From: Eric Dumazet Date: Wed, 28 Aug 2019 06:13:38 -0700 Subject: iommu/iova: Avoid false sharing on fq_timer_on Git-commit: 0d87308cca2c124f9bce02383f1d9632c9be89c4 Patch-mainline: v5.4-rc1 References: bsc#1151662 bsc#1151671 In commit 14bd9a607f90 ("iommu/iova: Separate atomic variables to improve performance") Jinyu Qi identified that the atomic_cmpxchg() in queue_iova() was causing a performance loss and moved critical fields so that the false sharing would not impact them. However, avoiding the false sharing in the first place seems easy. We should attempt the atomic_cmpxchg() no more than 100 times per second. Adding an atomic_read() will keep the cache line mostly shared. This false sharing came with commit 9a005a800ae8 ("iommu/iova: Add flush timer"). Signed-off-by: Eric Dumazet Fixes: 9a005a800ae8 ('iommu/iova: Add flush timer') Cc: Jinyu Qi Cc: Joerg Roedel Acked-by: Robin Murphy Signed-off-by: Joerg Roedel --- drivers/iommu/iova.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) --- a/drivers/iommu/iova.c +++ b/drivers/iommu/iova.c @@ -579,7 +579,9 @@ void queue_iova(struct iova_domain *iova spin_unlock_irqrestore(&fq->lock, flags); - if (atomic_cmpxchg(&iovad->fq_timer_on, 0, 1) == 0) + /* Avoid false sharing as much as possible. */ + if (!atomic_read(&iovad->fq_timer_on) && + !atomic_cmpxchg(&iovad->fq_timer_on, 0, 1)) mod_timer(&iovad->fq_timer, jiffies + msecs_to_jiffies(IOVA_FQ_TIMEOUT));