T1040 How do I gracefully shut down a running hypervisor?

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

T1040 How do I gracefully shut down a running hypervisor?

Jump to solution
529 Views
stephenmcintyre
Contributor II

Our system is designed around a T1040 series SoC and it will be running hypervisor with 2 to 4 partitions each running a version of linux. The design calls for accepting a data load request from any of the partitions (cpu's running linux). when the request is received the system needs to shut down gracefully and return to single cpu to data load the system. How can I shutdown gracefully without something like letting the WDT timeout and resetting the system? Is there another way to do this? Will one partition need to be in control and accept the request or can any partition accept the load request?

0 Kudos
1 Solution
444 Views
bpe
NXP Employee
NXP Employee

A complete shutdown to the Hypervisor is essentially a reboot (where would it
otherwise return control to?) However, shutting down all partitions
is effectively what you want - all secondary CPUs stop. Regarding
data load, it is not fully clear what was meant. Hypervisor by itself
does not own any IO device where it could load data from. Any IO should
be done by a guest owning the corresponding device. A manager partition
can then issue a HCALL to copy data to/from a managed partition.
There can be several manager/managed partitions. See QorIQ SDK 2.0
Documentation, Sections 11.1.2.4.9 and 11.1.2.6.5 for more details.

Visit the SDK download page to get the  Documentation in PDF:

https://www.nxp.com/support/developer-resources/run-time-software/linux-software-and-development-too...


Have a great day,
Platon

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

View solution in original post

1 Reply
445 Views
bpe
NXP Employee
NXP Employee

A complete shutdown to the Hypervisor is essentially a reboot (where would it
otherwise return control to?) However, shutting down all partitions
is effectively what you want - all secondary CPUs stop. Regarding
data load, it is not fully clear what was meant. Hypervisor by itself
does not own any IO device where it could load data from. Any IO should
be done by a guest owning the corresponding device. A manager partition
can then issue a HCALL to copy data to/from a managed partition.
There can be several manager/managed partitions. See QorIQ SDK 2.0
Documentation, Sections 11.1.2.4.9 and 11.1.2.6.5 for more details.

Visit the SDK download page to get the  Documentation in PDF:

https://www.nxp.com/support/developer-resources/run-time-software/linux-software-and-development-too...


Have a great day,
Platon

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------