Re: [TF-M] IPC and clang

Hi Andrej,

For you question, please see my comments: If I understand well, the Crypto, SST and Attestation services do not use IPC, so far. Right? - Yes.

Should the SST/Crypto/Attestation services be disabled when IPC is enabled? - No, we do not have to disable them.

May the Library and IPC APIs be used simultaneously? - Yes. When using " ConfigCoreIPC.cmake" configure file with enabling the " REGRESSION", you can see all the regression test can work.

What part of TFM is using IPC? - There are two IPC test partitions to use the IPC: trusted-firmware-m/test/test_services/tfm_ipc_client and trusted-firmware-m/test/test_services/tfm_ipc_service. They are used to do basic IPC function tests.

Thanks, Edison

-----Original Message----- From: TF-M tf-m-bounces@lists.trustedfirmware.org On Behalf Of Andrej Butok via TF-M Sent: Wednesday, March 20, 2019 3:36 PM To: tf-m@lists.trustedfirmware.org Subject: Re: [TF-M] IPC and clang

Hi Edison,

If I understand well, the Crypto, SST and Attestation services do not use IPC, so far. Right? Should the SST/Crypto/Attestation services be disabled when IPC is enabled? May the Library and IPC APIs be used simultaneously? What part of TFM is using IPC?

Thanks, Andrej

-----Original Message----- From: TF-M tf-m-bounces@lists.trustedfirmware.org On Behalf Of Edison Ai (Arm Technology China) via TF-M Sent: Wednesday, March 20, 2019 8:22 AM To: tf-m@lists.trustedfirmware.org Cc: nd nd@arm.com Subject: Re: [TF-M] IPC and clang

Hi Andrej,

We tested the IPC works on Musca A but not try it on Musca B yet. The current IPC related patches are used to enable IPC mechanism, but services such as crypto, protect storage and attestation are yet to make use of IPC.

Thanks, Edison

-----Original Message----- From: TF-M tf-m-bounces@lists.trustedfirmware.org On Behalf Of Andrej Butok via TF-M Sent: Tuesday, March 19, 2019 6:06 PM To: tf-m@lists.trustedfirmware.org Subject: Re: [TF-M] IPC and clang

Hi Edison,

OK. So, according to https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Freview.tru... the armclang IPC was added only to one platform (target/mps2/an521/armclang/mps2_an521_s.sct). What about Musca A and Musca B?

Thanks, Andrej

-----Original Message----- From: Edison Ai (Arm Technology China) Edison.Ai@arm.com Sent: Tuesday, March 19, 2019 9:52 AM To: Andrej Butok andrey.butok@nxp.com Cc: nd nd@arm.com Subject: RE: IPC and clang

Hi Andrej,

You can see the log history of master branch: https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgit.truste.... All the IPC patches had been existed in master branch.

You can use the master branch now, all the IPC functions had been ready for GCC and ARMCLANG.

Thanks, Edison

-----Original Message----- From: Andrej Butok andrey.butok@nxp.com Sent: Tuesday, March 19, 2019 4:43 PM To: Edison Ai (Arm Technology China) Edison.Ai@arm.com Cc: nd nd@arm.com Subject: RE: IPC and clang

Hi Edison,

https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgit.truste... master head the latest commit is still 4-day old (4 days Core: Retrieve extra parameter from correct positionHEADmaster Summer Qin). Should I wait some time till it will be propagated to the public git?

Thanks, Andrej

-----Original Message----- From: Edison Ai (Arm Technology China) Edison.Ai@arm.com Sent: Tuesday, March 19, 2019 9:26 AM To: Andrej Butok andrey.butok@nxp.com Cc: nd nd@arm.com Subject: RE: IPC and clang

Hi Andrej,

You are welcome.

Now, the "feature-ipc" branch had been merge into the master branch with the merge patch mentioned below. So all the patches in "feature-ipc" branch had been merge into master too. You can find the related IPC patch in the log history of master branch. The IPC can works rightly in GCC and ARMCLANG on master branch.

Thanks, Edison

-----Original Message----- From: Andrej Butok andrey.butok@nxp.com Sent: Tuesday, March 19, 2019 4:10 PM To: Edison Ai (Arm Technology China) Edison.Ai@arm.com Cc: nd nd@arm.com Subject: RE: IPC and clang

Thanks Adison,

Yes, we are using the master branch. When are you planning to merge the mentioned fix to the mainline?

Thanks, Andrej

-----Original Message----- From: TF-M tf-m-bounces@lists.trustedfirmware.org On Behalf Of Edison Ai (Arm Technology China) via TF-M Sent: Tuesday, March 19, 2019 9:00 AM To: tf-m@lists.trustedfirmware.org Cc: nd nd@arm.com Subject: Re: [TF-M] IPC and clang

Hi Andrej,

I think you mention the "Merge remote-tracking branch 'feature-ipc' into 'master" patch: https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Freview.tru.... This is a merge patch to fix the merge conflicts. The original patch to support to change the linker file is here: https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Freview.tru.... You can see both the linker files for GCC and ARMCLANG are changed.

IPC had been developed and tested on both the GCC and ARMLANG already.

Thanks for your question.

Edison

-----Original Message----- From: TF-M tf-m-bounces@lists.trustedfirmware.org On Behalf Of Andrej Butok via TF-M Sent: Tuesday, March 19, 2019 3:35 PM To: tf-m@lists.trustedfirmware.org Subject: [TF-M] IPC and clang

Hello,

I have noticed, that with adding the IPC feature to master branch, it were updated GCC linker files (#ifdef TFM_PSA_API sections), but ARMCLANG linker files are without any change. Does it mean that IPC was developed and tested only using GCC? Is there a plan to updated the armclang linker files?

Thanks, Andrej Butok