On 3/9/2024 4:46 AM, Borislav Petkov wrote:
> On Fri, Mar 08, 2024 at 05:43:26PM -0500, Yazen Ghannam wrote:
>>> diff --git a/drivers/ras/amd/atl/umc.c b/drivers/ras/amd/atl/umc.c
>>> index 08c6dbd44c62..65a0ab651ee2 100644
>>> --- a/drivers/ras/amd/atl/umc.c
>>> +++ b/drivers/ras/amd/atl/umc.c
>>> @@ -315,7 +315,7 @@ static u8 get_die_id(struct atl_err *err)
>>> * For CPUs, this is the AMD Node ID modulo the number
>>> * of AMD Nodes per socket.
>>> */
>>> - return topology_die_id(err->cpu) % amd_get_nodes_per_socket();
>>> + return topology_die_id(err->cpu) % topology_amd_nodes_per_pkg();
>>
>> "topology_die_id -> topology_amd_node_id" is also needed.
>
> Are you saying topology_die_id() was already wrong?
>
> Because even before the topo rewrite, this was
>
> - cpuinfo_x86.topo.die_id:
>
> The physical ID of the die. This information is retrieved via CPUID.
>
> while this code talks about the AMD node thing.
>
We used to save AMD Node ID into topology_die_id() before the following:
03fa6bea5a3e ("x86/cpu: Make topology_amd_node_id() use the actual node info")
>> Does this need to be fixed up in the RAS tree?
>
> I'll give a diff to Linus when I send the pull request.
>
Thanks,
Yazen
