Re: [reiserfs-dev] Re: Block I/O Enchancements, 2.5.1-pre2

Hans Reiser (reiser@namesys.com)
Sat, 01 Dec 2001 12:31:51 +0300


Nikita Danilov wrote:

>Hans Reiser writes:
>
>
>Having 64 bit inode numbers is good, but we *can* live without them:
>currently inode hash table can store inodes with identical inode
>numbers, provided they can be distinguished by find_actor. Inode numbers
>are just first fast guess during table scan, if they coincide,
>find_actor is used.
>

Nikita is entirely correct, 32 bits of inode number hash is plenty.

64k blocks remain a solution whose potential drain on memory bandwidth
worries me.

Hans

-
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/