By any chance do you have a list of drivers that assume this ?
What does it take to fix them ?
I think Jens BIO changes for 2.5 will fix this problem. But
2.4 needs a solution in this area too. This patch showed
significant improvement for database workloads.
If it is not reasonable to fix all the brokern drivers,
how about making this configurable (to do variable size IO) ?
Do you favour this solution ?
Regards,
Badari
>
> On Wed, Jan 09, 2002 at 10:12:11AM -0800, Badari Pulavarty wrote:
> > why ? could you explain ? I am not expecting that user buffer be aligned
> > to PAGE_SIZE.
>
> Okay, that part I misread from the message, but that leaves the question of
> "does it work?" Iirc, Jeff Merkey tested variable sized ios with nwfs, but
> found that triggered bugs in the low level drivers, some of which assume that
> all buffer heads within a request have the same block size. Given that
> concern, I really don't think this is a safe 2.4 patch.
>
> -ben
> --
> Fish.
-
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/