Re: [PATCH] Bluetooth: Pre-initialize variables in read_local_oob_ext_data_complete()

From: Geert Uytterhoeven <geert@linux-m68k.org>
To: Marcel Holtmann <marcel@holtmann.org>
Cc: "Gustavo F. Padovan" <gustavo@padovan.org>,
	Johan Hedberg <johan.hedberg@gmail.com>,
	BlueZ development <linux-bluetooth@vger.kernel.org>,
	netdev <netdev@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] Bluetooth: Pre-initialize variables in read_local_oob_ext_data_complete()
Date: Mon, 20 Apr 2015 10:47:03 +0200	[thread overview]
Message-ID: <CAMuHMdUXcAYe4KCqG9wV3O5QYzK9RGt+QZz7E_jk7v9yhtrkRQ@mail.gmail.com> (raw)
In-Reply-To: <8684664B-22C2-45E8-B547-48F173B3CA7A@holtmann.org>

Hi Marcel,

On Fri, Apr 17, 2015 at 10:38 PM, Marcel Holtmann <marcel@holtmann.org> wrote:
>>>> net/bluetooth/mgmt.c: In function ‘read_local_oob_ext_data_complete’:
>>>> net/bluetooth/mgmt.c:6474: warning: ‘r256’ may be used uninitialized in this function
>>>> net/bluetooth/mgmt.c:6474: warning: ‘h256’ may be used uninitialized in this function
>>>> net/bluetooth/mgmt.c:6474: warning: ‘r192’ may be used uninitialized in this function
>>>> net/bluetooth/mgmt.c:6474: warning: ‘h192’ may be used uninitialized in this function
>>>>
>>>> While these are false positives, the code can be shortened by
>>>> pre-initializing the hash table pointers and eir_len. This has the side
>>>> effect of killing the compiler warnings.
>>>
>>> can you be a bit specific on which compiler version is this. I fixed one occurrence that seemed valid. However in this case the compiler seems to be just plain stupid. On a gcc 4.9, I am not seeing these for example.
>>
>> gcc 4.1.2. As there were too many false positives, these warnings were
>> disabled in later versions (throwing away the children with the bad water).
>>
>> If you don't like my patch, just drop it. I only look at newly
>> introduced warnings
>> of this kind anyway.
>
> I really do not know what is the best solution here. This is a false positive. And I have been looking at this particular code for a warning that was valid, but we missed initially. But these warnings that you are fixing are clearly false positive.

I only sent patches to fix false positives if I think the patches improve the
code. As this is a subjective matter, it's up to you as the maintainer to
decide.

> If this only happens with an old compiler version, I would tend to leave the code as is. Then again, what is the general preferred approach here?

As this is a false positive, it's clearly up to the maintainer to
decide if the patch
improves the code or not.

Thanks!

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds