mempolicy-optimize-queue_folios_pte_range-by-pte-batching.patch removed from -mm tree
The quilt patch titled
Subject: mempolicy: optimize queue_folios_pte_range by PTE batching
has been removed from the -mm tree. Its filename was
mempolicy-optimize-queue_folios_pte_range-by-pte-batching.patch
This patch was dropped because it was merged into the mm-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
------------------------------------------------------
From: Dev Jain <[email protected]>
Subject: mempolicy: optimize queue_folios_pte_range by PTE batching
Date: Wed, 16 Apr 2025 11:00:48 +0530
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.
---
mm/mempolicy.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-)
mm/mempolicy.c renamed from mm/mempolicy.c~mempolicy-optimize-queue_folios_pte_range-by-pte-batching +10 -2
@@ -566,6 +566,7 @@ static void queue_folios_pmd(pmd_t *pmd,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_tpte_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 +588,9 @@ static int queue_folios_pte_range(pmd_t walk->action =ACTION_AGAIN ;return 0 ; } -for (; addr != end; pte++, addr +=PAGE_SIZE ) { +for (; addr != end; pte += nr, addr += nr *PAGE_SIZE ) { + max_nr = (end - addr) >>PAGE_SHIFT ; + nr =1 ; ptent =ptep_get (pte);if (pte_none (ptent))continue ;
@@ -598,6 +602,10 @@ static int queue_folios_pte_range(pmd_t folio =vm_normal_folio (vma, addr, ptent);if (!folio ||folio_is_zone_device (folio))continue ; +if (folio_test_large (folio) && max_nr !=1 ) + nr =folio_pte_batch (folio, addr, pte, ptent, + max_nr, fpb_flags, +NULL ,NULL ,NULL );/* * vm_normal_folio() filters out zero pages, but there might * still be reserved folios to skip, perhaps in a VDSO.
@@ -630,7 +638,7 @@ static int queue_folios_pte_range(pmd_tif (!(flags & (MPOL_MF_MOVE |MPOL_MF_MOVE_ALL )) || !vma_migratable (vma) || !migrate_folio_add (folio, qp->pagelist, flags)) { - qp->nr_failed++; + qp->nr_failed += nr;if (strictly_unmovable (flags))break ; }
_
Patches currently in -mm which might be from [email protected] are