Re: 2.4 vs 2.6 USB mass storage device performance

Alan Stern <[email protected]>
Newsgroups gmane.linux.usb.devel
Message-ID <[email protected]>
On Tue, 20 Nov 2007, syed khader wrote:

> Hi,
>   We are evaluating performance of USB on both 2.4.25
> and 2.6.18.2 kernels on an EHCI mass storage device
> with a Seagat SCSI device on a Conexant's ARM11
> platform.
>   We see significant performace difference w.r.t 2.4
> and 2.6 on the same device:
> 
> On 2.6:
> #sg_dd /dev/sda dio=1 bs=1M count=100 flag=direct
> time=1
>   reports 12MBps
> while on 2.4
> #sg_dd /dev/sda dio=1 bs=1M count=100 flag=direct
> time=1
>   reports just 5.1MBps

It's good that 2.6 is faster than 2.4 and not slower!  :-)

>   I understand that 2.6 is more capable, but does 2.4
> has any limitations? Are there any patches available
> to improve the performance in 2.4?

Not exactly limitations.  2.6 has code to speed up large bulk 
transfers; this code isn't present in 2.4.  That probably accounts for 
much of the speed difference.

The code has never been ported back to 2.4, as far as I know.  Of 
course, if you want to do the porting yourself, please feel free.  I'll 
be happy to tell you what needs to be added or changed.

Alan Stern


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.