On Thu, 24 Aug 2023 14:19:06 -0600
Ross Zwisler <zwisler@xxxxxxxxxx> wrote:
> > +int traceeval_iterator_remove(struct traceeval_iterator *iter)
> > +{
> > + struct traceeval *teval = iter->teval;
> > + struct hash_table *hist = teval->hist;
> > + struct entry *entry;
> > +
> > + if (iter->next < 1 || iter->next > iter->nr_entries)
> > + return 0;
> > +
> > + entry = iter->entries[iter->next - 1];
> > + if (!entry)
> > + return 0;
> > +
> > + hash_remove(hist, &entry->hash);
>
> Are we leaking 'entry' after we've removed it from the hash?
Yeah, I guess we are. Good catch!
>
> I think we need to call free_entry() in both traceeval_iterator_remove() as
> well as traceeval_remove(), just like we do in the loop in
> hist_table_release().
Yep, thanks Ross!
-- Steve
>
> > +
> > + /* The entry no longer exists */
> > + iter->entries[iter->next - 1] = NULL;
> > + teval->update_counter++;
> > +
> > + return 1;
> > }
> > --
> > 2.40.1
> >
|