> Ingo> definitely, because in the case of page buckets there are many locks
> Ingo> hashed in a mapping-neutral way. Ie. different parts of the same file will
> Ingo> likely map to different spinlocks.
>
> That's why it's likely to miss on each access.
yes, you are right.
> Ingo> In the radix tree case all pages in the inode will map to the
> Ingo> same spinlock.
>
> That's why it's likely to bounce on each access.
>
> So, is there any difference ? :)
no difference. I tried to create a testcase that shows the difference
(multiple processes read()ing a single big file on an 8-way box), but
performance was equivalent. So given the clear advantages of radix trees
in other areas, they win hands down. :)
Ingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/