Re: Verified IDE performance issues in kernels newer than 2.4.20

[Samuel Flory]

Herbert Poetzl wrote:

> On Thu, Sep 04, 2003 at 02:31:10AM +0300, K. Hampf wrote:
>  
>
> > BRIEF:
> > I discovered the 2.4.21 and  2.4.22 kernels give me roughly 15% of the 
> > troughput compared to 2.4.20. Anyone working on this?
> >    
>
> out of the blue, the following info could be very useful ...
> (for 2.4.20 and 2.4.22 on your systems)
>
> cat /proc/ide/*
> hdparm -i /dev/hd?
> hdparm /dev/hd?
>
> and try to make it available on a webpage
>  

 Also try "hdparm  -a 2048 <some device>" before running hdparm.  Also 
the ide section of dmesg woul dbe handy in addition to the above.

[root@goblin e2fsprogs-1.26]# hdparm  -t /dev/hdg

/dev/hdg:
Timing buffered disk reads:   70 MB in  3.04 seconds =  23.03 MB/sec
[root@goblin e2fsprogs-1.26]# hdparm  -a 2048 /dev/hdg

/dev/hdg:
setting fs readahead to 2048
readahead    = 2048 (on)
[root@goblin e2fsprogs-1.26]# hdparm  -t /dev/hdg

/dev/hdg:
Timing buffered disk reads:  122 MB in  3.01 seconds =  40.47 MB/sec




--
Once you have their hardware. Never give it back.
(The First Rule of Hardware Acquisition)
Sam Flory  <sflory@xxxxxxxxxxxx>


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/