Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
kdbus in the kernel
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3 ... 25, 26, 27  
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
Shamus397
Apprentice
Apprentice


Joined: 03 Apr 2005
Posts: 218
Location: Ur-th

PostPosted: Thu Dec 31, 2015 3:15 pm    Post subject: Reply with quote

Well, this is interesting. It seems that since kdbuswreck is meeting heavy resistance, they're trying to do an end run with this thing. LGPL-2, naturally. ;)
Back to top
View user's profile Send private message
Naib
Watchman
Watchman


Joined: 21 May 2004
Posts: 5689
Location: Removed by Neddy

PostPosted: Thu Dec 31, 2015 4:18 pm    Post subject: Reply with quote

Dupe
_________________
The best argument against democracy is a five-minute conversation with the average voter
Great Britain is a republic, with a hereditary president, while the United States is a monarchy with an elective king
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Fri Jan 01, 2016 7:49 pm    Post subject: Reply with quote

Shamus397 wrote:
Well, this is interesting. It seems that since kdbuswreck is meeting heavy resistance, they're trying to do an end run with this thing. LGPL-2, naturally. ;)
IDK, it looks like someone knocking up a basic "how to use memfd"; that and the barebones to clone a new namespace (which is already part of systemdbust.)

There isn't anything more: it certainly is nothing like a bus implementation; just minimal userland code to do the above, so far. (oh and someone's basic rb-tree, not very well implemented.)

My overwhelming impression is: are they really so stubborn that they cannot admit TIPC already does this much better? [1]

As for the C code, it's a shame they haven't done their time in ##c, never mind ##posix.
I'm not going to do a code review: they should go and learn the basics for themselves, including how to research in the first place.

About the only interesting part for me was that the S390 and CRIS architectures need the stack and flags parms to syscall(__NR_clone) passed the other way round. I certainly wouldn't use their "ABI" for it; it feels like another landgrab, since one would naturally just provide a static helper for the purpose, unused given that it's only going to be used in specific places that know it's a syscall.

Naming sucks as well, as per usual. IDK why "clever" people can't get their heads round the concept of notation.. they always seem to want to obfuscate, as if that proves how smart they are, instead of how dumb.

On the "UNIX sucks! POSIX sucks!" front, it's amusing to see the lofty comments about "a basic ISO-C11/POSIX environment"; the pretension in acting like they're progressive in pushing C11, rather than nubskull, is hilarious. Especially when you consider that all of the StdC headers they include were specified in C99.
And that their code cannot work with anything other than gcc, and will only ever compile, never mind run, on Linux.
And that their code is not standards-compliant in the first place.

[1] Note that Sievers and Poettering both received two copies of that email, in 2012.

Surely it's time to move past these chumps?
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Fri Jan 01, 2016 7:57 pm    Post subject: Reply with quote

truekaiser wrote:
Starting to think that GKH has mastered the art of passive-aggressiveness.

Definitely; he's a nasty piece of work, ime.
Quote:
Especially in light of what Landley wrote about recently in regards to getting mdev working, and having to deal with GKH and Sievers.

http://www.landley.net/notes-2015.html#05-07-2015

Yeah, I felt a lot of empathy for what he wrote; and I must confess, a certain sense of commiseration.
Quote:
And as i write that i glance once more at the Landley blog, and notice the crap being pulled about stable vs unstable. Where else have we seen that? Oh yeah, systemd. WTF is going on in the Linux world these days?! Is it yet another case of running out of external enemies and so we turn on each other?

digi_owl wrote:
I read that, and think. They are using instability as a form of lock in by only letting udev dev's know ahead of time what the api will change to.

Agreed; they've been using sysfs as a means of vendor (=team) lock-in for years now.
Such behaviour would never be tolerated within a software house.

But then, as Jenny Holzer said: abuse of power comes as no surprise.
Back to top
View user's profile Send private message
depontius
Advocate
Advocate


Joined: 05 May 2004
Posts: 3399

PostPosted: Fri Jan 01, 2016 8:35 pm    Post subject: Reply with quote

digi_owl wrote:
WTF is going on in the Linux world these days?!


Or perhaps someone thinks systemd is the way to desktop dominance, and doesn't realize that they're selling Linux's soul to do so.
_________________
.sigs waste space and bandwidth
Back to top
View user's profile Send private message
Shamus397
Apprentice
Apprentice


Joined: 03 Apr 2005
Posts: 218
Location: Ur-th

PostPosted: Mon Jan 04, 2016 5:37 pm    Post subject: Reply with quote

Naib wrote:
Dupe

Apologies, but I didn't see your topic. ;)
Back to top
View user's profile Send private message
Shamus397
Apprentice
Apprentice


Joined: 03 Apr 2005
Posts: 218
Location: Ur-th

PostPosted: Sat Jan 09, 2016 2:44 am    Post subject: Reply with quote

steveL wrote:
My overwhelming impression is: are they really so stubborn that they cannot admit TIPC already does this much better?

I think the reason that they're ignoring TIPC is that it's stable and works well, and thus doesn't fit into their worldview—not to mention the whole NIH syndrome that their project suffers from.

I have no idea what their gameplan is with BUS1 (isn't it cute, it almost looks like PID1, doesn't it?) but if I were to guess, I would say that they're going to push this as a replacement for DBUSt and you'll see the same metastasis with it that we've seen with DBUSt and other related projects. ;)
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Fri Jan 29, 2016 6:51 pm    Post subject: Reply with quote

steveL wrote:
My overwhelming impression is: are they really so stubborn that they cannot admit TIPC already does this much better?

Shamus397 wrote:
I think the reason that they're ignoring TIPC is that it's stable and works well, and thus doesn't fit into their worldview—not to mention the whole NIH syndrome that their project suffers from.

Heh, yeah, we can't be dealing with all that boring, reliable stuff that just works in the background, quietly minding its own business, and doing what it's supposed to do, and always has done.

No, we want innurvation, and interesting times, with conferences and kewl stuff to take home; insecure fanbois a must! That's how we keep score, well that and the money..
Quote:
I have no idea what their gameplan is with BUS1 (isn't it cute, it almost looks like PID1, doesn't it?) but if I were to guess, I would say that they're going to push this as a replacement for DBUSt and you'll see the same metastasis with it that we've seen with DBUSt and other related projects. ;)

IDK I have to admit I've lost all interest in what nuttiness they're going to come up with next.

I imagine they'll start banging on about "correctness" and "standards", just like McCreesh did when the much-vaunted speed of paludis turned out to be a load of hype, as a new way of brow-beating people, to avoid the embarrassing truth that even RedHat couldn't make their crapfest operate, when a) they have a shedload of resource to throw at it, and b) it's their in-house project.
Eventually, after another 5-10 years, they'll end up in the same place openrc was at 7 years ago, much like C++ ended up in the same place as LISP after 30 years of kitchen-sink, ad-hoc "throw it in and let the users patch it down for us."

Nonetheless, it's been good in terms of bumping into people I'd never have spoken with, like yourself, and being able to discuss cleaner ways of doing things, on the basis of forward planning in design from what sysadmins specify.
Rather than: landgrab and polish up the turd from self-proclaimed "kernel and userland experts", after we've crippled everyone's machines and wasted countless end-user hours.

The defence by the collective of its inheritance, has helped to remind us that Linux has always been about choice, and that dbust was a busted flush from the beginning.

Most of all, we have to conclude that we cannot trust anyone apart from ourselves when it comes to defending the ability to pass that legacy on to the next generation.
Certainly not corporations, and certainly not entitled nubskulls whose only real skill is the kind of PR a script would do better, and a burning belief that they're the one ("I'm a rockstar, right?")
Back to top
View user's profile Send private message
Tony0945
Advocate
Advocate


Joined: 25 Jul 2006
Posts: 3192
Location: Illinois, USA

PostPosted: Sat Jan 30, 2016 12:08 am    Post subject: Reply with quote

depontius wrote:
Or perhaps someone thinks systemd is the way to desktop dominance, and doesn't realize that they're selling Linux's soul to do so.


Don't realize? Or don't care?
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware All times are GMT
Goto page Previous  1, 2, 3 ... 25, 26, 27
Page 27 of 27

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum