[PATCH v2] mm/page_alloc: Add lockdep assertion for pageblock type change

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Since the migratetype hygiene patches [0], the locking here is
a bit more formalised.

For other stuff, it's pretty obvious that it would be protected by the
zone lock. But it didn't seem totally self-evident that it should
protect the pageblock type. So it seems particularly helpful to have it
written in the code.

[0] https://lore.kernel.org/lkml/20240320180429.678181-3-hannes@xxxxxxxxxxx/T/

Signed-off-by: Brendan Jackman <jackmanb@xxxxxxxxxx>
---
Note Vlastimil Acked the first version, I did not carry his ack since
I've changed the diff. Not sure if I'm being overly pedantic there?

Changes in v2:
- Fixed missing in_mem_hotplug() setup for !CONFIG_MEMORY_HOTPLUG
- Expanded commit message to include a bit more rationale
- Link to v1: https://lore.kernel.org/r/20250227-pageblock-lockdep-v1-1-3701efb331bb@xxxxxxxxxx
---
 include/linux/memory_hotplug.h | 5 +++++
 mm/memory_hotplug.c            | 5 +++++
 mm/page_alloc.c                | 4 ++++
 3 files changed, 14 insertions(+)

diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h
index eaac5ae8c05c8ee2f2868d5bc1b04d1f68235b3f..508a1d074527a3349c1c9789ecf35d5ab08a5ba6 100644
--- a/include/linux/memory_hotplug.h
+++ b/include/linux/memory_hotplug.h
@@ -175,6 +175,7 @@ void put_online_mems(void);
 
 void mem_hotplug_begin(void);
 void mem_hotplug_done(void);
+bool in_mem_hotplug(void);
 
 /* See kswapd_is_running() */
 static inline void pgdat_kswapd_lock(pg_data_t *pgdat)
@@ -223,6 +224,10 @@ static inline void put_online_mems(void) {}
 
 static inline void mem_hotplug_begin(void) {}
 static inline void mem_hotplug_done(void) {}
+static inline bool in_mem_hotplug(void)
+{
+	return false;
+}
 
 static inline bool movable_node_is_enabled(void)
 {
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index e3655f07dd6e33efb3e811cab07f240649487441..c44c2765b21f1e9e4b2d68abda9ac161fb2869ec 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -267,6 +267,11 @@ void mem_hotplug_done(void)
 	cpus_read_unlock();
 }
 
+bool in_mem_hotplug(void)
+{
+	return percpu_is_write_locked(&mem_hotplug_lock);
+}
+
 u64 max_mem_size = U64_MAX;
 
 /* add this memory to iomem resource */
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 579789600a3c7bfb7b0d847d51af702a9d4b139a..1ed21179676d05c66f77f9dbebf88e36bbe402e9 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -417,6 +417,10 @@ void set_pfnblock_flags_mask(struct page *page, unsigned long flags,
 
 void set_pageblock_migratetype(struct page *page, int migratetype)
 {
+	lockdep_assert_once(system_state == SYSTEM_BOOTING ||
+		in_mem_hotplug() ||
+		lockdep_is_held(&page_zone(page)->lock));
+
 	if (unlikely(page_group_by_mobility_disabled &&
 		     migratetype < MIGRATE_PCPTYPES))
 		migratetype = MIGRATE_UNMOVABLE;

---
base-commit: 83c34e70d8dd67525c2547e8c5ee1a4bf37ac06b
change-id: 20250227-pageblock-lockdep-9628c48d7e08

Best regards,
-- 
Brendan Jackman <jackmanb@xxxxxxxxxx>





[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux