00:01
<
clever >
SD cards work with a command/response scheme. For example, command "17" allows reading one sector (512 bytes) of the card memory. All communication is synchronous to a clock provided by the host (FPGA in our case). The clock should run below 400KHz at startup and can go faster after some card initialization.
00:02
<
samueldr >
I'm not looking into that right now though :)
00:02
<
samueldr >
that'll be for much later, I guess
00:03
<
clever >
2019-11-04 19:59:41 < emily> if you want an open source FPGA toolchain currently you should buy an iCE40 or ECP5 FPGA
00:03
<
clever >
2019-11-04 20:00:07 < emily> ofc, there is support in open source software for certain tasks (flashing itself, yosys has some existing Xilinx support to use with Vivado, etc.)
00:04
<
clever >
samueldr: if you want to avoid the 30gig blobs for other fpga dev
00:04
<
samueldr >
haha, the glasgow is based on the iCE40
00:04
<
samueldr >
so looks like that's about right
00:05
<
samueldr >
if you haven't looked, clever, the glasgow uses a thing that takes the python applet and compiles them for the FPGA
00:05
<
clever >
i would use clash if i was to do more fpga today
00:05
<
clever >
it does everything with haskell
00:22
orivej has quit [Ping timeout: 265 seconds]
00:22
LnL has quit [Ping timeout: 240 seconds]
00:24
LnL has joined #nixos-aarch64
00:24
LnL has joined #nixos-aarch64
00:24
LnL has quit [Changing host]
01:20
ris has quit [Ping timeout: 258 seconds]
02:36
h0m1 has quit [Ping timeout: 276 seconds]
02:38
h0m1 has joined #nixos-aarch64
02:41
<
samueldr >
I think we need a proper term to differentiate between actual cross-compilation and building through qemu-user
02:54
<
craige >
qemulation?
02:56
<
samueldr >
it's more about the fact it's factitious nativebuilding
02:56
<
samueldr >
rather than actual cross-compilation
03:24
cptchaos83 has joined #nixos-aarch64
03:26
ryantrinkle has quit [Ping timeout: 265 seconds]
03:40
ryantrinkle has joined #nixos-aarch64
05:01
Acou_Bass has quit [Ping timeout: 240 seconds]
05:03
Acou_Bass has joined #nixos-aarch64
05:14
ryantrinkle has quit [Ping timeout: 265 seconds]
06:05
zupo has joined #nixos-aarch64
06:57
Acou_Bass has quit [Ping timeout: 268 seconds]
07:02
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
07:08
Acou_Bass has joined #nixos-aarch64
07:11
zupo has joined #nixos-aarch64
07:29
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
07:44
orivej has joined #nixos-aarch64
07:58
sigtrm_ has quit [Ping timeout: 268 seconds]
08:08
zupo has joined #nixos-aarch64
08:20
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
08:49
FRidh has joined #nixos-aarch64
09:08
zupo has joined #nixos-aarch64
09:14
sigtrm has joined #nixos-aarch64
09:16
zupo_ has joined #nixos-aarch64
09:20
zupo has quit [Ping timeout: 276 seconds]
09:49
sphalerite has quit [Remote host closed the connection]
09:49
sphalerite has joined #nixos-aarch64
09:50
sphalerite has quit [Client Quit]
09:50
sphalerite has joined #nixos-aarch64
09:51
sphalerite has quit [Client Quit]
09:52
sphalerite has joined #nixos-aarch64
10:07
zupo_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
10:08
zupo has joined #nixos-aarch64
10:11
zupo has quit [Read error: Connection reset by peer]
10:12
orivej has quit [Ping timeout: 276 seconds]
10:54
tilpner_ has joined #nixos-aarch64
10:56
tilpner has quit [Ping timeout: 276 seconds]
11:01
tilpner_ is now known as tilpner
11:42
sigtrm has quit [Ping timeout: 265 seconds]
12:01
zupo has joined #nixos-aarch64
12:05
sigtrm has joined #nixos-aarch64
13:20
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
15:03
<
andi- >
Qcompiling?
15:11
<
gchristensen >
would be nice if we could get the Q to help us change th laws of physics to compile faster
15:24
<
andi- >
indeed. I'd like to get a copy of the binary cache +5y from now. Thanks.
15:25
<
gchristensen >
inb4 it suppotrs 18 architectures and takes up 250pb
15:30
ryantrinkle has joined #nixos-aarch64
17:43
<
samueldr >
this is fun and hard to find documentation for
17:43
<
samueldr >
"ERROR: Operation Denied" at the bootloader
17:43
<
samueldr >
no idea what it means
17:43
<
samueldr >
though once I flashed TWRP it went away
17:44
<
gchristensen >
nice.
17:44
<
samueldr >
seems to not be an issue at all in the end
17:44
<
samueldr >
simply that everyone seeing this has
*also* soft-bricked their devices
17:50
orivej has joined #nixos-aarch64
18:27
ryantrinkle has quit [Ping timeout: 268 seconds]
18:33
<
samueldr >
:/ don't know if my wiring is wrong device not configured right or the system not outputting to serial
18:36
<
samueldr >
using TWRP and default roms
18:36
<
samueldr >
so it's possible neither output to tty
19:05
<
samueldr >
okay, so fastboot oem uart enable does add console=ttyMSM0,115200,n8
19:13
<
samueldr >
looks like I need to break out the salaea clone
19:23
THFKA4 has joined #nixos-aarch64
19:23
THFKA4 has quit [Changing host]
19:27
<
samueldr >
great news, the information is not entirely wrong
19:27
<
samueldr >
pulseview sees the UART just right
19:27
<
samueldr >
(read in the green bubbles, Context u:object_r:perfetto_traces_data etc)
19:39
<
samueldr >
:/ shows up fine in the logic analyzer
19:39
<
samueldr >
nothing in picocom
19:43
vika_nezrimaya has joined #nixos-aarch64
19:43
zupo has joined #nixos-aarch64
19:45
<
samueldr >
thinking maybe the length of the serial cable's wires could affect something here
19:46
<
samueldr >
though the serial cable is a known good one, from adafruit (through newark)
19:47
<
samueldr >
or the other possibility I'm thinking of is I'm somehow using the wrong ground reference
19:47
<
samueldr >
but tried both
19:47
<
samueldr >
and it worked with the analyser
19:48
<
samueldr >
picocom's parameters match the analyser's
19:50
zupo has quit [Ping timeout: 240 seconds]
19:50
zupo has joined #nixos-aarch64
20:15
orivej has quit [Ping timeout: 265 seconds]
20:27
<
samueldr >
[ 2314.951449] c6 581 mdss_fb_set_backlight: backlight sent to panel :0
20:27
<
samueldr >
success!
20:27
Thra11 has quit [Ping timeout: 240 seconds]
20:27
<
samueldr >
used one with shorter wires
20:27
<
samueldr >
(and for anyone trying this: it
*is* orientation dependent)
20:27
<
samueldr >
(SBU1 and SBU2 will be switched around otherwise)
20:33
<
samueldr >
interesting!
20:33
<
samueldr >
when the device is powered off, you need to hold the power button to turn it on
20:33
<
samueldr >
it's all software based
20:33
<
samueldr >
the UEFI starts with a short press
20:37
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
20:38
FRidh has quit [Quit: Konversation terminated!]
20:39
zupo has joined #nixos-aarch64
20:41
Thra11 has joined #nixos-aarch64
20:41
<
samueldr >
just double checked and it does not want to work with the "known good" serial, so likely that
20:43
<
clever >
samueldr: if you dont care about serial from pc->phone, you can tie the rx pin to both SBU1 and SBU2
20:43
<
clever >
oh wait, no
20:43
<
clever >
then its a loopback
20:44
<
clever >
so the phone gets all of its output back in
20:44
<
clever >
would need diodes
20:44
<
samueldr >
though the "cable" is easily identifiable to put in the right way around
20:44
tilpner_ has joined #nixos-aarch64
20:44
<
clever >
better option, is to detect which way around the plug is (how does usb-c do it?) and then mosfet things around
20:44
<
samueldr >
type-c does it through a particular pin pair
20:44
<
samueldr >
this requires minimal soldering
20:45
<
samueldr >
and should be doable without at all by picking the right UART
20:45
tilpner has quit [Ping timeout: 276 seconds]
20:47
<
samueldr >
it's amazing to see that the battery icon thingy on android is just an android system
20:48
<
samueldr >
still, something is off with my setup
20:48
<
samueldr >
sometimes the uart hangs
20:48
<
samueldr >
thinking it's likely the new cheapo uart
20:58
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
21:24
<
samueldr >
*definitely* the UART used is not good
21:24
<
samueldr >
it wouldn't work
21:25
<
samueldr >
until I did a plug in-out cycle from usb
21:36
<
samueldr >
popped out the plastic covers and soldered the pins on the type-c breakout board since they would get off easily
21:36
<
samueldr >
and I
*think* ground not having a good connection could have been an issue
21:37
<
samueldr >
craige: pretty much confirmed that the instructions are good, though it could happen that some serial adapters will not work right
21:44
ris has joined #nixos-aarch64
21:46
<
samueldr >
once I circle back to porting to walleye, I'll be able to better debug issues
21:57
orivej has joined #nixos-aarch64
21:58
<
samueldr >
so, what I've learned today is great, that logic analyser and FLOSS is amazing
22:00
<
gchristensen >
nice!!
22:02
<
gchristensen >
ooooo!!!
22:02
<
samueldr >
a cheap analyser, a clone apparently (didn't know at purchase time) and software already in nixpkgs
22:02
<
gchristensen >
wow!!
22:02
<
samueldr >
just had to tag the right lines TX and RX
22:02
<
samueldr >
then add UART analysing
22:02
<
samueldr >
the software is pulseview
22:03
<
craige >
Big wins :-)
22:03
<
samueldr >
I wasn't
*that* surprised that it worked
22:03
<
samueldr >
but surprised that it worked without knowing much about it all
22:04
orivej has quit [Ping timeout: 240 seconds]
22:16
pbb has quit [Remote host closed the connection]
22:17
fooker has joined #nixos-aarch64
22:20
pbb has joined #nixos-aarch64
22:23
tilpner_ has quit [Remote host closed the connection]
22:24
tilpner_ has joined #nixos-aarch64
23:07
misuzu has quit [Quit: leaving]
23:15
misuzu has joined #nixos-aarch64