Domain is getting created more than once during asynchronous multiple
display heads(devices) probe. All the display heads share same SID and
are expected to be in same domain. As iommu_alloc_default_domain() call
is not protected, the group->default_domain and group->domain are ending
up with different domains and leading to subsequent IOMMU faults.
Fix this by protecting iommu_alloc_default_domain() call with group->mutex.
Signed-off-by: Ashish Mhetre <redacted>
---
drivers/iommu/iommu.c | 2 ++
1 file changed, 2 insertions(+)
From: Krishna Reddy <vdumpa@nvidia.com>
iommu_group is getting created more than once during asynchronous multiple
display heads(devices) probe on Tegra194 SoC. All the display heads share
same SID and are expected to be in same iommu_group.
As arm_smmu_device_group() is not protecting group creation across devices,
it is leading to multiple groups creation across devices with same SID and
subsequent IOMMU faults.
Fix this by protecting group creation with smmu->stream_map_mutex.
Signed-off-by: Krishna Reddy <vdumpa@nvidia.com>
---
drivers/iommu/arm/arm-smmu/arm-smmu.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
From: Will Deacon <will@kernel.org> Date: 2021-06-11 10:45:35
On Thu, Jun 10, 2021 at 09:46:53AM +0530, Ashish Mhetre wrote:
Domain is getting created more than once during asynchronous multiple
display heads(devices) probe. All the display heads share same SID and
are expected to be in same domain. As iommu_alloc_default_domain() call
is not protected, the group->default_domain and group->domain are ending
up with different domains and leading to subsequent IOMMU faults.
Fix this by protecting iommu_alloc_default_domain() call with group->mutex.
Can you provide some more information about exactly what the h/w
configuration is, and the callstack which exhibits the race, please?
@@ -273,7 +273,9 @@ int iommu_probe_device(struct device *dev)*supportdefaultdomains,sothereturnvalueisnotyet*checked.*/+mutex_lock(&group->mutex);iommu_alloc_default_domain(group,dev);+mutex_unlock(&group->mutex);
It feels wrong to serialise this for everybody just to cater for systems
with aliasing SIDs between devices.
Will
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
From: Robin Murphy <robin.murphy@arm.com> Date: 2021-06-11 12:49:25
On 2021-06-11 11:45, Will Deacon wrote:
On Thu, Jun 10, 2021 at 09:46:53AM +0530, Ashish Mhetre wrote:
quoted
Domain is getting created more than once during asynchronous multiple
display heads(devices) probe. All the display heads share same SID and
are expected to be in same domain. As iommu_alloc_default_domain() call
is not protected, the group->default_domain and group->domain are ending
up with different domains and leading to subsequent IOMMU faults.
Fix this by protecting iommu_alloc_default_domain() call with group->mutex.
Can you provide some more information about exactly what the h/w
configuration is, and the callstack which exhibits the race, please?
It'll be basically the same as the issue reported long ago with PCI
groups in the absence of ACS not being constructed correctly. Triggering
the iommu_probe_device() replay in of_iommu_configure() off the back of
driver probe is way too late and allows calls to happen in the wrong
order, or indeed race in parallel as here. Fixing that is still on my
radar, but will not be simple, and will probably go hand-in-hand with
phasing out the bus ops (for the multiple-driver-coexistence problem).
@@ -273,7 +273,9 @@ int iommu_probe_device(struct device *dev)*supportdefaultdomains,sothereturnvalueisnotyet*checked.*/+mutex_lock(&group->mutex);iommu_alloc_default_domain(group,dev);+mutex_unlock(&group->mutex);
It feels wrong to serialise this for everybody just to cater for systems
with aliasing SIDs between devices.
If two or more devices are racing at this point then they're already
going to be serialised by at least iommu_group_add_device(), so I doubt
there would be much impact - only the first device through here will
hold the mutex for any appreciable length of time. Every other path
which modifies group->domain does so with the mutex held (note the
"expected" default domain allocation flow in bus_iommu_probe() in
particular), so not holding it here does seem like a straightforward
oversight.
Robin.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
It feels wrong to serialise this for everybody just to cater for systems with
aliasing SIDs between devices.
Serialization is limited to devices in the same group. Unless devices share SID, they wouldn't be in same group.
Can you provide some more information about exactly what the h/w
configuration is, and the callstack which exhibits the race, please?
The failure is an after effect and is a page fault. Don't have a failure call stack here. Ashish has traced it through print messages and he can provide them.
From the prints messages, The following was observed in page fault case:
Device1: iommu_probe_device() --> iommu_alloc_default_domain() --> iommu_group_alloc_default_domain() --> __iommu_attach_device(group->default_domain)
Device2: iommu_probe_device() --> iommu_alloc_default_domain() --> iommu_group_alloc_default_domain() --> __iommu_attach_device(group->default_domain)
Both devices(with same SID) are entering into iommu_group_alloc_default_domain() function and each one getting attached to a different group->default_domain
as the second one overwrites group->default_domain after the first one attaches to group->default_domain it has created.
SMMU would be setup to use first domain for the context page table. Whereas all the dma map/unamp requests from second device would
be performed on a domain that is not used by SMMU for context translations and IOVA (not mapped in first domain) accesses from second device lead to page faults.
-KR
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
External email: Use caution opening links or attachments
On 2021-06-11 11:45, Will Deacon wrote:
quoted
On Thu, Jun 10, 2021 at 09:46:53AM +0530, Ashish Mhetre wrote:
quoted
Domain is getting created more than once during asynchronous multiple
display heads(devices) probe. All the display heads share same SID and
are expected to be in same domain. As iommu_alloc_default_domain() call
is not protected, the group->default_domain and group->domain are ending
up with different domains and leading to subsequent IOMMU faults.
Fix this by protecting iommu_alloc_default_domain() call with
group->mutex.
Can you provide some more information about exactly what the h/w
configuration is, and the callstack which exhibits the race, please?
It'll be basically the same as the issue reported long ago with PCI
groups in the absence of ACS not being constructed correctly. Triggering
the iommu_probe_device() replay in of_iommu_configure() off the back of
driver probe is way too late and allows calls to happen in the wrong
order, or indeed race in parallel as here. Fixing that is still on my
radar, but will not be simple, and will probably go hand-in-hand with
phasing out the bus ops (for the multiple-driver-coexistence problem).
For iommu group creation, the stack flow during race is like:
Display device 1:
iommu_probe_device -> iommu_group_get_for_dev -> arm_smmu_device_group
Display device 2:
iommu_probe_device -> iommu_group_get_for_dev -> arm_smmu_device_group
And this way it ends up in creating 2 groups for 2 display devices
sharing same SID.
Ideally for 2nd display device, iommu_group_get call from
iommu_group_get_for_dev should return same group as 1st display device.
But due to the race, it ends up with 2 groups.
For default domain, the stack flow during race is like:
Display device 1:
iommu_probe_device -> iommu_alloc_default_domain -> arm_smmu_domain_alloc
Display device 2:
iommu_probe_device -> iommu_alloc_default_domain -> arm_smmu_domain_alloc
Here also 2nd device should already have domain allocated and
'if(group->default_domain)' condition from iommu_alloc_default_domain
should be true for 2nd device.
Issue with this is IOVA accesses from 2nd device results in context faults.
@@ -273,7 +273,9 @@ int iommu_probe_device(struct device *dev)
* support default domains, so the return value is not yet
* checked.
*/
+ mutex_lock(&group->mutex);
iommu_alloc_default_domain(group, dev);
+ mutex_unlock(&group->mutex);
It feels wrong to serialise this for everybody just to cater for systems
with aliasing SIDs between devices.
If two or more devices are racing at this point then they're already
going to be serialised by at least iommu_group_add_device(), so I doubt
there would be much impact - only the first device through here will
hold the mutex for any appreciable length of time. Every other path
which modifies group->domain does so with the mutex held (note the
"expected" default domain allocation flow in bus_iommu_probe() in
particular), so not holding it here does seem like a straightforward
oversight.
Robin.
Serialization will only happen for the devices sharing same group. Only
the first device in group will hold this till domain is created. For
rest of the devices it will just check for existing domain in
iommu_alloc_default_domain and then return and release the mutex.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
From: Will Deacon <will@kernel.org> Date: 2021-06-17 17:49:08
On Thu, Jun 17, 2021 at 11:21:39AM +0530, Ashish Mhetre wrote:
On 6/11/2021 6:19 PM, Robin Murphy wrote:
quoted
External email: Use caution opening links or attachments
On 2021-06-11 11:45, Will Deacon wrote:
quoted
On Thu, Jun 10, 2021 at 09:46:53AM +0530, Ashish Mhetre wrote:
quoted
Domain is getting created more than once during asynchronous multiple
display heads(devices) probe. All the display heads share same SID and
are expected to be in same domain. As iommu_alloc_default_domain() call
is not protected, the group->default_domain and group->domain are ending
up with different domains and leading to subsequent IOMMU faults.
Fix this by protecting iommu_alloc_default_domain() call with
group->mutex.
Can you provide some more information about exactly what the h/w
configuration is, and the callstack which exhibits the race, please?
It'll be basically the same as the issue reported long ago with PCI
groups in the absence of ACS not being constructed correctly. Triggering
the iommu_probe_device() replay in of_iommu_configure() off the back of
driver probe is way too late and allows calls to happen in the wrong
order, or indeed race in parallel as here. Fixing that is still on my
radar, but will not be simple, and will probably go hand-in-hand with
phasing out the bus ops (for the multiple-driver-coexistence problem).
For iommu group creation, the stack flow during race is like:
Display device 1:
iommu_probe_device -> iommu_group_get_for_dev -> arm_smmu_device_group
Display device 2:
iommu_probe_device -> iommu_group_get_for_dev -> arm_smmu_device_group
And this way it ends up in creating 2 groups for 2 display devices sharing
same SID.
Ideally for 2nd display device, iommu_group_get call from
iommu_group_get_for_dev should return same group as 1st display device. But
due to the race, it ends up with 2 groups.
For default domain, the stack flow during race is like:
Display device 1:
iommu_probe_device -> iommu_alloc_default_domain -> arm_smmu_domain_alloc
Display device 2:
iommu_probe_device -> iommu_alloc_default_domain -> arm_smmu_domain_alloc
Here also 2nd device should already have domain allocated and
'if(group->default_domain)' condition from iommu_alloc_default_domain should
be true for 2nd device.
Issue with this is IOVA accesses from 2nd device results in context faults.
Thanks for the explanation (also Robin and Krishna). Please put some of this
in the commit message for the next version.
Will
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel