Host Controlled Mitigation of Panic Situation
By detecting panic situations and providing a host device detailed panic data and mitigation options before the panic situation condition occurs, failure of data storage devices may be reduced. Several mitigation options may be presented to the host device upon detection of a future panic situation, such as adjusting read performance; increasing device power; performing evacuation and management operations; and/or redirecting host commands to another data storage device for command completion. When the other data storage device is in the same PCIe tree and reachable, the command may be directed with PRPs/SGLs that point back to the same host device. In some embodiments, the data storage devices may have a submission queue between them. In some embodiments, the data storage device comprises a panic early detection module and a panic control module.
1 . A data storage device, comprising:
a memory device; and
a controller coupled to the memory device, wherein the controller is configured to:
detect a future panic situation of the data storage device;
analyze the future panic situation;
suggest at least one mitigation option to a host device, wherein the suggestion comprises a decision timeout;
perform a default mitigation option while waiting to receive a selected mitigation option from the host device;
receive from the host device the selected mitigation option selected from the suggested at least one mitigation options; and
perform the selected mitigation option.
2 . The data storage device of claim 1 , wherein the controller is further configured to perform a selected mitigation option based on the decision timeout.
3 . The data storage device of claim 1 , wherein receiving the selected mitigation option overrides the default mitigation option.
4 . The data storage device of claim 3 , wherein the controller is not exposed to additional future panic situations while waiting to receive the selected mitigation option from the host device.
5 . The data storage device of claim 1 , wherein a mitigation option of the at least one mitigation options redirects a host command received by the data storage device to a second data storage device.
6 . The data storage device of claim 5 , wherein the data storage device and second data storage device are peer-to-peer (P2P).
7 . The data storage device of claim 5 , wherein redirecting the host command comprises:
determining whether the second data storage device can perform the host command; and
queuing the host command into a submission queue of the second data storage device.
8 . The data storage device of claim 7 , wherein the submission queue of the second data storage device is a peer-to-peer (P2P) submission queue.
9 . The data storage device of claim 7 , wherein redirecting the host command further comprises:
parsing a completion entry to a completion queue of the second data storage device; and
writing a completion entry to a relevant host completion queue of the data storage device.
10 . The data storage device of claim 9 , wherein the completion queue of the second data storage device is a peer-to-peer (P2P) completion queue.
11 . The data storage device of claim 9 , wherein redirecting the host command further comprises interrupting the host device after writing the completion entry to the relevant host completion queue.
12 . A data storage device, comprising:
a memory device; and
a controller coupled to the memory device, wherein the controller is configured to:
detect a future panic situation of the data storage device based on a panic indicator;
analyze the future panic situation;
suggest at least one mitigation option to a host device, wherein a mitigation option of the at least one mitigation options comprises redirecting a host command to another location;
perform a mitigation option; and
determine that the future panic situation is resolved.
13 . The data storage device of claim 12 , wherein the another location is a second data storage device.
14 . The data storage device of claim 13 , wherein the second data storage device is in a same PCIe tree as the data storage device.
15 . The data storage device of claim 12 , wherein the controller is further configured to remove the panic indication and stop performing the mitigation option after the future panic situation is resolved.
16 . The data storage device of claim 12 , wherein the controller comprises a panic early detection module and a panic control module.
17 . A data storage device, comprising:
means to store data; and
a controller coupled to the means to store data, wherein the controller is configured to:
detect a future panic situation of the data storage device;
receive a host command queued in a submission queue of a host device;
redirect the host command to a second data storage device for completion, wherein the second data storage device is in a same PCIe tree as the data storage device; and
interrupt the host with a completion entry to a relevant host completion queue of the host device.
18 . The data storage device of claim 17 , wherein the controller is further configured to redirect the host command to the second data storage device with physical region pages (PRPs) that point back to a same host region.
19 . The data storage device of claim 17 , wherein the controller is further configured to redirect the host command to the second data storage device with scatter gather lists (SGLs) that point back to a same host region.
20 . The data storage device of claim 17 , wherein the controller is further configured to add an indicator that points to another location when the host command is received.