See below for link.

On 3/21/22 8:37 AM, Bill Mills wrote:
Everyone in the to: list,

Are you all free today at 3PM UTC (11 AM US East)?

Oliver,

See the DTE home page for a description of the meeting, the Zoom link and a google ics file.


It helps if I add the link:
https://linaro.atlassian.net/wiki/spaces/DTE/overview


However I just added you to the invite going forward.  Just delete after we are done if you do not want to continue.

The meeting is tied to UK time so right now appears 1 hour later than normal as US has done DST and the UK will not until next week.

Thanks,
Bill

On 3/7/22 11:00 AM, Olivier Masse wrote:
Hi Bill,

Could you send us some pointers to join your call ?

Sorry for this late request.

Olivier

On mar., 2022-02-22 at 15:00 -0500, Bill Mills wrote:
Caution: EXT Email

Oliver,

On 2/22/22 7:36 AM, Olivier Masse wrote:
Hi All,

Could we postpone to the next call ?


Yes we can discuss on March 7.

However I think you need to define what you want to discuss.

First of, there is nothing in your suggested DTS that tells Linux (or
other  OS) that it should not map the memory area.  It looks to me as
you should include the "no-map" property in your node.

How will the Linux video/graphics drivers refer to this node?  I
presume
by phandle, right?

Are you setting any suggested naming convention for the node name
(after
you fix the node name to include @address)?

You are defining a compatible string in the node.  The current
suggestion is not in the spec.  It could be "optee,sdp" if you get
optee
stakeholders to agree.  What should be the definition of "optee,sdp".
I presume it means secure data path but that could mean a ton of
things.
Will sdp always mean DRM protected media playback or could it include
more?

Thanks,
Bill

BR / Olivier

On mar., 2022-02-15 at 13:46 -0600, Rob Herring wrote:
Caution: EXT Email

On Fri, Feb 11, 2022 at 8:21 AM Bill Mills <bill.mi...@linaro.org

wrote:

Rob,

Can you confirm for the DT call on Feb 21?

I'm on holiday on the 21st.


Oliver,

On 2/11/22 4:54 AM, Olivier Masse wrote:
Hi Bill,

NXP had a discussion with Linaro about this optee os issue:



https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FOP-TEE%2Foptee_os%2Fissues%2F5133&amp;data=04%7C01%7Colivier.masse%40nxp.com%7C64218f2199dd4d71f84008d9f63e093a%7C686ea1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C637811568813890462%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&amp;sdata=Enk98Vs%2B1dDeoXceyiPnNhkSRksZ9YGvFN84DAQRnCI%3D&amp;reserved=0

Which is implemented by this first draft here:



https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FOP-TEE%2Foptee_os%2Fpull%2F5149&amp;data=04%7C01%7Colivier.masse%40nxp.com%7C64218f2199dd4d71f84008d9f63e093a%7C686ea1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C637811568813890462%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&amp;sdata=UYLsOYcqoRKyDWkAWjXeYD5l7IAYO%2Bx8YVMs8538n78%3D&amp;reserved=0

Could we be part of the next Device Tree call to discuss
about
adding
a reserved memory in optee os embedded DT ?


Thanks for bringing it up.
I have skimmed the PR threads and the discussion seems to be:
1) OP-TEE internal issues
2) DT standards questions and issues

For the DT call we need to focus on #2 above.
We will definitely need Rob for this discussion so we need to
do it
when
he can join.  (If we don't resolve the question in this email
thread
before then.)

Context for all:

SDP here is related to DRM protected playback of media streams.

Jens comment in PR:
   > So far we have managed to avoid defining our own bindings
in OP-
TEE,
   > instead we've been able to reuse already established
bindings.
With
   > this you're proposing something new. I'm not sure of the
best
way of
   > doing such a thing. Are we sure there is nothing to reuse?
   > If not: How should it be reviewed? Who should review it?

DTS in PR:
/*
    * Copyright (c) 2021, NXP. All rights reserved.
    *
    * SPDX-License-Identifier: BSD-3-Clause
    */

/dts-v1/;

/ {
          #address-cells = <1>;
          #size-cells = <0>;

          reserved-memory {
                  #address-cells = <1>;
                  #size-cells = <1>;

                  sdp_mem {
                          compatible = "optee-sdp";
                          reg = <0x3E800000 0x00400000>;
                  };
          };
};

However it was modified after that.

Oliver: please reply to this thread (on list please) with the
final
DTS
you are proposing. fix DT conventions and what are you doing
with
the
no-map property if anything.

Is the memory above meant to be visible to NS world AND S
world?  Or is
just for secure world.

Thanks,
Bill

Best regards,
Olivier Masse

--
Bill Mills
Principal Technical Consultant, Linaro
+1-240-643-0836
TZ: US Eastern
Work Schedule:  Tues/Wed/Thur
_______________________________________________
boot-architecture mailing list --
boot-architecture@lists.linaro.org
To unsubscribe send an email to
boot-architecture-le...@lists.linaro.org

--
Bill Mills
Principal Technical Consultant, Linaro
+1-240-643-0836
TZ: US Eastern
Work Schedule:  Tues/Wed/Thur

--
Bill Mills
Principal Technical Consultant, Linaro
+1-240-643-0836
TZ: US Eastern
Work Schedule:  Tues/Wed/Thur

--
Bill Mills
Principal Technical Consultant, Linaro
+1-240-643-0836
TZ: US Eastern
Work Schedule:  Tues/Wed/Thur
_______________________________________________
boot-architecture mailing list -- boot-architecture@lists.linaro.org
To unsubscribe send an email to boot-architecture-le...@lists.linaro.org

Reply via email to