Re: [PATCH] ASoC: amd: ps: Fix uninitialized ret in create_acp64_platform_devs()

From: Mukunda,Vijendar
Date: Thu Jan 05 2023 - 18:36:46 EST


On 05/01/23 23:04, Nathan Chancellor wrote:
> On Thu, Jan 05, 2023 at 10:11:09AM -0700, Nathan Chancellor wrote:
>> On Thu, Jan 05, 2023 at 10:34:10PM +0530, Mukunda,Vijendar wrote:
>>> On 05/01/23 21:23, Nathan Chancellor wrote:
>>>> Clang warns:
>>>>
>>>> sound/soc/amd/ps/pci-ps.c:218:2: error: variable 'ret' is used uninitialized whenever switch default is taken [-Werror,-Wsometimes-uninitialized]
>>>> default:
>>>> ^~~~~~~
>>>> sound/soc/amd/ps/pci-ps.c:239:9: note: uninitialized use occurs here
>>>> return ret;
>>>> ^~~
>>>> sound/soc/amd/ps/pci-ps.c:190:9: note: initialize the variable 'ret' to silence this warning
>>>> int ret;
>>>> ^
>>>> = 0
>>>> 1 error generated.
>>>>
>>>> Return -ENODEV, which matches the debug message's description of this block.
>>>>
>>>> Fixes: 1d325cdaf7a2 ("ASoC: amd: ps: refactor platform device creation logic")
>>>> Link: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FClangBuiltLinux%2Flinux%2Fissues%2F1779&data=05%7C01%7Cvijendar.mukunda%40amd.com%7Ce9133748553c4463824d08daef4317c1%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C638085368693977265%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=nDmZXCJR%2Fwt%2FautwTwcZostLvRtUrEbYV2l7M2W%2FTD0%3D&reserved=0
>>>> Signed-off-by: Nathan Chancellor <nathan@xxxxxxxxxx>
>>>> ---
>>>> sound/soc/amd/ps/pci-ps.c | 1 +
>>>> 1 file changed, 1 insertion(+)
>>>>
>>>> diff --git a/sound/soc/amd/ps/pci-ps.c b/sound/soc/amd/ps/pci-ps.c
>>>> index 401cfd0036be..f54b9fd9c3ce 100644
>>>> --- a/sound/soc/amd/ps/pci-ps.c
>>>> +++ b/sound/soc/amd/ps/pci-ps.c
>>>> @@ -217,6 +217,7 @@ static int create_acp63_platform_devs(struct pci_dev *pci, struct acp63_dev_data
>>>> break;
>>>> default:
>>>> dev_dbg(&pci->dev, "No PDM devices found\n");
>>>> + ret = -ENODEV;
>>> ACP PCI driver supports different configurations. As per design , even  when no child dev
>>> nodes are created , ACP driver probe should be successful. ACP PCI driver probe failure
>>> causes power state transition failures when no child device nodes are created.
>>> We should not return -ENODEV in this case.
>>> return 0 is enough in this case. No need to de-init the ACP.
>> Thank you for the review and input and sorry for getting it wrong! I
>> will send a v2 replacing 'break;' with 'return 0;' shortly.
> Actually, I re-read your message and looked at the code now that I am
> back in front of my keyboard and I am a little confused by your message.
> Should 'goto de_init' be replaced with 'return 0' to avoid the call to
> acp63_deinit() or should ret just be initialized to 0 in this block so
> that the call to acp63_deinit() still happens?
>
> Thanks again for your input!
> Nathan
'goto de-init' should be replaced with 'return 0' for default case.

-
Vijendar
>
>>>> goto de_init;
>>>> }
>>>>
>>>>
>>>> ---
>>>> base-commit: 03178b4f7e2c59ead102e5ab5acb82ce1eaefe46
>>>> change-id: 20230105-wsometimes-uninitialized-pci-ps-c-3b5725c6ed31
>>>>
>>>> Best regards,