iommu/amd: serialize sequence allocation under concurrent TLB invalidations
Linux
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
|
Linux
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
|
69d9339c02eddfc2a345be7220651068a3684383| No fix yet |
| Linux Linux |
| No fix yet |
In the Linux kernel, the following vulnerability has been resolved:
iommu/amd: serialize sequence allocation under concurrent TLB invalidations
With concurrent TLB invalidations, completion wait randomly gets timed out because cmd_sem_val was incremented outside the IOMMU spinlock, allowing CMD_COMPL_WAIT commands to be queued out of sequence and breaking the ordering assumption in wait_on_sem(). Move the cmd_sem_val increment under iommu->lock so completion sequence allocation is serialized with command queuing. And remove the unnecessary return.
From the CVE record · Linux
Concurrent TLB invalidations in the Linux kernel's AMD IOMMU code can queue completion-wait commands out of sequence, causing completion waits to time out. The input does not describe an attacker, an access requirement, or user interaction.
Engineers running Linux kernels with AMD IOMMU support should check their kernel versions; affected versions are not specified.