[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: More GC fun.

(Cc: back to the list - sorry)

nick@xxxxxxx.ru said:
>  Why not to make jffs_garbage_collect_now() treat those nodes as
> deleted and be happy instead? 

Because that's the symptomatic fix. It'd work, but it wouldn't help us deal 
with inum wraparound, which I'm going to have to deal with next.

jffs_create() at the moment doesn't even attempt to deal with what happens 
when c->next_ino wraps round to zero. I'd bet we don't deal with version 
numbers doing that either - that's going to be quite important for utmp, 
wtmp and the like.