Re: [PATCH 9/9] block: make blk-throttle preserve the issuing taskon delayed bios

[Vivek Goyal]

On Thu, Feb 16, 2012 at 02:37:58PM -0800, Tejun Heo wrote:
> Make blk-throttle call bio_associate_current() on bios being delayed
> such that they get issued to block layer with the original io_context.
> This allows stacking blk-throttle and cfq-iosched propio policies.
> bios will always be issued with the correct ioc and blkcg whether it
> gets delayed by blk-throttle or not.
> 
> Signed-off-by: Tejun Heo <tj@xxxxxxxxxx>
> Cc: Vivek Goyal <vgoyal@xxxxxxxxxx>
> ---
>  block/blk-throttle.c |    4 ++++
>  1 files changed, 4 insertions(+), 0 deletions(-)
> 
> diff --git a/block/blk-throttle.c b/block/blk-throttle.c
> index a7c8e0b..ad8e16b 100644
> --- a/block/blk-throttle.c
> +++ b/block/blk-throttle.c
> @@ -894,6 +894,9 @@ bool blk_throtl_bio(struct request_queue *q, struct bio *bio)
>  		goto out;
>  	}
>  
> +	/* bio_associate_current() needs ioc, try creating */
> +	create_io_context(GFP_ATOMIC, q->node);
> +
>  	/*
>  	 * A throtl_grp pointer retrieved under rcu can be used to access
>  	 * basic fields like stats and io rates. If a group has no rules,
> @@ -958,6 +961,7 @@ queue_bio:
>  			tg->io_disp[rw], tg->iops[rw],
>  			tg->nr_queued[READ], tg->nr_queued[WRITE]);
>  
> +	bio_associate_current(bio);

blk-throttle is just one of places where we lose submitter's context info.
There are others too.

As Kent said, doing this at more generic level will probably make sense.
I think dm layer also might benefit from it. We had issues with losing
this process context info while bio is being submitted with a helper thread.

https://lkml.org/lkml/2010/10/24/59

But I guess this will require some changes at dm layer too so that it
takes extra bio->ioc and bio->blkcg references at cloning time.

Thanks
Vivek
--
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/