Sven Strickroth <email@xxxxxxxxxx> writes:
> Am 05.08.2024 um 12:56 schrieb Sven Strickroth via GitGitGadget:
>> - strmap_for_each_entry(&repo->submodule_ref_stores, &iter, e)
>> + strmap_for_each_entry(&repo->submodule_ref_stores, &iter, e) {
>> ref_store_release(e->value);
>> + free(e->value);
>> + }
>> strmap_clear(&repo->submodule_ref_stores, 1);
>
> After further checking this does not seem to be necessary. The ref
> stores are already free'd in strmap_clear.
Is it "not necessary" or "actively harmful"? It sounds like the
latter?
