Re: [PATCH v5 10/11] block: remove bio_get_nr_vecs()

[Steven Whitehouse <swhiteho@redhat.com>]

Hi,


On 06/07/15 08:44, Ming Lin wrote:
> From: Kent Overstreet <kent.overstreet@gmail.com>
>
> We can always fill up the bio now, no need to estimate the possible
> size based on queue parameters.
[snip]
> diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c
> index 2c1ae86..64d3116 100644
> --- a/fs/gfs2/lops.c
> +++ b/fs/gfs2/lops.c
> @@ -261,18 +261,11 @@ void gfs2_log_flush_bio(struct gfs2_sbd *sdp, int rw)
>   static struct bio *gfs2_log_alloc_bio(struct gfs2_sbd *sdp, u64 blkno)
>   {
>   	struct super_block *sb = sdp->sd_vfs;
> -	unsigned nrvecs = bio_get_nr_vecs(sb->s_bdev);
>   	struct bio *bio;
>   
>   	BUG_ON(sdp->sd_log_bio);
>   
> -	while (1) {
> -		bio = bio_alloc(GFP_NOIO, nrvecs);
> -		if (likely(bio))
> -			break;
> -		nrvecs = max(nrvecs/2, 1U);
> -	}
> -
> +	bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES);
>   	bio->bi_iter.bi_sector = blkno * (sb->s_blocksize >> 9);
>   	bio->bi_bdev = sb->s_bdev;
>   	bio->bi_end_io = gfs2_end_log_write;

When I wrote this loop I understood that bio_alloc(GFP_NOIO, x) was only 
guaranteed to be successful in the case x = 1, and that for x > 1, it 
might return NULL. Does this now mean that for x = BIO_MAX_PAGES it will 
be guaranteed to never return NULL in this case?

Steve.