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;
Subject: Re: [PATCH] btrfs: Simplify conditional in assert
To:     dsterba@suse.cz, Wan Jiabing <wanjiabing@vivo.com>,
        Chris Mason <clm@fb.com>, Josef Bacik <josef@toxicpanda.com>,
        David Sterba <dsterba@suse.com>, linux-btrfs@vger.kernel.org,
        linux-kernel@vger.kernel.org, kael_w@yeah.net,
        johannes.thumshirn@wdc.com
References: <20211015103639.21838-1-wanjiabing@vivo.com>
 <20211015105154.GC30611@twin.jikos.cz>
From:   Nikolay Borisov <nborisov@suse.com>
Message-ID: <62b235b7-a99d-4244-89cf-e6649fbd3dd4@suse.com>
Date:   Fri, 15 Oct 2021 17:18:09 +0300
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.13.0
MIME-Version: 1.0
In-Reply-To: <20211015105154.GC30611@twin.jikos.cz>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Precedence: bulk


On 15.10.21 г. 13:51, David Sterba wrote:
> Adding Johannes to CC,
> 
> On Fri, Oct 15, 2021 at 06:36:39AM -0400, Wan Jiabing wrote:
>> Fix following coccicheck warning:
>> ./fs/btrfs/inode.c:2015:16-18: WARNING !A || A && B is equivalent to !A || B
>>
>> Signed-off-by: Wan Jiabing <wanjiabing@vivo.com>
>> ---
>>  fs/btrfs/inode.c | 3 +--
>>  1 file changed, 1 insertion(+), 2 deletions(-)
>>
>> diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
>> index e9154b436c47..da4aeef73b0d 100644
>> --- a/fs/btrfs/inode.c
>> +++ b/fs/btrfs/inode.c
>> @@ -2011,8 +2011,7 @@ int btrfs_run_delalloc_range(struct btrfs_inode *inode, struct page *locked_page
>>  		 * to use run_delalloc_nocow() here, like for  regular
>>  		 * preallocated inodes.
>>  		 */
>> -		ASSERT(!zoned ||
>> -		       (zoned && btrfs_is_data_reloc_root(inode->root)));
>> +		ASSERT(!zoned || btrfs_is_data_reloc_root(inode->root));
> 
> The short form is equivalent, but I'm not sure it's also on the same
> level of readability. Repeating the 'zoned' condition check makes it
> obvious on first sight, which is what I'd prefer.
> 
> Johannes if you'd like the new version I'll change it but otherwise I'm
> fine with what we have now.

Just my 2 cents:

The less code we have the better, i.e !zoned is obvious when it's true
i.e when zoned is false. So the way I read teh assert with the short
form is "we are not zoned OR we are (this is implicit) and this is the
data reloc root". Obviously this is personal preference as you deem it's
better  better to have the !zoned || zoned.


> 
>>  		ret = run_delalloc_nocow(inode, locked_page, start, end,
>>  					 page_started, nr_written);
>>  	} else if (!inode_can_compress(inode) ||
>> -- 
>> 2.20.1
>