Re: kernbench-16 on 2.5.59 vs 2.5.59-mm6

[Dipankar Sarma <dipankar@gamebox.net>]

On Mon, Jan 27, 2003 at 06:40:15PM +0100, Martin J. Bligh wrote:
> Going from 59 to 59-mm6, I get:
> 
> Kernbench-16:
>                                    Elapsed        User      System         CPU
>                         2.5.59       47.45      568.02      143.17     1498.17
>                     2.5.59-mm6       47.18      567.15      138.62     1495.50
> 
> Summary: Scheduler stuff seems like a wash (schedule -> do_schedule). 
> Seems to be some sort of rearrangement of the dcache stuff which 
> appears to be mildly beneficial (what's going in there?). 
> 
> diffprofile (+ gets worse, - gets better).
> 
> 2023 do_schedule
> 485 dentry_open
> 289 .text.lock.file_table

Looks like you are getting hit by contention on files_lock. I have
been messing around with some code to split up the files_lock, but
I can't seem to get the locking in the tty layer right.

Hmm.. .text.lock.namei is probably dcache_lock. -mms no longer has
dcache_rcu, so not quite sure what helped you here.


Thanks
Dipankar