Re: Adaptive read-ahead: benchmarks

[WU Fengguang <wfg@mail.ustc.edu.cn>]

On Thu, Sep 15, 2005 at 04:28:56PM +0200, Paolo Ornati wrote:
> On Thu, 15 Sep 2005 21:16:51 +0800
> WU Fengguang <wfg@mail.ustc.edu.cn> wrote:
> 
> > oprofile() {
> >         opcontrol --vmlinux=/temp/kernel/linux-2.6.13ra/vmlinux
> >         opcontrol --start
> >         opcontrol --reset
> >         echo $1 > /proc/sys/vm/readahead_ratio
> >         dd if=/temp/kernel/hugefile of=/dev/null bs=$bs
> >         opreport -l -o oprofile.$1.$bs /temp/kernel/linux-2.6.13ra/vmlinux
> >         opcontrol --stop
> > }
> 
> just a side note: shouldn't you umount and remount "/temp" to remove
> the pagecache? Or is the file so big that it doesn't matter?
Thanks for the reminding. I'm currently using huge files for testing:
# ll bigfile hugefile 
-rw-r--r--  2 root root 626M 2005-09-07 20:29 bigfile
-rw-r--r--  2 root root 3.7G 2005-09-11 11:24 hugefile
The umount tip seems nice and I'll use it for future testing :)
> 
> PS: instead of unmount/remount you can use "fadvise":
> http://www.zip.com.au/~akpm/linux/patches/stuff/ext3-tools.tar.gz
OK, I'll check it.
> 
> -- 
> 	Paolo Ornati
> 	Linux 2.6.14-rc1 on x86_64

-- 
WU Fengguang
Dept. of Automation
University of Science and Technology of China
Hefei, Anhui
-
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/