linux-kernel - Re: [PATCH v2] fs/buffer: Remove redundant assignment to err

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20230403161043.tecfvgmhacs4j3qp@quack3>
Date:   Mon, 3 Apr 2023 18:10:43 +0200
From:   Jan Kara <jack@...e.cz>
To:     Jiapeng Chong <jiapeng.chong@...ux.alibaba.com>
Cc:     viro@...iv.linux.org.uk, brauner@...nel.org,
        linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
        Abaci Robot <abaci@...ux.alibaba.com>
Subject: Re: [PATCH v2] fs/buffer: Remove redundant assignment to err

On Thu 23-03-23 10:32:59, Jiapeng Chong wrote:
> Variable 'err' set but not used.
> 
> fs/buffer.c:2613:2: warning: Value stored to 'err' is never read.
> 
> Reported-by: Abaci Robot <abaci@...ux.alibaba.com>
> Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4589
> Signed-off-by: Jiapeng Chong <jiapeng.chong@...ux.alibaba.com>

I don't think the patch is quite correct (Christian, please drop it if I'm
correct). See below:

> diff --git a/fs/buffer.c b/fs/buffer.c
> index d759b105c1e7..b3eb905f87d6 100644
> --- a/fs/buffer.c
> +++ b/fs/buffer.c
> @@ -2580,7 +2580,7 @@ int block_truncate_page(struct address_space *mapping,
>  	struct inode *inode = mapping->host;
>  	struct page *page;
>  	struct buffer_head *bh;
> -	int err;
> +	int err = 0;
>  
>  	blocksize = i_blocksize(inode);
>  	length = offset & (blocksize - 1);
> @@ -2593,9 +2593,8 @@ int block_truncate_page(struct address_space *mapping,
>  	iblock = (sector_t)index << (PAGE_SHIFT - inode->i_blkbits);
>  	
>  	page = grab_cache_page(mapping, index);
> -	err = -ENOMEM;
>  	if (!page)
> -		goto out;
> +		return -ENOMEM;
>  
>  	if (!page_has_buffers(page))
>  		create_empty_buffers(page, blocksize, 0);
> @@ -2609,7 +2608,6 @@ int block_truncate_page(struct address_space *mapping,
>  		pos += blocksize;
>  	}
>  
> -	err = 0;
>  	if (!buffer_mapped(bh)) {
>  		WARN_ON(bh->b_size != blocksize);
>  		err = get_block(inode, iblock, bh, 0);
> @@ -2633,12 +2631,11 @@ int block_truncate_page(struct address_space *mapping,
>  
>  	zero_user(page, offset, length);
>  	mark_buffer_dirty(bh);
> -	err = 0;

There is:

        if (!buffer_uptodate(bh) && !buffer_delay(bh) && !buffer_unwritten(bh))
                err = -EIO;

above this assignment. So now we'll be returning -EIO if
block_truncate_page() needs to read the block AFAICT. Did this pass fstests
with some filesystem exercising this code (ext2 driver comes to mind)?

								Honza
-- 
Jan Kara <jack@...e.com>
SUSE Labs, CR