Do not worry about the #Rust for Linux DMA drama that is making the rounds in the news and on social media: disagreements like that happen every now and then in #LinuxKernel land – and almost always some solution is found sooner or later that everybody can live with somehow. Either by the community itself, or by Linus himself.
https://www.heise.de/en/news/Is-a-kernel-developer-blocking-the-success-of-Rust-for-Linux-Yes-and-no-10269318.html #kernel #rustlang
2/ BTW, let me kinda repeat an impression I shared a few months ago[1] after participating in the #Rust for #Linux discussion on the invitation only #Kernel maintainer summit (https://lwn.net/Articles/991062/):
Many #LinuxKernel core maintainers including Linus really want Rust for Linux to succeed.
Sure, there are technical problems and hurdles that need to be solved – and humans to be motivated and convinced. But given such strong backing I'm pretty sure that will happen.
3/ Ohh, and if you haven't seen this, see also this post from @sima and the reply from @airlied about the #Rust for #Linux DMA drama:
https://lore.kernel.org/lkml/Z6OzgBYZN[email protected]/T/#u
'"The r4l team and drm maintainer team have this sort of thing in hand, it's not like we don't understand the community of the Linux kernel, and having this first reaction to blow shit up and dramatise it just isn't helpful."'
@kernellogger @sima @airlied kinda torn between the "trust in the institutions" and "the institutions have failed us" takes here.
im at least optimistic that the kernel (as an institution) has demonstrated a pretty good capability to change when change is justified... but still, that most of Marcan's points are more or less ignored in favour of shunning him for acting uncivilised doesn't fill me with optimism
@cas @kernellogger @sima @airlied the issue isn't bringing up those points. Most involved there not even denied that those are actual problems.
The issue is more, that you can't really have it both ways.
If you go full drama public and burn everything down, then people might just decide not wanting to deal with you any further. There are all human beings in the end. And sure, it's super frustrating, and everybody gets that.
But you also need to stay pragmatic and find your paths forward.
@karolherbst @kernellogger @sima @airlied yeah i suppose. but nobody speaking out (being "uncivil") is also how we end up with screwed up institutions
bweh idk,,
@cas @kernellogger @sima @airlied yeah.. I fully understand that dilemma.
If the goal was to just stir up shit and call it out, fine, that's kinda deserved, but you also have to stay realistic about the consequences.
It's a long ongoing issue and I don't want this to be a "yeah everybody has to stay civil" take either.
Sometimes it's entirely needed to call things out. Just sometimes people also perceive it as meddling and will have to deal with the fallout behind the scenes we don't see.
@karolherbst @cas @kernellogger @airlied yeah this isn't about "being civil" or "friendly" and even less about "don't call out"
this is about calling out in an ineffective and imprecise way, so that the people actually trying to change things are busy patching up your collateral damage instead of implementing real changes, while all you've achieved is airing your frustration for internet drama points
@karolherbst @cas @kernellogger @airlied when you're that harmful with your calling out, eventually I'm going to be fed up, and you get a live round shot across your bow
and if that then causes you to ragequit, because you can't actually deal with the heat you've been dishing out coming back around the corner:
fuck off
@karolherbst @cas @kernellogger @airlied or as Dave put it
"Being toxic on the right side of an argument is still toxic, [...]"
@karolherbst @cas @kernellogger @airlied so please do call out broken things, do change things, I've been doing it for over a decade in the linux kernel
but do it in a way that you're not just becoming part of the problem and making it bigger
@sima @karolherbst @cas @kernellogger @airlied Amen to that, but how are the r4l people supposed to feel when Linus doesn't reiterate that this work is welcomed?
I would have wanted a word to Hellwig saying something like he understands the frustration but it doesn't excuse the behaviour nor the outright refusal for rust drivers to share code. No one is asking him to maintain it further than any other driver using the subsystem, on the contrary.
@mupuf @karolherbst @cas @kernellogger @airlied Linus chimes in when he gets the separate rust dma abstraction pull requests, including Christoph's nack and an ack from a pile of other maintainers
and the people who know the process are 99% sure about what answer will come out of that interaction
there's a process for this stuff, and Linus generally doesn't preempt a discussion unless absolutely necessary
so the r4l people aren't worried here, just fed up with the internet drama points bs
@mupuf @karolherbst @cas @kernellogger @airlied the other aspect is the code of conduct enforcement, which yes is broken
but that is also not news
plus generally something that happens in private and takes a while - fdo coc also doesn't step in immediately. so again, maybe wait for a bit more until you judge how broken the process really is
not sure what my bet is, but could very well be that the lkml coc team weighs in after due deliberations too
@sima @karolherbst @cas @kernellogger @airlied Absolutely!
But Justice takes time, hence why I would have wanted a public word from Linus that would deescalate and show support to both sides, so that fair collaboration may be resumed.
It was just sad to see how antagonistic this thread ended up, and seeing some people taking in the punches without much public support... No surprise it made the buzz, even before the nuke was dropped on it. I feel sorry for everyone involved.
@sima @karolherbst @cas @kernellogger @airlied I agree that the final decision should be at pull time, and Linus should not preempt/meddle with discussions before that... but I think there are limits that should not be crossed by anyone and when it happens, it should be addressed by containing and deescalating rather than leaving it to fester. This is crisis communication 101.
The current process works, but given the heightened emotions, it is fair to say it can be improved.
@norbu @sima @karolherbst @cas @kernellogger @airlied Yep.
Seems like "being toxic on the wrong side of an argument" just to demonstrate the privilege the old guard has over the younglings.
@sima @karolherbst @cas @kernellogger @airlied
Sorry for the stupid question, but what harm did Marcan's posts outside the mailing list actually do? I find lkml threads difficult to navigate and I don't know off-hand who is or is not a common participant, but I couldn't see many drive-by responses in the thread itself; were there a bunch that I missed? Or was it just drama outside the mailing list, and if so, why does that matter?
Thorsten Leemhuis (acct. 1/4)
kcxt
☃️karolherbst☃️
🐧sima🐧
Martin Roukala (né Peres)
Norbert Preining
soc
Kyle Strand