Hi,
This is in context of earlier discussion about socialising TF-A more
widely, e.g. on LinkedIn and Twitter. I noticed that stackoverflow.com
didn't have a "tf-a" tag even though there are questions referring to
it. I have created it now.
Adding more questions and answers under this tag will also help in
socialising TF-A.
Best regards,
Okash
TL;DR: Attached is the TF-A LTS proposal doc. Please review and share
your thoughts.
Hello everyone,
Long term support for TF-A has a long history. As a community we have
flirted with it for a while, and like any flirtation, it started with
gossip -- in our case, over mailing list [1] -- more than two years
ago. Then Varun Wadekar did a wonderful tech forum presentation [2]
which demonstrated interest in the topic and raised interesting
questions about LTS.
After some time, the affair lost headline status, only to be revived
again [3], earlier this year. This time however, there is a formal
proposal (no pun intended). After the mailing list discussion and
feedback from TF-A tech forum [4], we have put together a draft which
attempts to give a concrete idea about what the TF-A LTS will look
like.
Please spare some time to read through it and share your feedback. The
plan is to put LTS into action this November.
Cheers!
Okash
[1] https://lists.trustedfirmware.org/archives/search?mlist=tf-a%40lists.truste…
[2] https://www.trustedfirmware.org/docs/TF-A-LTS.pdf
[3] https://lists.trustedfirmware.org/archives/list/tf-a@lists.trustedfirmware.…
[4] https://www.trustedfirmware.org/docs/TForg_LTS_proposal.pdf
Hi All,
Please find the minutes to the Sept 15 TSC.
Best regards,
Don - sent on behalf of the TSC co-chairs
======================================
Attnedees: David Brown (Linaro), Thomas Sanderson (Infineon), Kevin
Townsend (Linaro), Lionel D (ST), Eric Finco (ST), Andrej Butok (NXP), Dan
H (Arm), Antonio (Arm), Bill Peckham (Google), Kevin Oerton (NXM.Labs),
Julius Werner (Google), Okash (Google), Matteo (Arm)
Agenda items:
* LTS
* RMM
* FW handoff spec
Recurring:
* OpenCI update won't be held as Glen/Don are not available today. See
backup in the redacted board slides sent out by Don for Open CI monthly
update.
LTS sum-up from board meeting
* until we start, it's hard to gain momentum and hard to estimate costs in
advance.
* TF.org to evaluate direct funding but only from the next financial year
* Proposal: share the burden for 1 LTS release to be maintained for at
least one year. To evaluate the effort, gather data, see how it goes,
evaluate the engagement, and then have data to propose to the board for
long term funding. Share the burden between companies interested.
* Question: which platforms are tested and supported? The baseline is the
one tested by the official TF-A releases, with those platforms available in
the OpenCI
* Board requests to carefully iron out the messaging to be shared with the
community around the announce of the LTS so that expectations are clear in
the community
* Tech-wise, the proposal does not seem to have any objection so this is
mostly around funding/maintainership topics at the moment
* Companies interested are now to meet and agree next steps
TF-RMM component
* Upstreaming expected November 2022. Plans on integrating TF-RMM in the
OpenCI in the longer term as well.
Is the code already visible? It's still private. Arm Architecture group
has some private prototype that was shared with some partners under NDA.
But different significantly from what we're planning to upstream.
* Is it completely separate or depends on other TF.org projects? It's
coupled with TF-A in the same way as Hafnium is. It implements RMM
specification, so Linux and KVM and other clients they need to support the
same version of the spec of RMM to be compatible
Is the KVM counterpart of RMM going to be upstreamed? Yes, but Arm not
the maintainer so actual upstreaming will take longer as Arm does not
maintain those projects. But there will be public branches with these
changes in the meantime
* KVM, EDK2, etc altogether available in November, they will follow their
own destiny in each project.
Is there a plan-B if upstream does not accept them? There is the risk but
hopefully we will try to minimize and does not come to upstream maintainers
as a surprise (discussions under NDA for years already with non-Arm
maintainers). It does not mean it will be quick and easy but confident that
it will happen. For example one particular concern is the major work going
on in Android pKVM. Might risk some conflicts that slow down upstreaming
FW Handoff spec
* Several discussions around this happened so far. A number of interested
parties (e.g. uboot). Quite difficult to accommodate all requirements from
different parties. This is a tentative to summarise and centralise the
discussion so far to foster further collaboration and evolution of the
spec. We start on TF.org then we will re-assess if this the best place (or
needs to be moved to GitHub or other places).
* Contribution model: similar to other projects but with indepedent
maintainers, no strong links to other projects
* Creative commons license as it's more user friendly.
* Unless a big objection this is what is going to happen in the next weeks
* Is the format not pinned to devicetree? The spec just talks about
container format, the actual data can have different practical formats. It
tends to accommodate different implementation details. It's more focused on
the information itself rather than formats to organize this information
(i.e. registers to use, etc). It's quite a simple spec
* The spec will allow to specify different formats for the practical
structures that hold the information
Are there plans to add also tftf-tests for RMM? Yes there are some
additional tests that will be upstreamed.
* There are some changes in TF-A for example that are already happening
upstream in preparation for November 2022
Agenda items finished, no further objections.
Discussion about what will happen in subsequent meetings.
1. Roadmap sharing: Start again with TF-M (action item: inform Shebu and
invite for October (last presentation was in February)
2. Members inform about their usage (e.g. Kevin from NXM did very good
presentation last round) of TF.org projects, we believe it can be
beneficial for the community. Call to major partners in the call if there
is something you might be considering.
TSC interested especially in hearing about any particular challenge that
you need to overcome to use the projects commercially, any
suggestions/lesson learnt that can benefit the different projects in terms
of technical direction. Not necessarily to public TSC, can be a members
only TSC (and share public/open info only)
Feel free to reach out (directly to us in private if not comfortable)
STM: We can do that for some of the projects of TF.org (what we use).
Some direction, what we think it's difficult, what can be changed in terms
of direction.
Is there anything else members would like to have in future TSC?
No particular feedback.
Last clarification about the structure of mailing lists. TSC-Private is the
members only invited to the meeting.
TSC is for public attendance (no filter). Don has re-organized recently so
hopefully minutes of meetings will have been correctly archived in the list
manager now.
No particular other business, meeting close.
IMPORTANT NOTICE: The contents of this email and any attachments are
confidential and may also be privileged. If you are not the intended
recipient, please notify the sender immediately and do not disclose the
contents to any other person, use it for any purpose, or store or copy the
information in any medium. Thank you.
