On 2022/7/6 01:08, Robin Murphy wrote:
Currently we rely on registering all our instances before initially
allowing any .probe_device calls via bus_set_iommu(). In preparation for
phasing out the latter, make sure we won't inadvertently return success
for a device associated with a known but not yet registered instance,
otherwise we'll run straight into iommu_group_get_for_dev() trying to
use NULL ops.
That also highlights an issue with intel_iommu_get_resv_regions() taking
dmar_global_lock from within a section where intel_iommu_init() already
holds it, which already exists via probe_acpi_namespace_devices() when
an ANDD device is probed, but gets more obvious with the upcoming change
to iommu_device_register(). Since they are both read locks it manages
not to deadlock in practice, so I'm leaving it here for someone with
more confidence to tackle a larger rework of the locking.
Thanks for highlighting this. I will look into it later.
Best regards,
baolu
Signed-off-by: Robin Murphy <robin.murphy@xxxxxxx>
---
v3: New
drivers/iommu/intel/iommu.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c
index 44016594831d..3e02c08802a0 100644
--- a/drivers/iommu/intel/iommu.c
+++ b/drivers/iommu/intel/iommu.c
@@ -4600,7 +4600,7 @@ static struct iommu_device *intel_iommu_probe_device(struct device *dev)
u8 bus, devfn;
iommu = device_to_iommu(dev, &bus, &devfn);
- if (!iommu)
+ if (!iommu || !iommu->iommu.ops)
return ERR_PTR(-ENODEV);
info = kzalloc(sizeof(*info), GFP_KERNEL);
