Hi,
Attached is a small patch, against 2.4.0-test5, for
fs/inode.c/find_inode().
Previous, find_inode() first checked against i_sb, and then
i_ino. This patch reverses the test order.
Reasoning;
o i_sb is more likely to give "false positives" than i_ino, so
best to test i_ino first.
o i_sb is on a different L1 cache line than the inode hash linkage
(while i_ino is on the same L1 line as the linkage), so avoid
accessing it when not necessary.
OK, the i-node cache is a slave to the dcache, so this isn't a real
hot-spot, but the patch is valid.
Mark
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Mon Aug 07 2000 - 21:00:12 EST