►
From YouTube: CentOS Automotive SIG meeting - 2022-02-02
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Okay,
good
morning,
everybody
or
good
evening,
wherever
it
is.
However,
it
is
we're
in
your
part
of
the
world
and
welcome
to
the
centos
automotive
sig
meeting
for
february
2nd.
We
did
not
have
a
meeting
in
january
and
I
was
not
present
at
the
one
in
december.
I
appreciate
ian
and
I
think
pierre,
I'm
not
sure
exactly
who
ran
that
one.
A
But
well
that's
good.
That's
good!
It's
good
good
to
have
a
community
effort,
so
so
let's
go
ahead
and
dive
in
because
we
have
a
lot
to
cover,
particularly
in
the
technical
status.
I
made
a
brief
agenda.
A
If
anybody
has
any
desire
to
to
add
to
this,
please
just
drop
a
note
in
the
chat.
Keep
an
eye
on
that.
You
have
any
agenda
items
that
you
would
like
to
add.
A
A
And
I
do
want
to
make
sure
that
we
discuss
the
future
meeting
schedule
because
it
does
seem
to
kind
of
be
a
perpetual
thorn
in
our
side
that
that
we
don't
really
have
a
set
meeting
time,
because
there
isn't
a
single
time
where
everybody
is
available,
and
hopefully
we
can
get
that
worked
out.
A
So
what
I'd
like
to
do
first
is
just
launch
into
the
technical
status
update,
because
that
I
think,
is
going
to
be
the
most
interesting
for
everybody
and
pierre.
I
will
hand
it
over
to
you.
B
Thank
you
jeffrey.
So
I,
as
you
may
have
followed
from
the
the
weekly
newsletter
that
we're
trying
to
send
on
a
on
a
weekly
basis
to
the
automotive,
steve
sieg.
So
if
you've
sold
them,
there
will
probably
be
a
little
new,
but
to
come
back
to
where
we
were
before
the
you
know.
B
The
end
of
the
year
break
the
place
where
we
mostly
work
is
the
automotive
sigripo
on
gitlab,
so
gitlab.com
for
that
slash,
automotive,
slash,
automotive,
seek
this
is
where
we're
building
some
of
our
toolings,
but
also
where
we
are
hosting
our
image
sample
image
definitions.
B
You
can
find
instruction
on
how
to
build
these
images
on
six.central.org,
slash,
automotive
and
you
have
to
generic
documentation
there.
But
then
you
have
a
you
know
a
bill,
a
page,
that's
specific!
To
building
these
images.
We
have
a
ci
pipeline.
That
is
work
in
progress.
It's
actually
it
is
actually
running.
We
still
have
some
quirks
here
and
there,
but
it
is,
it
is
present.
If
you
look
at
merch
requests
that
are
in
this
gitlab
project,
you
will
see
a
number
of
tests
coming
back.
B
It's
it
tries
to
create
a
young
repository.
Then
it
tries
to
use
that
game
repository
to
build
images
and
it
lets
you
know
how
well
it
did.
We've
moved
from
center
stream
a
to
center
stream
nine
some
time
ago.
Now,
so
all
the
sample
images
you'll
find
on
the
repository
here
are
all
c9
cs9
based
as
part
of
the
ci
pipeline.
We
we're
trying
to
get
some
testing
done.
B
Currently
a
lot
of
them
are.
We
have
some
places
that
are
placeholders.
So
if
you
look
at
some
of
the
merch
requests
you'll,
you
may
see
something
related
to
fusa
for
the
quality
gate.
This
is
for,
for
example,
one
of
the
placeholder
that
we
have
for
the
pipeline.
There
is
no
actual
test
being
done
yet,
but
we
are
building
the
infrastructure
too
low
as
to
to
have
some
tests.
A
B
Images
so
something
where
you
don't
have
to
manually
build
images.
You
can
just
grab
one
from
from
server
flash
it
on
your
sd
card,
get
it
to
boot
on
your
raspberry,
pi
or
load
it
onto
a
kvm
or
qmu.
Whichever
format
you
do
you
desire,
we
have
actually
top
in
the
same
vein.
We
have
now
sample
images
with
different
types,
so
we
used
to
have
qk02
images
which
are
usable
with
qmu
with
kvm,
but
we
also
have
what
is
it
yeah,
dot
img
file
which
you
can
flash
on
your
sd
card?
B
Nowadays,
you
can
also
get
oci
containers,
so
you're
able
to
look
at.
So
we
have
a
developer,
a
developer
environment
image
and
you
can
build
it
as
an
oci
table,
and
then
you
can
just
put
man
import.
That
archive.
C
A
B
A
ready-to-use
container,
with
all
all
the
pre-installed
the
prerequisite
installed
for
and
reading
the
kernel
source
rpm,
that's
one
of
the
target
that
we
were
looking
at
as
well
as
actually
building
images,
although
this
is
not
the
best
and
the
best
example,
because
you
can't
actually
build
images
within
a
container.
Yet
we
have
other
means
to
do
that
which
are
mentioned
in
the
documentation
we
just
spoke
about.
B
You
can
get
a
table
just
a
table
of
the
the
root
file
system.
You
can
get
an
os3
repository
if
you've
been
playing
with
ostrich,
then
it
allows
you
to
build
a
repo
for
the
next
update
of
your
image.
All
of
this
is
present
in
the
make
help.
B
So
if
you
download
the
automative
sig
repo,
if
you
clone
it,
sorry
you
go
to
the
os,
build
manifest
folder
and
you
just
type
make
help
and
you
have
all
the
instructions
using
this
make
file,
which
is
actually
also
something
new.
Since
the
last
meeting,
we
have
wrapped
up
all
the
the
commands
that
we
add
into
an
easy-to-use
mac
file,
which
explains
a
number
of
options
that
are
available.
B
Which
are
specific
to
the
to
the
to
the
studio
automotive
sig?
One
of
them
is
the
kernel
auto,
which
has
actually
been
renamed
to
kernel
automotive,
simply
because,
apparently,
when
kernel
also
sounded
too
much
like
an
automated
kernel
or
something
that
builds
automatically,
and
so
people
had
assumptions
with
auto
kernel,
auto
that
were
not
matched
with
what
actually
is
the
kind
of
auto,
so
it
was
renowned
to
canal
automotive.
B
B
For
example,
it
has
some.
Some
patches
have
been
merged
there
for
a
better
support
of
the
raspberry
pi
4,
which
are
not
present
in
the
rel
kernel
or
the
center
stream
kernel,
and
the
idea
is
of
course,
to
to
enable
more
hardware,
as
as
we
progress
and
find
new
address,
to
enable
hinder
so
overall
we're
progressing.
B
D
Specifically,
I'd
like
to
hear
people's
thoughts
on
various
configs,
like
I'd
really
like
to
hear,
if
somebody
says
oh
you're
missing
my
favorite
device.
D
You
know:
we've
made
a
good
what
we
thought
was
a
good
stab
at
what
the
configs
should
be,
but
we
know
they're
not
all
there.
So
don't
be
shy
about
letting
us
know
that
it's
funny,
I'm
sitting
here
in
an
open
source
meeting,
saying
don't
be
shy,
but
anyway,
please.
I
guess
the
solicitation
is,
let's
take
a
look
at
the
kernel
config
and
make
sure
it
meets
your
needs
and
if
it
doesn't
we'll
try
and
adjust.
E
I
saw
a
slide
where
it
said
things
were
going
to
cs9
and
I
went
back
to
the
build
page
and
can
I
will
I
be
able
to
take
the
centos
stream
eight
instructions
and
then
paste
in
a
euro
that
works
for
nine
and
it
will
work
like
I've
got
a
I've
got
a
cs9
machine.
I
could
or
just
do.
The
docs
need
to
be
changed.
B
Yeah,
no,
you
need
to
look
at
the
well,
the
reload,
so
it's
the
oh
yeah.
I
see
what
you
mean.
The
installing
os
build
part
as
something
that's
specific
to
center
stream.
Eight.
That
thing
is
not
specific
to
censor
stream,
nine,
you
can
use
the
rel
command
and
that
will
work
that
will
work
fine.
It's
just
that
for
some
reason,
the
dnf
plug-in
on
sensor
stream
8
is
confused
between
apple
8
and
center
stream
8..
B
That
is
not
a
problem
that
exists
in
center
stream
9,
as
and
as
the
there
is
no
centos
linux
nine.
So
that
problem
is
actually
going
away
there.
So
you
can
just
use
the
the
fedora
or
develop
the
red
host
commands.
They
will
work,
they
will
work
fine,
I'll
I'll,
choose
the
document
too.
To
say,
though,.
E
A
A
I
think
that
was
up
from
december
just
to
continue
building
infrastructure,
which
sounds
like
a
lot
of
work
has
been
done
on
that
to
find
a
better
time
to
meet,
which
has
not
been
accomplished
and
to
encourage
all
interested
parties
to
join
this
community,
which
I
have
noticed
around
that
there
has
been
more
discussion
about
the
centaur,
sig
and
more
interest,
and
I
know
that
a
lot
of
people
will
be
watching
this
recording
that,
because
of
the
previous
issue,
that
finding
a
better
time
to
meet
is
actually
really
a
challenge,
but
but
yeah
let's
go
through
some
of
the
goals
that
we
actually
had
in
2021.
A
We
for
we
started
this
sig
in
september.
I
believe
our
first
official
meeting
was
in
september,
so
it
really
is
only
about
four
months
old
and
in
that
time
I
think
we
have
done
quite
a
lot
to
build
a
foundational
community
to
establish
working
relationships
among
all
of
the
participants
as
much
as
we
possibly
can
and,
of
course,
we've
been
encouraging
others
to
join.
A
We
are
still
in
the
process
of
creating
documentation
around
the
community,
so
I
put
an
x
there
to
make
sure
that
we
noticed
that
we
still
need
to
do
that,
including
some
detailed
contribution
guidelines
pierre
and
team
have
built.
Obviously
we
were
just
looking
at
some
of
the
the
great
documentation
that
exists
already.
A
I
want
to
make
sure
that
we
get
our
community
guidelines
down
and
make
it
as
welcoming
a
place
to
join
as
possible
and
also
to
get
a
contributing
file
into
the
repo
itself
because
kind
of
a
standard
in
open
source.
A
We
have
been
developing
plans
for
the
automotive
variant.
Clearly
there
there
are
plans
that
are
in
code
and
in
the
repo
now
there
is
build
and
test
infrastructure
we're
working
on
a
ci
pipeline.
I
I
think
that
it's
all
going
extremely
well
and
very
quickly,
and
we
did
do
some
presentations
at
automotive
linux
summit.
A
Gabrielle
actually
did
a
talk
about
our
functional
safety
methodology,
which
was
great.
It
was
a
a
follow-on
from
the
talk
he
a
similar
talk.
He
gave
the
previous
month
at
the
elisa
workshop
and
then
it
was
supposed
to
be
pierre
er.
Sorry
supposed
to
be
gabrielle
and
myself
giving
a
functional
safety
buff,
and
since
I
was
out
sick,
it
ended
up
just
being
gabrielle,
but
evidently
there
were
29
people
there
at
the
automotive
linux
summit,
buff
for
functional
safety,
and
that
was
all
going
well.
A
G
Out
yeah.
A
A
Things
are
going
quite
well.
This
is
a
basically
a
wall
of
text.
I
might
split
it
in
here
at
some
point
goals
and
questions.
So
these
are
some
questions
that
I
have
for
the
group.
I
don't
remember
how
many
people
we
have
on
20
something
we
have
23
people
on.
It
looks
like
and.
A
A
I
think
that
we
do
need
a
contributing
file
in
the
repo.
As
I
had
said,
I
think
we
need
to
make
sure
that
the
repo
follows
all
the
best
open
source
guidelines
and
is
as
as
open
and
welcoming
as
possible.
I
think
we've.
Certainly
the
ci
pipeline
is
going
to
do
a
great
job
with
that
and
the
available
images
is
going
to
be
are
going
to
be
very
popular.
I
think.
A
I
think
we
need
to
do
some
documentation,
work
and
develop
pages
on
the
wiki.
That
would
be
the
centos
wiki,
and
one
of
the
things
that
we've
talked
about
in
previous
meetings
is
creating
a
page
so
that
people
can
post
hardware
ideas
right
now.
We
are
really
using
the
raspberry
pi
4
only
because
it
is
ubiquitous
and
readily
available,
or
it
would
be
if
it
weren't
for
the
chip
shortage.
A
We
want
to
to
have
one
basic
configuration
and
no
bsps.
I
remember
writing
that
one
down
we
want
to
have
supporting
pages
for
other
hardware,
so
that
it
is
an
easy
process
to
get
this
to
get
this.
This
operating
system
running
on
alternative
hardware
and
the
other
question
I
have
about
the
wiki
is:
should
we
be
listing
the
technical
leads
for
the
sig?
I
think
that
I'm
listed
as
the
acting
chair.
A
One
other
thing
that
I
wanted
to
do
also
is
is
ask
we'll
talk
about
it
in
a
second
is:
do
we
want
to
elect
a
new
chair?
It
is
largely
an
administrative
role.
A
I
don't
have
any
problem
running
the
meetings
and
setting
up
times
and
such,
but
I
wanted
to
give
other
people
the
opportunity,
if
possible,
but
I
know
the
pr
has
been
doing
a
great
job
and
I
wanted
to
make
sure
that
he
gets
some
recognition,
but
also
I
wanted
to
make
sure
that
people
know
where
they
can
go
for
problems
and
questions,
and
things
feel
free
to
jump
in
anytime,
too.
I'm
basically
just
kind
of
going
over
these
things,
one
by
one.
A
One
thing
that
I
had
had
forgotten
to
add
on
this
also
is:
there
has
been
a
request
to
start
a
synchronous,
discussion
channel
other
than
irc,
and
so
I've
been
exploring
the
possibility
of
using
either
matrix
or
discourse,
and
if
people
have
opinions
on
that,
I'd
be
glad
to
hear.
There
are
some
some
challenges
with
each
of
those,
particularly
in
terms
of
creating
a
bridge
back
to
irc,
so
that
people
can
continue
to
use
irc
ir.
A
D
A
new
one
for
me
so
once
again,
okay.
H
G
H
D
A
So
one
of
the
purposes
for
this
was
to
was
to
enable
the
ability
to
share
code
snippets,
and
I
don't
know
how
easy
that
is
in
in
certainly
not
very
easy
in
irc
matrix
is
it
makes
it
possible,
but
I
believe
it's
a
little
bit
painful.
Somebody
with
more
experience
can
no.
H
A
H
Well,
there
I
mean
the
the
issue
is:
is
I
think
we
are
going
to
end
up
with
islands
of
discussion
yeah
we
some
of
us
aren't
going
to.
I
don't.
I
have
a
discourse
account,
but
I
never
am
on
it.
I,
the
matrix
irc,
is
useful
for
me,
but
now
it
is
also
not
zero
cost.
Let
us
be
clear
here
right.
F
C
H
Both
the
services,
it's
just
coming
out
of
a
different
pool
that
we
might
if
we
were
to
use
it,
a
lot,
might
need
to
also
chip
in
on.
F
A
A
What
has
mentioned
to
me
is
accessibility,
some
concerns
about
security,
which
I
would
it
didn't
quite
follow,
and
the
ability
to
share
code
snippets
they
basically
one
that's
more
feature
driven
than
irc
yeah.
Not
just
your
discussion
got.
F
D
Element
for
their
matrix,
client
or
is
there
a
I
guess
I
was
just
soliciting
alternatives.
A
H
Either
element
or
the
web
client,
you
can
also
there's
a
program
called
neon.
H
I
believe
you
can
use
pigeon
as
well,
quite
likely.
The
pigeon
experience,
though,
is
like
everything
else.
A
A
So
one
just
just
for
for
completion,
one
thing
that
has
come
up
in
discussion
is
slack
and
slack
is
both
slack
is
problematic
in
a
number
of
ways.
We're
not
not
really
using
that
as
a
viable
option.
F
Jefford,
does
it
make
sense
to
sort
of?
I
don't.
I
don't
think
we'll
make
a
decision
in
the
moment
here,
but
I
think
it's
also
important.
If
we
feel
like
this
is
a
is
a
way
that
we
might
get
some
broader
interest
and
and
be
more
accessible.
Maybe
we
should
have
a
working
group
and
just
say:
let's
decide
on
something
and
then
I
think
whatever
it
is,
we
should
throw
our
weight
behind
it.
You
know
we
should
say
this:
we're
we're
camping
here
and
rob
or
rob.
Please
come.
A
F
H
H
A
Meet
that
yeah,
I
would
agree
with
that.
Rich
has
pointed
out
also
that
discourse
does
have
a
traditional
mailing
list.
Interface.
H
Yeah,
I
would
say
that's
like
saying
that
you,
you
could
also
have
a
program
that
sends
irc
to
email
once
a
day.
I
have
not
found
it
very
useful,
my
it
garbages
it
the
stuff.
I
have
to
deal
with
on
code,
wise,
that's,
but
for
other
communication
it
probably
works
fine,
but
the
the
the
part
that
we
were
looking
for,
which
was
that
we
wanted
to
be
able
to
share
code
snippets
and.
A
A
Okay,
clark
can't
be
quick,
we
don't
actually
have
a
matrix
channel,
yet
hyperscale
sig
does
have
a
matrix
channel
and
they've
been
working
out
the
bugs
with
things
like
a
bridge
over
to
irc,
okay,.
A
D
A
Invented
here
I
think
that
we'll
probably
end
up
with
a
matrix
channel,
and
I
will
try
to
get
that
set
up
over
the
next
before.
D
I
would
suggest
we
try
it,
and
just
I
mean
I
don't
see,
I
don't
know.
No
one
here
has
said:
oh
discourse
is
better
than
matrix,
or
vice
versa.
Let's
just
pick
one
and
see
if
we
can
get
some
discussion
going
and
if
it
doesn't
work
out,
move.
D
You
know
I
I
got
my
irc
channels
and
I
I
never
use
them
on
my
phone,
which
is
kind
of
a
detriment
that
that
sometimes
I'm
away.
D
Of
course,
it's
sometimes
good
that
some
of
those
colonel.org
guys
can't
get
a
hold
of
me.
So
so
I
would,
I
would
say:
let's
do
something
modern,
I
I
don't.
I
don't
think
we
need
to
propagate
rrc,
even
though
I
like
it.
E
B
H
D
H
And
we'll
get
a
tattoo
we're
going
to
the
communication
in
a
channel
is
two-way.
There
has
to
be
people,
communicating,
oh
for
reason
there
and
there
hasn't
been
any
real
reason
to
communicate.
Here.
Are
our
fixes
and
patches
go
get
done
elsewhere?
Our.
H
There
hasn't
been
a
widespread
product
that
people
could
try
out
or
a
use
case
and
that
kind
of
cuts
down
the
need
to
have
any
communication,
because
it's
there's
not
a
lot
of
well.
I
had
a
problem
getting
this
to
work
on
my
raspberry
pi
or
I'm
trying
to
put
this
in
my
ford,
ref
50,
and
it's
it's
crashing
at
this
point.
How
do
I
fix
it?.
A
Synchronous
communication
tends
to
also
be
question
and
answer
so
it's
it
usually
is
people
who
are
having
difficulty
with
the
artifacts
that
exist
as
opposed
to
having
difficulty
creating
those
artifacts.
I
find
those
tend
to
be
either
happening
in
tickets
or
on
the
mailing
list,
as
they
are
yeah
so.
D
Well,
it'll
be
a
faster
communication
channel
once
we
we
start
to
have
multiple
folks
working
on
similar
systems
and
you
have
to
make
decisions.
What
are
we
going
to
name
this?
What
apis
do
we
need?
I
don't
like
your
api.
I
want
you
to
use
my
api,
so
let's
have
a
fight.
You
know
I
mean
you
know
the
typical
developer
discussion
and
I
think
a
a
synchronous
channel
can
facilitate
a
lot
of
that.
A
A
A
What
would
we
want
to
see
supported
and,
more
importantly,
how
do
you
think
we
might
be
able
to
fund
such
a
thing.
A
F
I
think
this
actually
came
up
in
the
december
meeting
a
little
bit
and
yeah
and
it
seems
a
great
idea.
I
think
the
feedback
in
december
was
the
nature
of
the
reference
hardware
in
this
space
makes
it
less
likely
that
other
than
other
than
a
handful
of
well-known,
truly
publicly
available
things
like
the
the
raspberry,
pi
and
nvidia
has
relatively
easily
acquired
reference
platforms
for
some
of
their
iot
stuff,
which
is
also
the
foundation
for
some
of
their
automotive
work.
Bruce
benson.
F
Thank
you,
gosh
that
this
is
worth
trying,
but
we
shouldn't
really
count
on
this
that
a
lot
of
hardware
in
the
space
is
relatively
tightly
controlled.
So
the
notion
of
having
a
public
board
farm
is
not.
You
know.
We
could
certainly
try
and
change
hearts
and
minds,
and
that
would
be
wonderful,
but.
D
It
it's
really
useful
for
generic
parts
of
the
framework,
like
you
know,
ensuring
that
the
kernel
you've
got
is
is
doing
what
it's
supposed
to
do.
Is
it's
interfacing
properly
with
the
systemd,
and
you
know,
services
up
in
user
space,
but
it
gets
really
cost
prohibitive
and
labor
intensive.
When
you
start
doing
things
like
saying,
let's
add
in
this
particular
gpu
that
we
can
use
for
a
a
neural
net,
you
know
for
for
onboard
eight
ass
calculations.
D
So
I
would
love
to
see
us
have
a
board
form
so
that
we
could
do
things
like
ci
and
if
we
could
get
the
right
mix
of
devices
in
there,
it's
very
nice
for
somebody
who
doesn't
have
access
to
a
qualcomm
snapdragon
or
you
know,
an
nxp
board
or
something
like
that
to
be
able
to
say.
Oh,
let
me
reserve
that
provision
it.
D
A
A
And
I
will
go
back
and
I
hope
would
you
happen
to
know
if
the
december
meeting
was
recorded.
F
A
D
D
Well,
I
guess
what
I'm
saying
is:
let's,
let's
go
a
little
further
until
we
have,
you
know
a
a
solid
community.
Okay,
I
think
you're
doing
a
fine
job.
I
don't
know
what
anybody
else
thinks,
but.
E
I
wanted
to
use
this
as
a
bruce
learns
about
the
incubation
of
open
source
projects
to
be
part
of
a
centos.
Sig
does
centos
want
this
to
grow
and
have
like
I'm
going
to
say,
scary
stuff,
like
bylaws
and
succession
and
stuff,
like
that,
almost
like
it's
incorporated
to
shepherd
any
to
provide
assurance
that
source
code
is
shepherded
along
through
some
some
kind
of
life,
or
it
would
maybe
automatically
subsume
under
all
of
the
things
that
said,
toss,
I'm
not
saying
anything
is
or
isn't
being
done.
G
The
centos
project
does
not
expect
cigs
to
have
their
own
bylaws.
The
centers
project
does
expect
sigs
to
abide
by
the
overarching
sentence
code
of
conduct
and
existing
bylaws,
which
are
fairly
minimal.
But
beyond
that,
we
just
want
you
to
be
functional
and
play
nice.
E
Okay,
yeah
and
I'm
not
a
big
bylaws
kind
of
guy.
I'm
I
like,
even
though
I
raise
my
hand,
I'm
more
of
a
wild
west
character.
F
It's
meant
to
my
understanding
by
the
way
is
ours
is
among
the
more
rigorously
run
already
in
this
space.
Richard
nodding
vigorously
right
we're
we're
already
engaging
in
sig
best
practices.
E
A
A
A
I
I
make
mention
of
the
centos
stick
wherever
it's
no,
but
one
thing
I
would
like
to
know
is:
how
closely
does
this
want
to
follow
those,
and
vice
versa,
do
we
are
we?
Are
we
hopeful
that,
for
example,
sophie
will
be
of
our
centos
sig.
A
Maybe
nobody
this
may
be.
I
may
be
talking
to
nobody,
I'm
not
sure
anybody
has
an
opinion
about,
but
the
eclipse
software
defined
edge
project
in
particular.
I
think
opportunity
there
for
our
our
eventual
architecture
to
follow
the
architecture
by
the
group
and
and
that
we
get
that
process
and
vice
versa.
E
I'm
glad
you
brought
that
up,
I
ran
across
there,
the
eclipse
stuff
and
it
there
seemed
some.
You
know
I
got
the
vibe
there
something
synergistic
there
and
I
didn't
have
time
to
look
into
it
deeper
and
I've
I've
I've
enjoyed
using
some
eclipse
products
at
a
deep
level,
not
that
that
directly
relates
to
a
foundation
project.
A
Yeah
you
can
get
a
clue
about
how
eclipse
runs
open
source
initiatives
by
having
with
the
the
ide,
for
example,
has
grown
to
be
quite
a
bit
larger
than
just
the
id
they're
similar
to
the
linux
foundation,
in
that
they
host
initiatives
based
on
specific
things,
there's
a
link
for
it
and
since
red
hat
is
involved
with
and
and
has
a
number
of
other
things
on
radar.
Even
if
we
aren't
directly
involved
red-headed,
I
just
wondered
how
much
of
a
how
much
feedback
we
should
we
sent
us
with
these.
E
I
you
know
I
get
when
I
when
I
look
at
eclipse.
I
can
kind
of
paint
a
sketch
in
my
my
mind
of
oh
there's,
a
a
pattern
fly
ui
where
you
kind
of
like
the
cockpit
os
tree
builder
thing
and
then
something
to
slurp
up
code
like
microshift.
Perhaps
and
then,
since
we
do
projects
that
use
che,
you
could
have
an
iede
pop
up
and
then
you
have
like
here's.
The
target
push
the
button
poof
it
shoots
it
into
the
board
and
and
off
they
go.
E
So
there
seem
to
be
synergistic
elements,
so
it
does
seem
interesting
to
track
these
guys.
B
A
Okay,
well,
I
can
make
sure
to
kind
of
mention
in
these
formal
meetings,
what
I
have
seen
going
on
publicly
in
the
software
defined
edge
project,
and
maybe
vice
versa,
create
a
link
there.
A
Okay
and
then,
of
course,
the
larger
question
is:
what
do
we
want
to
see
in
2022
from
this
group?
We've
got
about
12
minutes
left.
F
I'll
I'll
throw
out
there
something
that
you
know
this
is
either
at
the
risk
of
stating
the
obvious
or
exciting
news.
I
don't
know
how
it
will
land
for
everybody.
I
think
we're
going
we're
going
to
put
more
emphasis
on
red
hat's
contribution
to
the
sig,
providing
something
that
looks
more
and
more
like
the
sort
of
upstream
version
of
the
automotive
os
product.
F
So
we're
already
started
some
tentative
steps
in
that
direction
in
terms
of
where
they
differ,
that
clark
and
francisco
and
other
members
of
the
the
kernel
effort
associated
with
our
program
are
working
on
so
and
a
big
part
of
that
will
be
essentially
trying
to
mirror
some
of
the
stream
like
workflow
that
occurs
with
rel
for
anything
that
is
associated
with
the
red
hat
automotive
product.
F
F
Further
to
the
point
we
were
making
with
bruce
earlier.
These
are
kind
of
very
open-ended
low-key
groups
in
the
sig,
so
it
you
know
other
things.
You
know
there
could
be
initiatives
to
sort
of
facilitate
hardware
enablement.
There
could
be
work
on.
You
know
larger
frameworks,
I'm
not
entirely
sure,
but
we're.
Certainly,
I
guess
I'm
just
announcing
that
you'll
see
more
and
more
some
efforts
to
formalize
how
we
have
an
upstream
for
the
in
vehicle
os
product
very
much
inspired
by
how
centos
stream
works.
F
A
The
intention
for
this
sig
has
been
to
do
this
within
centos
as
a
as
a
public
base,
as
opposed
to
being
just
an
upstream
for
red
hat,
and,
and
so
I
would
love
to
see
that
I
would
love
to
see
us
bring
in
maybe
mirror
some
code
or
or
be
a
sort
of
collecting
repository
for
enabling
software
like
middleware.
A
You
know
ross
is
most
relevant
right
now
to
automated
driving
and
driving,
but
but
it's
a
very
robust
community,
it's
a
great
operating
system,
but
it's
a
great
set
of
middleware
and
I
would
love
to
see
us
support
it
to
some
degree.
A
C
D
F
F
C
C
F
I
know
we've
seen
that
there
have
been
various
efforts
at
this.
It's
usually
something
that
someone
really
ruminates
on
it
for
quite
some
time
and
shares
what
they've
done.
There's
not
a
lot
of
commonality
in
it.
So
like
rapid
boot,
saying
that
the
automotive
kernel,
which
is
the
first
thing
that
we've
really
surfaced
as
a
unique
artifact,
you
know
the
first
thing
to
work
on
that
would
be
to
well.
F
F
F
D
I
don't
I
don't
know
what
the
current
pi
numbers
are,
but
so
suddenly
we're
down
to
1.2
seconds
to
have
the
kernel,
functional
and
user
space
running
and
drivers
initialized
and
the
camera
producing
frames
and
the
lcd
displaying
them.
So
I
mean
it's
a
challenge,
I'm
not
saying
we
can't
do
it,
but
I
kind
of
suspect,
at
least
in
early
days
we're
going
to
be
cooperating
a
lot
closer
with
firmware
than
you
know.
We're
used
to.
F
F
D
We're
looking
at
things
like
trying
to
parallelize
initialization,
you
know
I
mean
we
do
have
this
cool
facility
of
kernel
threads
that
we
we
could
kick
off,
because
we
have
multiple
cores
and
do
device
initializations,
but
then
there's
a
rendezvous
problem.
You
have
to
deal
with
independency
problem
so,
but
I
think
it's
all
good
and
I
would
I
I
would
love
to
see
the
sig
dive
deep
into
that.
To
say
how
can
we
restructure
things
and
and
the
other
side
of
it
is?
C
Clearly
everybody
wants
that,
but
we
have
to
be
careful
that,
as
it
was
said
before,
some
people
have
done
proof
of
concept,
but
mostly
with
security,
disabled
and,
as
today,
with
a
new
cyber
security
regulation
coming
in
it
has
to
handle
the
security
as
well
and,
for
example,
just
having
a
a
nice
configuration
with
you
know
the
root
partition
in
read-only
and
all
that
stuff
is
very
specific
to
the
embedded
space.
But
it's
very
very
important.
So
I
agree
it's
it's
it's
closely
related
to
the
kernel.
C
D
C
And
cyber
security
regulation
is
my
data
from
june
up
this
year
for
anybody
so
155
and
156
for
u.s
united
nations.
Well,
I
don't
know
what
155
means.
I'm
sorry,
it's
you
know
more
or
less.
It
all
says
the
same
thing.
You
have
to
look
for
all
your
assets.
You
have
to
describe
how
your
set
can
be
attacked
and
you
have
to
describe
how
you're
going
to
remediate
if
one
of
your
asset
is
attacked
more
or
less
they
all
have.
C
You
know
this
general
idea
in
mind
and
you
need
an
external
authority
to
say.
Yes,
you've
done
the
the
work
that
you
should
done.
So
it's
it's
okay
to
go
to
go
on
the
road,
but
it's
a
big
big
thread
in
the
automotive
sector
today,
not
only
in
the
automotive
but
especially
in
the
automotive
sector,
and
it's
going
to
be
mandatory
from
june
in
u.s
and
from
december
in
europe
and
december
and
japan.
E
The
draft
of
the
cyber
for
automotive
from
iso
and
it
there
were
a
lot
I
am.
I
am
a
former
information
security
auditor
and
I'm
an
embedded
practitioner,
so
I
try
to
stay
abreast
of
these
things.
So
if
you've
got
something
on
the
horizon,
I'd
I'd
love
to
hear
it
is
where
I'm
going
with
that.
But
what
I'd
seen
so
far
had
a
pretty
standardized
kind
of
thing
and
the
automotive
specific
stuff
seemed
like
it
was
still
evolving.
C
A
C
A
A
What
I'm
going
to
do,
maybe
is
set
up
the
next
meeting
in
this
time
same
time,
space
as
a
as
a
an
office
hour,
so
we
can
just
kind
of
come
into
general
discussion
and
we'll
do
that
in
two
to
three
weeks.
A
So,
thank
you
all
for
coming
and
thank
you
for
this
very,
very
interesting
stuff.
Here,
I'm
gonna
post
this.
I
will
make
these
slides
available.
Obviously,.