NishanthMenon changed the topic of #openocd to: this is the place to discuss all things OpenOCD | Logs: https://libera.irclog.whitequark.org/openocd/
tsal_ has joined #openocd
tsal has joined #openocd
tsal_ has quit [Ping timeout: 252 seconds]
boru` has joined #openocd
boru has quit [Killed (NickServ (GHOST command used by boru`))]
boru` is now known as boru
tsal has quit [Ping timeout: 264 seconds]
tsal has joined #openocd
zjason` is now known as zjason
Steffanx has quit [*.net *.split]
flatmush has quit [*.net *.split]
olerem has quit [*.net *.split]
captain_morgan has quit [*.net *.split]
flatmush has joined #openocd
olerem has joined #openocd
Steffanx has joined #openocd
captain_morgan has joined #openocd
sbach has quit [*.net *.split]
sbach has joined #openocd
sbach has joined #openocd
sbach has quit [Changing host]
mithro has joined #openocd
sys64738 has joined #openocd
<sys64738> hi, im working on some code that turns the rpi pico into an ft2232 clone (without knowing much about the low level details about jtag), and, is it ok to pause the clock signal in the middle of transferring a command, or does the entire command have to be sent in order for things to work? (usb has a max packet size of much less than the 65536 supported by the mpsse protocol, so, these would have to
<sys64738> be split up, and keeping the timing consistent between multiple usb packets sounds like itll be annoying)
<sys64738> (asking here because #openjtag doesnt seem to exist yet? and ##openjtag is practically empty)
<PaulFertser> sys64738: hi
<PaulFertser> sys64738: I can't answer your question (I can't outright remember any reason for it to be not ok) but I'm interested why you decided to implement MPSSE and not cmsis-dap or jlink protocol.
<sys64738> i have to admit that the decision wasnt very informed
<PaulFertser> As MPSSE was meant to be implemented in ASIC right from the beginning, and the other protocols are implemented by MCUs.
<PaulFertser> sys64738: btw, have you checked openocd src/jtag/target/mpsse.c? It reads out the maximum packet size from the USB descriptor so you should be able to specify it as your hardware permits.
<PaulFertser> descriptor->endpoint[i].wMaxPacketSize;
<sys64738> i mean it was mostly "ok it seems to have been done before (eg. https://github.com/Disasm/f042-ftdi/), it doesnt have to do 'smart' stuff but its just pushing bits, and also flashrom has an ftdi driver as well so thatd also come for free"
<PaulFertser> flashrom has jlink driver too
<sys64738> ah i see, thanks. now lets hope libftdi does that too, for flashrom
<sys64738> also with the pio in the pico/rp2040, i dont think itll be much of an issue to turn it into nicely clocked signals, its mostly the delay that was bothering me. but i guess if openocd limits it itself, it isnt that much of an issue then
<sys64738> hm i just discovered that https://github.com/raspberrypi/picoprobe/issues/2 is a thing already
<PaulFertser> Nice link
<sys64738> ok so theres a version taht runs code on one core of the rp2040 to debug the other core; and theres a version to use swd to debug other hardware. but only swd and no jtag, and i dont think my ecp5 has swd support?
<sys64738> sooo still not that useful for my case :/
<PaulFertser> sys64738: cmsis-dap protocol has jtag too
<sys64738> yes but the pico implementation doesnt
<PaulFertser> But you can extend it with code from daplink.
<sys64738> though it doesnt seem that difficult to extend