linux-kernel - Re: [PATCH v2] mempolicy: Optimize queue_folios_pte

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <37d70e38-5da7-401a-a7d3-62b5d1aa4407@arm.com>
Date: Wed, 16 Apr 2025 10:28:29 +0530
From: Dev Jain <dev.jain@....com>
To: David Hildenbrand <david@...hat.com>, akpm@...ux-foundation.org
Cc: ryan.roberts@....com, willy@...radead.org, linux-mm@...ck.org,
 linux-kernel@...r.kernel.org, hughd@...gle.com, vishal.moola@...il.com,
 yang@...amperecomputing.com, ziy@...dia.com
Subject: Re: [PATCH v2] mempolicy: Optimize queue_folios_pte_range by PTE
 batching



On 15/04/25 10:49 pm, David Hildenbrand wrote:
> On 15.04.25 16:57, Dev Jain wrote:
>> After the check for queue_folio_required(), the code only cares about the
>> folio in the for loop, i.e the PTEs are redundant. Therefore, optimize
>> this loop by skipping over a PTE batch mapping the same folio.
>>
>> With a test program migrating pages of the calling process, which 
>> includes
>> a mapped VMA of size 4GB with pte-mapped large folios of order-9, and
>> migrating once back and forth node-0 and node-1, the average execution
>> time reduces from 7.5 to 4 seconds, giving an approx 47% speedup.
>>
>> v1->v2:
>>   - Follow reverse xmas tree declarations
>>   - Don't initialize nr
>>   - Move folio_pte_batch() immediately after retrieving a normal folio
>>   - increment nr_failed in one shot
>>
>> Signed-off-by: Dev Jain <dev.jain@....com>
>> ---
>>   mm/mempolicy.c | 12 ++++++++++--
>>   1 file changed, 10 insertions(+), 2 deletions(-)
>>
>> diff --git a/mm/mempolicy.c b/mm/mempolicy.c
>> index b28a1e6ae096..ca90cdcd3207 100644
>> --- a/mm/mempolicy.c
>> +++ b/mm/mempolicy.c
>> @@ -566,6 +566,7 @@ static void queue_folios_pmd(pmd_t *pmd, struct 
>> mm_walk *walk)
>>   static int queue_folios_pte_range(pmd_t *pmd, unsigned long addr,
>>               unsigned long end, struct mm_walk *walk)
>>   {
>> +    const fpb_t fpb_flags = FPB_IGNORE_DIRTY | FPB_IGNORE_SOFT_DIRTY;
>>       struct vm_area_struct *vma = walk->vma;
>>       struct folio *folio;
>>       struct queue_pages *qp = walk->private;
>> @@ -573,6 +574,7 @@ static int queue_folios_pte_range(pmd_t *pmd, 
>> unsigned long addr,
>>       pte_t *pte, *mapped_pte;
>>       pte_t ptent;
>>       spinlock_t *ptl;
>> +    int max_nr, nr;
>>       ptl = pmd_trans_huge_lock(pmd, vma);
>>       if (ptl) {
>> @@ -586,7 +588,8 @@ static int queue_folios_pte_range(pmd_t *pmd, 
>> unsigned long addr,
>>           walk->action = ACTION_AGAIN;
>>           return 0;
>>       }
>> -    for (; addr != end; pte++, addr += PAGE_SIZE) {
>> +    for (; addr != end; pte += nr, addr += nr * PAGE_SIZE) {
>> +        nr = 1;
>>           ptent = ptep_get(pte);
>>           if (pte_none(ptent))
>>               continue;
>> @@ -598,6 +601,11 @@ static int queue_folios_pte_range(pmd_t *pmd, 
>> unsigned long addr,
>>           folio = vm_normal_folio(vma, addr, ptent);
>>           if (!folio || folio_is_zone_device(folio))
>>               continue;
>> +        if (folio_test_large(folio) &&
>> +           (max_nr = ((end - addr) >> PAGE_SHIFT)) != 1)
> 
> That's real nasty :)
> 
> Let's simply do at the beginning of the loop:
> 
>      max_nr = (end - addr) >> PAGE_SHIFT;
>      nr = 1;
> 
> Then here
> 
> if (folio_test_large(folio) && max_nr != 1)
>      nr = ...
> 
> The compiler is smart enough to optimize the computation of values where 
> really required.

If that's the case, I'll change it, thanks.

> 
> With that
> 
> Acked-by: David Hildenbrand <david@...hat.com>

Thanks!

> 
> Thanks!
>