NACK: [SRU][B/F:linux-kvm][[PATCH 1/1] UBUNTU: [Config] kvm: Add support for modifying LDT

Roxana Nicolescu roxana.nicolescu at canonical.com
Thu Mar 16 16:38:51 UTC 2023


On 15/03/2023 10:20, Stefan Bader wrote:
> On 13.03.23 17:28, Roxana Nicolescu wrote:
>> BugLink: https://bugs.launchpad.net/bugs/2009442
>>
>> Enables CONFIG_MODIFY_LDT_SYSCALL, adding support for modifying
>> the local descriptor table. This allows glibs autopkgtests to run
>> properly.
>>
>> Config matches main kernel as well as newer versions of linux-kvm.
>>
>> Signed-off-by: Roxana Nicolescu <roxana.nicolescu at canonical.com>
>> ---
>
> I am geared towards NACK here because it would be better to have this 
> documented/enforced in annotations. Even with the risk that you then 
> need 2 patches due to context.
> Also, that is an issue but was the related bug report generated with 
> the script for ADT failures? Cause in that case, once you decide this 
> is to be solved on the kernel side, you should invalidate (best with a 
> note) the other source which initially gets marked as affected. In 
> this case autopkgtest. There was already a question about this.
>
> -Stefan
>
>>   debian.kvm/config/config.common.ubuntu | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/debian.kvm/config/config.common.ubuntu 
>> b/debian.kvm/config/config.common.ubuntu
>> index 8bedc2793145..441e71ed7d90 100644
>> --- a/debian.kvm/config/config.common.ubuntu
>> +++ b/debian.kvm/config/config.common.ubuntu
>> @@ -1387,7 +1387,7 @@ CONFIG_MISC_FILESYSTEMS=y
>>   # CONFIG_MMIOTRACE is not set
>>   CONFIG_MMU=y
>>   CONFIG_MMU_NOTIFIER=y
>> -# CONFIG_MODIFY_LDT_SYSCALL is not set
>> +CONFIG_MODIFY_LDT_SYSCALL=y
>>   CONFIG_MODULES=y
>>   CONFIG_MODULES_TREE_LOOKUP=y
>>   CONFIG_MODULES_USE_ELF_RELA=y
>
>

I agree. I will send a v2 where I update the annotations file.

Regarding the ticket, it is now properly updated. When I created it, I 
did not know the root cause and because the failure was in adt, I put 
autopkgtest there. Now it's invalid as you suggested.

Thanks,
Roxana




More information about the kernel-team mailing list