[PATCH] iommu/rockchip: Remove redundant DMA syncs
From: Robin Murphy <robin.murphy@arm.com>
Date: 2021-05-21 13:52:01
Also in:
linux-iommu, linux-rockchip
Subsystem:
arm/rockchip soc support, iommu subsystem, the rest · Maintainers:
Heiko Stuebner, Joerg Roedel, Will Deacon, Linus Torvalds
When we allocate new pagetable pages, we don't modify them between the initial dma_map_single() call and the dma_sync_single_for_device() call via rk_iommu_flush(), so the latter is entirely pointless. Signed-off-by: Robin Murphy <robin.murphy@arm.com> --- drivers/iommu/rockchip-iommu.c | 3 --- 1 file changed, 3 deletions(-)
diff --git a/drivers/iommu/rockchip-iommu.c b/drivers/iommu/rockchip-iommu.c
index 7a2932772fdf..e5116d6a6260 100644
--- a/drivers/iommu/rockchip-iommu.c
+++ b/drivers/iommu/rockchip-iommu.c@@ -682,7 +682,6 @@ static u32 *rk_dte_get_page_table(struct rk_iommu_domain *rk_domain, dte = rk_mk_dte(pt_dma); *dte_addr = dte; - rk_table_flush(rk_domain, pt_dma, NUM_PT_ENTRIES); rk_table_flush(rk_domain, rk_domain->dt_dma + dte_index * sizeof(u32), 1); done:
@@ -1004,8 +1003,6 @@ static struct iommu_domain *rk_iommu_domain_alloc(unsigned type) goto err_free_dt; } - rk_table_flush(rk_domain, rk_domain->dt_dma, NUM_DT_ENTRIES); - spin_lock_init(&rk_domain->iommus_lock); spin_lock_init(&rk_domain->dt_lock); INIT_LIST_HEAD(&rk_domain->iommus);
--
2.21.0.dirty
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel