Re: [PATCH v3 0/3] Encapsulate PTE contents from non-arch code

[Ryan Roberts]

On 13/06/2023 03:16, Muchun Song wrote:
> 
> 
>> On Jun 12, 2023, at 23:15, Ryan Roberts <ryan.roberts@xxxxxxx> wrote:
>>
>> Hi All,
>>
>> (Including wider audience this time since changes touch a fair few subsystems)
>>
>> This is the second half of v3 of a series to improve the encapsulation of pte
>> entries by disallowing non-arch code from directly dereferencing pte_t pointers.
>> Based on earlier feedback, I split the series in 2; the first part, fixes for
>> existing bugs, was already posted at [3] and merged into mm-stable. This second
>> part contains the conversion from direct dereferences to instead use
>> ptep_get()/ptep_get_lockless().
>>
>> See the v1 cover letter at [1] for rationale for this work.
>>
>> Based on feedback at v2, I've removed the new ptep_deref() helper I originally
>> added, and am now using the existing ptep_get() and ptep_get_lockless() helpers.
> 
> When I first saw the name of ptep_get()/ptep_get_lockless(), I thought
> the pte seems like to be protected by the refcount mechanism (Why I have
> this though? Because Qi Zheng has proposed a approach to free pte page tables
> by using the refcount mechanism [1]). And your proposed name of ptep_deref()
> is intuitive for me, so I have another thought, should we rename ptep_get()
> to ptep_deref()? Just a thought from me, I'd like to hear if others object.

I see your point, but I think any renaming exercise should be discussed and
applied in the context of a separate patch series, given that ptep_get() and
ptep_get_lockless() already exist in the code base. This would be a much bigger
job, since it would need to cover all the arch code too.

> 
> Thanks.
> 
> [1] https://lore.kernel.org/lkml/20211110105428.32458-7-zhengqi.arch@xxxxxxxxxxxxx/