Home » Mailing list archives » linux-kernel » 2008 » March » 13

hackbench regression since 2.6.25-rc

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
[view in full thread]
From: Zhang, Yanmin
Subject: hackbench regression since 2.6.25-rc
Date: Thursday, March 13, 2008 - 12:46 am

Comparing with 2.6.24, on my 16-core tigerton, hackbench process mode has about
40% regression with 2.6.25-rc1, and more than 20% regression with kernel
2.6.25-rc4, because rc4 includes the reverting patch of scheduler load balance.

Command to start it.
#hackbench 100 process 2000
I ran it for 3 times and sum the values.

I tried to investiagte it by bisect.
Kernel up to tag 0f4dafc0563c6c49e17fe14b3f5f356e4c4b8806 has the 20% regression.
Kernel up to tag 6e90aa972dda8ef86155eefcdbdc8d34165b9f39 hasn't regression.

Any bisect between above 2 tags cause kernel hang. I tried to checkout to a point between
these 2 tags for many times manually and kernel always paniced.

All patches between the 2 tags are on kobject restructure. I guess such restructure
creates more cache miss on the 16-core tigerton.

Any idea?

-yanmin


--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
Messages in current thread:
hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Thu Mar 13, 12:46 am)
Re: hackbench regression since 2.6.25-rc, Andrew Morton, (Thu Mar 13, 1:48 am)
Re: hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Thu Mar 13, 2:28 am)
Re: hackbench regression since 2.6.25-rc, Andrew Morton, (Thu Mar 13, 2:52 am)
Re: hackbench regression since 2.6.25-rc, Greg KH, (Thu Mar 13, 8:14 am)
Re: hackbench regression since 2.6.25-rc, Randy Dunlap, (Thu Mar 13, 9:19 am)
Re: hackbench regression since 2.6.25-rc, Greg KH, (Thu Mar 13, 10:12 am)
Re: hackbench regression since 2.6.25-rc, Christoph Lameter, (Thu Mar 13, 5:16 pm)
Re: hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Thu Mar 13, 5:50 pm)
Re: hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Thu Mar 13, 8:04 pm)
Re: hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Thu Mar 13, 8:30 pm)
Re: hackbench regression since 2.6.25-rc, Greg KH, (Thu Mar 13, 10:01 pm)
Re: hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Thu Mar 13, 10:28 pm)
Re: hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Thu Mar 13, 10:32 pm)
Re: hackbench regression since 2.6.25-rc, Christoph Lameter, (Thu Mar 13, 11:32 pm)
Re: hackbench regression since 2.6.25-rc, Christoph Lameter, (Thu Mar 13, 11:34 pm)
Re: hackbench regression since 2.6.25-rc, Christoph Lameter, (Thu Mar 13, 11:39 pm)
Re: hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Fri Mar 14, 12:14 am)
Re: hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Fri Mar 14, 12:23 am)
Re: hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Fri Mar 14, 12:29 am)
Re: hackbench regression since 2.6.25-rc, Christoph Lameter, (Fri Mar 14, 2:05 pm)
Re: hackbench regression since 2.6.25-rc, Christoph Lameter, (Fri Mar 14, 2:06 pm)
Re: hackbench regression since 2.6.25-rc, Christoph Lameter, (Fri Mar 14, 2:08 pm)
Re: hackbench regression since 2.6.25-rc, Christoph Lameter, (Fri Mar 14, 5:15 pm)
Re: hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Sun Mar 16, 8:05 pm)
Re: hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Sun Mar 16, 8:35 pm)
Re: hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Mon Mar 17, 12:50 am)
Re: hackbench regression since 2.6.25-rc, Christoph Lameter, (Mon Mar 17, 10:27 am)
Re: hackbench regression since 2.6.25-rc, Christoph Lameter, (Mon Mar 17, 10:32 am)
Re: hackbench regression since 2.6.25-rc, Zhang, Yanmin, (Mon Mar 17, 8:28 pm)
Re: hackbench regression since 2.6.25-rc, Christoph Lameter, (Mon Mar 17, 9:07 pm)