[dm-devel] Re: [PATCH] dm-rwlock.patch (Re: 2.6.4-rc1-mm1: queue-congestion-dm-implementation patch)

[Miquel van Smoorenburg]

On 2004.03.09 10:44, Joe Thornber wrote:
> On Tue, Mar 09, 2004 at 12:31:39AM +0100, Miquel van Smoorenburg wrote:
> > I didn't test for DMF_SUSPENDED since it is only set when DMF_BLOCK_IO
> > is set too. And "r = bdi_bits" is the return value for "congested".
> > So I think that FIXME can be removed ?
> 
> Yep, I thought bdi_bits meant that it wasn't congested.
> 
> I also noticed that I'm not protecting the if(!md->map) in
> __dm_request properly, so this patch pushes that check down into
> __split_bio(), and removes __dm_request() completely.

Looks good. I've been doing heavy I/O on an LVM partition while in another
window running a continuous lvextend/lvreduce loop and didn't see anything
strange happening.

Mike.