Hi James, I recently got a chance to look into the ARM CCA specifications. And if I understood correctly then in Realm world PSCI is managed by RMM but the Realm vCPUs are taken care by the host.
As per our recent changes we handle the SMC/HVC hyper call exits in user-space. If the VMs are to run in Realm world and vCPU Hotplug needs to be supported then do we need to consider any change there?
Thanks Salil
Hi Salil,
On 7/17/23 18:17, Salil Mehta wrote:
I recently got a chance to look into the ARM CCA specifications. And if I understood correctly then in Realm world PSCI is managed by RMM but the Realm vCPUs are taken care by the host. As per our recent changes we handle the SMC/HVC hyper call exits in user-space. If the VMs are to run in Realm world and vCPU Hotplug needs to be supported then do we need to consider any change there?
It would need a call between EL2 and R-EL2 to set which CPUs the realm is allowed to bring online via PSCI. I don't think would be too much of a problem.
We didn't do that for the kernel because it was better to avoid creating more in-kernel state that has to be migrated.
Thanks,
James
Hi James, Thanks for replying.
-----Original Message----- From: James Morse james.morse@arm.com Sent: Monday, July 24, 2023 4:43 PM To: Salil Mehta salil.mehta@huawei.com; Suzuki K Poulose suzuki.poulose@arm.com; linaro-open-discussions@op-lists.linaro.org Cc: Joyce Qi joyce.qi@linaro.org; Jonathan Cameron jonathan.cameron@huawei.com; Salil Mehta salil.mehta@opnsrc.net Subject: Re: [Question - ARM CCA] Regarding vCPU Hotplug Support in ARM Realm world
Hi Salil,
On 7/17/23 18:17, Salil Mehta wrote:
I recently got a chance to look into the ARM CCA specifications. And if I
understood correctly then
in Realm world PSCI is managed by RMM but the Realm vCPUs are taken care
by the host.
As per our recent changes we handle the SMC/HVC hyper call exits in user-
space. If the VMs
are to run in Realm world and vCPU Hotplug needs to be supported then do
we need
to consider any change there?
It would need a call between EL2 and R-EL2 to set which CPUs the realm is allowed to bring online via PSCI. I don't think would be too much of a problem.
We didn't do that for the kernel because it was better to avoid creating more in-kernel state that has to be migrated.
Please follow this conversation and reply here (sorry for duplicating):
https://lore.kernel.org/all/7da93c6e-1cbf-8840-282e-f115197b80c4@arm.com/
Thanks Salil.
linaro-open-discussions@op-lists.linaro.org
-
James Morse -
Salil Mehta