<thejpster[m]>
I think `&mut 64` will convert to `*mut u64` automatically, but other references (like `&str` or `&[8]`) are larger than a pointer and thus won't. I might cast the reference to a pointer first just to be sure.
<thejpster[m]>
TIL: `clobber_abi("C")`
<dirbaio[m]>
yep it's valid
<dirbaio[m]>
* yep it's a valid thing to do
<dirbaio[m]>
you must respect the invariants of the reference though. for example don't keep it around beyond the borrow's lifetime, don't write to it if it's not mut...
<jason-kairos[m]>
Makes sense. I also don't know if my clobber_abi is really valid or necessary for a cortex SVC since the processor already saves those registers I think
<dirbaio[m]>
in this case it seems fine
<dirbaio[m]>
(assuming svc #123 immediately writes the time to the pointer then returns)
<jason-kairos[m]>
I wondered about that, but never tested it. Good to know the compiler has my back with regards to preventing nonsensical behavior.
<jason-kairos[m]>
* I wondered about that, but never dared to test it. Good to know the compiler has my back with regards to preventing nonsensical behavior.
jiande2020 has quit [Quit: Ping timeout (120 seconds)]
jiande2020 has joined #rust-embedded
pcs38 has joined #rust-embedded
pcs38 has quit [Quit: leaving]
sirhcel[m] has quit [Quit: Idle timeout reached: 172800s]
starblue has quit [Ping timeout: 276 seconds]
ouilemur has quit [Quit: WeeChat 4.5.2]
ouilemur has joined #rust-embedded
starblue has joined #rust-embedded
pcs38 has joined #rust-embedded
<thejpster[m]>
I don't have time to attend the meeting today. Please take a moment to review robamu's excellent work on derive-mmio and cortex-ar. Lots of good progress.
<thejpster[m]>
I hope to see some of you Nuremberg next week where you can come and see demos of analysing defmt and RTIC performance on STM32 using a SEGGER J-Trace tracing probe and running Rust on the NXP S32Z270 SoC using our new cortex-ar crates and a Lauterbach TRACE32 debug probe - or anything else Rust related.
<thejpster[m]>
Also I'd like to highlightly the open job I posted earlier for a German-speaking Embedded Rust trainer (it's not just training, it's these demos too).
<thejpster[m]>
Oh, one more thing - I think I unblocked defmt-1.0 with an approach to defmt-on-std that doesn't require any API changes, and that should be moving ahead again.
<adamgreig[m]>
ok, let's begin! only announcement from me is that the new arm team now officially exists; i'll update the repo and crates.io permissions as appropriate tonight
<adamgreig[m]>
chrysn: wanna go first?
<adamgreig[m]>
or is it chrysn...
chrysn[m] has joined #rust-embedded
<chrysn[m]>
Yeah sorry, feel free to kick out the old me, that server is dead.
<chrysn[m]>
It's really short: We are here, we try to integrate a lot of existing work into ready-to-use embedded RTOS usability.
<chrysn[m]>
We're active, we try to contribute to all the projects that we use to make the ecosystem better, and already have good examples that, without per-board configuration, run on a wide variety of boards, be it cortex-m, RISC-V or Xtensa.
<jannic[m]>
Yes, it has been open for half a year and has been kept up to date by the author, but didn't get a review yet. In the issue being solved by that PR, https://github.com/rust-embedded/heapless/issues/501, @Dirbaio discussed the possible solution strategies with the author and if I understand the topic correctly, the approach is valid.
bartmassey[m] has joined #rust-embedded
<bartmassey[m]>
Great find — thanks for bringing it!
<adamgreig[m]>
tricky issue..
<adamgreig[m]>
well a few minutes reading through it hasn't helped me any
<jannic[m]>
I can't claim that I understand what's going on there :-) and it surely takes some time to review.
<adamgreig[m]>
😅 hopefully libs team can have a look at the pr after the meeting, i don't think we'll make any headway during the meeting unless someone is around
<adamgreig[m]>
thanks for bringing it up!
<adamgreig[m]>
bartmassey then, wanna talk about the new disco book?
<bartmassey[m]>
it looks to me like a valid merge, in that the folks who have thought it through all like it. Might want to macroize it later I guess.
<bartmassey[m]>
Yeah, I've embarrassingly found that since losing an active collaborator the Disco Book has fallen way down in my stack. I am looking for somebody to work with me regularly on it, help organize new Book Sprint, etc.
<bartmassey[m]>
Lame, but there it is. Anyone interested feel free to message me here or whatever.
<adamgreig[m]>
i am hoping to be out of house-restoration hell by rustweek so if nothing else let's chat then, but if anyone is keen on helping bring the books up to date please get in touch!
<adamgreig[m]>
ok, that's everything from the agenda, is there anything else anyone wanted to discuss this week?
<jannic[m]>
Not directly for the meeting, but in case someone has hints on how to properly manage DMA buffers in rust: I wasn't able to give a really good answer to https://github.com/rp-rs/rp-hal/issues/905
rmsyn[m] has joined #rust-embedded
<rmsyn[m]>
since there seems to be a number of OS-related projects, does it make sense to propose a separate OS working group, would it still fall under the rust-embedded WG?
therealprof[m] has joined #rust-embedded
<therealprof[m]>
Those projects are all independent. I think there would have to be some significant interest from those projects to organise within the WG.
<bartmassey[m]>
As somebody currently working on an educational OS project, I'd be happy to collaborate with anybody and everybody. Not sure what the collaboration would be though.
<adamgreig[m]>
yea, we could add a team if there was interest but the question would be what they work on. it could also be a matrix channel to start with and see if enough interested people want to get together?
<bartmassey[m]>
I'll try to answer the DMA thing a little bit at some point soon.
<bartmassey[m]>
There's already a Rust OS matrix channel I think?
<bartmassey[m]>
I'll look for it and post it.
<bartmassey[m]>
Got to go to class. Peace all.
<thejpster[m]>
<rmsyn[m]> "since there seems to be a number..." <- They should all join The Rust Society
<JamesMunns[m]>
JamesMunns[m]: I need to write a follow up with the notes from the most recent council meeting. It's fairly wishy washy at the current moment. Any words of feedback or support welcome, as always.
<rmsyn[m]>
I think that is probably what I had in mind. some sort of umbrella for collaboration across projects for sharing ideas, techniques, tools, etc.
<thejpster[m]>
Tock, Ariel, Linux, Zephyr - all should be talking to each other but there’s no venue for it
<therealprof[m]>
Ooh, secret societies...
<therealprof[m]>
TIL I learned about typst
<therealprof[m]>
* TIL about typst
<JamesMunns[m]>
therealprof[m]: a+ would recommend
<therealprof[m]>
I immediately fell in love. 😉
<JamesMunns[m]>
using a gist to host the pdf too is a neat trick - you can't upload files, but you CAN post the source as a gist, then clone the gist as a repo, add the pdf, and push
<JamesMunns[m]>
(gotta run, have a good week everyone! I'll also be at Embedded World with a fun Poststation demo, and won't be able to attend the next wg meeting)
<adamgreig[m]>
thanks, enjoy embedded world!
<adamgreig[m]>
if there's nothing else that anyone wants to discuss then let's finish here, thanks everyone!
<adamgreig[m]>
thejpster: if you have a minute of jorge's time, could you see if he could add me as an owner on cortex-m-rt-macros? then i can also add the new arm team to it
<adamgreig[m]>
and for arm-dcc, cortex-ar, and cortex-r-rt, could you add the arm team? cargo owner --add github:rust-embedded:arm
<adamgreig[m]>
that was quick, thank you!
<thejpster[m]>
<adamgreig[m]> "and for arm-dcc, cortex-ar..." <- Is that possible on an iPhone?
<thejpster[m]>
I have to manually go to /owvers as I can’t find the link
starblue has quit [Ping timeout: 252 seconds]
<thejpster[m]>
Oh ffs I was on the wrong GitHub account.
<adamgreig[m]>
no rush
<thejpster[m]>
“Only members of a team can add the team”
<adamgreig[m]>
oh, lol
<adamgreig[m]>
well feel free to add me, or wait til you're a member of the team and then add it
<thejpster[m]>
I invited you
starblue has joined #rust-embedded
jiande2020 has quit [*.net *.split]
kline has quit [*.net *.split]
wose has quit [*.net *.split]
jiande2020 has joined #rust-embedded
kline has joined #rust-embedded
wose has joined #rust-embedded
<JamesMunns[m]>
By the way, is anyone interested in HIL testing, and know what they would need external to your target board to do the testing you'd want to do?
<JamesMunns[m]>
I'm looking into using Poststation for cases like this: using standard boards like rp2040/2350 as the i/o interfaces, for either driving some serial comms, or lowish sample rate analog/digital sniffing
<JamesMunns[m]>
Very interested what real world needs are, in terms of timing requirements, bandwidth for sampling, the level of automation people want/need, etc.
<JamesMunns[m]>
Hoping to build some basic remote control interface firmwares as open source components, as part of the poststation-util repo
<JamesMunns[m]>
(happy to chat here, or send me a DM if it needs to be private)