Re: [PATCH 1/2] arm64/mm: Change THP helpers to comply with generic MM semantics

[Anshuman Khandual <anshuman.khandual@arm.com>]

On 08/18/2020 05:56 PM, Jonathan Cameron wrote:
> On Tue, 18 Aug 2020 15:11:58 +0530
> Anshuman Khandual <anshuman.khandual@arm.com> wrote:
> 
>> On 08/18/2020 02:43 PM, Jonathan Cameron wrote:
>>> On Mon, 17 Aug 2020 14:49:43 +0530
>>> Anshuman Khandual <anshuman.khandual@arm.com> wrote:
>>>   
>>>> pmd_present() and pmd_trans_huge() are expected to behave in the following
>>>> manner during various phases of a given PMD. It is derived from a previous
>>>> detailed discussion on this topic [1] and present THP documentation [2].
>>>>
>>>> pmd_present(pmd):
>>>>
>>>> - Returns true if pmd refers to system RAM with a valid pmd_page(pmd)
>>>> - Returns false if pmd does not refer to system RAM - Invalid pmd_page(pmd)
>>>>
>>>> pmd_trans_huge(pmd):
>>>>
>>>> - Returns true if pmd refers to system RAM and is a trans huge mapping
>>>>
>>>> -------------------------------------------------------------------------
>>>> |	PMD states	|	pmd_present	|	pmd_trans_huge	|
>>>> -------------------------------------------------------------------------
>>>> |	Mapped		|	Yes		|	Yes		|
>>>> -------------------------------------------------------------------------
>>>> |	Splitting	|	Yes		|	Yes		|
>>>> -------------------------------------------------------------------------
>>>> |	Migration/Swap	|	No		|	No		|
>>>> -------------------------------------------------------------------------
>>>>
>>>> The problem:
>>>>
>>>> PMD is first invalidated with pmdp_invalidate() before it's splitting. This
>>>> invalidation clears PMD_SECT_VALID as below.
>>>>
>>>> PMD Split -> pmdp_invalidate() -> pmd_mkinvalid -> Clears PMD_SECT_VALID
>>>>
>>>> Once PMD_SECT_VALID gets cleared, it results in pmd_present() return false
>>>> on the PMD entry. It will need another bit apart from PMD_SECT_VALID to re-
>>>> affirm pmd_present() as true during the THP split process. To comply with
>>>> above mentioned semantics, pmd_trans_huge() should also check pmd_present()
>>>> first before testing presence of an actual transparent huge mapping.
>>>>
>>>> The solution:
>>>>
>>>> Ideally PMD_TYPE_SECT should have been used here instead. But it shares the
>>>> bit position with PMD_SECT_VALID which is used for THP invalidation. Hence
>>>> it will not be there for pmd_present() check after pmdp_invalidate().
>>>>
>>>> A new software defined PMD_PRESENT_INVALID (bit 59) can be set on the PMD
>>>> entry during invalidation which can help pmd_present() return true and in
>>>> recognizing the fact that it still points to memory.
>>>>
>>>> This bit is transient. During the split process it will be overridden by a
>>>> page table page representing normal pages in place of erstwhile huge page.
>>>> Other pmdp_invalidate() callers always write a fresh PMD value on the entry
>>>> overriding this transient PMD_PRESENT_INVALID bit, which makes it safe.
>>>>
>>>> [1]: https://lkml.org/lkml/2018/10/17/231
>>>> [2]: https://www.kernel.org/doc/Documentation/vm/transhuge.txt  
>>>
>>> Hi Anshuman,
>>>
>>> One query on this.  From my reading of the ARM ARM, bit 59 is not
>>> an ignored bit.  The exact requirements for hardware to be using
>>> it are a bit complex though.
>>>
>>> It 'might' be safe to use it for this, but if so can we have a comment
>>> explaining why.  Also more than possible I'm misunderstanding things!   
>>
>> We are using this bit 59 only when the entry is not active from MMU
>> perspective i.e PMD_SECT_VALID is clear.
>>
> 
> Understood. I guess we ran out of bits that were always ignored so had
> to start using ones that are ignored in this particular state.

Right, there are no more available SW PTE bits.

#define PTE_DIRTY               (_AT(pteval_t, 1) << 55)
#define PTE_SPECIAL             (_AT(pteval_t, 1) << 56)
#define PTE_DEVMAP              (_AT(pteval_t, 1) << 57)
#define PTE_PROT_NONE           (_AT(pteval_t, 1) << 58) /* only when !PTE_VALID */

Earlier I had proposed using PTE_SPECIAL at PMD level for this purpose.
But Catalin prefers these unused bits as the entry is anyway invalid
and which also leaves aside PTE_SPECIAL at mapped PMD for later use.
There is already one comment near PMD_PRESENT_INVALID definition which
explains this situation.

+/*
+ * This help indicate that the entry is present i.e pmd_page()
+ * still points to a valid huge page in memory even if the pmd
+ * has been invalidated.
+ */
+#define PMD_PRESENT_INVALID	(_AT(pteval_t, 1) << 59) /* only when !PMD_SECT_VALID */