PK11_MakeString can fail because of oom, and in both cases tmp isn't null checked which means whatever the goal is, it probably isn't working.
Assignee: nobody → rrelyea
Summary: tmp = PK11_MakeString(...) is a bad pattern → callers of PK11_MakeString don't check result for NULL
* line 1145 -- tmp = PK11_MakeString(NULL,slot->token_name, * line 1377 -- tmp = PK11_MakeString(NULL,slot->slot_name, In these cases, PK11_MakeString returns its second argument. I guess it's remotely conceivable that the value of that argument could be NULL, in which case PK11_MakeString will return the address of an allocated string, which will then be leaked because these callers do NOTHING with tmp.
You need to log in before you can comment on or make changes to this bug.