Received-SPF: pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <65f66ce6-7695-2884-23a9-378b5c9ca04f@linux.alibaba.com>
Date:   Tue, 6 Sep 2022 09:06:24 +0800
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:91.0)
 Gecko/20100101 Thunderbird/91.13.0
Subject: Re: [Ocfs2-devel] [PATCH v2 08/14] ocfs2: replace ll_rw_block()
Content-Language: en-US
To:     Zhang Yi <yi.zhang@huawei.com>, linux-ext4@vger.kernel.org,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
        cluster-devel@redhat.com, ntfs3@lists.linux.dev,
        ocfs2-devel@oss.oracle.com, reiserfs-devel@vger.kernel.org,
        jack@suse.cz
Cc:     axboe@kernel.dk, hch@infradead.org, tytso@mit.edu,
        agruenba@redhat.com, almaz.alexandrovich@paragon-software.com,
        viro@zeniv.linux.org.uk, yukuai3@huawei.com, rpeterso@redhat.com,
        dushistov@mail.ru, chengzhihao1@huawei.com
References: <20220901133505.2510834-1-yi.zhang@huawei.com>
 <20220901133505.2510834-9-yi.zhang@huawei.com>
From:   Joseph Qi <joseph.qi@linux.alibaba.com>
In-Reply-To: <20220901133505.2510834-9-yi.zhang@huawei.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk


On 9/1/22 9:34 PM, Zhang Yi via Ocfs2-devel wrote:
> ll_rw_block() is not safe for the sync read path because it cannot
> guarantee that submitting read IO if the buffer has been locked. We
> could get false positive EIO after wait_on_buffer() if the buffer has
> been locked by others. So stop using ll_rw_block() in ocfs2.
> 
> Signed-off-by: Zhang Yi <yi.zhang@huawei.com>

Looks good to me.
Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com>

> ---
>  fs/ocfs2/aops.c  | 2 +-
>  fs/ocfs2/super.c | 4 +---
>  2 files changed, 2 insertions(+), 4 deletions(-)
> 
> diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
> index af4157f61927..1d65f6ef00ca 100644
> --- a/fs/ocfs2/aops.c
> +++ b/fs/ocfs2/aops.c
> @@ -636,7 +636,7 @@ int ocfs2_map_page_blocks(struct page *page, u64 *p_blkno,
>  			   !buffer_new(bh) &&
>  			   ocfs2_should_read_blk(inode, page, block_start) &&
>  			   (block_start < from || block_end > to)) {
> -			ll_rw_block(REQ_OP_READ, 1, &bh);
> +			bh_read_nowait(bh, 0);
>  			*wait_bh++=bh;
>  		}
>  
> diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
> index e2cc9eec287c..26b4c2bfee49 100644
> --- a/fs/ocfs2/super.c
> +++ b/fs/ocfs2/super.c
> @@ -1764,9 +1764,7 @@ static int ocfs2_get_sector(struct super_block *sb,
>  	if (!buffer_dirty(*bh))
>  		clear_buffer_uptodate(*bh);
>  	unlock_buffer(*bh);
> -	ll_rw_block(REQ_OP_READ, 1, bh);
> -	wait_on_buffer(*bh);
> -	if (!buffer_uptodate(*bh)) {
> +	if (bh_read(*bh, 0) < 0) {
>  		mlog_errno(-EIO);
>  		brelse(*bh);
>  		*bh = NULL;