iommu/amd/iommu_v2: Fix pasid_state refcount dec hit 0 warning on pasid unbind
Linux
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
| No fix yet |
Linux
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
| No fix yet |
| Linux Linux |
|
|
In the Linux kernel, the following vulnerability has been resolved:
iommu/amd/iommu_v2: Fix pasid_state refcount dec hit 0 warning on pasid unbind
When unbinding pasid - a race condition exists vs outstanding page faults.
To prevent this, the pasid_state object contains a refcount. * set to 1 on pasid bind * incremented on each ppr notification start * decremented on each ppr notification done * decremented on pasid unbind
Since refcount_dec assumes that refcount will never reach 0: the current implementation causes the following to be invoked on pasid unbind: REFCOUNT_WARN("decrement hit 0; leaking memory")
Fix this issue by changing refcount_dec to refcount_dec_and_test to explicitly handle refcount=1.
From the CVE record · Linux
A race between PASID unbinding and outstanding page faults in the Linux kernel's AMD IOMMU code can cause a refcount warning and memory leak. The input does not describe how an attacker could trigger it or specify network access, authentication, or user-interaction requirements.
Engineers running Linux kernels with the AMD IOMMU PASID code should check their deployments; affected versions are not specified.