linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] mm: release the spinlock on zap_pte_range
@ 2019-07-29  7:10 Minchan Kim
  2019-07-29  7:45 ` Michal Hocko
  2019-08-06  7:05 ` [mm] 755d6edc1a: will-it-scale.per_process_ops -4.1% regression kernel test robot
  0 siblings, 2 replies; 25+ messages in thread
From: Minchan Kim @ 2019-07-29  7:10 UTC (permalink / raw)
  To: Andrew Morton
  Cc: LKML, linux-mm, Minchan Kim, Miguel de Dios, Wei Wang,
	Michal Hocko, Johannes Weiner, Mel Gorman

In our testing(carmera recording), Miguel and Wei found unmap_page_range
takes above 6ms with preemption disabled easily. When I see that, the
reason is it holds page table spinlock during entire 512 page operation
in a PMD. 6.2ms is never trivial for user experince if RT task couldn't
run in the time because it could make frame drop or glitch audio problem.

This patch adds preemption point like coyp_pte_range.

Reported-by: Miguel de Dios <migueldedios@google.com>
Reported-by: Wei Wang <wvw@google.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Minchan Kim <minchan@kernel.org>
---
 mm/memory.c | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)

diff --git a/mm/memory.c b/mm/memory.c
index 2e796372927fd..bc3e0c5e4f89b 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1007,6 +1007,7 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb,
 				struct zap_details *details)
 {
 	struct mm_struct *mm = tlb->mm;
+	int progress = 0;
 	int force_flush = 0;
 	int rss[NR_MM_COUNTERS];
 	spinlock_t *ptl;
@@ -1022,7 +1023,16 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb,
 	flush_tlb_batched_pending(mm);
 	arch_enter_lazy_mmu_mode();
 	do {
-		pte_t ptent = *pte;
+		pte_t ptent;
+
+		if (progress >= 32) {
+			progress = 0;
+			if (need_resched())
+				break;
+		}
+		progress += 8;
+
+		ptent = *pte;
 		if (pte_none(ptent))
 			continue;
 
@@ -1123,8 +1133,11 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb,
 	if (force_flush) {
 		force_flush = 0;
 		tlb_flush_mmu(tlb);
-		if (addr != end)
-			goto again;
+	}
+
+	if (addr != end) {
+		progress = 0;
+		goto again;
 	}
 
 	return addr;
-- 
2.22.0.709.g102302147b-goog


^ permalink raw reply related	[flat|nested] 25+ messages in thread

end of thread, other threads:[~2019-08-27 18:41 UTC | newest]

Thread overview: 25+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-07-29  7:10 [PATCH] mm: release the spinlock on zap_pte_range Minchan Kim
2019-07-29  7:45 ` Michal Hocko
2019-07-29  8:20   ` Minchan Kim
2019-07-29  8:35     ` Michal Hocko
2019-07-30 12:11       ` Minchan Kim
2019-07-30 12:32         ` Michal Hocko
2019-07-30 12:39           ` Minchan Kim
2019-07-30 12:57             ` Michal Hocko
2019-07-31  5:44               ` Minchan Kim
2019-07-31  7:21                 ` Michal Hocko
2019-08-06 10:55                   ` Minchan Kim
2019-08-09 12:43                     ` [RFC PATCH] mm: drop mark_page_access from the unmap path Michal Hocko
2019-08-09 17:57                       ` Mel Gorman
2019-08-09 18:34                       ` Johannes Weiner
2019-08-12  8:09                         ` Michal Hocko
2019-08-12 15:07                           ` Johannes Weiner
2019-08-13 10:51                             ` Michal Hocko
2019-08-26 12:06                               ` Michal Hocko
2019-08-27 16:00                                 ` Johannes Weiner
2019-08-27 18:41                                   ` Michal Hocko
2019-07-30 19:42     ` [PATCH] mm: release the spinlock on zap_pte_range Andrew Morton
2019-07-31  6:14       ` Minchan Kim
2019-08-06  7:05 ` [mm] 755d6edc1a: will-it-scale.per_process_ops -4.1% regression kernel test robot
     [not found]   ` <20190806080415.GG11812@dhcp22.suse.cz>
2019-08-06 11:00     ` Minchan Kim
2019-08-06 11:11       ` Michal Hocko

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).