Re: Very fast DIO synchronously

On 02/09/2005 09:30, SA wrote:
> However the read / write methods are much faster than the ioctl()s regardless of the 
> datasize transferred. I don't know where the additional ioctl overhead comes from 
> and I found it very surprising that there was such a difference.

The only thing that springs to my mind is that the BKL (big kernel lock) 
is grabbed when processing an ioctl using the regular method.  I don't 
think it should make that much difference though.

-- 
-=( Ian Abbott _at_ MEV Ltd.    E-mail: <abbotti_at_mev.co.uk>        )=-
-=( Tel: +44 (0)161 477 1898   FAX: +44 (0)161 718 3587         )=-

Received on 2005-09-02Z08:29:32