Re: [CFT] Re: VFS deadlock ?

[Linus Torvalds]

On Thu, Mar 21, 2013 at 10:18 PM, Al Viro <viro@xxxxxxxxxxxxxxxxxx> wrote:
>>
>> This seems too simple, but I don't see why iget_locked() would be any
>> better. It just wastes time hashing the inode that we aren't really
>> interested in hashing. The inode is always filled by the caller
>> anyway, so we migth as well just get a fresh pseudo-filesystem inode
>> without any crazy hashing..
>
> Umm...
> static int proc_delete_dentry(const struct dentry * dentry)
> {
>         return 1;
> }
>
> static const struct dentry_operations proc_dentry_operations =
> {
>         .d_delete       = proc_delete_dentry,
> };
>
> IOW, dcache retention in procfs is inexistent and the damn thing tries
> to cut down on the amount of inode allocation/freeing/filling.

Ok, that's kind of ugly, but shouldn't be a correctness issue. It
should still work - just cycle through inodes quite aggressivelydue to
no longer re-using them - so I suspect Dave could test it (with the
extra line removed I pointed out just a moment ago).

And I wonder how big of a deal the aggressive dentry deletion is.
Maybe it's even ok to allocate/free the inodes all the time. The whole
"get the inode hash lock and look it up there" can't be all that
wonderful either. It takes the inode->i_lock for each entry it finds
on the hash list, which looks horrible. I suspect our slab allocator
isn't much worse than that, although the RCU freeing of the inodes
could end up being problematic.

                    Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/