bitcoin#12680 on future backports since it basically replicates (only) a part of it. |
|
If you wish @UdjinM6 I can update the entire PeerLogicValidation and add NetEventsInterface. EDIT: That would update signals and schedular use too. |
|
You mean backporting bitcoin#10756 first (or is there anything else)? 10756 looks pretty straightforward on the first sight (and I also like the fact that it drops some boost dependencies), though there might be quite a few merge conflicts since it changes the code all over the place actually. I'm not sure if we really want to touch any critical parts atm just because of some warnings tbh, probably better to postpone non-critical/non-functional fixes like this to future "backport packs". Thoughts @codablock ? |
|
Agree with @UdjinM6. Not the right time to backport this. Also, if we want to backport cosmetical stuff from Bitcoin, then I'd always prefer to backport in the same order as stuff was merged into Bitcoin. When I backported the changes from Bitcoin 0.13+0.14, the merge resolution which was needed due to previous out-of-order backporting was about 90% of the work (and it took weeks). |
peerLogic.reset(); in init.cpp calls on delete on PeerLogicValidation in net_processing and PeerLogicValidation is non-final.