Re: [PATCH v3 3/3] xen/cppcheck: globally suppress unusedStructMember

[Luca Fancellu <Luca.Fancellu@xxxxxxx>]

> On 3 Mar 2023, at 20:55, Stefano Stabellini <sstabellini@xxxxxxxxxx> wrote:
> 
> On Fri, 3 Mar 2023, Luca Fancellu wrote:
>> unusedStructMember warnings from cppcheck are not reliable and
>> are causing a lot of false positives, suppress the checker
>> globally for now.
>> 
>> Signed-off-by: Luca Fancellu <luca.fancellu@xxxxxxx>
> 
> Reviewed-by: Stefano Stabellini <sstabellini@xxxxxxxxxx>
> 
> This is great! Results are much cleaner. With this series accepted, I
> think we can start looking into how to "diff" cppcheck results to spot
> regressions in new patches.

Yes indeed! I have some ideas about the diff, I’ll present something at the
FuSa meeting

> 
> 
>> ---
>> Changes from v2:
>> - New patch
>> ---
>> xen/scripts/xen_analysis/cppcheck_analysis.py | 3 +++
>> 1 file changed, 3 insertions(+)
>> 
>> diff --git a/xen/scripts/xen_analysis/cppcheck_analysis.py 
>> b/xen/scripts/xen_analysis/cppcheck_analysis.py
>> index e385e2c8f54a..ab52ce38d502 100644
>> --- a/xen/scripts/xen_analysis/cppcheck_analysis.py
>> +++ b/xen/scripts/xen_analysis/cppcheck_analysis.py
>> @@ -133,6 +133,8 @@ def generate_cppcheck_deps():
>>     # - Explicitly suppress warnings on compiler-def.h because cppcheck 
>> throws
>>     #   an unmatchedSuppression due to the rule we put in 
>> suppression-list.txt
>>     #   to skip every finding in the file.
>> +    # - Explicitly suppress findings for unusedStructMember that is not very
>> +    #   reliable and causes lots of false positives.
>>     #
>>     # Compiler defines are in compiler-def.h which is included in config.h
>>     #
>> @@ -145,6 +147,7 @@ def generate_cppcheck_deps():
>>  --inline-suppr
>>  --suppressions-list={}/suppression-list.txt
>>  --suppress='unmatchedSuppression:*'
>> + --suppress='unusedStructMember:*'
>>  --include={}/include/xen/config.h
>>  -DCPPCHECK
>> """.format(settings.outdir, CPPCHECK_BUILD_DIR, settings.xen_dir,
>> -- 
>> 2.34.1
>> 
>