On 2/5/24 15:36, zhangpeng (AS) wrote:
> On 2024/2/5 15:31, David Hildenbrand wrote:
>
>> On 05.02.24 08:24, zhangpeng (AS) wrote:
>>> On 2024/2/5 14:52, Huang, Ying wrote:
>>>
>>>> "zhangpeng (AS)" <[email protected]> writes:
>>>>> On 2024/2/5 10:56, Huang, Ying wrote:
>>>>>> Peng Zhang <[email protected]> writes:
>>>>>>> From: ZhangPeng <[email protected]>
>>>>>>>
>>>>>>> The major fault occurred when using mlockall(MCL_CURRENT | MCL_FUTURE)
>>>>>>> in application, which leading to an unexpected performance issue[1].
>>>>>>>
>>>>>>> This caused by temporarily cleared PTE during a read/modify/write update
>>>>>>> of the PTE, eg, do_numa_page()/change_pte_range().
>>>>>>>
>>>>>>> For the data segment of the user-mode program, the global variable area
>>>>>>> is a private mapping. After the pagecache is loaded, the private anonymous
>>>>>>> page is generated after the COW is triggered. Mlockall can lock COW pages
>>>>>>> (anonymous pages), but the original file pages cannot be locked and may
>>>>>>> be reclaimed. If the global variable (private anon page) is accessed when
>>>>>>> vmf->pte is zeroed in numa fault, a file page fault will be triggered.
>>>>>>>
>>>>>>> At this time, the original private file page may have been reclaimed.
>>>>>>> If the page cache is not available at this time, a major fault will be
>>>>>>> triggered and the file will be read, causing additional overhead.
>>>>>>>
>>>>>>> Fix this by rechecking the PTE without acquiring PTL in filemap_fault()
>>>>>>> before triggering a major fault.
>>>>>>>
>>>>>>> Testing file anonymous page read and write page fault performance in ext4
>>>>>>> and ramdisk using will-it-scale[2] on a x86 physical machine. The data
>>>>>>> is the average change compared with the mainline after the patch is
>>>>>>> applied. The test results are within the range of fluctuation, and there
>>>>>>> is no obvious difference. The test results are as follows:
>>>>>>> processes processes_idle threads threads_idle
>>>>>>> ext4 file write: -1.14% -0.08% -1.87% 0.13%
>>>>>>> ext4 file read: 0.03% -0.65% -0.51% -0.08%
>>>>>>> ramdisk file write: -1.21% -0.21% -1.12% 0.11%
>>>>>>> ramdisk file read: 0.00% -0.68% -0.33% -0.02%
>>>>>>>
>>>>>>> [1] https://lore.kernel.org/linux-mm/[email protected]/
>>>>>>> [2] https://github.com/antonblanchard/will-it-scale/
>>>>>>>
>>>>>>> Suggested-by: "Huang, Ying" <[email protected]>
>>>>>>> Suggested-by: Yin Fengwei <[email protected]>
>>>>>>> Signed-off-by: ZhangPeng <[email protected]>
>>>>>>> Signed-off-by: Kefeng Wang <[email protected]>
>>>>>>> ---
>>>>>>> RFC->v1:
>>>>>>> - Add error handling when ptep == NULL per Huang, Ying and Matthew Wilcox
>>>>>>> - Check the PTE without acquiring PTL in filemap_fault(), suggested by
>>>>>>> Huang, Ying and Yin Fengwei
>>>>>>> - Add pmd_none() check before PTE map
>>>>>>> - Update commit message and add performance test information
>>>>>>>
>>>>>>> mm/filemap.c | 18 ++++++++++++++++++
>>>>>>> 1 file changed, 18 insertions(+)
>>>>>>>
>>>>>>> diff --git a/mm/filemap.c b/mm/filemap.c
>>>>>>> index 142864338ca4..b29cdeb6a03b 100644
>>>>>>> --- a/mm/filemap.c
>>>>>>> +++ b/mm/filemap.c
>>>>>>> @@ -3238,6 +3238,24 @@ vm_fault_t filemap_fault(struct vm_fault *vmf)
>>>>>>> mapping_locked = true;
>>>>>>> }
>>>>>>> } else {
>>>>>>> + if (!pmd_none(*vmf->pmd)) {
>>>>>>> + pte_t *ptep;
>>>>>>> +
>>>>>>> + ptep = pte_offset_map_nolock(vmf->vma->vm_mm, vmf->pmd,
>>>>>>> + vmf->address, &vmf->ptl);
>>>>>>> + if (unlikely(!ptep))
>>>>>>> + return VM_FAULT_NOPAGE;
>>>>>>> + /*
>>>>>>> + * Recheck pte as the pte can be cleared temporarily
>>>>>>> + * during a read/modify/write update.
>>>>>>> + */
>>>>>> I think that we should add some comments here about the racy checking.
>>>>> I'll add comments in a v2 as follows:
>>>>> /*
>>>>> * Recheck PTE as the PTE can be cleared temporarily
>>>>> * during a read/modify/write update of the PTE, eg,
>>>>> * do_numa_page()/change_pte_range(). This will trigger
>>>>> * a major fault, even if we use mlockall, which may
>>>>> * affect performance.
>>>>> */
>>>> Sorry, my previous words aren't clear enough. I mean some comments as
>>>> follows,
>>>>
>>>> We don't hold PTL here, so the check is still racy. But acquiring PTL
>>>> hurts performance and the race window seems small enough.
>>>
>>> Got it. I'll add comments in a v2 as follows:
>>> /*
>>> * Recheck PTE as the PTE can be cleared temporarily
>>> * during a read/modify/write update of the PTE.
>>> * We don't hold PTL here as acquiring PTL hurts
>>> * performance. So the check is still racy, but
>>> * the race window seems small enough.
>>> */
>>
>> It'd be worth spelling out what happens when we lose the race.
>>
> I'll add what happens when we lose the race as follows:
> /*
> * Recheck PTE as the PTE can be cleared temporarily
> * during a read/modify/write update of the PTE, eg,
> * do_numa_page()/change_pte_range(). This will trigger
> * a major fault, even if we use mlockall, which may
> * affect performance.
> * We don't hold PTL here as acquiring PTL hurts
> * performance. So the check is still racy, but
> * the race window seems small enough.
> */
>
I believe David was asking to add:
"...but the race window seems small enough.
If we lose the race during the check, the page_fault will
be triggered. Butthe page table entry lock still make sure
the correctness:
- If the page cache is not reclaimed, the page_fault will
work like the page fault was served already and bail out.
- If the page cache is reclaimed, the major fault will be
triggered, page cache is filled, page_fault also work
like the page fault was served already and bail out.
"