On 07/22/2013 01:27 AM, Hanjun Guo wrote:
Hi Graeme,
On 2013-7-19 23:54, Graeme Gregory wrote:
Hi Andrea,
Currently the #ifdef are due to the completely missing API support on the arm/arm64 platform.
Even with PSCI support we will need to write the infrastructure for it then link it in with processor_idle. But arm said PSCI is not compulsary so we we also need to deal with the case without PSCI.
In theory, PSCI will be pretty close to compulsory on ARMv8 -- or at least that appears to be the consensus and the desire on the server side of things. I have no idea where the mobile side is going to go but I would hope they standardize on PSCI.
ARMv7 is of course completely different and PSCI will likely not be -- nor can it be -- compulsory there. There are too many SoCs already on the market that do not use PSCI.
How about deal with the case without PSCI for upstreaming code first? If the this version of ACPI code is accepted, we can send another patch set for PSCI. In this way we can reduce the complexity for upstream.
I like this approach; the cpu idle code for ARM seems to be getting more straightforward (at least according to the session I was in at LCE13). This would also give us the flexibility of working with either ARMv7 or v8 as availability of HW allows.
Unless someone has a better idea, I suggest we follow Hanjun's approach for now and adjust our cards as needed.
If anything I can do for upstream, I would be happy for that :)
Thanks Hanjun
Basically we have a moderate to large chunk of work to do on the backend API for idling on arm/arm64.
Thanks
Graeme
On 19/07/13 16:32, Andrea Gallo wrote:
Graeme,
are the #ifdef due to lack of PSCI support? so will be solved in the coming months? or are you referring to other #ifdef? could you clarify?
I remember we also had #ifdef as we were not supporting DMI and dmidecode at that time but now it is working.
I would suggest that you propose a sorting / priority list of #ifdef and related tech topics to look into, then we can take actions and plan accordingly.
thanks for your initiative!
Andrea
On 19 July 2013 17:20, Graeme Gregory graeme.gregory@linaro.org wrote:
Hi Guys,
Ok, a serious impediment to us upstreaming code is the state of CPU idle, currently we have large chunks of code #defined out for our build.
We need to come up with a plan to solve this!
Graeme
Linaro-acpi mailing list Linaro-acpi@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-acpi
Linaro-acpi mailing list Linaro-acpi@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-acpi
Linaro-acpi mailing list Linaro-acpi@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-acpi