| 1 |
Am 19.10.2013 17:02, schrieb Daniel Campbell: |
| 2 |
> On 10/17/2013 11:27 PM, Mark David Dumlao wrote: |
| 3 |
>> https://www.linux.com/news/featured-blogs/200-libby-clark/733595-all-about-the-linux-kernel-cgroups-redesign |
| 4 |
>> |
| 5 |
>> Not sure if I read that just right... but since nobody is doing cgroup |
| 6 |
>> management besides systemd, in practice the cgroups implementation in |
| 7 |
>> Linux wasn't very consistent. So since systemd is doing it, their work |
| 8 |
>> is helping shape the kernel's cgroups api? |
| 9 |
>> |
| 10 |
>> Interesting... |
| 11 |
>> |
| 12 |
> >From my perspective it looks like systemd developers are trying to push |
| 13 |
> their ideas into the kernel, almost like they intend to merge systemd |
| 14 |
> *with* the kernel. |
| 15 |
|
| 16 |
from what I read in the article cgroups are a mess and are cleaned up |
| 17 |
anyway. The only real user of cgroups at the moment is systemd. |
| 18 |
Others are welcome to make use of cgroups too. But in the current state |
| 19 |
nobody blames them for not jumping in. |
| 20 |
|
| 21 |
> If systemd is the only implementation of cgroups and |
| 22 |
> their developers are working on cgroup support in the kernel, it spells |
| 23 |
> calamity given their history of evangelism and zealotry. |
| 24 |
|
| 25 |
well, going over some old ml threads on fedora mailing lists all I could |
| 26 |
find was that Poettering and Sievers DID listen and DID make changes if |
| 27 |
the demand was high enough. |
| 28 |
|
| 29 |
Sure, I dislike systemd. Sure what happened with udev was a dick move. |
| 30 |
But their 'zealotry' is a lot less developed than the zealotry of those |
| 31 |
who exploded about using an 'init-thingy' in the future. |
| 32 |
|
| 33 |
> |
| 34 |
> I truly wish I understood why a single userland program and its |
| 35 |
> developers are being given the keys to an entire subsystem of the |
| 36 |
> kernel. |
| 37 |
they aren't. |
| 38 |
|
| 39 |
> Their changes to udev have proven to be a headache for users, |
| 40 |
|
| 41 |
yes? which ones? |
| 42 |
|
| 43 |
> and the kernel is held to a much higher standard of stability and |
| 44 |
> interoperability. In addition, the top-level developers of systemd (and |
| 45 |
> GNOME, and the now-deprecated consolekit/polkit/udisks/etc) are employed |
| 46 |
> by a for-profit company (Red Hat), which has a vested interest in |
| 47 |
> shaping Linux as a platform. They and other corporations cannot be |
| 48 |
> trusted with stuff like this... |
| 49 |
|
| 50 |
hm, Redhat is one of the companies investing the most money into linux |
| 51 |
kernel, userland, graphics... if you 'don't trust them' you are pretty |
| 52 |
much 20 years too late. |
| 53 |
|
| 54 |
> |
| 55 |
> I'd like to see what Linus has to say about this if/when he finds out. |
| 56 |
> He's not impressed with Sievers or Poettering. Personally I'd like to |
| 57 |
> see them ostracized from the community and contained to their own |
| 58 |
> distro, where they belong. |
| 59 |
> |
| 60 |
so much about zealotry. |
Archives