Re: [PATCH 0/2] mm: memory-failure: fix HWPoison flag race with non-atomic page flag ops

[David Hildenbrand (Arm)]

On 6/29/26 23:54, Michael S. Tsirkin wrote:
> On Mon, Jun 29, 2026 at 11:22:11PM +0200, David Hildenbrand (Arm) wrote:
>> [...]
>>
>>>
>>> And again, I'm really not sure fixing a theoretical race when memory
>>> is failing is worth slowing the world by 0.1-1% for.
>>>
>>
>> Fully agreed.
> 
> Let's say, we just send every bad page through MF machinery X times,
> and hope the chances of a race are reduced enough not to matter.
> WDYT?

My personal opinion is that we either fix it properly or leave it broken. Making
something unlikely a bit less likely is probably not worth the churn.

Making something unlikely impossible OTOH (with no measurable overhead) seems
tempting, as long as it's not too complicated.

Let's discuss the other options we have in the other subthread.

-- 
Cheers,

David