From: hch@infradead.org (Christoph Hellwig)
Date: Mon, 19 Nov 2018 00:05:44 -0800
Subject: [PATCH 6/7] block: make blk_poll() take a parameter on whether
 to spin or not
In-Reply-To: <20181117214354.822-7-axboe@kernel.dk>
References: <20181117214354.822-1-axboe@kernel.dk>
 <20181117214354.822-7-axboe@kernel.dk>
Message-ID: <20181119080544.GF9622@infradead.org>

> -bool blk_poll(struct request_queue *q, blk_qc_t cookie)
> +bool blk_poll(struct request_queue *q, blk_qc_t cookie, bool spin)

I find the paramter name a little confusing.  Maybe wait_for_request,
although I don't particularly like that one either.  But we really need
to document the parameter well here, no matter what we end up naming
it.  And we should use a consistent name through the whole stack.

> index c1ec3475a140..f6971b45bc54 100644
> --- a/drivers/nvme/target/io-cmd-bdev.c
> +++ b/drivers/nvme/target/io-cmd-bdev.c
> @@ -116,7 +116,7 @@ static void nvmet_bdev_execute_rw(struct nvmet_req *req)
>  
>  	cookie = submit_bio(bio);
>  
> -	blk_poll(bdev_get_queue(req->ns->bdev), cookie);
> +	blk_poll(bdev_get_queue(req->ns->bdev), cookie, true);

This opportunistic poll is pretty bogus now as we never set the HIPRI
flag and it should probably be removed in a prep patch.  We should then
later try to use a scheme similar to your aio polling for the nvme
target as well.