Re: [PATCH] f2fs: Fix type of single bit bitfield in f2fs_io_info

From: Chao Yu <chao@kernel.org>
To: Nathan Chancellor <nathan@kernel.org>
Cc: jaegeuk@kernel.org, ndesaulniers@google.com, trix@redhat.com,
	linux-f2fs-devel@lists.sourceforge.net,
	linux-kernel@vger.kernel.org, llvm@lists.linux.dev,
	patches@lists.linux.dev, kernel test robot <lkp@intel.com>
Subject: Re: [PATCH] f2fs: Fix type of single bit bitfield in f2fs_io_info
Date: Thu, 2 Feb 2023 15:11:18 +0800	[thread overview]
Message-ID: <e4dec7de-40d8-5e66-2e83-17e1540042ff@kernel.org> (raw)
In-Reply-To: <Y9tU+woHY23zHpmv@dev-arch.thelio-3990X>

On 2023/2/2 14:15, Nathan Chancellor wrote:
> On Thu, Feb 02, 2023 at 02:13:18PM +0800, Chao Yu wrote:
>> On 2023/2/2 0:40, Nathan Chancellor wrote:
>>> Clang warns:
>>>
>>>     ../fs/f2fs/data.c:995:17: error: implicit truncation from 'int' to a one-bit wide bit-field changes value from 1 to -1 [-Werror,-Wsingle-bit-bitfield-constant-conversion]
>>>             fio->submitted = 1;
>>>                            ^ ~
>>>     ../fs/f2fs/data.c:1011:15: error: implicit truncation from 'int' to a one-bit wide bit-field changes value from 1 to -1 [-Werror,-Wsingle-bit-bitfield-constant-conversion]
>>>                             fio->retry = 1;
>>>                                        ^ ~
>>>
>>>     ../fs/f2fs/segment.c:3320:16: error: implicit truncation from 'int' to a one-bit wide bit-field changes value from 1 to -1 [-Werror,-Wsingle-bit-bitfield-constant-conversion]
>>>                     fio->in_list = 1;
>>>                                  ^ ~
>>>
>>> There is not a bug here because the value of these fields is never
>>> explicitly compared against (just whether it is zero or non-zero) but
>>> it is easy to silence the warning by using an unsigned type to allow
>>> an assignment of 0 or 1 without implicit conversion.
>>
>> Nathan, thanks a lot for catching this, do you mind letting I merge this fix
>> into original patch? as the original patch has not been upstreamed yet.
> 
> No worries, do whatever you need to! I just care that the problem gets
> resolved one way or another :)

Thank you! Updated a v3 patch. :)

Thanks,

> 
> Cheers,
> Nathan