diff mbox series

[v4,1/7] iommu: Move iotlb_sync_map out from __iommu_map

Message ID 20210107122909.16317-2-yong.wu@mediatek.com (mailing list archive)
State New, archived
Headers show
Series MediaTek IOMMU improve tlb flush performance in map/unmap | expand

Commit Message

Yong Wu (吴勇) Jan. 7, 2021, 12:29 p.m. UTC 
In the end of __iommu_map, It alway call iotlb_sync_map.

This patch moves iotlb_sync_map out from __iommu_map since it is
unnecessary to call this for each sg segment especially iotlb_sync_map
is flush tlb all currently. Add a little helper _iommu_map for this.

Signed-off-by: Yong Wu <yong.wu@mediatek.com>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
---
 drivers/iommu/iommu.c | 23 ++++++++++++++++++-----
 1 file changed, 18 insertions(+), 5 deletions(-)

Comments

Doug Anderson Feb. 2, 2021, 1:07 a.m. UTC | #1 
Hi,

On Thu, Jan 7, 2021 at 4:31 AM Yong Wu <yong.wu@mediatek.com> wrote:
>
> @@ -2438,18 +2435,31 @@ static int __iommu_map(struct iommu_domain *domain, unsigned long iova,
>         return ret;
>  }
>
> +static int _iommu_map(struct iommu_domain *domain, unsigned long iova,
> +                     phys_addr_t paddr, size_t size, int prot, gfp_t gfp)
> +{
> +       const struct iommu_ops *ops = domain->ops;
> +       int ret;
> +
> +       ret = __iommu_map(domain, iova, paddr, size, prot, GFP_KERNEL);

The above is broken.  Instead of GFP_KERNEL it should be passing "gfp".


> +       if (ret == 0 && ops->iotlb_sync_map)
> +               ops->iotlb_sync_map(domain);
> +
> +       return ret;
> +}
> +
>  int iommu_map(struct iommu_domain *domain, unsigned long iova,
>               phys_addr_t paddr, size_t size, int prot)
>  {
>         might_sleep();
> -       return __iommu_map(domain, iova, paddr, size, prot, GFP_KERNEL);
> +       return _iommu_map(domain, iova, paddr, size, prot, GFP_KERNEL);
>  }
>  EXPORT_SYMBOL_GPL(iommu_map);
>
>  int iommu_map_atomic(struct iommu_domain *domain, unsigned long iova,
>               phys_addr_t paddr, size_t size, int prot)
>  {
> -       return __iommu_map(domain, iova, paddr, size, prot, GFP_ATOMIC);
> +       return _iommu_map(domain, iova, paddr, size, prot, GFP_ATOMIC);

Specifically the above bug means we drop the "GFP_ATOMIC" here.

It means we trigger a warning, like this (on a downstream kernel with
the patch backported):

 BUG: sleeping function called from invalid context at mm/page_alloc.c:4726
 in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 9, name: ksoftirqd/0
 CPU: 0 PID: 9 Comm: ksoftirqd/0 Not tainted 5.4.93-12508-gc10c93e28e39 #1
 Call trace:
  dump_backtrace+0x0/0x154
  show_stack+0x20/0x2c
  dump_stack+0xa0/0xfc
  ___might_sleep+0x11c/0x12c
  __might_sleep+0x50/0x84
  __alloc_pages_nodemask+0xf8/0x2bc
  __arm_lpae_alloc_pages+0x48/0x1b4
  __arm_lpae_map+0x124/0x274
  __arm_lpae_map+0x1cc/0x274
  arm_lpae_map+0x140/0x170
  arm_smmu_map+0x78/0xbc
  __iommu_map+0xd4/0x210
  _iommu_map+0x4c/0x84
  iommu_map_atomic+0x44/0x58
  __iommu_dma_map+0x8c/0xc4
  iommu_dma_map_page+0xac/0xf0

---

A quick (but not very tested) fix at:

https://lore.kernel.org/r/20210201170611.1.I64a7b62579287d668d7c89e105dcedf45d641063@changeid/


-Doug
diff mbox series

Patch

diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index ffeebda8d6de..c304a6a30d42 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -2426,9 +2426,6 @@  static int __iommu_map(struct iommu_domain *domain, unsigned long iova,
 		size -= pgsize;
 	}
 
-	if (ops->iotlb_sync_map)
-		ops->iotlb_sync_map(domain);
-
 	/* unroll mapping in case something went wrong */
 	if (ret)
 		iommu_unmap(domain, orig_iova, orig_size - size);
@@ -2438,18 +2435,31 @@  static int __iommu_map(struct iommu_domain *domain, unsigned long iova,
 	return ret;
 }
 
+static int _iommu_map(struct iommu_domain *domain, unsigned long iova,
+		      phys_addr_t paddr, size_t size, int prot, gfp_t gfp)
+{
+	const struct iommu_ops *ops = domain->ops;
+	int ret;
+
+	ret = __iommu_map(domain, iova, paddr, size, prot, GFP_KERNEL);
+	if (ret == 0 && ops->iotlb_sync_map)
+		ops->iotlb_sync_map(domain);
+
+	return ret;
+}
+
 int iommu_map(struct iommu_domain *domain, unsigned long iova,
 	      phys_addr_t paddr, size_t size, int prot)
 {
 	might_sleep();
-	return __iommu_map(domain, iova, paddr, size, prot, GFP_KERNEL);
+	return _iommu_map(domain, iova, paddr, size, prot, GFP_KERNEL);
 }
 EXPORT_SYMBOL_GPL(iommu_map);
 
 int iommu_map_atomic(struct iommu_domain *domain, unsigned long iova,
 	      phys_addr_t paddr, size_t size, int prot)
 {
-	return __iommu_map(domain, iova, paddr, size, prot, GFP_ATOMIC);
+	return _iommu_map(domain, iova, paddr, size, prot, GFP_ATOMIC);
 }
 EXPORT_SYMBOL_GPL(iommu_map_atomic);
 
@@ -2533,6 +2543,7 @@  static size_t __iommu_map_sg(struct iommu_domain *domain, unsigned long iova,
 			     struct scatterlist *sg, unsigned int nents, int prot,
 			     gfp_t gfp)
 {
+	const struct iommu_ops *ops = domain->ops;
 	size_t len = 0, mapped = 0;
 	phys_addr_t start;
 	unsigned int i = 0;
@@ -2563,6 +2574,8 @@  static size_t __iommu_map_sg(struct iommu_domain *domain, unsigned long iova,
 			sg = sg_next(sg);
 	}
 
+	if (ops->iotlb_sync_map)
+		ops->iotlb_sync_map(domain);
 	return mapped;
 
 out_err: