►
From YouTube: Diagnostics WG meeting March 24 2021
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
B
C
C
C
B
Yeah,
let's
keep
it,
then
I
mean
if
it
is
something
which
will
get
executed
in
in
a
reasonable
time
frame.
Then
it's
worthwhile
to
keep
it
so
that
we
can
track
the
progress.
C
Just
been
a
bit
chaotic
originally
should
hopefully
have
time
this.
B
Michael,
I
don't
see
the
item
which
you
are
putting
in
the
minutes.
That
is
report
idle
time
correctly.
Do
you
mean
to
have
a
discussion
around
that.
C
Yeah,
that's
just
I'd
like
to
get
some
people
looking
at
that.
It's
fixing
the
cpu
profiler
doesn't
correctly
report
idle
time.
This
is
actually
like
restoring
some
old
functionality
that
was
removed.
She
was
incomplete.
C
So
basically,
the
the
change
before
it
had
like
between
the
prepare
and
check
phases
of
libby
b.
It
would
set
idle
to
true,
which
is
mostly
true
other
than
in
the
I
o
phase.
Any
callbacks
that
happen
should
be
idle
false
and
so
before
it
didn't
do
anything
about
callbacks,
so
the
functionality
was
removed.
So
I've
restart
all
that,
and
then
it's
like
literally
two
lines,
just
added
to
the
internal
callback
scope
to
turn
the
idle
off
until
it's
done
the
scope.
C
Yeah,
so
the
the
reason
it
was
removed
before
is
because,
like
there's
a
whole
bunch
of
stuff
that
is
like
clearly
not
idle
activity
that
would
be
attributed
as
idle
activity
and,
like
that's
just
clearly
wrong.
With
with
this,
we
would
still
have
like
anything
that
doesn't
actually
use
make
kodak
would
be
attributed
as
idle,
which
is
not
ideal,
but
anything
that
doesn't
use
make
callback
is
already
broken
for
async
hooks
as
well.
C
C
E
C
So
native
modules
would
essentially
report
the
same
as
they
already
do.
It'd
just
look
like
idle
time
like
it
always
has,
or
no
it
it's.
C
It
would
become
idle
time
now
before
it
would
have
looked
like.
Basically,
the
process
is
never.
C
But
now
it
will
look
like
has
a
bunch
of
idle
around
when
the
event
loop
is
waiting
for
stuff,
but
any
anything
that
doesn't
be
a
function
call
directly
without
make
callback.
C
I'm
not
really
sure,
but
I
I
don't
know
if
that
necessarily
matters
here
since,
like
what
the
cpu
profiler
reports
in
the
first
place
is
not
really
a
reflection
of
the
entire
process.
It's
a
reflection
of
the
context
or
I
guess
the
isolate.
A
B
B
B
C
C
C
B
B
Michael
last
week
we
had.
In
the
last
instance,
we
had
richard
yep,
who
presented
the
idea
of
breaking
down
the
diagnostic
best
practice,
documentation
into
fine
grained
work
items
or
to
see
what
is
the
gap
with
respect
to
the
progress
in
the
documentation
and
see
how
can
he,
I
wasn't
sure
the
aspect
was,
how
can
he
contribute
or
how
can
he
make
an
assessment
about
what
is
the
actual
work
required.
E
I
think
it
was.
It
was
to
start
trying
to
get
it
moving
forward
and
to
like
he
was
going
to
try
and
pick
up
one
or
two
of
the
sections
to
move
it
forward.
So
the
first
step
was
to
kind
of
like
break
it
down
into
like
what
the
next
steps
would
be
from
where
we
are
yeah
and
I'm
just
looking
to
see
if
he'd
created
those
issues
here,
because
I
thought
he
for
some
reason.
I
thought
he.
B
B
E
B
B
We
did
not
have
the
complete
user
journey
document
yeah,
but
whereas
at
this
point
we
have
it
so
probably
it's
worthwhile,
instead
of
or
apart
from
looking
at
the
current
structure
in
the
documentation,
folder
and
see
what
is
in
the
and
what
is
not
in
there.
B
B
E
E
E
So
yeah
I'll
I'll
follow
up
with
richard
to
see
I
for
some
reason
I
thought
he'd
created
something,
but
I
do
agree.
I
can't
see
it
there
and
it
should
be
fairly
new
in
the
the
list
right.
B
E
B
Okay,
then
that's
it
for
today,
let's
regroup
in
two
weeks
time.