starblue1 has quit [Ping timeout: 240 seconds]
starblue1 has joined #rust-embedded
tokomak has joined #rust-embedded
tokomak has quit [Ping timeout: 240 seconds]
emerent_ has joined #rust-embedded
emerent has quit [Killed (mercury.libera.chat (Nickname regained by services))]
<re_irc> <omar_u8> 9names: Im actually curious to know whats the problem. It'd keep on buggin me if I leave it like that :)
<re_irc> <omar_u8> xiretza: I guess its not meant to be a smooth sail :) ```
<re_irc> Copyright (C) 2022 Free Software Foundation, Inc.
<re_irc> License GPLv3+: GNU GPL version 3 or later http://gnu.org/licenses/gpl.html
<re_irc> GNU gdb (GDB) 11.2
<re_irc> <omar_u8> I guess there's a different problem now.
<re_irc> This is free software: you are free to change and redistribute it.
<re_irc> <omar_u8> GNU gdb (GDB) 11.2
<re_irc> License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
<re_irc> Copyright (C) 2022 Free Software Foundation, Inc.
Guest2 has joined #rust-embedded
<re_irc> <omar_u8> xiretza: I couldnt get to capture the backtrace using gdb as I keep getting that the stack is empty. I did however, obtain the debug output from openocd using the -d option. I honestly dont see anything obvious that casuses it to seg fault. The output is quite lengthy so I didnt want to post it all here.
<re_irc> <omar_u8> These are the last few outputs, not sure if it says much or if im looking at the right place even
<re_irc> Debug: 147 2 hla_transport.c:173 hl_transport_init(): current transport hla_swd
<re_irc> Debug: 148 2 hla_interface.c:42 hl_interface_open(): hl_interface_open
<re_irc> Debug: 146 2 hla_transport.c:156 hl_transport_init(): hl_transport_init
<re_irc> <9names (@9names:matrix.org)> which version of openocd are you running? maybe it is this:
<re_irc> <omar_u8> lemme check one sec
<re_irc> <omar_u8> its 0.11.0
<re_irc> <omar_u8> 9names: Thanks, I'll take a look.
<re_irc> <9names (@9names:matrix.org)> looks like that was before this patch! latest release might just fix your problems
<re_irc> <omar_u8> Isnt 0.11.0 the latest? :S
<re_irc> <9names (@9names:matrix.org)> latest official release, sure. you've got me there.
<re_irc> <9names (@9names:matrix.org)> for the longest time no-one ran official openocd releases since they were so old. i guess that's a habit that dies hard :/
<re_irc> <omar_u8> I see. So that means I can install a more recent unofficial release?
<re_irc> <9names (@9names:matrix.org)> yeah, you can grab a binary release from https://xpack.github.io/blog/2021/12/07/openocd-v0-11-0-3-released/ or just build it from source.
<re_irc> it's not too hard, just make sure you enable stlink support
<re_irc> <omar_u8> Awesome, thanks! Im gonna give it a try.
<re_irc> <ryan-summers> Yeah the failure of openocd releases is what lead me to use probe-rs instead
<re_irc> <omar_u8> ryan-summers: Thing is with probe-rs, I managed to flash but I still had issues setting up the debug in VSCode
<re_irc> <ryan-summers> Yeah but openocd just has a raw GDB interface, doesn't it?>
<re_irc> <ryan-summers> Probe-rs also has that
<re_irc> <omar_u8> True, through VSCode setup is different
<re_irc> <ryan-summers> Does openocd support VSCode? I didn't know that
<re_irc> <ryan-summers> If Openocd supports VSCode via the GDB server, you can use the exact same implementation from openocd for probe-rs btw
<re_irc> <omar_u8> No it doesnt, but there are available setups that I found more straightforward for now
<re_irc> <omar_u8> I ran into a new set of issues since the extension was an alpha and still not integrated
<re_irc> <ryan-summers> In any case, I'd highly recommend giving probe-rs a go since you'll likely find more support for it here, plus the devs are active in helping people out here
<re_irc> <ryan-summers> But up to you (this is coming from someone who continued using openocd for a year or so before switching over)
<re_irc> <omar_u8> Is there something similar to cargo-generate though targeted for the probe-rs?
<re_irc> <omar_u8> Quite frankly I'm finding myself lost in the different options and finding the right combintaion that can give me a working integrated setup to code and debug in VSCode.
<re_irc> <omar_u8> While I find tools thatn make things easier in one aspect, i seem to run into a new issue when trying to get to my end goal.
<re_irc> <omar_u8> If anything, Im finding it a good learning experience as I try to work through the issues. Thanks to everybody's help here as well.
<re_irc> <Noah> omar_u8: jack is working hard to make vscode a way better experience and he has also set this up: https://github.com/probe-rs/probe-rs-debugger-test
<re_irc> <Noah> There is two drawbacks in our vscode plugin atm. no asm view and no symbol stepping. only asm stepping.
<re_irc> <Noah> but you can get around this with breakpoints ofc.
<re_irc> <omar_u8> Thanks, I've been trying to get that to work, but problems Im facing now are with regards to VSCode recognising the probe-rs-debugger extension. I followed the instructions here (https://probe.rs/docs/tools/vscode/) but still no luck yet.
starblue1 has quit [Ping timeout: 240 seconds]
starblue1 has joined #rust-embedded
<re_irc> <Noah> omar_u8: yo will get better luck in finding help in #probe-rs:matrix.org (https://matrix.to/#/#probe-rs:matrix.org) :)
<re_irc> <omar_u8> Noah: Got it, thanks!
ni has quit [Quit: WeeChat 3.0]
ni has joined #rust-embedded
Guest2 has quit [Quit: Client closed]
Guest261 has joined #rust-embedded
cr1901 has quit [Read error: Connection reset by peer]
cr1901 has joined #rust-embedded
Guest261 has quit [Quit: Client closed]
Guest215 has joined #rust-embedded
<re_irc> <Mehmet Ali> korken89: I had asked you some questions about the accelerometer crate, I have a new one. Is it crazy to implement "Vector" there for x,y,z and t?
<re_irc> <Mehmet Ali> Or,anyone, really.
<re_irc> <Mehmet Ali> timestamp is critical in my case.
<re_irc> <Mehmet Ali> And could the types of x,y,z be different than the type of t?
<re_irc> <ryan-summers> Why don't you just wrap the "Vector"in a container type with the timestamp as another member?
<re_irc> <ryan-summers> Is there any reason it couldn't just be:
<re_irc> struct TimedMeasurement {
<re_irc> stamp: TimeStamp,
<re_irc> measurement: accelerometer::vector::F32x3
<re_irc> };
<re_irc> <ryan-summers> Are you trying to pass an "impl Vector" object around instead?
<re_irc> <Mehmet Ali> Actually, this might be saner.
<re_irc> <Mehmet Ali> I am trying to implement this;
<re_irc> <Mehmet Ali> It expects Vector, to my understanding.
<re_irc> <ryan-summers> Can't you just create your own custom type and impl Vector for it?
<re_irc> <ryan-summers> In your app, you know concrete details of the type so you can use the non-trait portions of the type
<re_irc> <Mehmet Ali> Therefore I thought that I have to implement Vector first in my crate and use that
<re_irc> <ryan-summers> But you wouldn't be allowed to pass around any information about the time in a generic manner
<re_irc> <Mehmet Ali> Yep, that is what I had in ming.
<re_irc> <ryan-summers> Yeah, you would if you want to include timestamp based on my reading
<re_irc> <Mehmet Ali> Sort of a newbie, this bur for my type, right?
<re_irc> <Mehmet Ali> Of course not with a macro, explicitly.
<re_irc> <ryan-summers> Yeah, I'd imagine something similar, but obviously referring to your internal "self.measurement" type. Also, "from_iter()" might be hard unless your component also contains the timestamp
<re_irc> <ryan-summers> Or just default the timestamp to unset in that case
<re_irc> <Mehmet Ali> Oh you mean that having non equivalent types is an issue for the impl of FromIterator, right?>
<re_irc> <ryan-summers> Possibly?
<re_irc> <ryan-summers> It's just something I noticed
<re_irc> <Mehmet Ali> I think so.
<re_irc> <Mehmet Ali> Well, it is pretty horrendous, but tstamp LSB, MSB might also work
<re_irc> <Mehmet Ali> But I can also omit it, since implementation is internal
<re_irc> <Mehmet Ali> and I won't use that.
<re_irc> <ryan-summers> Actually yeah, you could serialize it somehow. But yeah, gross
<re_irc> <Mehmet Ali> And would unset that if it ever happens.
<re_irc> <ryan-summers> If you don't support it with your implementation, just do an "unimplemented!()" so code panics if it ever tries to use it
<re_irc> <ryan-summers> Which I'd say is better than half-implementing it and fighting bugs later
<re_irc> <Mehmet Ali> Ah perfect.
<re_irc> <Mehmet Ali> Good tip/.
Guest215 has quit [Quit: Client closed]
creich_ has joined #rust-embedded
creich has quit [Ping timeout: 240 seconds]
crabbedhaloablut has quit [Quit: No Ping reply in 180 seconds.]
crabbedhaloablut has joined #rust-embedded