[v2,2/5] fs/proc/task_mmu: Reduce scope of lazy mmu region

Commit Message

Ryan Roberts March 3, 2025, 2:15 p.m. UTC

Update the way arch_[enter|leave]_lazy_mmu_mode() is called in
pagemap_scan_pmd_entry() to follow the normal pattern of holding the ptl
for user space mappings. As a result the scope is reduced to only the
pte table, but that's where most of the performance win is.

While I believe there wasn't technically a bug here, the original scope
made it easier to accidentally nest or, worse, accidentally call
something like kmap() which would expect an immediate mode pte
modification but it would end up deferred.

Acked-by: David Hildenbrand <[e-mail]>
Signed-off-by: Ryan Roberts <[e-mail]>
---
 fs/proc/task_mmu.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

Patch

diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index c17615e21a5d..b0f189815512 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -2459,22 +2459,19 @@  static int pagemap_scan_pmd_entry(pmd_t *pmd, unsigned long start,
 	spinlock_t *ptl;
 	int ret;
 
-	arch_enter_lazy_mmu_mode();
-
 	ret = pagemap_scan_thp_entry(pmd, start, end, walk);
-	if (ret != -ENOENT) {
-		arch_leave_lazy_mmu_mode();
+	if (ret != -ENOENT)
 		return ret;
-	}
 
 	ret = 0;
 	start_pte = pte = pte_offset_map_lock(vma->vm_mm, pmd, start, &ptl);
 	if (!pte) {
-		arch_leave_lazy_mmu_mode();
 		walk->action = ACTION_AGAIN;
 		return 0;
 	}
 
+	arch_enter_lazy_mmu_mode();
+
 	if ((p->arg.flags & PM_SCAN_WP_MATCHING) && !p->vec_out) {
 		/* Fast path for performing exclusive WP */
 		for (addr = start; addr != end; pte++, addr += PAGE_SIZE) {
@@ -2543,8 +2540,8 @@  static int pagemap_scan_pmd_entry(pmd_t *pmd, unsigned long start,
 	if (flush_end)
 		flush_tlb_range(vma, start, addr);
 
-	pte_unmap_unlock(start_pte, ptl);
 	arch_leave_lazy_mmu_mode();
+	pte_unmap_unlock(start_pte, ptl);
 
 	cond_resched();
 	return ret;