Re: [PATCHv3 5/6] block/bounce: count bytes instead of sectors

[Christoph Hellwig <hch@lst.de>]

On Mon, May 23, 2022 at 02:01:18PM -0700, Keith Busch wrote:
> From: Keith Busch <kbusch@kernel.org>
> 
> Individual bv_len's may not be a sector size.
> 
> Signed-off-by: Keith Busch <kbusch@kernel.org>
> ---
>  block/bounce.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/block/bounce.c b/block/bounce.c
> index 8f7b6fe3b4db..20a43c4dbdda 100644
> --- a/block/bounce.c
> +++ b/block/bounce.c
> @@ -207,17 +207,18 @@ void __blk_queue_bounce(struct request_queue *q, struct bio **bio_orig)
>  	struct bvec_iter iter;
>  	unsigned i = 0;
>  	bool bounce = false;
> -	int sectors = 0;
> +	int sectors = 0, bytes = 0;
>  
>  	bio_for_each_segment(from, *bio_orig, iter) {
>  		if (i++ < BIO_MAX_VECS)
> -			sectors += from.bv_len >> 9;
> +			bytes += from.bv_len;
>  		if (PageHighMem(from.bv_page))
>  			bounce = true;
>  	}
>  	if (!bounce)
>  		return;
>  
> +	sectors = ALIGN_DOWN(bytes, queue_logical_block_size(q)) >> 9;

Same comment about SECTOR_SHIFT and a comment here.  That being said,
why do we even align here?  Shouldn't bytes always be setor aligned here
and this should be a WARN_ON or other sanity check?  Probably the same
for the previous patch.