* [PATCH 1/4] mm/compaction: use "spinlock_t *" to record held lock in compact [un]lock functions
[not found] <20230719113001.2023703-1-shikemeng@huaweicloud.com>
@ 2023-07-19 11:29 ` Kemeng Shi
2023-07-19 11:29 ` [PATCH 2/4] mm/compaction: use "spinlock_t *" to record held lock in isolate_migratepages_block Kemeng Shi
2023-07-19 11:30 ` [PATCH 4/4] mm/compaction: add compact_unlock_irqrestore to remove repeat code Kemeng Shi
2 siblings, 0 replies; 4+ messages in thread
From: Kemeng Shi @ 2023-07-19 11:29 UTC (permalink / raw)
To: akpm, linux-mm, linux-kernel; +Cc: shikemeng
Make compact_lock_irqsave and compact_unlock_should_abort use
"spinlock_t *" to record held lock.
This is a preparation to use compact_unlock_should_abort in
isolate_migratepages_block to remove repeat code.
Signed-off-by: Kemeng Shi <shikemeng@huaweicloud.com>
---
mm/compaction.c | 23 +++++++++++------------
1 file changed, 11 insertions(+), 12 deletions(-)
diff --git a/mm/compaction.c b/mm/compaction.c
index 9641e2131901..dfef14d3ef78 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -523,22 +523,22 @@ static bool test_and_set_skip(struct compact_control *cc, struct page *page)
* abort when the current block is finished regardless of success rate.
* Sync compaction acquires the lock.
*
- * Always returns true which makes it easier to track lock state in callers.
+ * Always returns lock which makes it easier to track lock state in callers.
*/
-static bool compact_lock_irqsave(spinlock_t *lock, unsigned long *flags,
+static spinlock_t *compact_lock_irqsave(spinlock_t *lock, unsigned long *flags,
struct compact_control *cc)
__acquires(lock)
{
/* Track if the lock is contended in async mode */
if (cc->mode == MIGRATE_ASYNC && !cc->contended) {
if (spin_trylock_irqsave(lock, *flags))
- return true;
+ return lock;
cc->contended = true;
}
spin_lock_irqsave(lock, *flags);
- return true;
+ return lock;
}
/*
@@ -553,12 +553,12 @@ static bool compact_lock_irqsave(spinlock_t *lock, unsigned long *flags,
* Returns true if compaction should abort due to fatal signal pending.
* Returns false when compaction can continue.
*/
-static bool compact_unlock_should_abort(spinlock_t *lock,
- unsigned long flags, bool *locked, struct compact_control *cc)
+static bool compact_unlock_should_abort(spinlock_t **locked,
+ unsigned long flags, struct compact_control *cc)
{
if (*locked) {
- spin_unlock_irqrestore(lock, flags);
- *locked = false;
+ spin_unlock_irqrestore(*locked, flags);
+ *locked = NULL;
}
if (fatal_signal_pending(current)) {
@@ -586,7 +586,7 @@ static unsigned long isolate_freepages_block(struct compact_control *cc,
int nr_scanned = 0, total_isolated = 0;
struct page *cursor;
unsigned long flags = 0;
- bool locked = false;
+ spinlock_t *locked = NULL;
unsigned long blockpfn = *start_pfn;
unsigned int order;
@@ -607,8 +607,7 @@ static unsigned long isolate_freepages_block(struct compact_control *cc,
* pending.
*/
if (!(blockpfn % COMPACT_CLUSTER_MAX)
- && compact_unlock_should_abort(&cc->zone->lock, flags,
- &locked, cc))
+ && compact_unlock_should_abort(&locked, flags, cc))
break;
nr_scanned++;
@@ -673,7 +672,7 @@ static unsigned long isolate_freepages_block(struct compact_control *cc,
}
if (locked)
- spin_unlock_irqrestore(&cc->zone->lock, flags);
+ spin_unlock_irqrestore(locked, flags);
/*
* There is a tiny chance that we have read bogus compound_order(),
--
2.30.0
^ permalink raw reply related [flat|nested] 4+ messages in thread* [PATCH 2/4] mm/compaction: use "spinlock_t *" to record held lock in isolate_migratepages_block
[not found] <20230719113001.2023703-1-shikemeng@huaweicloud.com>
2023-07-19 11:29 ` [PATCH 1/4] mm/compaction: use "spinlock_t *" to record held lock in compact [un]lock functions Kemeng Shi
@ 2023-07-19 11:29 ` Kemeng Shi
2023-07-19 11:30 ` [PATCH 4/4] mm/compaction: add compact_unlock_irqrestore to remove repeat code Kemeng Shi
2 siblings, 0 replies; 4+ messages in thread
From: Kemeng Shi @ 2023-07-19 11:29 UTC (permalink / raw)
To: akpm, linux-mm, linux-kernel; +Cc: shikemeng
Use "spinlock_t *" instead of "struct lruvec *" to record held lock in
isolate_migratepages_block.
This is a preparation to use compact_unlock_should_abort in
isolate_migratepages_block to remove repeat code.
Signed-off-by: Kemeng Shi <shikemeng@huaweicloud.com>
---
mm/compaction.c | 21 ++++++++++-----------
1 file changed, 10 insertions(+), 11 deletions(-)
diff --git a/mm/compaction.c b/mm/compaction.c
index dfef14d3ef78..638146a49e89 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -840,7 +840,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
unsigned long nr_scanned = 0, nr_isolated = 0;
struct lruvec *lruvec;
unsigned long flags = 0;
- struct lruvec *locked = NULL;
+ spinlock_t *locked = NULL;
struct folio *folio = NULL;
struct page *page = NULL, *valid_page = NULL;
struct address_space *mapping;
@@ -911,7 +911,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
*/
if (!(low_pfn % COMPACT_CLUSTER_MAX)) {
if (locked) {
- unlock_page_lruvec_irqrestore(locked, flags);
+ spin_unlock_irqrestore(locked, flags);
locked = NULL;
}
@@ -946,7 +946,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
if (PageHuge(page) && cc->alloc_contig) {
if (locked) {
- unlock_page_lruvec_irqrestore(locked, flags);
+ spin_unlock_irqrestore(locked, flags);
locked = NULL;
}
@@ -1035,7 +1035,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
if (unlikely(__PageMovable(page)) &&
!PageIsolated(page)) {
if (locked) {
- unlock_page_lruvec_irqrestore(locked, flags);
+ spin_unlock_irqrestore(locked, flags);
locked = NULL;
}
@@ -1120,12 +1120,11 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
lruvec = folio_lruvec(folio);
/* If we already hold the lock, we can skip some rechecking */
- if (lruvec != locked) {
+ if (&lruvec->lru_lock != locked) {
if (locked)
- unlock_page_lruvec_irqrestore(locked, flags);
+ spin_unlock_irqrestore(locked, flags);
- compact_lock_irqsave(&lruvec->lru_lock, &flags, cc);
- locked = lruvec;
+ locked = compact_lock_irqsave(&lruvec->lru_lock, &flags, cc);
lruvec_memcg_debug(lruvec, folio);
@@ -1188,7 +1187,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
isolate_fail_put:
/* Avoid potential deadlock in freeing page under lru_lock */
if (locked) {
- unlock_page_lruvec_irqrestore(locked, flags);
+ spin_unlock_irqrestore(locked, flags);
locked = NULL;
}
folio_put(folio);
@@ -1204,7 +1203,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
*/
if (nr_isolated) {
if (locked) {
- unlock_page_lruvec_irqrestore(locked, flags);
+ spin_unlock_irqrestore(locked, flags);
locked = NULL;
}
putback_movable_pages(&cc->migratepages);
@@ -1236,7 +1235,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
isolate_abort:
if (locked)
- unlock_page_lruvec_irqrestore(locked, flags);
+ spin_unlock_irqrestore(locked, flags);
if (folio) {
folio_set_lru(folio);
folio_put(folio);
--
2.30.0
^ permalink raw reply related [flat|nested] 4+ messages in thread* [PATCH 4/4] mm/compaction: add compact_unlock_irqrestore to remove repeat code
[not found] <20230719113001.2023703-1-shikemeng@huaweicloud.com>
2023-07-19 11:29 ` [PATCH 1/4] mm/compaction: use "spinlock_t *" to record held lock in compact [un]lock functions Kemeng Shi
2023-07-19 11:29 ` [PATCH 2/4] mm/compaction: use "spinlock_t *" to record held lock in isolate_migratepages_block Kemeng Shi
@ 2023-07-19 11:30 ` Kemeng Shi
2 siblings, 0 replies; 4+ messages in thread
From: Kemeng Shi @ 2023-07-19 11:30 UTC (permalink / raw)
To: akpm, linux-mm, linux-kernel; +Cc: shikemeng
Add compact_unlock_irqrestore to remove repeat code. This also make
compact lock functions sereis complete as we can call
compact_lock_irqsave/compact_unlock_irqrestore in pair.
Signed-off-by: Kemeng Shi <shikemeng@huaweicloud.com>
---
mm/compaction.c | 43 ++++++++++++++++---------------------------
1 file changed, 16 insertions(+), 27 deletions(-)
diff --git a/mm/compaction.c b/mm/compaction.c
index c1dc821ac6e1..eb1d3d9a422c 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -541,6 +541,14 @@ static spinlock_t *compact_lock_irqsave(spinlock_t *lock, unsigned long *flags,
return lock;
}
+static inline void compact_unlock_irqrestore(spinlock_t **locked, unsigned long flags)
+{
+ if (*locked) {
+ spin_unlock_irqrestore(*locked, flags);
+ *locked = NULL;
+ }
+}
+
/*
* Compaction requires the taking of some coarse locks that are potentially
* very heavily contended. The lock should be periodically unlocked to avoid
@@ -556,10 +564,7 @@ static spinlock_t *compact_lock_irqsave(spinlock_t *lock, unsigned long *flags,
static bool compact_unlock_should_abort(spinlock_t **locked,
unsigned long flags, struct compact_control *cc)
{
- if (*locked) {
- spin_unlock_irqrestore(*locked, flags);
- *locked = NULL;
- }
+ compact_unlock_irqrestore(locked, flags);
if (fatal_signal_pending(current)) {
cc->contended = true;
@@ -671,8 +676,7 @@ static unsigned long isolate_freepages_block(struct compact_control *cc,
}
- if (locked)
- spin_unlock_irqrestore(locked, flags);
+ compact_unlock_irqrestore(&locked, flags);
/*
* There is a tiny chance that we have read bogus compound_order(),
@@ -935,10 +939,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
}
if (PageHuge(page) && cc->alloc_contig) {
- if (locked) {
- spin_unlock_irqrestore(locked, flags);
- locked = NULL;
- }
+ compact_unlock_irqrestore(&locked, flags);
ret = isolate_or_dissolve_huge_page(page, &cc->migratepages);
@@ -1024,10 +1025,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
*/
if (unlikely(__PageMovable(page)) &&
!PageIsolated(page)) {
- if (locked) {
- spin_unlock_irqrestore(locked, flags);
- locked = NULL;
- }
+ compact_unlock_irqrestore(&locked, flags);
if (isolate_movable_page(page, mode)) {
folio = page_folio(page);
@@ -1111,9 +1109,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
/* If we already hold the lock, we can skip some rechecking */
if (&lruvec->lru_lock != locked) {
- if (locked)
- spin_unlock_irqrestore(locked, flags);
-
+ compact_unlock_irqrestore(&locked, flags);
locked = compact_lock_irqsave(&lruvec->lru_lock, &flags, cc);
lruvec_memcg_debug(lruvec, folio);
@@ -1176,10 +1172,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
isolate_fail_put:
/* Avoid potential deadlock in freeing page under lru_lock */
- if (locked) {
- spin_unlock_irqrestore(locked, flags);
- locked = NULL;
- }
+ compact_unlock_irqrestore(&locked, flags);
folio_put(folio);
isolate_fail:
@@ -1192,10 +1185,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
* page anyway.
*/
if (nr_isolated) {
- if (locked) {
- spin_unlock_irqrestore(locked, flags);
- locked = NULL;
- }
+ compact_unlock_irqrestore(&locked, flags);
putback_movable_pages(&cc->migratepages);
cc->nr_migratepages = 0;
nr_isolated = 0;
@@ -1224,8 +1214,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
folio = NULL;
isolate_abort:
- if (locked)
- spin_unlock_irqrestore(locked, flags);
+ compact_unlock_irqrestore(&locked, flags);
if (folio) {
folio_set_lru(folio);
folio_put(folio);
--
2.30.0
^ permalink raw reply related [flat|nested] 4+ messages in thread
* [PATCH 4/4] mm/compaction: add compact_unlock_irqrestore to remove repeat code
[not found] <20230725180456.2146626-1-shikemeng@huaweicloud.com>
@ 2023-07-25 18:04 ` Kemeng Shi
0 siblings, 0 replies; 4+ messages in thread
From: Kemeng Shi @ 2023-07-25 18:04 UTC (permalink / raw)
To: akpm, linux-mm, linux-kernel; +Cc: willy, baolin.wang, david, shikemeng
Add compact_unlock_irqrestore to remove repeat code. This also make
compact lock functions sereis complete as we can call
compact_lock_irqsave/compact_unlock_irqrestore in pair.
Signed-off-by: Kemeng Shi <shikemeng@huaweicloud.com>
---
mm/compaction.c | 43 ++++++++++++++++---------------------------
1 file changed, 16 insertions(+), 27 deletions(-)
diff --git a/mm/compaction.c b/mm/compaction.c
index c1dc821ac6e1..eb1d3d9a422c 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -541,6 +541,14 @@ static spinlock_t *compact_lock_irqsave(spinlock_t *lock, unsigned long *flags,
return lock;
}
+static inline void compact_unlock_irqrestore(spinlock_t **locked, unsigned long flags)
+{
+ if (*locked) {
+ spin_unlock_irqrestore(*locked, flags);
+ *locked = NULL;
+ }
+}
+
/*
* Compaction requires the taking of some coarse locks that are potentially
* very heavily contended. The lock should be periodically unlocked to avoid
@@ -556,10 +564,7 @@ static spinlock_t *compact_lock_irqsave(spinlock_t *lock, unsigned long *flags,
static bool compact_unlock_should_abort(spinlock_t **locked,
unsigned long flags, struct compact_control *cc)
{
- if (*locked) {
- spin_unlock_irqrestore(*locked, flags);
- *locked = NULL;
- }
+ compact_unlock_irqrestore(locked, flags);
if (fatal_signal_pending(current)) {
cc->contended = true;
@@ -671,8 +676,7 @@ static unsigned long isolate_freepages_block(struct compact_control *cc,
}
- if (locked)
- spin_unlock_irqrestore(locked, flags);
+ compact_unlock_irqrestore(&locked, flags);
/*
* There is a tiny chance that we have read bogus compound_order(),
@@ -935,10 +939,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
}
if (PageHuge(page) && cc->alloc_contig) {
- if (locked) {
- spin_unlock_irqrestore(locked, flags);
- locked = NULL;
- }
+ compact_unlock_irqrestore(&locked, flags);
ret = isolate_or_dissolve_huge_page(page, &cc->migratepages);
@@ -1024,10 +1025,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
*/
if (unlikely(__PageMovable(page)) &&
!PageIsolated(page)) {
- if (locked) {
- spin_unlock_irqrestore(locked, flags);
- locked = NULL;
- }
+ compact_unlock_irqrestore(&locked, flags);
if (isolate_movable_page(page, mode)) {
folio = page_folio(page);
@@ -1111,9 +1109,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
/* If we already hold the lock, we can skip some rechecking */
if (&lruvec->lru_lock != locked) {
- if (locked)
- spin_unlock_irqrestore(locked, flags);
-
+ compact_unlock_irqrestore(&locked, flags);
locked = compact_lock_irqsave(&lruvec->lru_lock, &flags, cc);
lruvec_memcg_debug(lruvec, folio);
@@ -1176,10 +1172,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
isolate_fail_put:
/* Avoid potential deadlock in freeing page under lru_lock */
- if (locked) {
- spin_unlock_irqrestore(locked, flags);
- locked = NULL;
- }
+ compact_unlock_irqrestore(&locked, flags);
folio_put(folio);
isolate_fail:
@@ -1192,10 +1185,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
* page anyway.
*/
if (nr_isolated) {
- if (locked) {
- spin_unlock_irqrestore(locked, flags);
- locked = NULL;
- }
+ compact_unlock_irqrestore(&locked, flags);
putback_movable_pages(&cc->migratepages);
cc->nr_migratepages = 0;
nr_isolated = 0;
@@ -1224,8 +1214,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
folio = NULL;
isolate_abort:
- if (locked)
- spin_unlock_irqrestore(locked, flags);
+ compact_unlock_irqrestore(&locked, flags);
if (folio) {
folio_set_lru(folio);
folio_put(folio);
--
2.30.0
^ permalink raw reply related [flat|nested] 4+ messages in thread