Re: [PATCH -V2] tdx, memory hotplug: Check whole hot-adding memory range for TDX

[David Hildenbrand <david@redhat.com>]

On 11.10.24 11:48, David Hildenbrand wrote:
> On 11.10.24 10:51, Huang, Ying wrote:
>> David Hildenbrand <david@redhat.com> writes:
>>
>>> On 11.10.24 03:27, Huang, Ying wrote:
>>>> David Hildenbrand <david@redhat.com> writes:
>>>>
>>>>>>      extern u64 max_mem_size;
>>>>>>        extern int mhp_online_type_from_str(const char *str);
>>>>>> diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
>>>>>> index 621ae1015106..c4769f24b1e2 100644
>>>>>> --- a/mm/memory_hotplug.c
>>>>>> +++ b/mm/memory_hotplug.c
>>>>>> @@ -1305,6 +1305,11 @@ int try_online_node(int nid)
>>>>>>      	return ret;
>>>>>>      }
>>>>>>      +int __weak arch_check_hotplug_memory_range(u64 start, u64 size)
>>>>>> +{
>>>>>> +	return 0;
>>>>>> +}
>>>>>
>>>>> BTW, I remember that "__weak" doesn't always behave the way it would
>>>>> seem, which is the reason we're usually using
>>>>>
>>>>> #define arch_check_hotplug_memory_range arch_check_hotplug_memory_range
>>>>>
>>>>> #ifndef arch_check_hotplug_memory_range
>>>>> ...
>>>>> #endif
>>>>>
>>>>>
>>>>> Not that I remember the details, just that it can result in rather
>>>>> surprising outcomes (e.g., the wrong function getting called).
>>>> I can replace __weak with #define/#ifndef.
>>>> However, it appears that "__weak" is still widely used now.
>>>
>>> Probably better to avoid new ones.
>>
>> Sure.  Will do that in the future versions.
>>
>>> See also
>>> Documentation/dev-tools/checkpatch.rst
>>>
>>> I assume checkpatch.pl should complain as well?
>>
>> Double checked again.  It doesn't complain for that.
> 
> Indeed, it only checks for usage of "weak" for *declarations*. So maybe
> it's fine after all and I am misremembering things. So just leave it as
> is for the time being.
> 

For completeness, this is the issue I remembered:

commit 65d9a9a60fd71be964effb2e94747a6acb6e7015
Author: Naveen N Rao <naveen@kernel.org>
Date:   Fri Jul 1 13:04:04 2022 +0530

     kexec_file: drop weak attribute from functions
     
     As requested
     (http://lkml.kernel.org/r/87ee0q7b92.fsf@email.froward.int.ebiederm.org),
     this series converts weak functions in kexec to use the #ifdef approach.
     
     Quoting the 3e35142ef99fe ("kexec_file: drop weak attribute from
     arch_kexec_apply_relocations[_add]") changelog:
     
     : Since commit d1bcae833b32f1 ("ELF: Don't generate unused section symbols")
     : [1], binutils (v2.36+) started dropping section symbols that it thought
     : were unused.  This isn't an issue in general, but with kexec_file.c, gcc
     : is placing kexec_arch_apply_relocations[_add] into a separate
     : .text.unlikely section and the section symbol ".text.unlikely" is being
     : dropped.  Due to this, recordmcount is unable to find a non-weak symbol in
     : .text.unlikely to generate a relocation record against.



-- 
Cheers,

David / dhildenb