Re: [PATCH 3/5] mm/parisc: constify ptep_get() argument

[John David Anglin <dave.anglin@bell.net>]

On 2026-07-25 12:27 p.m., Helge Deller wrote:
> On 7/24/26 17:48, Usama Arif wrote:
>>
>>
>> On 24/07/2026 16:32, Pedro Falcato wrote:
>>> On Fri, Jul 24, 2026 at 04:03:44PM +0100, Usama Arif wrote:
>>>>
>>>>
>>>> On 24/07/2026 15:14, Matthew Wilcox wrote:
>>>>> On Fri, Jul 24, 2026 at 02:36:59PM +0100, Usama Arif wrote:
>>>>>>
>>>>>>
>>>>>> On 24/07/2026 11:20, Pedro Falcato wrote:
>>>>>>> ptep_get() does not need write access to the PTE.
>>>>>>>
>>>>>>> Signed-off-by: Pedro Falcato <pfalcato@suse.de>
>>>>>>> ---
>>>>>>>   arch/parisc/include/asm/pgtable.h | 2 +-
>>>>>>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>>>>>>
>>>>>> hmm I think you might break build bisectibility if
>>>>>> you separate out the patches, you should squash patch 3 and 4,
>>>>>> with patch 1.
>>>>>
>>>>> I don't see how this breaks bisectability.  Can you elaborate on what
>>>>> you think would break?
>>>>
>>>> Patch 1 does:
>>>>
>>>> -static inline pte_t ptep_get_lockless(pte_t *ptep)
>>>> +static inline pte_t ptep_get_lockless(const pte_t *ptep)
>>>>   {
>>>>       return ptep_get(ptep);
>>>>   }
>>>>
>>>>
>>>> ptep_get() takes a non-const arg till patch 3 for example for parsic.
>>>> I don't think you can pass a const variable to non const function arg?
>>>>
>>>> So parsic won't compile in patch 1 and 2, powerprc wont compile
>>>> for patches 1, 2 and 3..
>>>
>>> Aha, yes, nice catch! I'll have to rethink that. Maybe squashing the patches
>>> would be the cleanest way forward.
>>>
>>
>> Yes, squashing is the simplest way forward.
>>
>>> FWIW, the vast majority of architectures aren't doing funny things on
>>> ptep_get(). E.g PA-RISC, loongarch only define these so they can use it in
>>> their own asm/pgtable.h. I'd really like to delete these but I can't tell
>>> if they are *actually* required.
>>>
>>
>> Ah maybe there is something in git history on why it was needed?
>>
>> If not, hopefully someone from the arch lists could clarify..
> I tried to remove ptep_get() and ptep_test_and_clear_young() from
> arch/parisc/include/asm/pgtable.h, but this then triggers compile errors:
> 
>  CC      arch/parisc/kernel/asm-offsets.s
> In file included from /home/cvs/parisc/git-kernel/linus-linux-2.6/include/linux/mm.h:31,
>                  from /home/cvs/parisc/git-kernel/linus-linux-2.6/include/linux/pid_namespace.h:7,
>                  from /home/cvs/parisc/git-kernel/linus-linux-2.6/include/linux/ptrace.h:10,
>                  from /home/cvs/parisc/git-kernel/linus-linux-2.6/arch/parisc/kernel/asm-offsets.c:21:
> /home/cvs/parisc/git-kernel/linus-linux-2.6/include/linux/pgtable.h: In function ‘clear_young_dirty_ptes’:
> /home/cvs/parisc/git-kernel/linus-linux-2.6/include/linux/pgtable.h:683:25: error: implicit declaration of function ‘ptep_test_and_clear_young’;
> did you mean ‘pmdp_test_and_clear_young’? [-Wimplicit-function-declaration]
>   683 |                         ptep_test_and_clear_young(vma, addr, ptep);
>       |                         ^~~~~~~~~~~~~~~~~~~~~~~~~
>       |                         pmdp_test_and_clear_young
> 
> So, I think the answer is, that the local copy of ptep_get() is used in the parisc implementation of
> ptep_test_and_clear_young() function.
> 
> Switching to the default include/linux/pgtable.h seems not easy doable....

Did you remove define for __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG in parisc version of pgtable.h?

Dave
-- 
John David Anglin  dave.anglin@bell.net