From: stephano <stephano.cetola@linux.intel.com>
To: "edk2-devel@lists.01.org" <edk2-devel@lists.01.org>
Subject: [edk2-announce] Community Meeting Minutes
Date: Tue, 19 Feb 2019 22:23:59 -0800 [thread overview]
Message-ID: <6b070757-61b6-c41e-bfd6-aff48a7bf62b@linux.intel.com> (raw)
An HTML version is available here:
https://www.tianocore.org/minutes/Community-2019-02.html
Github Pull Requests
---------------------
We are still considering Github as a possible platform for patch review.
There are two issues we'd like to overcome:
1. comprehensive email notifications or backup/archival functionality
2. workflow for users that do not have a consistent internet connection
The notification issue degrades our current ability to archive the
history of a patch review. Github notifications do not provide:
1. the subject line of the patch
2. trailing code context of comments (code that comes AFTER the comment)
3. the commit message
In this way, it is hard to avoid losing the meaning of the pull request
conversation if we were to move to a different system some day. The
longevity of PR branches is also a concern, and a workaround would need
to be found and maintained. We need to be sure that PR branches can be
easily archived so that if a Github user deletes their account, even if
their PR was rejected, the code would be available.
Jeremiah and I are both looking for work around to these issues,
hopefully without having to maintain more code. See here for details on
Laszlo's thorough Github evaluation:
https://lists.01.org/pipermail/edk2-devel/2018-December/033509.html
To be clear, these hurtles do not *have to be* overcome. They were
brought up by several maintainers and community members. They are simply
barriers that we want to discuss fully before committing to a
transition. It is impossible to "change our mind" on this transition
without some loss of data, so best we be sure before we switch.
Working in a community means providing consistent messaging, clear
expectations, and the understanding that each member is valuable.
Working in the open means moderating, evaluating, and distilling input
from community members and companies without bias or prejudice. I take
this transition seriously and do not plan to rush it.
Gerrit Code Review
---------------------
Gerrit code review is viable platform, but Intel's implementation of the
'repo' tool is still working on being open sourced. Also, there would be
a lot of work that would go into implementing a proper Gerrit solution.
If someone would like to volunteer to carry this out, please feel free
to contact me, but I'm going to postpone this for the moment as my
schedule is rather full.
Groups.io
---------------------
Laszlo and I will evaluate Groups.io, however initial impressions is
that this will work as a communication platform going forward. We'd like
to use this for design discussions and as a replacement for our current
mailing list as 01.org does not allow attachments or whistling. Also,
Groups.io allows for online search of the list, chats, and uploads which
is much easier for some than searching through emails. Assuming this is
successful, I will work with 01.org to archive all (if possible) of the
edk2-devel mailing list into groups.io for a (mostly) seamless
transition. I will see how long 01.org is willing to store archives, and
will notify the community of how long that archive will be available.
Community Bug Triage
---------------------
Community bug triage meetings will occur every two weeks, and we will
have 2 meetings to accommodate both sets of timezones. See here for details:
https://www.tianocore.org/bug-triage/
I will be working to develop bugzilla reports and researching possible
platforms that we could add to a community CI. Maintaining these
platforms would be a great way to add to the list of community bugs and
encourage open development.
Community Design
---------------------
We will be starting the community design meetings in March and holding
them every 2 weeks. If we find that there are enough submissions we can
meet every week. I will hold two meetings, much like the community
meetings. Any submission will be discussed in both meeting and notes
sent out in the meeting minutes. Discussion can then continue on the
mailing list. I will send out an RFC for folks to submit possible topics
the day before each meeting.
Rust in EDK II Exploration Notes
------------------------------------------
I have been working with the Rust community, as well as members of
Intel's Firmware Security teams, to explore the benefits of using Rust
in EDK II. For those of you who have never heard of Rust, please take
some time to look into it:
https://en.wikipedia.org/wiki/Rust_(programming_language)
Here is a brief overview of the pros/cons:
Benefits of Rust
Rust is a modern language with features like counted buffers and
strings. It can call C code and be called by C code. It requires no
calls to "free", and does so without garbage collection by statically
inserting calls to "free" for you. Rust types keep track of which
structs own which memory chunks so that developers do not have to keep
track of ownership of struct memory. Rust includes a lot of
functionality not found in C. These features include Unicode support, an
ecosystem library, structural types, and matching (it is very easy to
wrap types as a "SUCCESS" or "FAILURE WITH ERROR CODE"). Rust has no
NULL pointers as that concept has its own type, which makes for cleaner
semantics. It has a robust built in macro system. For example, when you
get a compilation error in code calling a macro, the compiler will be
able to show you where in the macro the failure occurred.
Drawbacks of Rust
Rust is a younger language, though there are many examples of where it
has been used in production. As such, many features one might take for
granted in C still do not exist in Rust, or are newly added. For
example, variadic functions (varargs) were just recently added. There
are other corner cases of "things you can do in C" that still need to be
added. Rust does have a great package management system (called cargo -
crates are packages), but in a system like EDK2, we will want to limit
those outside dependencies. Sub-command cargo-vendor is a tool where you
can pin package versions and update them only as needed.
TianoCore in the News
---------------------
Here's my presentation from FOSDEM:
https://fosdem.org/2019/schedule/event/uefi_boot_for_mere_mortals/
I collaborated with Alexander Graf from SUSE. I'll be giving this talk
again at LinuxFest Northwest (http://linuxfestnorthwest.org).
Thank you all for joining. As always, please feel free to email the list
or contact me directly with any questions or comments.
Kind Regards,
Stephano Cetola
TianoCore Community Manager
next reply other threads:[~2019-02-20 6:24 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-20 6:23 stephano [this message]
2019-02-20 6:45 ` [edk2-announce] Community Meeting Minutes stephano
2019-02-20 7:49 ` Rebecca Cran
-- strict thread matches above, loose matches on Subject: below --
2019-01-11 19:26 stephano
2019-01-13 3:59 ` Rebecca Cran
2019-01-14 9:28 ` Laszlo Ersek
2019-01-14 17:06 ` stephano
2019-02-07 17:52 ` Jeremiah Cox
2019-02-07 18:30 ` stephano
2019-02-08 6:41 ` Rebecca Cran
2019-02-08 9:01 ` Laszlo Ersek
2019-02-08 17:33 ` Rebecca Cran
2019-02-08 17:52 ` Andrew Fish
2019-02-22 11:52 ` Rebecca Cran
2019-02-08 20:33 ` Laszlo Ersek
2019-02-08 13:58 ` Ard Biesheuvel
2019-02-14 19:07 ` Jeremiah Cox
2019-02-14 20:27 ` Rebecca Cran
2019-02-14 22:13 ` Kinney, Michael D
2019-02-15 2:56 ` Rebecca Cran
2019-02-15 14:30 ` Laszlo Ersek
2019-02-15 17:55 ` stephano
2019-02-15 8:43 ` Ard Biesheuvel
2019-02-15 14:23 ` Laszlo Ersek
2019-02-15 19:54 ` Felix Polyudov
2019-02-15 22:53 ` Laszlo Ersek
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-list from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=6b070757-61b6-c41e-bfd6-aff48a7bf62b@linux.intel.com \
--to=devel@edk2.groups.io \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox