nvme-pci: ensure we're polling a polled queue
Linux
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
|
Linux
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
|
≥ 4b04cc6a8f86c4842314def22332de1f15de8523 · < b96c7b25eb1b748f3e3b1832ebf028b0b223d7e3| No fix yet |
| Linux Linux |
|
|
In the Linux kernel, the following vulnerability has been resolved:
nvme-pci: ensure we're polling a polled queue
A user can change the polled queue count at run time. There's a brief window during a reset where a hipri task may try to poll that queue before the block layer has updated the queue maps, which would race with the now interrupt driven queue and may cause double completions.
From the CVE record · Linux
The Linux kernel's nvme-pci driver can race when a user changes the polled queue count at runtime. During a reset, a high-priority task may poll a queue that has become interrupt-driven before the block layer updates its queue maps, potentially causing double completions. This requires the queue-count change and reset to coincide; the input does not specify authentication requirements.
Operators running Linux kernels with the nvme-pci driver should check their versions; affected versions are not specified.