Re: [tip:x86/urgent] BUILD SUCCESS WITH WARNING 064855a69003c24bd6b473b367d364e418c57625

[Reinette Chatre <reinette.chatre@intel.com>]

Hi Babu,

On 8/20/2021 7:16 AM, Babu Moger wrote:
> 
> 
> On 8/19/21 4:05 PM, Borislav Petkov wrote:
>> On Thu, Aug 19, 2021 at 01:39:46PM -0700, Reinette Chatre wrote:
>>> I can confirm that the removed comment explains why m would be initialized
>>> when used in the code that follows.
>>>
>>> How would you prefer to address this? We could add just the comment back in
>>> support of future reports or perhaps by adding the default case back with
>>> the same error that would be returned earlier when there is an invalid
>>> EVENT_ID. Something like:
>>>
>>> ---8<---
>>> diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c
>>> b/arch/x86/kernel/cpu/resctrl/monitor.c
>>> index 57e4bb695ff9..05b99e4d621c 100644
>>> --- a/arch/x86/kernel/cpu/resctrl/monitor.c
>>> +++ b/arch/x86/kernel/cpu/resctrl/monitor.c
>>> @@ -304,6 +304,12 @@ static u64 __mon_event_count(u32 rmid, struct rmid_read
>>> *rr)
>>>   	case QOS_L3_MBM_LOCAL_EVENT_ID:
>>>   		m = &rr->d->mbm_local[rmid];
>>>   		break;
>>> +	default:
>>> +		/*
>>> +		 * Code would never reach here because
>>> +		 * an invalid event id would fail the __rmid_read.
>>> +		 */
>>> +		return RMID_VAL_ERROR;
>>>   	}
>>>
>>>   	if (rr->first) {
>>
>> Right, I would normally not take a patch just to fix a tool because it
>> cannot see it correctly.
>>
>> But Babu has another use case which breaks the build so I guess that's
>> serious enough to make an exception.
>>
>> Babu, can you please explain?
> 
> Details:
> 
> The patch applies cleanly to RHEL8.5 tree but the build fails with an
> uninitialized variable warning treated as an error. The RHEL8.5 Makefile
> uses '-Werror=maybe-uninitialized' to force uninitialized variable
> warnings to be treated as errors. The build error was found while using
> the redhat/configs/kernel-x86_64.config kernel config. BTW, the 5.14-rc6
> build passes and it does not build using '-Werror=maybe-uninitialized'.
> The error from the build is below:
> 
> arch/x86/kernel/cpu/resctrl/monitor.c: In function ‘__mon_event_count’:
> arch/x86/kernel/cpu/resctrl/monitor.c:261:12: error: ‘m’ may be used
> uninitialized in this function [-Werror=maybe-uninitialized]
>    m->chunks += chunks;
>              ^~
> 
> The following patch fixes the problem.
> 
> diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c
> b/arch/x86/kernel/cpu/resctrl/monitor.c
> index 57e4bb695ff9..553cc6410442 100644
> --- a/arch/x86/kernel/cpu/resctrl/monitor.c
> +++ b/arch/x86/kernel/cpu/resctrl/monitor.c
> @@ -304,6 +304,8 @@ static u64 __mon_event_count(u32 rmid, struct
> rmid_read *rr)
>          case QOS_L3_MBM_LOCAL_EVENT_ID:
>                  m = &rr->d->mbm_local[rmid];
>                  break;
> +       default:
> +               return RMID_VAL_ERROR;
>          }
> 
>          if (rr->first) {
> 
> 
> Reinette,
> Are you going to send the official patch or let me know I will send. You
> can add signoff from me after adding above details. We probably need Fixes
> and Cc: stable@vger.kernel.org.
> 

Could you please go ahead and send a patch with these details you 
collected?

Thank you very much

Reinette