Skip to contentSkip to conversation

2.5: push BKL out of llseek

PATCH John Hawkes 1 participant 1 message Feb 1, 2002

lore
Share & tools
lore
#0 lore raw

From: "Dave Jones" <[email protected]>

Quoted 1 line

did you benchmark with anything other than dbench ?

I've done substantial AIM7 benchmarking on a 28p ia64 NUMA system, and llseek's BKL usage is a significant contributor to poor scaling. For 500 AIM7 "tasks" and ext2 filesystems, waiting on the BKL consumes about half of the available CPU cycles, and sys_lseek()'s usage is the most significant cycle waster, followed by ext2_get_block() and ext2_write_inode(). Anton's llseek patch from last November does make a measurable improvement in AIM7 throughput.

--
John Hawkes
[email protected]