diff mbox series

[v2] iommu: Streamline iommu_iova_to_phys()

Message ID f564f3f6ff731b898ff7a898919bf871c2c7745a.1626354264.git.robin.murphy@arm.com (mailing list archive)
State New, archived
Headers show
Series [v2] iommu: Streamline iommu_iova_to_phys() | expand

Commit Message

Robin Murphy July 15, 2021, 1:04 p.m. UTC 
If people are going to insist on calling iommu_iova_to_phys()
pointlessly and expecting it to work, we can at least do ourselves a
favour by handling those cases in the core code, rather than repeatedly
across an inconsistent handful of drivers.

Since all the existing drivers implement the internal callback, and any
future ones are likely to want to work with iommu-dma which relies on
iova_to_phys a fair bit, we may as well remove that currently-redundant
check as well and consider it mandatory.

Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>

---

v2: Lowered the priority of the ops check per Baolu's suggestion, only
    even further to the point of non-existence :)
---
 drivers/iommu/amd/io_pgtable.c              | 3 ---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 3 ---
 drivers/iommu/arm/arm-smmu/arm-smmu.c       | 3 ---
 drivers/iommu/iommu.c                       | 5 ++++-
 4 files changed, 4 insertions(+), 10 deletions(-)

Comments

Christoph Hellwig July 15, 2021, 2:07 p.m. UTC | #1 
On Thu, Jul 15, 2021 at 02:04:24PM +0100, Robin Murphy wrote:
> If people are going to insist on calling iommu_iova_to_phys()
> pointlessly and expecting it to work,

Maybe we need to fix that?
Robin Murphy July 15, 2021, 2:16 p.m. UTC | #2 
On 2021-07-15 15:07, Christoph Hellwig wrote:
> On Thu, Jul 15, 2021 at 02:04:24PM +0100, Robin Murphy wrote:
>> If people are going to insist on calling iommu_iova_to_phys()
>> pointlessly and expecting it to work,
> 
> Maybe we need to fix that?

Feel free to try, but we didn't have much luck pushing back on it 
previously, so playing whack-a-mole against netdev now is a game I'm 
personally happy to stay away from ;)

Robin.
Christoph Hellwig July 16, 2021, 6:19 a.m. UTC | #3 
On Thu, Jul 15, 2021 at 03:16:08PM +0100, Robin Murphy wrote:
> On 2021-07-15 15:07, Christoph Hellwig wrote:
> > On Thu, Jul 15, 2021 at 02:04:24PM +0100, Robin Murphy wrote:
> > > If people are going to insist on calling iommu_iova_to_phys()
> > > pointlessly and expecting it to work,
> > 
> > Maybe we need to fix that?
> 
> Feel free to try, but we didn't have much luck pushing back on it
> previously, so playing whack-a-mole against netdev now is a game I'm
> personally happy to stay away from ;)

One thing I've done with symbols I want people to not use it to
unexport them.  But what about vfio?

While we're talking about iommu_iova_to_phys: __iommu_dma_unmap_swiotlb
calls it unconditionally, despite only needed ing the physical address.
Can we optimize that somehow by splitting out the bounce buffering case
out?
Robin Murphy July 16, 2021, 12:01 p.m. UTC | #4 
On 2021-07-16 07:19, Christoph Hellwig wrote:
> On Thu, Jul 15, 2021 at 03:16:08PM +0100, Robin Murphy wrote:
>> On 2021-07-15 15:07, Christoph Hellwig wrote:
>>> On Thu, Jul 15, 2021 at 02:04:24PM +0100, Robin Murphy wrote:
>>>> If people are going to insist on calling iommu_iova_to_phys()
>>>> pointlessly and expecting it to work,
>>>
>>> Maybe we need to fix that?
>>
>> Feel free to try, but we didn't have much luck pushing back on it
>> previously, so playing whack-a-mole against netdev now is a game I'm
>> personally happy to stay away from ;)
> 
> One thing I've done with symbols I want people to not use it to
> unexport them.  But what about vfio?

Yeah, it's not like they shouldn't be calling it at all - I see it as 
primarily intended for use by drivers managing their own domains, but I 
don't entirely disagree with using it on DMA domains either in niche 
cases - it's that they blindly grab the default domain without even 
checking whether DMA mappings are actually translated or not (and thus 
whether they even need to make that call every time they pull a 
descriptor back out of a ringbuffer). IIRC the argument was essentially 
that checking the domain type was an IOMMU API detail that those driver 
shouldn't have to know about and the abstraction should just take care 
of it, despite the fact that they're punching through 2 layers of 
abstraction to even reach that point. And apparently keeping track of 
their own descriptor addresses would be too much work, but expensive 
indirect calls to either return the address they already have or go off 
and do a software table walk with atomic synchronisation and everything 
are fine :/

> While we're talking about iommu_iova_to_phys: __iommu_dma_unmap_swiotlb
> calls it unconditionally, despite only needed ing the physical address.
> Can we optimize that somehow by splitting out the bounce buffering case
> out?

Indeed, as I think I mentioned recently on another thread, all the 
bounce-buffering stuff is fairly ugly because it's basically the old 
intel-iommu code dropped in with as few changes as possible for ease of 
review, since Tom was no longer able to spend time refining it, and 
nobody else has got round to cleaning it up yet either. In fact the 
whole flow through iommu_dma_unmap_page() flow might be the worst-hit - 
reusing the iommu_dma_sync op made perfect sense when it was just cache 
maintenance, but now means that at worst we do iova_to_phys *twice* plus 
a pointless swiotlb_sync :(

Robin.
Joerg Roedel July 26, 2021, 11:38 a.m. UTC | #5 
On Thu, Jul 15, 2021 at 02:04:24PM +0100, Robin Murphy wrote:
> ---
> 
> v2: Lowered the priority of the ops check per Baolu's suggestion, only
>     even further to the point of non-existence :)

Err, missed this newer version. Replaced the previous patch with this
one.
diff mbox series

Patch

diff --git a/drivers/iommu/amd/io_pgtable.c b/drivers/iommu/amd/io_pgtable.c
index bb0ee5c9fde7..182c93a43efd 100644
--- a/drivers/iommu/amd/io_pgtable.c
+++ b/drivers/iommu/amd/io_pgtable.c
@@ -493,9 +493,6 @@  static phys_addr_t iommu_v1_iova_to_phys(struct io_pgtable_ops *ops, unsigned lo
 	unsigned long offset_mask, pte_pgsize;
 	u64 *pte, __pte;
 
-	if (pgtable->mode == PAGE_MODE_NONE)
-		return iova;
-
 	pte = fetch_pte(pgtable, iova, &pte_pgsize);
 
 	if (!pte || !IOMMU_PTE_PRESENT(*pte))
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index 3e87a9cf6da3..c9fdd0d097be 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -2481,9 +2481,6 @@  arm_smmu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
 {
 	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
 
-	if (domain->type == IOMMU_DOMAIN_IDENTITY)
-		return iova;
-
 	if (!ops)
 		return 0;
 
diff --git a/drivers/iommu/arm/arm-smmu/arm-smmu.c b/drivers/iommu/arm/arm-smmu/arm-smmu.c
index 0f181f76c31b..0d04eafa3fdb 100644
--- a/drivers/iommu/arm/arm-smmu/arm-smmu.c
+++ b/drivers/iommu/arm/arm-smmu/arm-smmu.c
@@ -1317,9 +1317,6 @@  static phys_addr_t arm_smmu_iova_to_phys(struct iommu_domain *domain,
 	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
 	struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
 
-	if (domain->type == IOMMU_DOMAIN_IDENTITY)
-		return iova;
-
 	if (!ops)
 		return 0;
 
diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index 43a2041d9629..d4ba324fb0bc 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -2371,7 +2371,10 @@  EXPORT_SYMBOL_GPL(iommu_detach_group);
 
 phys_addr_t iommu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
 {
-	if (unlikely(domain->ops->iova_to_phys == NULL))
+	if (domain->type == IOMMU_DOMAIN_IDENTITY)
+		return iova;
+
+	if (domain->type == IOMMU_DOMAIN_BLOCKED)
 		return 0;
 
 	return domain->ops->iova_to_phys(domain, iova);