linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Zi Yan <zi.yan@sent.com>
To: David Hildenbrand <david@redhat.com>, linux-mm@kvack.org
Cc: linux-kernel@vger.kernel.org,
	Michael Ellerman <mpe@ellerman.id.au>,
	Christoph Hellwig <hch@lst.de>,
	Marek Szyprowski <m.szyprowski@samsung.com>,
	Robin Murphy <robin.murphy@arm.com>,
	linuxppc-dev@lists.ozlabs.org,
	virtualization@lists.linux-foundation.org,
	iommu@lists.linux-foundation.org,
	Vlastimil Babka <vbabka@suse.cz>,
	Mel Gorman <mgorman@techsingularity.net>,
	Eric Ren <renzhengeek@gmail.com>, Zi Yan <ziy@nvidia.com>
Subject: [PATCH v4 1/7] mm: page_alloc: avoid merging non-fallbackable pageblocks with others.
Date: Wed, 19 Jan 2022 14:06:17 -0500	[thread overview]
Message-ID: <20220119190623.1029355-2-zi.yan@sent.com> (raw)
In-Reply-To: <20220119190623.1029355-1-zi.yan@sent.com>

From: Zi Yan <ziy@nvidia.com>

This is done in addition to MIGRATE_ISOLATE pageblock merge avoidance.
It prepares for the upcoming removal of the MAX_ORDER-1 alignment
requirement for CMA and alloc_contig_range().

MIGRARTE_HIGHATOMIC should not merge with other migratetypes like
MIGRATE_ISOLATE and MIGRARTE_CMA[1], so this commit prevents that too.
Also add MIGRARTE_HIGHATOMIC to fallbacks array for completeness.

[1] https://lore.kernel.org/linux-mm/20211130100853.GP3366@techsingularity.net/

Signed-off-by: Zi Yan <ziy@nvidia.com>
---
 include/linux/mmzone.h | 11 +++++++++++
 mm/page_alloc.c        | 37 ++++++++++++++++++++-----------------
 2 files changed, 31 insertions(+), 17 deletions(-)

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index aed44e9b5d89..71b77aab748d 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -83,6 +83,17 @@ static inline bool is_migrate_movable(int mt)
 	return is_migrate_cma(mt) || mt == MIGRATE_MOVABLE;
 }
 
+/*
+ * Check whether a migratetype can be merged with another migratetype.
+ *
+ * It is only mergeable when it can fall back to other migratetypes for
+ * allocation. See fallbacks[MIGRATE_TYPES][3] in page_alloc.c.
+ */
+static inline bool migratetype_is_mergeable(int mt)
+{
+	return mt < MIGRATE_PCPTYPES;
+}
+
 #define for_each_migratetype_order(order, type) \
 	for (order = 0; order < MAX_ORDER; order++) \
 		for (type = 0; type < MIGRATE_TYPES; type++)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 8dd6399bafb5..15de65215c02 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1117,25 +1117,24 @@ static inline void __free_one_page(struct page *page,
 	}
 	if (order < MAX_ORDER - 1) {
 		/* If we are here, it means order is >= pageblock_order.
-		 * We want to prevent merge between freepages on isolate
-		 * pageblock and normal pageblock. Without this, pageblock
-		 * isolation could cause incorrect freepage or CMA accounting.
+		 * We want to prevent merge between freepages on pageblock
+		 * without fallbacks and normal pageblock. Without this,
+		 * pageblock isolation could cause incorrect freepage or CMA
+		 * accounting or HIGHATOMIC accounting.
 		 *
 		 * We don't want to hit this code for the more frequent
 		 * low-order merging.
 		 */
-		if (unlikely(has_isolate_pageblock(zone))) {
-			int buddy_mt;
+		int buddy_mt;
 
-			buddy_pfn = __find_buddy_pfn(pfn, order);
-			buddy = page + (buddy_pfn - pfn);
-			buddy_mt = get_pageblock_migratetype(buddy);
+		buddy_pfn = __find_buddy_pfn(pfn, order);
+		buddy = page + (buddy_pfn - pfn);
+		buddy_mt = get_pageblock_migratetype(buddy);
 
-			if (migratetype != buddy_mt
-					&& (is_migrate_isolate(migratetype) ||
-						is_migrate_isolate(buddy_mt)))
-				goto done_merging;
-		}
+		if (migratetype != buddy_mt
+				&& (!migratetype_is_mergeable(migratetype) ||
+					!migratetype_is_mergeable(buddy_mt)))
+			goto done_merging;
 		max_order = order + 1;
 		goto continue_merging;
 	}
@@ -2484,6 +2483,7 @@ static int fallbacks[MIGRATE_TYPES][3] = {
 	[MIGRATE_UNMOVABLE]   = { MIGRATE_RECLAIMABLE, MIGRATE_MOVABLE,   MIGRATE_TYPES },
 	[MIGRATE_MOVABLE]     = { MIGRATE_RECLAIMABLE, MIGRATE_UNMOVABLE, MIGRATE_TYPES },
 	[MIGRATE_RECLAIMABLE] = { MIGRATE_UNMOVABLE,   MIGRATE_MOVABLE,   MIGRATE_TYPES },
+	[MIGRATE_HIGHATOMIC] = { MIGRATE_TYPES }, /* Never used */
 #ifdef CONFIG_CMA
 	[MIGRATE_CMA]         = { MIGRATE_TYPES }, /* Never used */
 #endif
@@ -2795,8 +2795,8 @@ static void reserve_highatomic_pageblock(struct page *page, struct zone *zone,
 
 	/* Yoink! */
 	mt = get_pageblock_migratetype(page);
-	if (!is_migrate_highatomic(mt) && !is_migrate_isolate(mt)
-	    && !is_migrate_cma(mt)) {
+	/* Only reserve normal pageblocks (i.e., they can merge with others) */
+	if (migratetype_is_mergeable(mt)) {
 		zone->nr_reserved_highatomic += pageblock_nr_pages;
 		set_pageblock_migratetype(page, MIGRATE_HIGHATOMIC);
 		move_freepages_block(zone, page, MIGRATE_HIGHATOMIC, NULL);
@@ -3545,8 +3545,11 @@ int __isolate_free_page(struct page *page, unsigned int order)
 		struct page *endpage = page + (1 << order) - 1;
 		for (; page < endpage; page += pageblock_nr_pages) {
 			int mt = get_pageblock_migratetype(page);
-			if (!is_migrate_isolate(mt) && !is_migrate_cma(mt)
-			    && !is_migrate_highatomic(mt))
+			/*
+			 * Only change normal pageblocks (i.e., they can merge
+			 * with others)
+			 */
+			if (migratetype_is_mergeable(mt))
 				set_pageblock_migratetype(page,
 							  MIGRATE_MOVABLE);
 		}
-- 
2.34.1



  reply	other threads:[~2022-01-19 19:06 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-19 19:06 [PATCH v4 0/7] Use pageblock_order for cma and alloc_contig_range alignment Zi Yan
2022-01-19 19:06 ` Zi Yan [this message]
2022-01-24 14:02   ` [PATCH v4 1/7] mm: page_alloc: avoid merging non-fallbackable pageblocks with others Mel Gorman
2022-01-24 16:12     ` Zi Yan
2022-01-24 16:43       ` Mel Gorman
2022-01-19 19:06 ` [PATCH v4 2/7] mm: page_isolation: move has_unmovable_pages() to mm/page_isolation.c Zi Yan
2022-01-25  6:23   ` Oscar Salvador
2022-01-19 19:06 ` [PATCH v4 3/7] mm: page_isolation: check specified range for unmovable pages Zi Yan
2022-01-24  9:55   ` Oscar Salvador
2022-01-24 17:17     ` Zi Yan
2022-01-25 13:19       ` Oscar Salvador
2022-01-25 13:21         ` Oscar Salvador
2022-01-25 16:31           ` Zi Yan
2022-02-02 12:18   ` Oscar Salvador
2022-02-02 12:25     ` David Hildenbrand
2022-02-02 16:25       ` Zi Yan
2022-02-02 16:35       ` Oscar Salvador
2022-01-19 19:06 ` [PATCH v4 4/7] mm: make alloc_contig_range work at pageblock granularity Zi Yan
2022-02-04 13:56   ` Oscar Salvador
2022-02-04 15:19     ` Zi Yan
2022-01-19 19:06 ` [PATCH v4 5/7] mm: cma: use pageblock_order as the single alignment Zi Yan
2022-01-19 19:06 ` [PATCH v4 6/7] drivers: virtio_mem: use pageblock size as the minimum virtio_mem size Zi Yan
2022-01-19 19:06 ` [PATCH v4 7/7] arch: powerpc: adjust fadump alignment to be pageblock aligned Zi Yan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220119190623.1029355-2-zi.yan@sent.com \
    --to=zi.yan@sent.com \
    --cc=david@redhat.com \
    --cc=hch@lst.de \
    --cc=iommu@lists.linux-foundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=m.szyprowski@samsung.com \
    --cc=mgorman@techsingularity.net \
    --cc=mpe@ellerman.id.au \
    --cc=renzhengeek@gmail.com \
    --cc=robin.murphy@arm.com \
    --cc=vbabka@suse.cz \
    --cc=virtualization@lists.linux-foundation.org \
    --cc=ziy@nvidia.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox