Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Mon, 30 Nov 2020 09:22:18 -0800
From:   "Darrick J. Wong" <darrick.wong@oracle.com>
To:     chenlei0x@gmail.com
Cc:     hch@infradead.org, linux-xfs@vger.kernel.org,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
        Lei Chen <lennychen@tencent.com>
Subject: Re: [PATCH] fs: iomap: Replace bio_add_page with __bio_add_page in
 iomap_add_to_ioend
Message-ID: <20201130172218.GA143045@magnolia>
References: <1606721331-4211-1-git-send-email-lennychen@tencent.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1606721331-4211-1-git-send-email-lennychen@tencent.com>
Precedence: bulk

On Mon, Nov 30, 2020 at 03:28:51PM +0800, chenlei0x@gmail.com wrote:
> From: Lei Chen <lennychen@tencent.com>
> 
> iomap_add_to_ioend append page on wpc->ioend->io_bio. If io_bio is full,
> iomap_chain_bio will allocate a new bio. So when bio_add_page is called,
> pages is guaranteed to be appended into wpc->ioend->io_bio. So we do not
> need to check if page can be merged. Thus it's a faster way to directly
> call __bio_add_page.

How much faster?

--D

> Signed-off-by: Lei Chen <lennychen@tencent.com>
> ---
>  fs/iomap/buffered-io.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c
> index 10cc797..7a0631a 100644
> --- a/fs/iomap/buffered-io.c
> +++ b/fs/iomap/buffered-io.c
> @@ -1310,7 +1310,7 @@ static void iomap_writepage_end_bio(struct bio *bio)
>  			wpc->ioend->io_bio =
>  				iomap_chain_bio(wpc->ioend->io_bio);
>  		}
> -		bio_add_page(wpc->ioend->io_bio, page, len, poff);
> +		__bio_add_page(wpc->ioend->io_bio, page, len, poff);
>  	}
>  
>  	wpc->ioend->io_size += len;
> -- 
> 1.8.3.1
>