►
From YouTube: CHIPS Alliance TSC Meeting - 2021-03-24
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
All
right
welcome.
Everybody
today
is
march
24th.
This
is
the
tsc
meeting
for
the
chips
alliance
got
a
number
of
folks
here
on
on
the
call
I've
got
rob,
maines
have
henry
cook
and
michael
giada.
This
is
brian
warner
henry.
You
want
to
call
us
to
order
here.
A
Sounds
good
we
have
the
the
meeting
agenda
is
in
github,
I've
got
ships
alliance,
slash
tsc,
and
it's
in
the
meetings
directory.
A
I
have
a
couple
things
on
the
agenda
for
today,
so
we
recently
had
a
an
email
vote
to
establish
an
analog
working
group,
so
just
reviewing
the
results
of
that,
there
were
two
out
of
two
votes
in
favor
of
it.
That
was
on
march
5th.
So
maybe
the
first
thing
we
can
do
in
today's
meeting
is
discuss
what
exactly
this
means
and
what
the
next
steps
are.
A
I
had
put
in
a
draft
document
which
basically
has
you
know
some
some
basic
biographical
information
about
the
working
group
and
we
need
to
maybe
figure
out
where
to
go
from
there.
We've
had
one
application
come
in
for
sandbox
from
aib.
We
can
review
that
and
hopefully
vote
on
that
as
well
and
then.
Finally,
I
think
there
are
we'll
probably
want
to
maybe
just
discuss
a
little
bit
about
next
week.
The
chips
alliance
workshop
which
is
coming
up.
A
A
C
C
C
A
There
we
go
just
one
moment
here:
we
can
bring
this
up.
A
So
here
are
the
meeting
minutes
typically,
for
you
know
for
the
future
whoever's
capturing
the
minutes,
whether
it's
me
or
somebody
else.
You
know
it's
good
to
get
these
up
kind
of
early
and
then
just
wait
for
people
to
take
a
look
through
them
and
approve
them,
and
once
we
get
a
enough
approvals
on
them,
then
we
can
just
merge
them
directly
in,
but
last
meeting
was
pretty
straightforward.
A
We
went
through
the
sandbox
applications
for
rocket
chip
and
for
chisel
ended
up
voting
to
accept
the
sandbox
applications
went
through
the
cla
and
then
also
discussed
the
chips
alliance.
Commons.
A
So
if
you
guys
are
both
good
with
this,
then
just
go
ahead
and
approve
the
pr.
We
can
merge
that
in.
A
The
next
thing
on
the
agenda
was
to
discuss
the
analog
working
group
so
just
kind
of
recapping
some
of
the
discussions
which
have
happened
in
various
places,
both
the
last
tsc
meeting
and
then
also
in
the
governing
board
meeting.
A
We
discussed
forming
a
a
an
analog
working
group,
which
is
basically
to
be
a
place
where
we
can
do
work
in
advance
of
having
analog
projects
merged
into
chipset
lines.
The
idea
here
is
that
their
you
know,
our
existing
governance
basically
says
work
groups
are
formed
either
by
the
tsc
voting
that
this
is
something
they
need
to
focus
on
and
want
to
have
a
place
to
to
be
able
to
collaborate
with
each
other.
A
A
That
said,
michael
you
and
you
and
henry
are
both
certainly
better
versed
in
this
need
than
I
am
so
I
put
in
a
pr
with
an
initial
with
an
initial
file.
Initial
description
of
the
working
group,
but
we
could
really
use
somebody
to
contribute,
maybe
a
mission
and
some
other
you
know
other
information
on
what
exactly
the
scope
of
this
would
be
and
how
often
you
want
to
meet
and
so
forth.
So
it'd
be
great.
A
If
somebody
could
just
maybe
push
some
patches
against
this
against
this
document,
and
we
can
start
using
this
as
the
base
for
for
the
work
around
the
the
working
group.
Does
that
make
sense.
D
To
multitask
so
that
we
don't
get
installed
on
like
so
with
those
descriptions.
Definitely
at
some
point
we'll
probably
have
to
have
those
descriptions
correspond
to
the
stuff
we
have
on
the
website
as
well,
which
is
always
a
pain
to
keep
up
to
date.
So
we're
probably
eventually
going
to
distribute
that
between
the
working
groups
themselves,
just
to
make
sure
that
these
things
are
today.
D
But
I
think
yes,
we
should
have
the
description
of
the
working
groups
and,
of
course,
the
analog
group
that
we
just
established
is
a
special
case
where
we're
trying
to
like
create
that
ecosystem
and
the
enthusiasm
around
it.
But
I
think
it's
it's
worthwhile
to
like
keep
a
list
and
an
explanation
of
you
know
what
do
the
groups
actually
do
and
the
repository
is
probably
easier
to
do
really
for,
for
you
know,
technical
people
and
then
the
the
website
we
could
base
the
information
there
on
whatever
is
in
the
repo.
D
It
will
be
a
good
source
of
information
that
we
could
ask
the
working
groups
to
keep
up
to
date,
and
we
should
just
have
to
care
about
the
marketing
message
reflecting
the
what
the
technical
groups
are
doing,
but
as
it
is
today,
it's
like
marketing
content
that
actually
kind
of
depends
on
technical
people.
A
Yeah,
that's
a
really
good
point,
michael,
and
it
does
also
underscore
the
reason
why
we're
asking
existing
chips
alliance
projects
to
go
through
this
whole
process
as
well.
I
mean
there's
the
best
way
to
know
how
to
describe
the
projects
that
are
in
the
chips
alliance
is
to
have
the
projects
actually
put
that
together
themselves
and
to
do
it
in
a
consistent
way.
A
So
that's
kind
of
a
nice
segue
into
the
aib
application
which
came
in
for
the
sandbox
projects,
so
we're
cruising
right
along
today.
Do
you
want
to
move
into
that.
A
That'd
be
awesome,
yeah,
I've,
I've
put
in
the
basics
here
I
can
put
in
we're
probably
going
to
need
a
zoom
link
and,
let's
see
anything
else,
it's
probably
the
main
thing
I'll
I'll
generate
a
custom,
zoom
link
once
you
decide
when
you
want
to
meet,
let
me
know
and
I'll
put
an
entry
on
the
calendar
as
well.
I
mean
these
are
all
things
which
are
available
to
chips
alliance
projects,
so
the
same
things
that
are
available
to
the
project
should
also
be
available
to
the
working
group.
D
Yeah,
the
meetings
will
probably
start
happening,
not
immediately
in
the
sense
that
we
need
to
first
kind
of
get
back
to
the
people
already
in
the
chips
alliance
and
and
get
them
to
get
involved.
I
I
mean
specifically
intel
and
berkeley
and
then-
and
you
know
a
lot
of
our
planning
around
how
to
get
people
to
you
know,
put
their
analog
stuff
in
ships
right.
D
This
could
happen
in
this
forum
instead
of
like
the
tsc
meetings
or
our
marketing.
So
to
establish
those
meetings,
I
I
don't
think
we
should,
like.
You
know,
like
first
establish
a
meeting
and
then
watch
nobody
come
right
like
we
have
to
come
and
ask
a
bunch
of
people
that
we
know
are
interested
in
this
to
set
up
that
meeting
and
and
then
we
can
have
you
know,
external
people
also
join
like
they.
D
Don't
necessarily
need
to
even
be
part
of
chips
alliance,
because
our
technical
meetings
are
open
and
once
those
meetings
are
happening
regularly,
then
we'll
have
this
data,
but
like
just
just
finding
a
time
without
actually
having
people
to
meet
is
not
possible.
So
you
know
the
calendar
entry
for
this
will
be
empty
for
now
right.
D
D
Faith
they've
shown
a
lot
of
interest,
but
also
kind
of
they've
been
very
kind
and
and
supporting
this
initiative,
so
I
think
hopefully,
they'll
be
interested
to
to
to
join
such
a
meeting
and
then,
of
course,
we
can
also
join
it
and
like
help
it
tag
along
with
the
hope
of
you
know
getting
more
members,
and
even
if
you
know
some
people
don't
join
us
members,
they
might
start
joining
our
meetings
when
those
meetings
become
regular.
D
E
That's
why
that's
why
that's
yeah
that
that's
why
I
asked
my
question
was
a
loaded
question.
I
guess,
but
you
know
in
terms
of
like
seeding
initial
agendas
that
you
know
like.
I
could
share
with
professor
pan
at
ut
austin,
to
see
if
he
would
be
interested
in
either
himself
participating
or
having
one
of
his
students
participate
and
the
same
thing
for
professor
kong
at
ucsd.
D
D
So
if,
if
there's
people
interested
in
generally
speaking,
making
analog
design
more
reusable
and
open
source
and
not
just
working
away
in
their
own
corner
but
like
potentially
collaborating
with
others
and
integrating
different
tools
together-
and
I
think
reusing
concepts
and
and
potentially
code
between
their
projects,
they
should
join
proactively
and
we
can't
up
front
tell
them
what
the
agenda's
gonna
to
be
because
it's
going
to
be
dictated
by.
You
know
the
stuff
that
happens.
Look
at
the
tools,
work
group,
for
example.
D
D
So,
of
course
I
don't
know
what
your
discussions
with
with
those
people
have
been,
but
hopefully.
E
I
haven't
had
interest,
I
haven't
had
any
discussions
at
all,
I'm
just
also
just
I'm
thinking
from
this
perspective,
right
that
I
guess
maybe
from
a
business
perspective
or
company
perspective
right,
but
if
we
want
someone
to
set
time
aside
on
their
calendar,
it
helps
to
have
an
a
initial
idea
as
to
what
the
meeting
is
going
to
be
about
see
what
I'm
saying
so
so,
let's
you
know
right
so,
let's
say
like
where
the
idea
is
to
develop
a
common
or
shared
analog
tool
framework
right,
that's
broken
down
to
different
into
different
component
architectures
right,
so
perhaps
borah
could
leave
lead
like
an
introductory
conversation
on
that.
E
You
know.
Andrew
khan
could
talk
a
bit
about
what
open
road
is,
and
then
maybe
professor
pan
could
talk
about
how
some
of
his
work
could
plug
into
such
a
framework
and
then
start
driving
towards
an
open
contributory
framework.
That's
an
idea
yeah
anyway,.
D
D
D
You're
right,
I'm
not
knowledgeable
enough
about
analog
to
to
to
try
to
dictate
that
and
of
course,
borah
and
and
dave
can
can
help
but
yeah
it.
D
Of
topics
it
should
be
like
one
is
like,
generally
speaking,
tools,
analog
tools
and
methodologies
and
another
should
be
just
analog
ip
and
making
people
aware
of
all
the
ip
that
exists
and
potentially
looking
at
how
to
do
effective
tape.
Outs
of
existing
analog
ip
using
skywater.
You
know
like
that,
I
think,
is
the
kind
of
topics
that
should
be
covered
in
in
the
analog
work
group
and
in
general,
like
just
just
changing
people's
approach
to
analog
and
trying
to
think
how
things
can
be
automated
better
and
how
costs
can
be
shared.
D
D
And
again,
we'll
have
to
explain
that
you
don't
need
to
put
your
iap
interchips
up
front
right
like
that's,
not
what
we're
asking
you
to
do
because,
like
all
of
those
people
we're
trying
to
talk
to,
these
are,
of
course,
people
with
some
kind
of
ip
or
most
of
them,
especially
the
universities.
So
they
might
think
chips
is
coming
at
them
to
like
hijack
their
ip.
D
D
A
Okay,
so
we'll
just
leave
that
pr
open.
Then,
when
everybody's
comfortable
with
it,
you
can
just
merge
it
in
it's
great
to
see
this
moving
forward,
though,
thank
you
for
all
the
help.
On
all
that,
that's
context
I
wouldn't
have
been
able
to
add
myself.
A
Okay,
so
moving
forward
here,
the
just
one
moment
to
bring
my
screen
back
up
there
we
go.
The
only
other
thing
that
I
had
for
today
was
to
go
through.
A
A
However,
it
would
be
good
for
the
two
of
you
as
the
tsc
members
to
review
these
sandbox
applications,
and
then
we
can
provide
any
feedback
that
there
may
need
to
be,
and
then
hopefully
vote
to
accept
these
projects
in
so
first
project
was
aib.
This
one
came
in
just
a
few
days
ago.
Here
it
came
in
from
dave
kellet
at
intel.
A
There
are
a
couple
of
things
here:
yeah
the
this
specific
application
looks
like.
It
covers
both
specification
and
some
of
the
code
and
utility
repos
for
aib.
I
had
responded
back
to
dave
and
had
asked
if
he
could
also
enumerate
out
the
other
open
source
licenses
that
are
involved
here.
Given
the
specification
license
is
owfa
1.0
currently,
so
we
should
probably
have
that
here
in
the
application
as
well.
A
A
couple
of
other
things
we
ought
to
include
in
here
looks
like
need
some
links
to
the
issue
trackers
and
then
maybe
a
statement
of
why
the
project
wants
to
join
the
the
chips
alliance.
Once
we
have
that,
I
think,
though
we
could
probably
handle
this
offline
instead
of
waiting
until
the
next
tsc
meeting.
Does
anybody
have
anything
else
they
would
want
to
add
on
that.
B
No,
that
seems
fine.
I
think
we
should
maybe
make
it
clear
to
people
that,
like
you,
can
use
github
itself
as
an
issue
tracker
like
you,
don't
necessarily
need
to
have
a
separate
thing,
set
up
good
point
and
maybe
same
for
the
website.
In
the
absence
of
a
separate
website.
To
like
read
me
on
your
top
level,
repo
is
kind
of
your
become
your
your
landing
landing
page
yeah.
A
That's
that's
a
good
point
henry.
I
think
in
general,
we'd
want
to
communicate,
don't
don't
over
complicate
this
yeah.
B
A
Yeah
I
can
I'll
do
that
I'll
update
the
template,
for
that.
I
think
we
can
probably
fast
track
that
that
seems
like
it's
pretty
yeah
useful,
but
also
trivial
change,
so
yeah.
We
can
take
an
action
and
do
that.
A
Okay,
so
I'm
going
to
say
on
this
one,
we're
probably
not
quite
ready
to
go
through
and
review
it
just
yet.
I
did
see
a
pr
to
revert
this,
so
it
may
be
that
it
was
landed
prematurely
anyhow.
So
we
can
I'll
follow
up
with
dave
on
that
one
there's
also
the
application
for
omni
extend
this
came
in
from
western
digital.
A
D
Oh
you're
right,
I
didn't
see
that
this
is
the
spec
right.
D
Yeah,
and
that,
I
think,
is
why
they
they
chose
like
oh
wait.
I
need
to
open
a
new
window-
creative
commons,
which
is
not
ideal,
of
course,
but
you
don't
typically
choose
it
for
code.
D
A
Okay,
well
so,
let's,
let's
maybe
go
back
to
to
dijon
here
and
just.
D
A
Okay,
all
right
I'll
I'll,
take
an
action,
I'll
respond
back
to
dejan
here
and-
and
we
can
get
this
corrected
and
then
maybe
again
for
this
one
do
a
vote
out
of
band
from
the
tsc
meetings.
Are
you
both
comfortable
with
email?
Yeah
sounds
good
and
then
we'll
just
maybe
follow
the
same
process.
A
We
used
for
the
last
vote,
the
one
that
we
used
to
set
up
the
the
analog
working
group,
where
I'll
just
I'll
send
an
email
out
to
the
tsc
record
the
results
and
then
next
tsc
meeting
will
just
go
through
and
review
any
votes
which
happened
over
email
just
so
that
it's
on
the
record.
A
Okay,
okay,
great
henry
and
michael:
do
you
have
anything
else
for
today.
E
I
did
have
one
quick
question
myself.
Oh
rob
sure
I
just
this
is
just
on
the
on
the
artificial
intelligence
work
group
does
that
exist
yet
and
anything
you
know
of
note
on
that.
The
reason
I'm
asking
is
because
I
was
approached
by
the
ai
hardware
summit
about
having
chips
alliance
participate
in
that,
at
least
as
a
you
know.
As
a
you
know,
friendly
mention
of
what
our
involvement
and
then
also
asked
about.
E
If
I
would
be
willing
to
host
a
panel
conversation
and
then
also
I
was
considering
approaching
a
contact
at
ibm
in
the
artificial
intelligence
group,
there
he's
a
ibm
fellow
and
then
also
a
microsoft,
distinguished
engineer
in
artificial
intelligence
as
well.
So
I
just
wanted
to
see
if
there
was
anything
I
should
be
aware
of
relative
to
that
topic.
E
These
are
the
other
panelists.
I
didn't
catch
that!
No,
I'm
sorry,
so
this
is
about
they're,
not
panelists.
So
the
one
gentleman
is
in
the
art
of
is
a
ibm
fellow
in
the
artificial
intelligence
area.
He
is
at
mit
and
he
does
work
with
the
ibm
chip
development
work
in
artificial
intelligence,
which
is
in
albany
new
york.
E
C
E
D
D
You
know
ai
stuff,
that's
strictly
dependent
on
their
own
proprietary
tools,
which
kind
of
makes
it
difficult
to
like
build
a
community
around
it
because,
like
obviously
anyone
using
that
will
need
to
be
also
using
a
specific
set
of
tools.
We
just
don't
have
an
ip
in
that
area.
Now.
We'd,
of
course
like
to
be
proactive,
and
definitely
it
makes
sense
to
reach
out
to
these
people.
D
But-
and
there
are
some
interesting
projects
out
there
in
machine
learning,
but
some
of
them
are,
for
example
like
like
tvm
as
an
example,
it's
a
framework
which
also
has
vta,
which
is
ip
for
for
machine
learning
on
fpga,
but
that
is
part
of
apache
foundation.
I
think
right,
for
example,
so
I
I
don't
think
they're
looking
to
join
something
like
alliance.
D
Instead,
they
could
collaborate
and
I
they
should
collaborate
and
would
love
to
collaborate
with
them,
but
if
we
can
identify,
you
know
like
projects,
be
it
accelerator,
ip
or
or
some
kind
of
a
machine
learning
related
tooling
for
for
asic
development
or
related
to
ac
development.
D
That
would
be
a
great
you
know
kernel
to
build
around
such
a
working
group,
and
certainly
it's
worthwhile
talking
to
people.
You
know
in
this
area
to
try
to
make
that
group
happen.
Some
of
the
work
we
could
also
connect
to
and
potentially
even
incorporate,
is
you
know
the
the
cfu
work
we're
doing
with
risk.
Five
like
fpga,
based
acceleration
of.
D
There
isn't
much
we're
in
contact
with
at
least
today.
Okay,
there
might
be
you
know,
we
should
definitely
talk
to
digital
about
it,
because
I
know
the
interest
to
form
such
a
group,
and
so
on
has
been
coming
also
from
them
and
there
is
somewhere
they're
involved
with
which
potentially
they'd
be
interested
in
seeding.
This
group
with
but
yeah
nothing,
nothing
very
concrete,
like
we'd,
have
to
talk
to
people
and
like
come
up
with
a
plan
like
as
compared
to
the
analog
ware
group,
for
example.
D
There
it's
like
a
concrete
group
of
companies
that
seemed
at
least
to
be
willing
to
collaborate
about.
You
know
around
specific
projects,
also
tentative
and
also
you
know
proactive
on
our
from
us,
but
at
least
there's.
You
know
some
concrete
actions
to
be
taken,
like
you
know,
make
those
people
talk
to
these
people
and
like
let's
have
this
project
join
chips.
Alliance
et
cetera,
it's
it's
something
we
that
make
happen,
but
at
least
we
know
what
in
in
the
ai
space.
D
A
Okay
and
last
thing,
so
would
the
three
of
you
like
to
say
something
about
the
workshop
is
coming
up
next
week.
E
A
Okay,
I
don't
have
anything
else
for
today.
Henry
michael
rob
anything
else,
you'd
like
to
add
not
from
my
side.
A
All
right
henry:
would
you
like
to
close
this
up,
then.
B
Yep.
This
concludes
this
meeting
of
the
technical
steering
committee.