►
From YouTube: ORI FPGA Standup 12 July 2022
Description
DMA memory address bus error, comparison to PLUTO connections, and Remote Lab South construction and bacteriophage R&D (cross-posted)
A
A
So
so
we
can
write
code
in
the
ide
for
the
zc706,
but
what
we
found
out
yesterday
and
the
day
before
the
day
before
we
really
didn't
know
what
was
going
on,
but
definitely
yesterday
it
was.
It
became
very
clear
that
the
zc
on
the
zc706
of
the
arv
9371
that
the
encoder
once
it's
hooked
up
in
the
hdl.
A
We
can't
access
it
through
dma.
So
we
get
a
bus
error.
This
bus
error
message
is
not
obvious.
Invite
us,
which
is
disappointing,
but
we
can
see
it
clearly
when
we
log
in
to
the
target
and
we're
not
really
sure
what's
going
on,
so
we
thought.
Maybe
it
was
because
the
the
encoder
was
in
the
amba
pl
part
of
the
device
tree
instead
of
under
axi,
but
it
looks
like
that.
A
A
A
A
A
We
definitely
need
some
help
to
get
the
to
get
the
excellent
results.
The
amazing
results
from
the
pluto
and
the
936
61..
We
want
to
duplicate
that
for
the
9371
on
the
zc706,
and
we
also
want
to
carry
that
forward
onto
the
ultrascale,
and
I
haven't
even
started
looking
at
the
zcu
106
or
any
of
the
ultrascale
devices.
A
B
A
B
A
C
A
B
A
A
A
A
A
A
Oh
sorry,
32
bits
for
channel
zero,
32
bits
for
channel
one
16
bit
iq,
and
so
our
job
was
to
figure
out
how
to
do
the
correct
bus
sizing
and
then
fit
the
encoder
in
and
so
that
that's
how
we
chose
to
do
it
on
the
bottom.
So
we
dropped
the
encoder
in
between
the
transmit
dma
and
the
dac
fifo,
which
left
the
master.
Axi
stream
transfer
request
line
connected
from
the
transmit
dma
controller
off
to
the
dac
fifo,
because
I
wasn't
really
sure
what
else
to
do
with
it
in
the
pluto
design.
A
This
is
not
connected,
the
signal
doesn't
exist
or
the
signal
exists.
The
the
pins
exist,
but
it's
not
connected
in
the
pluto,
probably
because
there
isn't
a
dac
fifo.
The
dvb
encoder
connects
directly
up
to
rrc
interpolator,
so
we
dropped
the
dbb
encoder
in
here,
and
this
is
the
connections
and
then
the
this
particular
bit
file
was
exported
and
this
xa
xsa
is
what
we're
using
and
now.
I
have
no
idea
why
the
exported
board
system
board
diagram
doesn't
show
this
and
so
I'll.
Maybe
that's
the
problem.
A
I
don't
know,
but
in
any
case
the
problem
that
we're
having
doesn't
have
anything
to
do
with
any
of
the
stream
connections.
That
has
everything
to
do
with
the
light
x
or
light
connection.
The
axi
light
connection
is
from
the
processor
to
the
encoder,
and
that
is
that's
set
up
for
sure
in
the
tackle
script
in
the
board
treckle
script,
so
we
should
be
able
to
talk
to
the
encoder
from
the
processor
side,
just
like
anything
else,
with
any
sort
of
other
axi
interface.
A
A
B
A
B
B
A
B
A
A
B
B
B
B
C
A
C
A
Because
that's
in
the
tackle
the
that
connection,
the
light,
the
axa
light,
the
and
one
clock
and
and
one
reset-
and
we
see
that
like
when
you
make
this
and
you
call
it
up
in
vivato-
all
of
that
is
magically
there.
And
then
I
drew
the
rest
of
them.
But
it
didn't
take
because
it
goes
back
to
the
the
board
tackle.
A
B
B
And
then
I
I
I
put
make
then
I
have
an
xpr.
I
I
open
it
on
vivaldo
and
be
sure
that
all
the
connection
I
did
on
antiqual
is
okay.
If
not,
if
there
is
some
missing
lines,
then
I
plugged
this
missing
line,
see
how
the
tickle
is
well.
What
what
is
the
line
to
to
modify
well
on
vivado?
When
you
put
the
line,
then
you
have
some
typical
truss?
A
C
B
B
A
B
A
C
C
A
A
A
A
B
A
A
C
A
C
A
B
B
C
C
A
C
A
C
A
Scratch
register
and
everything
it
works
fine,
but
like
every,
we
get
a
bus
error
every
time
for
our
own
ip,
but
we
have
this
axi
light
interface
and
I'm
pretty
sure
that
that's
the
interface
that
communicates
to
the
control
registers
and
config
registers.
So
we've
done
so
we've
done
something
wrong
like
in
maybe
in
pedal,
linux
or
x.
Exporting
maybe
I
mean,
but
I
can't
find
anything.
A
C
B
A
B
B
A
C
C
A
I
just
don't
know
how
to
I
don't
know
how
to
to
get
the
files
to
show
up,
but
I
think
I
can
probably
edit
the
the
hard-coded
directory
things
and
we
can
and
now
we
know
the
connections
that
we
need
to
do
from
the
diagram.
So
if
we
add
the
board
connections
into
that
particular
tucle
script
and
then
redo
all
of
this,
if
we
still
get
the
same
problem,
then
it's
some
other
problem.
C
C
C
A
C
C
So
bef
just
want
to
express
that
instead
of
that,
can
I
make
this
new
tackle
script
based
on
the
connections
that
we
have
made
yeah
test
it
out,
and
then
I
can
improve
all
on
the
changes
that
we
have
that
we
are
talking
on
the
slack.
That
way,
we
will
be
sure
that,
because
we
both
are
at
the
same
page,
we
have
our
connections
and
everything
repository
is
building.
C
B
It
could
help
also
other
person
to
replate
replicate
that's
what
I
try
to
do
with
the
pluto.
I
think
that
it's
working,
so
maybe
you
can
you
can
inspire
of
how
I
I
did
it.
Maybe
it's
not
that
way,
but
I
think
that's
the
the
firmware
is
complete,
even
with
hdl
and
all
the
yes
all
the
stuff
so
yeah.
I
think
that
tickle
could
be
a
good
idea
as
soon
as
you
put
your
firmware
and
I
yeah
I
I
can.
I
can
review
on
your
pc
remotely.
A
Awesome,
thank
you.
Okay.
Is
there
any
any
opinions
about
the
the
master,
xi
stream
x,
f,
e
r
request
line
that
comes
from
d
from
tx
dma
and
it
goes
off
to
the
dac
fifo.
This
is
not
connected
up
in
the
pluto
design,
but
it,
but
we
have
it.
I've
left
it
in
place,
but
I'm
concerned
that
this
signal
is
going
to
screw
us
up.
Does
anybody
know
anything
about
it?
C
A
A
A
A
This
do
and,
and
it
turns
out,
it
goes
high
when
there's
a
dma
request
and
it
stays
high
as
long
as
there's
gma
requests,
but
now
that
we
have
logic
in
the
middle,
I'm
worried
that
it'll
it'll
de-assert
before
we're
done
with
the
the
encoder,
and
you
don't
have
the
signal
hooked
up
at
all
in
the
pluto
and
it
seems
to
work
just
fine
so
like,
but
you
go
from
dma
to
encoder
to
the
rrc
interpolator.
You
don't
go
through
a
fifo
for
dac.
B
A
A
It
looked
important
and
I
don't
know
what
to
do
with
it
or
if
we
have
to
manipulate
the
signal
or
incorporate
it
in
any
way,
I'd
like
to
use
it
if
we
have
to
so
I
it's
not
responsible.
The
signal
is
not
responsible
for
the
problems
we're
currently
having,
but
I
I'm
looking
forward
a
couple
of
days
from
now.
You
know
when
we're
triumphant
and
victorious
and
everything
was
working
and
then
we
go.
Why
is
it
cutting
off
our
transmission
too
early?
A
C
A
A
C
A
I
think
I've
cut
and
pasted
some
some
notes
just
slack,
but
yeah.
That's
just
something.
This
is
one
of
the
things
I'm
trying
to
track
and
worry
about
cool.
Okay,
I
think
we
have
a
plan.
This
is
great,
so
yeah,
if
you,
if
you
need
to
drop
off
the
call
from
from
here
fine.
So
so
what
I'd
like
to
talk
about
next
is
remote
lab
south
fpga
stations
and
development
and
capital
expenses
and
planning
for
for
the
for
that
particular
lab
with
james.
A
A
Yeah,
no
thank
you.
This
is
fantastic.
It's
a
privilege
to
work
with
all
of
you,
wonderful,
competent
people.
Thank
you
so
much
for
your
time
and
very
soon
we're
going
to
be
working
with
a
working
downlink
and
then
we'll
have
to
work
on
the
scheduler,
which
is
how
do
we
feed
it
and
what
quality
of
service
decisions
do
we
make,
and
all
of
that?
Wonderful,
adaptive,
coding
and
modulation
magic
will
happen
and
it
will
be
just
as
hard
and
there
will
be
all
sorts
of
problems
to
happen
there.
A
It's
known
there
are
dozens
of
people
complaining
about
it
on
the
web
and
there's
a
workaround.
So
instead
of
using
the
graphical
user
interface,
you
have
to
use
the
command
line,
which
is
platform
create
and
then
a
whole
bunch
of
stuff
that
you
have
to.
So
what
I'm
going
to
do
is
add
it
to
our
repository
readme
for
how
to
use
fpgas.
A
C
A
Creating
an
application-
and
so
thank
you
to
engineer
zone
stack
overflow,
google,
for
google
search
for
for
all
the
different
items,
so
you
can
get
through
with
a
with
a
command
line
thing
and
it.
It
just
goes
to
show
you
that
big
companies
with
their
proprietary
tools
have
bad
bugs
that
you
know
we
can't
help
books
and
you
know
it's
it's
it's
just
a
a
note.
A
So
what
I'll
do
is
I'll
make
sure
that
that
is
in
the
that's
in
our
documentation,
so
that
you're,
not
flummoxed
or
or
sidelined
by
that,
because
and
oh
if
you
do
go
straight
to
the
create
application
first,
which
is
what
I
did
the
other
day.
Thinking
that
you
just
have
to
create
an
application
on
the
processor
side,
if
you
don't
do
the
project
platform,
first
you'll
crash
the
zc706
and
it
will
be
bricked
for
some.
B
A
So
that's
what
happened!
Anshul
when
you
said
that
you
couldn't
find
the
target.
It's
because
I
tried
to
run
an
application
and
it
just
crashed,
and
I
didn't
understand
why
and
didn't
think
of
it.
So
I
repeated
it
after
you
talked
to
me
and
sure
enough:
you
can.
You
can
completely
brick
the
zc706
by
going
straight
to
application.
So
all
of
this
all
I'll
make
sure
it's
written
down
just
a
little
just
yet
another
learning
opportunity
for
us.
You
know,
but
once
you.
C
A
Platform
creation,
if
you
just
do
the
whole
thing,
then
you're
all
of
a
sudden
you're
in
a
debugger,
and
you
got
you
got
access
to
the
cc706
and
that's
what
we'll
probably
be
using
to
write
the
scheduler
and
all
of
the
multiplexing
software
to
connect
the
uplink
and
downlink.
So,
okay.
So
anyway,
that's
the
that's
the
big
win
for
this
week.
It
was
a
lot
of
work,
but
we're
pretty
confident.
That's
the
way
to
do
it
and
yeah
see
you
on
slack
and
and
next
week
same
time,.
C
A
All
in
the
host,
it's
all
cross
compilation,
it's
all
it
all
magically
works
because
you've
included
the
xsa
file.
You've
incorporated
all
of
the
what
you
need
to
know
so
you're
not
having
to
write
code
on
the
host.
Now
you
can,
if
you
want
to,
but
you
have
to
go
back
to
pedal
linux
and
include
a
c
environment
and
then
in
pedal
linux
and
then
reboot.
It.
B
C
C
A
C
A
A
C
C
C
C
C
A
There
is
a
way
to
show
it.
I
haven't
found
it
out
yet
so
I
apologize
to
anybody
out
there
listening
to
this.
That's
a
big
fan
of
vitus
and
knows
how
to
do
this.
If
you
have
advice
for
us,
we
would
like
to
hear
it.
You
know,
so
let
us
know
where
we're
on,
but
it
it
just
stops.
It
just
halts
so
and
you
have
to
kind
of
figure
it
out
from
looking
at
the
registers,
which
you.
C
C
A
C
A
All
right,
james
thanks
so
much
for
hanging
in
there.
Hey
tell
us,
I
know
it's,
it's
probably
hot
there
and
have
you
had
any
thunderstorms
and
we
are.
We
have
funding
lined
up
and
I
think
we're
just
waiting
for
documentation
for
just
to
make
it
really
clean
and
easy
for
the
501c3
to
fund
the
lab
work.
C
That
is,
that
is
all
correct.
It's
been
pretty
hot
here,
but
you
have
the
advantage
of
architecture
and
air
conditioning
all
built
for
this
kind
of
weather.
So
it's
been
pretty
good
until
the
dog
goes,
but
it's
been
15
minutes
since
I
went
outside
so
now.
I
must
go
outside
for
the
next
half
hour
like
dog,
please
no
yeah.
C
A
C
A
C
A
So
if
you
want
to
tackle
universities,
you
should
because
that's
one
of
the
missions
that
we
have
you
know,
but
like
independent
researchers
and
hobbyists
and
enthusiasts,
and
and
also
people
that
want
to
use
this
opportunity
to
work
on
open
source
software
in
order
to
like
build
up
a
better
career.
We
are
all
in
on
that
too,
but
the
outreach
and-
and
you
know,
communications
required
to
to
talk
to
people
that
are
interested
in
that
is
is,
is,
can
be
very
time
consuming.
A
C
A
A
So
so
it's
you
know
it's
just
you
approach
it
with
some
joy
and
enthusiasm,
and
I
think
I
think
you've
already
got
those
things
down
and
and
you're
you're,
already
well-versed
and
and
competent
in
that
yeah,
so
yeah.
I
think
we're
going
to
be
able
to
build
not
just
a
fpga
sort
of
nexus.
Another
a
station
very
similar
to
what
we
have
already
in
the
uk
and
in
san
diego,
but
the
the
remote
labs
south
also
has
a
biomedical
lab
planned
as
well.
A
So
that's
a
completely
different
type
of
endeavor,
but
we're
we're
very
committed
to
seeing
that
happen
and
very
excited
about
it.
So
there's
been
some
discussion
about
the
physical
physical
plant
requirements
for
a
bio
for
a
bacteriophage
or
any
other
biological
or
or
agricultural
work,
and
that
brings
us
to
fundraising.
So
there
are
some
some
funds,
some
some
grants
that
we
are
targeting.
A
And
it's
at
least
several
million
just
to
get
off
the
ground
with
a
team
of
three
people
doing
the
work.
So
we
don't
have
a
million
dollars
for
that,
but
we
do
have
enough
money
to
to
tackle
a
problem
that
is
arguably
just
as
pressing
and
that
is
feeding
people
and
feeding
vulnerable
people.
So
fish
farms
around
the
world
can
really
feed
a
lot
of
people.
A
So
back
you
know,
looking
at
the
problems
of
fish
farms
and
and
whether
or
not
a
bacteriophage
cocktail
can
work,
we
can
actually
set
up
an
aquaponics
greenhouse
or
you
know,
a
very
small
commercial
style
fish
farm,
especially
in
arkansas.
I
think
that
the
climate
is
really
well
suited
to
this
and
there's
already
existing
farms
that
we
might
be
able
to
collaborate
with
and
uams
might
be
interested
in
this,
and
we
do
have
a
bacteriophage.
A
Initiative
or
institute
here
at
ucsd
at
the
university
of
california
at
san
diego,
they
went
on
hiatus
with
kovid
and
have
not
really
come
out
of
it
yet,
but
they
know
about
us
and
any
day
we're
hoping
that
it
becomes
a
little
more
active.
So
so
anyway,
that's
one
of
the
things
that
that
we
are
interested
in
in
seeing
if
it
can
happen
at
remote
love
south,
is
to
also
add,
essentially
a
wet
lab
for
bacteriophage
work.
A
C
A
Yeah,
I'm
excited
about
it,
the
there's
at
least
two
fda
grants
that
are
open
right
now
and
we
don't
have
to
to
leap
for
these.
It's
just
proof
that
we're
on
the
right
track.
There's
there's
one
on
innovative
farming
that
might
help
fund
some
work
to
get
to
kickstart,
the
bacteriophage
for
for
fish
farms,
but
there's
another
one.
That's
these
are.
A
These
are
grants
that
happen
fairly
routinely
but
for
a
grant
available
for
a
workshop
or
conference,
and
so
I'd
like
to
to
propose
that
we
really
seriously
think
about
either
a
workshop
or
a
small
conference
either
in
little
rock
or
somewhere
else.
I
think
little
rock
would
work
for
for
gathering
together
scientists
and
and
engineers
that
want
to
work
on
this.
So
if
we
have
a
focused
sort
of
sort
of
scientific
conference
or
scientific
workshop,
there
there's
money
available
through
right.
A
It
looks
like
either
department
of
agriculture,
fda
or
or
nsf,
so
as
soon
as
possible.
What
I'll
do
is
I'll?
Send
you
the
proposal
or
the
request
for
for
proposals
for
the
the
conference
and
see
what
you
think
and
if
we
can't
hit
this
one
by
the
deadline,
then
we
should
look
for
the
next
one
and
and
try
to
start
putting
together
a
conference
or
workshop.
It
does
not
have
to
be
large
to
be
effective,
but
I
think
we
could
could
make
a
good
yeah.
A
A
Okay,
well
today,
I
will
send
you
the
the
details
on
those
two
requests,
their
solicitations
for
for
grants
and
then
we'll
we'll
see
what
we
can
do
to
try
to
to
put
together
a
decent
proposal.
It'll
probably
take
at
least
the
in
order
to
really
do
this
right,
we'll
need
to
have
some
some
regular
meetings
in
the
short
term
to
to
hammer
it
out
and
I'll
help.
All
that
I
can
so
that'll
be
something
that'll
I'll
turn
around
to
you
guys
today.
A
A
Thank
you.
Thank
you
very
much.
Your
your
expertise
and
and
and
all
of
your
help
is
deeply
appreciated.
Could
not
we
would
not
be
here
without
you
any
any
thoughts
on
the
uplink
or
work
there.
That
needs
to
be
so
like.
I
did.
I
think
we're
probably
gonna
have
to
start
talking
about
what
is
receiving
the
uplink
on
the
on
the
fpga
side.
So
the
we've
talked
about
polyphase
analyzers
and
it's
been
a
while,
since
we
demonstrated
the
poly
phase,
filter
bank
stuff,
any
thoughts
there
or
too
early.
A
All
right,
I'm
thinking
about
it,
but
I
I
think
we're
in
decent
shape.
I
know
that
the
thesis
course
will
have
to
be
modified
because
it
targets
the
70
45
instead
of
what
we've
got,
but
the
code's
good.
There
is
a
known
bug
with
the
number
of
channels,
so
the
implementation.
The
last
time
I
talked
to
the
thesis
course
people
this
will
for
those
of
you
listening.
This
is
what
receives
the
uplink.
A
It's
a
frequency
division,
multiple
access,
uplink
and
it
needs
to
be
received
that
way
and
then
digitized
or
decoded,
demodulated,
decoded
and
then
multiplexed.
But
you
know
you
first,
you
have
to
receive
it
and
we
believe
that
the
best
way
to
do
that
is
with
the
polyface
channelizer
and
we
do
have
a
solid,
open
source
base
to
work
from,
but
it's
going
to
have
to
be
ported
to
our
our
system
and
there's
an
interesting
bug.
A
So
we
should
be
aware
that
there
might
be
a
raw
channel
limit
that
we
need
to
address
and
it
might
require
some
backing
up
and
looking
at
what
happened
so
it
could
have
already
been
fixed,
but
I
that
was
the
it
was
a
very
interesting
and
somewhat
subtle
thing
that
really
kind
of
wrecked
anything
above
eight
channels,
and
we
will
definitely
be
requiring
more
than
eight
channels.
So
that's
all
I
know
about
that.
So
it's
that's
well
well,
past.
A
A
Okay,
thank
you,
everybody!
Thank
you
for
everybody.
That's
that's!
Listening
for
your
support
and
positive
vibes.
It
makes
all
the
difference
in
the
world
and
we're
looking
forward
to
being
able
to
share
videos
of
things
working
and
it's
a
delightful
project
to
work
on,
and
we
hope
that
it
brings
a
lot
of
joy
to
people
that
use
it
all
right,
see
you
on
slack.