sgcarnaval has quit [Ping timeout: 250 seconds]
gnufan_home has quit [Quit: Leaving.]
dj_pi has joined ##openfpga
gsi__ has joined ##openfpga
gsi_ has quit [Ping timeout: 245 seconds]
emeb_mac has quit [Ping timeout: 258 seconds]
flaviusb has quit [Quit: Leaving.]
OmniMancer has joined ##openfpga
futarisIRCcloud has joined ##openfpga
emeb_mac has joined ##openfpga
ZombieChicken has joined ##openfpga
dj_pi has quit [Ping timeout: 248 seconds]
<Sprite_tm> Well fuck me.... seems the tinyfpga bootloader is just plain ole broken :/ I get a sync of 00000111 instead of 00000001
<tnt> Sprite_tm: really ? I mean I know it's not spec-complient, but that seems a bit much.
<Sprite_tm> On second thought, no it's not. I think my drivers are somehow not driving the USB port hard enough, and the LA stumbles because the edges are all wonky.
<tnt> Sprite_tm: What value of series resistor are you using ? I know I had issue when I was using ~100R (that's all I had at the time). Now I'm using 33R
<Sprite_tm> It's still a bit odd, though... Ks seem to last a lot longer than Js.
<Sprite_tm> tnt: 27R I think.
<Sprite_tm> Yes, 27R.
<Sprite_tm> Hm, maybe my zeners are fucking stuff up... lemme desolder them...
<Sprite_tm> Nope, still horrible edges and an unhappy laptop.
<tnt> Note that you need to probe "on the laptop side" to see signals received by the laptops. If you probe tx packets on the fpga side, they won't look good because you'll also see the reflected signal. (At least that's been my experience when probing usb)
<tnt> Although usually it's not bad enough to prevent decoding, it just looks bad on the analog channels.
<TD-Linux> ugh I keep on making the kicad default solder mask setting mistake :(
<TD-Linux> this time was more minor as I only forgot to change 1 of the 2 settings
flaviusb has joined ##openfpga
rohitksingh has joined ##openfpga
<Sprite_tm> tnt: Hmm, that's a thought... means I have to find an usb male+female plug somewhere, or slaughter an USB cable or so.
<Sprite_tm> I'm just a bit frustrated... the USB bit is step x in an elaborate yak shaving excercise... I want to get on with everything, dammit!
<tnt> Sprite_tm: whar are you building ? (sorry, I think I might have asked, but can't find it anymore ...)
<Sprite_tm> Badge for Hackaday Supercon 2019.
<Sprite_tm> Going to be an FPGA-based one. Point is, I need to get the BOM finished so we can get a quote, and for that I need to see if my proto has everything (inc flash) working, and for that I need a bootloader, and for that (and also because I need to test USB) I need an USB bootloader to work.
<Sprite_tm> TinyFPGA-Bootloader seems like a short and sweet one that's easily portable, so I did, and now it just doesn't work. USB is not that hard hardware wise and the bootloader seems well-tested, the damn thing should just plug in and do something. I'm frustrated because it feels like debugging this feels like going down *yet another* rabbit hole.
<Sprite_tm> Sorry for venting here.
<tnt> Sprite_tm: well it's well tested on one particular build of one particular hw. I mean until not that long ago, it had inferred latches in it when built with yosys. I think there is a reason luke decided to rewrite the core from scratch :p
<tnt> I can give it a quick shot on an ice40.
<Sprite_tm> Mmm, fair nuf.
<tnt> you're based off master from https://github.com/tinyfpga/TinyFPGA-Bootloader.git ?
<Sprite_tm> Yes.
<Sprite_tm> Huh, I think timings are fucky somehow... http://j0h.nl/ctkB/full
<Sprite_tm> That's a sync word, it's supposed to be j-k-j-k-j etc. 48KHz on top. Notice that the Js are 4 48MHz clocks, Ks are 3.
<tnt> yeah, weird indeed :/
<Sprite_tm> Ah, ffs :X seems the device needs the rx inputs hardwired to the idle state when it sends.
<Sprite_tm> If not, the rx tries to sync to its own outgoing signal and it hickups, I think.
<Sprite_tm> Yep, got 'em.
<Sprite_tm> [551472.536711] usb 1-2: New USB device found, idVendor=1d50, idProduct=6130, bcdDevice= 0.00
<Sprite_tm> [551472.536724] usb 1-2: New USB device strings: Mfr=0, Product=0, SerialNumber=0
<Sprite_tm> [551472.538844] cdc_acm 1-2:1.0: ttyACM0: USB ACM device
<tnt> Ah well there you go :)
<tnt> Funny, when doing the port right now, I saw that masking and wondered if it was needed and in doubt, I left it :p
<Sprite_tm> Kinda obscure that it results in a seemingly ok working stack, aside from that timing error.
<Sprite_tm> Hm, it still ain't happy, but that may be the flash side of things.
<tnt> yup, works here too [1334216.084034] usb 2-1: New USB device found, idVendor=1d50, idProduct=6130, bcdDevice= 0.00
<Sprite_tm> What happens if you open the serial port? Just cat /dev/ttyACM0 is enough.
<Sprite_tm> I get a few seconds (5-10) of nothingness, then an USB reset.
rohitksingh_ has joined ##openfpga
rohitksingh has quit [Ping timeout: 248 seconds]
<tnt> Sprite_tm: mmm, here it seems to just wait forever ?
<tnt> Ah, but I don't have the "warmboot" primitive wired so it won't ever try to reload.
<Sprite_tm> I don't have that hooked up here either.
<Sprite_tm> Does your setup work with tinyprog?
<tnt> I don't have any metadata flashed or even have the flash wired to the bootloader ... (I sure don't want it to touch my DFU bootloader ...)
<Sprite_tm> Hey, the -l flag works for me too.
<Sprite_tm> Strange, most other things seem to fail.
emeb_mac has quit [Ping timeout: 272 seconds]
<Sprite_tm> Well, guess I can start figuring out if the signals make it to my flash.
<Sprite_tm> tnt: I can imagine you not wanting to mess with your flash; thanks for helping me so far at least :)
<tnt> I think you need to flash the metadata before tinyprog will do anything.
<Sprite_tm> Weird thing is, I get no activity on the SPI bus at all when I try to flash anything.
<Sprite_tm> Tbh, my flash chip isn't connected right either, that's the 2nd issue. I get no sck, for starters.
<tnt> Maybe it read the metadata on boot ?
<Sprite_tm> My scope says no.
<tnt> Sprite_tm: I wired the flash (after reading what tiny prog does when doing -l) and it can at least read the flash id.When doing -l at least
<Sprite_tm> tnt: is -l supposed to show that by default?
<tnt> No, I added a print
<Sprite_tm> [239, 64, 24]
<Sprite_tm> If any, it isn't nothing :)
<Sprite_tm> Took me a bodge wire... seemingly, there's a break in the sck on my board.
<Sprite_tm> Yeah, seems I have a connection. Yay :)
<Sprite_tm> Now I just need to figure out how to flash stuff without the metadata... it's kinda a shame tinyprog doesn't have a --force option.
<Sprite_tm> Ah, some hacking in tinyprog later and it works, yay!
sgcarnaval has joined ##openfpga
gnufan_home has joined ##openfpga
<Sprite_tm> Wow, default boot-from-gflash is pretty slow on this thing... 4 sec from power-on to image running.
<Sprite_tm> *boot-from-flash
<Sprite_tm> Probably need to enable qpi and higher clock rates somehow.
<daveshah> ecppack should allow configuring both
<Sprite_tm> Yes, just got it.
<Sprite_tm> Btw, you may want to document the allowable frequencies somewhere in the command-line help... took me some grepping through the source to find them.
<Sprite_tm> Oooh, qspi and 38.8MHz makes the boot nigh instantly.
Asu has joined ##openfpga
rohitksingh_ has quit [Ping timeout: 258 seconds]
rohitksingh has joined ##openfpga
<Sprite_tm> daveshah: So you know if the programn pin is supposed to be capable of being lowered internally, or would I have to pull it low using an external ciruit? Configuring it as an iobuf does nothing :/
<daveshah> There's no way of triggering a reboot internally, afaik
<daveshah> tinyfpga had to connect it to another output pin
<Sprite_tm> Blergh :/ the ecp5 is pretty damn janky when it comes to multiboot it seems.
<daveshah> Yes, iCE40 is much better
<Sprite_tm> Thanks for that info, I should be able to bodge it on my current proto and I'll add a proper connection on tmy next one.
<daveshah> I think ECP5 multiboot is mostly designed as a failsafe rather than for switching at runtime
<Sprite_tm> Yah, looks like it, with the messing around with gold images etc.
futarisIRCcloud has quit [Quit: Connection closed for inactivity]
<Sprite_tm> Ah, I read something about that multiboot was handled by an internal CPU core btw? Is more known about that already?
jcreus has quit [Ping timeout: 244 seconds]
rohitksingh has quit [Ping timeout: 248 seconds]
wpwrak has quit [Ping timeout: 272 seconds]
rohitksingh has joined ##openfpga
futarisIRCcloud has joined ##openfpga
Laksen has joined ##openfpga
rohitksingh has quit [Ping timeout: 248 seconds]
ZombieChicken has quit [Quit: WeeChat 2.5]
ZombieChicken has joined ##openfpga
ZombieChicken has quit [Client Quit]
dj_pi has joined ##openfpga
dj_pi has quit [Client Quit]
<Sprite_tm> Hot diggity, it may take an entire sunday and some interactions between my head and the desk, but in the end that multiboot-usb-bootloader-thing works pretty well.
<Sprite_tm> daveshah: I can also confirm the USRMCLK macro works in practice as well.
futarisIRCcloud has quit [Quit: Connection closed for inactivity]
<daveshah> Sprite_tm: I don't think there's an internal CPU core
<daveshah> I know there was a rumour that some mystery bytes in multi boot bitstreams were instructions for it
<daveshah> But it turns out they were just an ASCII comment bit reversed for some reason
<whitequark> lol
s_frit has quit [Remote host closed the connection]
s_frit has joined ##openfpga
<Sprite_tm> daveshah: Well, that was unexpected :)
rohitksingh has joined ##openfpga
AndrevS has joined ##openfpga
rohitksingh has quit [Ping timeout: 244 seconds]
rohitksingh has joined ##openfpga
lutsabound has joined ##openfpga
gsi__ is now known as gsi_
cr1901_modern1 has joined ##openfpga
cr1901_modern has quit [Ping timeout: 246 seconds]
noobineer has quit [Ping timeout: 248 seconds]
emeb has joined ##openfpga
cr1901_modern has joined ##openfpga
cr1901_modern1 has quit [Ping timeout: 272 seconds]
mumptai has joined ##openfpga
rohitksingh has quit [Ping timeout: 248 seconds]
OmniMancer has quit [Quit: Leaving.]
uovo has joined ##openfpga
oeuf has quit [Ping timeout: 268 seconds]
Asu has quit [Ping timeout: 246 seconds]
Asu has joined ##openfpga
Dolu1 has joined ##openfpga
Asu` has joined ##openfpga
Asu has quit [Ping timeout: 268 seconds]
Laksen has quit [Quit: Leaving]
emeb has quit [Quit: Leaving.]
emeb_mac has joined ##openfpga
emeb_mac has quit [Ping timeout: 245 seconds]
Asu has joined ##openfpga
Asu` has quit [Ping timeout: 245 seconds]
Asu` has joined ##openfpga
Asu has quit [Ping timeout: 272 seconds]
Asu` has quit [Quit: Konversation terminated!]
gnufan_home has quit [Ping timeout: 248 seconds]
gnufan_home has joined ##openfpga
gnufan_home1 has joined ##openfpga
gnufan_home has quit [Ping timeout: 248 seconds]
biostar has joined ##openfpga
biostar has quit [Client Quit]
biostar has joined ##openfpga
biostar has quit [Ping timeout: 256 seconds]
futarisIRCcloud has joined ##openfpga
Dolu1 has quit [Quit: Leaving]
gnufan_home1 has quit [Ping timeout: 246 seconds]
AndrevS has quit [Remote host closed the connection]
gnufan_home has joined ##openfpga
sgcarnaval has quit [Ping timeout: 258 seconds]
mumptai has quit [Quit: Verlassend]
dj_pi has joined ##openfpga
s_frit has quit [Remote host closed the connection]
s_frit has joined ##openfpga