mm/mmu_gather: replace IPI with synchronize_rcu() when batch allocation fails
PATCH v3 1/1 mm/mmu_gather: replace IPI with synchronize_rcu() when batch allocation fails
From: Lance Yang <[email protected]>
When freeing page tables, we try to batch them. If batch allocation fails (GFP_NOWAIT), __tlb_remove_table_one() immediately frees the one without batching.
On !CONFIG_PT_RECLAIM, the fallback sends an IPI to all CPUs via tlb_remove_table_sync_one(). It disrupts all CPUs even when only a single process is unmapping memory. IPI broadcast was reported to hurt RT workloads[1].
tlb_remove_table_sync_one() synchronizes with lockless page-table walkers (e.g. GUP-fast) that rely on IRQ disabling. These walkers use local_irq_disable(), which is also an RCU read-side critical section.
This patch introduces tlb_remove_table_sync_rcu() which uses RCU grace period (synchronize_rcu()) instead of IPI broadcast. This provides the same guarantee as IPI but without disrupting all CPUs. Since batch allocation already failed, we are in a slow path where sleeping is acceptable - we are in process context (unmap_region, exit_mmap) with only mmap_lock held.
tlb_remove_table_sync_one() is retained for other callers (e.g., khugepaged after pmdp_collapse_flush(), tlb_finish_mmu() when tlb->fully_unshared_tables) that are not slow paths. Converting those may require different approaches such as targeted IPIs.
[1] https://lore.kernel.org/linux-mm/[email protected]/
---
v2 -> v3:
- Remove explicit might_sleep() as synchronize_rcu() already has it (per Peter)
- Add changelog explanation for why tlb_remove_table_sync_one() is retained (per Peter)
- Collect Acked-by from David and Peter, thanks!
- https://lore.kernel.org/linux-mm/[email protected]/
v1 -> v2:
- Wrap synchronize_rcu() in tlb_remove_table_sync_rcu() with proper kerneldoc (per David)
- Add might_sleep() to make sleeping constraint explicit (per Dave)
- Clarify this is for synchronization, not memory freeing (per Dave)
- https://lore.kernel.org/linux-mm/[email protected]/
include/asm-generic/tlb.h | 4 ++++ mm/mmu_gather.c | 21 ++++++++++++++++++++- 2 files changed, 24 insertions(+), 1 deletion(-)
include/asm-generic/tlb.h +4 -0
@@ -251,6 +251,8 @@ static inline void tlb_remove_table(struct mmu_gather *tlb, void *table)void tlb_remove_table_sync_one (void ); +void tlb_remove_table_sync_rcu (void ); +#else #ifdef tlb_needs_table_invalidate
@@ -259,6 +261,8 @@ void tlb_remove_table_sync_one(void);static inline void tlb_remove_table_sync_one (void ) { } +static inline void tlb_remove_table_sync_rcu (void ) { } +#endif /* CONFIG_MMU_GATHER_RCU_TABLE_FREE */
mm/mmu_gather.c +20 -1
@@ -296,6 +296,25 @@ static void tlb_remove_table_free(struct mmu_table_batch *batch) call_rcu(&batch->rcu, tlb_remove_table_rcu); } +/** +* tlb_remove_table_sync_rcu - synchronize with software page-table walkers +* +* Like tlb_remove_table_sync_one() but uses RCU grace period instead of IPI +* broadcast. Use in slow paths where sleeping is acceptable. +* +* Software/Lockless page-table walkers use local_irq_disable(), which is also +* an RCU read-side critical section. synchronize_rcu() waits for all such +* sections, providing the same guarantee as tlb_remove_table_sync_one() but +* without disrupting all CPUs with IPIs. +* +* Do not use for freeing memory. Use RCU callbacks instead to avoid latency +* spikes. +*/ +void tlb_remove_table_sync_rcu (void ) +{ +synchronize_rcu (); +} +#else /* !CONFIG_MMU_GATHER_RCU_TABLE_FREE */ static void tlb_remove_table_free (struct mmu_table_batch *batch)
@@ -339,7 +358,7 @@ static inline void __tlb_remove_table_one(void *table)#else static inline void __tlb_remove_table_one (void *table) { -tlb_remove_table_sync_one (); +tlb_remove_table_sync_rcu ();__tlb_remove_table (table); }#endif /* CONFIG_PT_RECLAIM */
Signature · 1 line
2.49.0