[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <4cfd2806-e461-4aa8-bae3-516b309bf260@nurfuerspam.de>
Date: Tue, 24 Jun 2025 05:36:11 +0200
From: Stefan Herdler <herdler@...fuerspam.de>
To: Hans Verkuil <hans@...erkuil.nl>, lkp@...el.com
Cc: hverkuil@...all.nl, linux-kernel@...r.kernel.org,
oe-kbuild-all@...ts.linux.dev
Subject: Re: [PATCH] media: av7110: Workaround a compiler warning
Hi Hans,
On 20/06/25 15:33 Hans Verkuil wrote:
> Hi Stefan,
>
> On 26/04/2025 23:54, Stefan Herdler wrote:
>> Avoid compiler warning caused by an empty block statement.
>>
>> Signed-off-by: Stefan Herdler <herdler@...fuerspam.de>
>> Reported-by: kernel test robot <lkp@...el.com>
>> Closes: https://lore.kernel.org/oe-kbuild-all/202504262127.FkkWHzfs-lkp@intel.com/
>> ---
>>
>> I could not reproduce this warning with gcc10 here, but I already had similar
>> warnings with gcc12 on other projects.
>> I think it has most probably been triggered by the empty block statement.
>>
>> I'm pretty confident this patch will fix the warning, but not 100% sure.
>>
>> The kernel test robot will pick this patch and test is anyway, if I'm correct.
>> Let's wait, if it complains again.
>> That's the best I can do right now.
>
> Is this patch still needed? I noticed that it wasn't CC-ed to the linux-media
> mailinglist, so it never appeared in patchwork, and was never picked up.
Thanks for the hint, next time I CC the media-list too.
Meanwhile I think the patch isn't necessary.
In haven't found any other evidence that there is something wrong.
- I wasn't able to reproduce this warning.
- Debian has recent builds of kernel 6.15.3 with gcc 14.
The kernel-package contains the "dvb-ttpci.ko" file[1] and there are no
warnings (for this module) in the build log[2].
The warning still doesn't make sense to me -there is no function statement-
and the code looks o.k. to me too.
The only debatable thing is the empty function block, which is generally
suggested to avoid.
Meanwhile I think a null statement ";" is maybe a more elegant solution here.
Is there a recommend practice for this? I haven't found anything in the
documentation so far.
Regards
Stefan
[1]: https://packages.debian.org/search?searchon=contents&keywords=dvb-ttpci.ko&mode=path&suite=experimental&arch=any
[2]: https://buildd.debian.org/status/logs.php?pkg=linux&ver=6.15.3-1%7Eexp1&arch=amd64&suite=experimental
>
> If it is still needed, then please post it to linux-media and it will be
> picked up for v6.17.
>
> Regards,
>
> Hans
>
>>
>> Regards
>> Stefan
>>
>>
>> drivers/staging/media/av7110/av7110.c | 3 ++-
>> 1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/staging/media/av7110/av7110.c b/drivers/staging/media/av7110/av7110.c
>> index bc9a2a40a..962536ed7 100644
>> --- a/drivers/staging/media/av7110/av7110.c
>> +++ b/drivers/staging/media/av7110/av7110.c
>> @@ -159,7 +159,7 @@ static void init_av7110_av(struct av7110 *av7110)
>> * some special handling for the Siemens DVB-C cards...
>> */
>> } else if (av7110_init_analog_module(av7110) == 0) {
>> - /* done. */
>> + goto init_skip_analog_module;
>> } else if (dev->pci->subsystem_vendor == 0x110a) {
>> pr_info("DVB-C w/o analog module @ card %d detected\n", av7110->dvb_adapter.num);
>> av7110->adac_type = DVB_ADAC_NONE;
>> @@ -168,6 +168,7 @@ static void init_av7110_av(struct av7110 *av7110)
>> pr_info("adac type set to %d @ card %d\n", av7110->adac_type, av7110->dvb_adapter.num);
>> }
>>
>> +init_skip_analog_module:
>> if (av7110->adac_type == DVB_ADAC_NONE || av7110->adac_type == DVB_ADAC_MSP34x0) {
>> // switch DVB SCART on
>> ret = av7110_fw_cmd(av7110, COMTYPE_AUDIODAC, MainSwitch, 1, 0);
>
Powered by blists - more mailing lists