Re: [PATCH 3/3] mm: read remote memory without the mmap lock where possible

[David Hildenbrand (Arm)]

On 6/19/26 16:03, Suren Baghdasaryan wrote:
> On Fri, Jun 19, 2026 at 6:47 AM Rik van Riel <riel@xxxxxxxxxxx> wrote:
>>
>> On Tue, 2026-06-16 at 23:19 -0700, Suren Baghdasaryan wrote:
>>>
>>> I don't think we should be using per-VMA locks if the read spans
>>> multiple VMAs. Doing that would risk a possibility of reading
>>> inconsistent data since we are locking one VMA at a time. While we
>>> load and read VMA, its neighboring VMA can be unmapped and another
>>> one
>>> can be mapped in its place.
>>
>> How is that different from userspace overwriting
>> data while we are reading it, and us reading half
>> new, and half old contents?
>>
>> We already have nothing synchronizing the contents
>> today.
>>
>> All our locking does is protect the metadata we
>> use to find the memory.
> 
> Current locking ensures that when reading across VMAs, the structure
> of the VMA tree stays consistent during that read. I think that
> *structural consistency* of the area we are reading is important to
> preserve.

I tend to agree: if more than one VMA is involved, it's best to fallback to the
mmap lock for now.

-- 
Cheers,

David