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, 4, 5, 6 ... 25, 26, 27  Next  
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
ct85711
Veteran
Veteran


Joined: 27 Sep 2005
Posts: 1696

PostPosted: Thu Jan 08, 2015 12:56 am    Post subject: Reply with quote

For me, I stopped using any kind of ramfs, back in either 2001 or 02. So I remember using the old initrd quite well, and all the trouble I was encountering with it. Surprisingly enough, once I recompiled the kernel so that it didn't need a initrd/initramfs (on a original Redhat linux distro); all those troubles stopped and never had to worry about it since. The key thing is, Why should I be forced to use a initramfs when my system hasn't ever needed one to begin with? I have NEVER needed LUKS, LVM, or any other special thing. When my system dies, I download a random linux live cd to in the worse case to fix my problems. Either way from how I've seen it, if my drive fails, it doesn't matter what was on it, as what's dead is still dead, not going to change that. Note, that's why you back up your important information (I don't bother backing up anything, but nothing on my system is that critical if I loose it either).
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


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

PostPosted: Thu Jan 08, 2015 6:33 pm    Post subject: Reply with quote

steveL wrote:
Consider that you're using some disk tool in initramfs. Now that tool is upgraded on your main machine (ie the rootfs.) Your initramfs is using older software.

Now we hope that's not an issue, but sometimes it is, or might be. I'd rather not have to worry about it.

Similarly we were told on the list not to worry about building an initramfs since they're small, and quick to build. At the same time we were told that an "initramfs is the new root", so you can do all your recovery from there. Personally I'd rather not have to think about what to build in, but just use the rootfs, since I configure my kernel to start the rootfs without modules, when I install.

tclover wrote:
There are valid reasons... but I don't see a _valid_ to reject an initramfs as de facto (inflexible like) stance.

I'm not the one taking an inflexible stance though; the nub-skool are.

I'm perfectly content for anyone to use w/e TF they want, since it makes zero difference to me. I just don't see the point in pretending regressions are "innovative progress."

If I'm inflexible about anything, it is about not accepting snakeoil-salesmen at face-value.
Quote:
I do understand why those who did _not_ need an initramfs took such an opposition to such matter/sabotage.
However, the worries you don't want to think of are no different to other sensible updates (such as OpenRC when a major update hit the tree that require care that can render a system unbootable.)

No they're not. I never forget to run dispatch-conf, since update runs it for me automatically, and shouts at me if I didn't run it the last time against its advice.

Why should I add another item to my kernel upgrade routine, in order to do a load of unnecessary crap that I never needed before, and I don't need now, so that I can leave myself open to breakage I don't even have to think about now, all so that some twat who would've been kicked out of Computing within his first two years in my day, can strut around posturing about how everyone's machines are his?

Hand-waving about how everything is the same at the end of the day, doesn't add anything, it merely clouds the issue, which to me is essentially about diversity in an ecosystem as opposed to an unpleasantly-enforced monoculture.
Quote:
Such updates are worrisome, and there is no way to avoid them.

Really? Yet here you are saying I can do exactly that for a whole class of worry:
Quote:
I guess trying to avoid useless hassles is a sane stance.

But, I've been using an initramfs since 2011... and I did not bump into that system update which require rebuilding the initramfs.

So what? Don't extrapolate from your experience to the ROTW; that's just insane, especially when it comes to computing, imo.

It's not "a worrisome burden" for you; good for you. Use what you want, no-one minds or really even cares, apart from the neo-fascists who want to turn your computer into their device.

As for the part about init=/bin/bash is was just a minimal example; keep a sysresccd disk around, or download one from another box when things fsck up. Live-disks are two-a-penny, with GUI and everything.. But I don't see any major difference between relying on the /boot partition and relying on a small root on the same drive, apart from less work for me ;-)
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


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

PostPosted: Thu Jan 08, 2015 6:46 pm    Post subject: Reply with quote

ct85711 wrote:
Either way from how I've seen it, if my drive fails, it doesn't matter what was on it, as what's dead is still dead, not going to change that.
Note, that's why you back up your important information.

++
Back to top
View user's profile Send private message
MustrumR
n00b
n00b


Joined: 15 Nov 2011
Posts: 71
Location: Right here

PostPosted: Thu Jan 08, 2015 7:00 pm    Post subject: Reply with quote

About separate usr without initramfs this is what I would do:
- Have sys-apps/busybox on rootfs

Code:

cat > /init << EOF
#!/bin/bb
busybox mount /usr &&
exec /sbin/init || exec /bin/busybox sh
EOF


And boot with init=/init

This is fully compatible with the usr-gentoo overlay which tries to move some stuff from / to /usr.
Back to top
View user's profile Send private message
khayyam
Watchman
Watchman


Joined: 07 Jun 2012
Posts: 6228
Location: Room 101

PostPosted: Thu Jan 08, 2015 7:03 pm    Post subject: Reply with quote

steveL wrote:
I just don't see the point in pretending regressions are "innovative progress."

steve ... just to be clear, the use of an initramfs is a "regression"? Certainly they should not be required, but for LVM on a LUKS encrypted volume they are necessary. No, they are not the "new root", and no you shouldn't need them to have a separate /usr, they may even have to be maintained (which is actually less of an issue than its being made out to be ... as not all initramfs are created equal) ... but those questions seem to be unrelated to the initramfs facility itself. As tclover stated "there are valid reasons" so your "not being inflexible" seems rather ... well, inflexible (indeed, rather intolerant and dictatorial).

best ... khay
Back to top
View user's profile Send private message
saellaven
Guru
Guru


Joined: 23 Jul 2006
Posts: 502

PostPosted: Thu Jan 08, 2015 7:19 pm    Post subject: Reply with quote

khayyam wrote:
steveL wrote:
I just don't see the point in pretending regressions are "innovative progress."

steve ... just to be clear, the use of an initramfs is a "regression"? Certainly they should not be required, but for LVM on a LUKS encrypted volume they are necessary. No, they are not the "new root", and no you shouldn't need them to have a separate /usr, they may even have to be maintained (which is actually less of an issue than its being made out to be ... as not all initramfs are created equal) ... but those questions seem to be unrelated to the initramfs facility itself. As tclover stated "there are valid reasons" so your "not being inflexible" seems rather ... well, inflexible (indeed, rather intolerant and dictatorial).

best ... khay


Requiring an initramfs solely for a separate /usr IS indeed a regression... it wasn't broken until WilliamH intentionally broke it and the patches exist, though he won't accept them because he wants it broken. Ergo, it is a regression.

There's nothing wrong with an initramfs if you need one for technical reasons, but a separate /usr is not one of those technical reasons, despite what WilliamH and the rubber stamp Council claim, as we've proven,
Back to top
View user's profile Send private message
asturm
Developer
Developer


Joined: 05 Apr 2007
Posts: 6995
Location: Austria

PostPosted: Thu Jan 08, 2015 7:45 pm    Post subject: Reply with quote

saellaven wrote:
Requiring an initramfs solely for a separate /usr IS indeed a regression... it wasn't broken until WilliamH intentionally broke it and the patches exist, though he won't accept them because he wants it broken. Ergo, it is a regression.

Isn't that utter BS?

From the wiki:

Quote:
When the /usr partition is on a separate file system, tools and drivers that have files stored within /usr cannot be used unless /usr is available. If those tools are needed to make /usr available, then we cannot boot up the system.

Unless your boot needs something from /usr, you're set.
_________________
backend.cpp:92:2: warning: #warning TODO - this error message is about as useful as a cooling unit in the arctic
Back to top
View user's profile Send private message
khayyam
Watchman
Watchman


Joined: 07 Jun 2012
Posts: 6228
Location: Room 101

PostPosted: Thu Jan 08, 2015 9:13 pm    Post subject: Reply with quote

saellaven wrote:
Requiring an initramfs solely for a separate /usr IS indeed a regression... it wasn't broken until WilliamH intentionally broke it and the patches exist, though he won't accept them because he wants it broken. Ergo, it is a regression.

saellaven ... please do not conflate the "initramfs facility itself" with the rationale provided by developers for its "requirement", that wasn't the line of reasoning I was following, nor was it the subject of the quote from tclover and steveL's reply to which I was responding.

saellaven wrote:
There's nothing wrong with an initramfs if you need one for technical reasons, but a separate /usr is not one of those technical reasons, despite what WilliamH and the rubber stamp Council claim, as we've proven,

Fair enough, but it has nothing to do with what I was responding to, or the context in which steve responded to tclover.

best ... khay
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 09, 2015 4:24 am    Post subject: Reply with quote

saellaven wrote:
Requiring an initramfs solely for a separate /usr IS indeed a regression... it wasn't broken until WilliamH intentionally broke it and the patches exist, though he won't accept them because he wants it broken. Ergo, it is a regression.

khayyam wrote:
saellaven ... please do not conflate the "initramfs facility itself" with the rationale provided by developers for its "requirement", that wasn't the line of reasoning I was following, nor was it the subject of the quote from tclover and steveL's reply to which I was responding.

Hmm it was actually; you're changing my words or the context they were put in, to mean something else. FTR I'm not saying you're doing so maliciously (just in case someone who doesn't know us thinks we're arguing..)
steveL wrote:
I just don't see the point in pretending regressions are "innovative progress."

khayyam wrote:
steve ... just to be clear, the use of an initramfs is a "regression"?

No; when did I say the use of anything is a regression? Regressions is a wider point about various changes we've seen that are hyped for years as "essential" and then quietly not mentioned any more once they've served their purpose, which is in fact pretext to justify arm-twisting us into accepting bad ideas. And meantime our machines are clogged down with more and more crap, all inturgrated since ofc you want to feel that your pid1 is capable of introspecting glib; I mean, who wouldn't, right? Right? ;)

In this case, I see it as a regression in Gentoo that support for a working model is deliberately crippled, and people are fed utter nonsense to justify it. When we even mention it, we get accused of all sorts of sins, but no-one questions the people who fed them the lies in the first place. And they are lies, pure and simple; the technical discussion was had back then, the requirements were listened to, and proof of concept was delivered. Then the hand-waving and the "we don't want to listen to you, but here's some vague crap to make it sound reasonable" began, as it has so many times before.

That's a regression in the same way that dropping openrc and moving to systemd would be a regression: in the capability of the end-product, which in Gentoo's case is a from-source distro that you can do whatever you want with.

So much for "just do the work and prove your point, and we'll work with you." It's hard to keep up "good faith" when the people you're dealing with don't show any, and just use the very idea as yaf stick to beat you with, should you have the temerity to point out the emperor's clothes they're all raving about, don't seem to cover much.

But then, the developers are way behind the users, as a collective. Still, we only want them to keep churning out ebuilds, even if they're not very capable with bash, and have far too inflated an opinion of themselves. Arrogance in young males is nothing new; we've all been there. It's the older ones who nod and pretend to be wise, while doing sweet fanny adams about any of the messes they see in front of them, that annoy me; principally because they know better, but also because they are so smug about doing nothing to show leadership.

For all those still clinging to some ridiculous notion that Gentoo (and Linux more broadly) is "not about choice", prove it. Argue it out and show us how it's not; as atm all you've done is provide a propaganda^W marketing slogan with zero actual substance. An application of Linux that is locked-down doesn't prove it either: it just proves you can choose to do that with a Linux.
Back to top
View user's profile Send private message
saellaven
Guru
Guru


Joined: 23 Jul 2006
Posts: 502

PostPosted: Fri Jan 09, 2015 4:31 am    Post subject: Reply with quote

genstorm wrote:
saellaven wrote:
Requiring an initramfs solely for a separate /usr IS indeed a regression... it wasn't broken until WilliamH intentionally broke it and the patches exist, though he won't accept them because he wants it broken. Ergo, it is a regression.

Isn't that utter BS?

From the wiki:

Quote:
When the /usr partition is on a separate file system, tools and drivers that have files stored within /usr cannot be used unless /usr is available. If those tools are needed to make /usr available, then we cannot boot up the system.

Unless your boot needs something from /usr, you're set.


Correction... if you need something from /usr to mount /usr, then you need an initramfs. There is no technical reason why you can't mount /usr separately as long as you (can) mount /usr first otherwise... which is exactly what SteveL's patch does. The only reason that patch is not a part of openrc right now is because WilliamH is abusing his position to keep it out and further abused his position on the Council to enforce this political (because it isn't a technical) decision on everyone.

Thus, you need an initramfs if you use LUKS or whatever... and you need an initramfs if you buy the RedHat BS that "/usr is the new /" where we just haphazardly install anything and everything to /usr out of ignorance or laziness. You do not need an initramfs if everything you need to boot is on / and never have until WilliamH brought his agenda to Gentoo (and again, a trivial patch fixes it but he won't accept it).
Back to top
View user's profile Send private message
Ant P.
Watchman
Watchman


Joined: 18 Apr 2009
Posts: 5761

PostPosted: Fri Jan 09, 2015 5:04 am    Post subject: Reply with quote

saellaven wrote:
You do not need an initramfs if everything you need to boot is on / and never have until WilliamH brought his agenda to Gentoo (and again, a trivial patch fixes it but he won't accept it).

Heh, I now know where you're coming from. I got to deal with that brain damage first-hand when I was fixing the runit ebuild.

Unprivileged-uid-usable binaries dumped into /usr/sbin/, other (boot-required) stuff needlessly thrown in /usr/bin/ when /bin/ would have been fine... 4 months to apply one-line patches for things like, oh, having consoles *work properly* on boot... no prize for guessing the maintainer.
Back to top
View user's profile Send private message
khayyam
Watchman
Watchman


Joined: 07 Jun 2012
Posts: 6228
Location: Room 101

PostPosted: Fri Jan 09, 2015 9:02 am    Post subject: Reply with quote

steveL wrote:
khayyam wrote:
steve ... just to be clear, the use of an initramfs is a "regression"?

No; when did I say the use of anything is a regression? Regressions is a wider point about various changes we've seen that are hyped for years as "essential" and then quietly not mentioned any more once they've served their purpose, which is in fact pretext to justify arm-twisting us into accepting bad ideas. And meantime our machines are clogged down with more and more crap, all inturgrated since ofc you want to feel that your pid1 is capable of introspecting glib; I mean, who wouldn't, right? Right? ;)

steve ... well, context here is important ... tclover asked that the question of boot failure when using initramfs be elaborated, s/he then agrees with you wrt "inaccurate information being used to justify dropping support", and all through the exchange s/he is focused on the initramfs facility and not the rational provided by developers/upstream for dropping support for separate /usr (which, again, s/he agrees is not a valid reason for requiring an initramfs ... in fact s/he calls it "moronic"). You then respond in a roughshod manner taking what was said as somehow a justification for what s/he's already disavowed, and so the "valid reasons", and "reject[ion] of the initramfs de facto" is made a whipping boy for "regressions". In short, you constructed a stawman in place of the argument, or discussion, tclover was having, hence my question: is the use of an initramfs a regression ... as that is all tclover was defending, the facility itself and "valid reasons" for its use.

steveL wrote:
In this case, I see it as a regression in Gentoo that support for a working model is deliberately crippled, and people are fed utter nonsense to justify it. When we even mention it, we get accused of all sorts of sins, but no-one questions the people who fed them the lies in the first place. And they are lies, pure and simple; the technical discussion was had back then, the requirements were listened to, and proof of concept was delivered. Then the hand-waving and the "we don't want to listen to you, but here's some vague crap to make it sound reasonable" began, as it has so many times before.

But tclover had already described those justifications as "moronic", so you weren't even arguing against a position s/he'd taken ... if s/he'd argued such a case then your "regression" would have made sense.

best ... khay
Back to top
View user's profile Send private message
Anon-E-moose
Advocate
Advocate


Joined: 23 May 2008
Posts: 3925
Location: Dallas area

PostPosted: Fri Jan 09, 2015 11:01 am    Post subject: Reply with quote

It's kind of funny how the argument "I don't need or use an initramfs so you shouldn't or it's moronic"
sounds an awful lot like the justifications that LP uses for his choices in systemd "I don't need it so no one else should".

Seriously folks.
We all (pretty much) agree that the /usr thing re. udev wasn't a good idea. Let it go. Move forward instead of being stuck somewhere in the past.
_________________
Asus m5a99fx, FX 8320 - nouveau, oss4, rx550 for qemu passthrough
Acer laptop E5-575, i3-7100u - i965, alsa
---both---
5.0.13 zen kernel, profile 17.0 (no-pie) amd64-no-multilib
gcc 8.2.0, eudev, openrc, openbox, palemoon
Back to top
View user's profile Send private message
saellaven
Guru
Guru


Joined: 23 Jul 2006
Posts: 502

PostPosted: Fri Jan 09, 2015 12:46 pm    Post subject: Reply with quote

Anon-E-moose wrote:
It's kind of funny how the argument "I don't need or use an initramfs so you shouldn't or it's moronic"
sounds an awful lot like the justifications that LP uses for his choices in systemd "I don't need it so no one else should".


To be clear, I have no problem with the people that want or need an initramfs and I believe it is certainly a valuable tool for those who need it. I'm not against the facility existing, just the mentality that we HAVE to use it. The Council's decision to mandate it was the moment I got involved in the systemd debate precisely because I'm all about choice and they're trying to remove choice.

Quote:

Seriously folks.
We all (pretty much) agree that the /usr thing re. udev wasn't a good idea. Let it go. Move forward instead of being stuck somewhere in the past.


I only brought it up because someone asked and I tried to give the short version rather than give all the minutia. Further, the simple fact is the problem still exists primarily because WilliamH isn't out for the most robust openrc possible.
Back to top
View user's profile Send private message
depontius
Advocate
Advocate


Joined: 05 May 2004
Posts: 3383

PostPosted: Fri Jan 09, 2015 12:48 pm    Post subject: Reply with quote

Anon-E-moose wrote:
It's kind of funny how the argument "I don't need or use an initramfs so you shouldn't or it's moronic"
sounds an awful lot like the justifications that LP uses for his choices in systemd "I don't need it so no one else should".

Seriously folks.
We all (pretty much) agree that the /usr thing re. udev wasn't a good idea. Let it go. Move forward instead of being stuck somewhere in the past.


Hmmmm.... Something of a middle-ground. Some of the /usr-related complaints were being directed toward the OpenRC developer, suggesting that he's "gone over to systemd" and is therefore not sufficiently interested any more. Normally when one loses interest in an owned project, you hand the reins over to others who still are. To put the tin-foil hat on for a moment, the argument could be made that OpenRC is being gently ridden into the ground.

That's perhaps a tin-foil hat theory, (I find it fun now and then to keep one laying around.) but it is worth remembering that the time may come when it becomes necessary to fork OpenRC. Some may argue that one such time was back with the /usr issue. That time/opportunity missed, the /usr fix can be queued as one of the first enhancements, shortly followed by start-stop-daemon enhancements being discussed in other threads.
_________________
.sigs waste space and bandwidth
Back to top
View user's profile Send private message
tclover
Guru
Guru


Joined: 10 Apr 2011
Posts: 516

PostPosted: Fri Jan 09, 2015 1:11 pm    Post subject: Reply with quote

Thanks Khayyam, for clarifying a few things which is _more_ than necessary seeing how SteveL, and a few others, like to keep that confusion between the _facility_ and being _forced_ to use the "facility" (initramfs) going as far as _never_ use an initramfs (because it's, _"obviously"_, too much a maintenance _burden_.)

After reading SteveL response and subsequents other answers, I decided to not waste unnecessary times reading or posting subsequent posts which would be, for sure, _mere_ waste of time.
_________________
home/:mkinitramfs-ll/:supervision/:e-gtk-theme/:overlay/
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 09, 2015 5:54 pm    Post subject: Reply with quote

khayyam wrote:
steve ... just to be clear, the use of an initramfs is a "regression"?

steveL wrote:
No; when did I say the use of anything is a regression? Regressions is a wider point about various changes we've seen that are hyped for years as "essential" and then quietly not mentioned any more once they've served their purpose, which is in fact pretext to justify arm-twisting us into accepting bad ideas. And meantime our machines are clogged down with more and more crap, all inturgrated since ofc you want to feel that your pid1 is capable of introspecting glib; I mean, who wouldn't, right? Right? ;)

khayyam wrote:
steve ... well, context here is important ... tclover asked that the question of boot failure when using initramfs be elaborated, s/he then agrees with you wrt "inaccurate information being used to justify dropping support", and all through the exchange s/he is focused on the initramfs facility and not the rational provided by developers/upstream for dropping support for separate /usr (which, again, s/he agrees is not a valid reason for requiring an initramfs ... in fact s/he calls it "moronic").

No s/he also in fact asserted that I am taking an "inflexible stance" in rejecting initramfs across the board which I never did. That is the strawman here, and you are merely repeating it, despite being told twice "no, that's not what I am saying".

I am not sure what I can do beyond reiterate that both of you are spinning my position up in to something it is not, and then attacking that supposed position of mine.

As usual, everyone focusses on the people discussing the issue, rather than the actual issue. And we get the usual cognitive dissonance crap about "can't we all just get along? move on.." implying as ever that the problem is the people, not the actual issue.

Instead of personalising it, just talk about the issue. Since we've all agreed that there was no technical reason for the drop of support, despite the pretext of several being presented, it was a bad move. Period.

Now by all means move on: just don't pretend that issue has anything whatsoever to do with how we discussed things here, now.
Back to top
View user's profile Send private message
khayyam
Watchman
Watchman


Joined: 07 Jun 2012
Posts: 6228
Location: Room 101

PostPosted: Fri Jan 09, 2015 7:18 pm    Post subject: Reply with quote

steveL wrote:
khayyam wrote:
steve ... well, context here is important ... tclover asked that the question of boot failure when using initramfs be elaborated, s/he then agrees with you wrt "inaccurate information being used to justify dropping support", and all through the exchange s/he is focused on the initramfs facility and not the rational provided by developers/upstream for dropping support for separate /usr (which, again, s/he agrees is not a valid reason for requiring an initramfs ... in fact s/he calls it "moronic").

No s/he also in fact asserted that I am taking an "inflexible stance" in rejecting initramfs across the board which I never did. That is the strawman here, and you are merely repeating it, despite being told twice "no, that's not what I am saying".

steve ... no, that conflation begins with you, you move from "boot failure when using initramfs" to the whole question of separate /usr, and what "your told on the list [...] initramfs is the new root", etc, so you come down as practically dismissive of any suggestion of an initramfs being in any way useful.

Additionally, when you return with your "I just don't see the point in pretending regressions are innovative progress", this is in reply to the following ... which you'd sniped:

tclover wrote:
There are valid reasons... but I don't see a _valid_ to reject an initramfs as de facto (inflexible like) stance. I do understand why those who did _not_ need an initramfs took such an opposition to such matter/sabotage.

To me that reads altogether different to an "assert[ion] that [you are] taking an "inflexible stance" in rejecting initramfs across the board", if you read what its in response to you have a slew of reasons most of which don't relate to what tclover had asked for ... so, a barrage of reasons (again, mostly about separate /usr) that hardly comes across as anything but down on initramfs. Again, your judgement was that this discussion is *all* about separate /usr ... but that was far from the question asked, and there was no disagreement there.

steveL wrote:
I am not sure what I can do beyond reiterate that both of you are spinning my position up in to something it is not, and then attacking that supposed position of mine.

You could take it as an object lesson in how not to rub people the wrong way, myself included. If you want to now state that we are "spinning your position" I'll remind you that this "position" was focused on what you wanted to bang on about, and paid little attention to what tclover had actually asked ... so, hows that for spin?

steveL wrote:
As usual, everyone focusses on the people discussing the issue, rather than the actual issue. And we get the usual cognitive dissonance crap about "can't we all just get along? move on.." implying as ever that the problem is the people, not the actual issue.

It might focus on "people" because someone might wonder why you feel the need to be so aggressive about such a minor point (particularly as the major point you wanted to make re "regression" wasn't even contested) ... you're acting like people we're actually in substantive disagreement. Novel idea, maybe I said something because you *were* acting kind of boorish.

steveL wrote:
Instead of personalising it, just talk about the issue. Since we've all agreed that there was no technical reason for the drop of support, despite the pretext of several being presented, it was a bad move. Period.

Honestly, good advice sometimes comes from the wrong person ...

steveL wrote:
Now by all means move on: just don't pretend that issue has anything whatsoever to do with how we discussed things here, now.

... but so does bad, or unintelligible, advice ...

best ... khay
Back to top
View user's profile Send private message
WWWW
Tux's lil' helper
Tux's lil' helper


Joined: 30 Nov 2014
Posts: 143

PostPosted: Sat Jan 10, 2015 12:14 pm    Post subject: Reply with quote

I have to say this to support SteveL.

Yes, initramfs does have its drawbacks.

I think what SteveL says or aims is for simplicity in a process in general. Cut off all the unnecessary stuff as much as possible to have something working.

initramfs did become somewhat bloated over time. I remember the time the kernel itself was able to boot itself alone.

I had to give up this idea due to using zfs, or when using lvm. Now I have to resort to genkernel which is immensely bloated and grabs tons of stuff to build the initramfs.

The difference between one building something and knowing where all the pieces fall in contrast to build something blindly adding everything in the hopes this way it'll boot. The former has it's merit.

Here the conflict is taking away the former option without apparent reason.

Take as an example GRUB vs UEFI booting. Grub has taken away the option to boot in a simple way and favored placing files all over the place.

Before everything needed used to reside under /boot/grub/ now the files are scattered all over the place. Even the more intuitive grub file in /etc has a big fat warning on top "DON'T CHANGE ME HERE IT WILL BE OVERWRITTEN!!".

Grub has evolved to a configuration completely automated but it fails in many scenarios where somebody doesn't use the 'ubuntu linux' experience.

Try to set a boot entry manually in grub2. Plain schizophrenic. Crazy order of things, insane syntax, modules dependencies, etc. And prolly you end up with a borked entry that needs further fixing.

Now try UEFI boot entry.

1- kernel in /boot (this is an extra step due to initramfs)
2- boot kernel in /esi
3- boot the machine

Without an initramfs it get's reduced to 2 steps:

1- kernel in /esi
2- boot the machine

No fucking around with insane amount of files, syntax, etc whatsoever. I never got around using grub2 due to this.

It's been a while since I don't use separate /usr anymore thanks to systemd isis terrorism borking my system. So I can't go into detail about this.

In other words, I supposed it comes down to defending the "simplicity of a process", nothing more and nothing less. Sure initramfs can solve or work in many scenarios, but if it can be done in a more concise way? Why not? Why destroy, impede, discourage the 'simplicity of a process'?

thanks
Back to top
View user's profile Send private message
ct85711
Veteran
Veteran


Joined: 27 Sep 2005
Posts: 1696

PostPosted: Sat Jan 10, 2015 5:25 pm    Post subject: Reply with quote

The part I am waiting for is when we get to the point that initramfs is no longer needed at all, even for zfs and lvm and stuff. Yes, I know it's been said the kernel devs said they will not add support to the kernel, but who said anything that a package adds onto the kernel such that you install the package that adds an addition to the kernel (not meaning as a module, actual addition to the kernel). You recompile the kernel adding the support for the package into the kernel. I'm not meaning adding a kernel module that gets loaded after the kernel is running (from what I've seen/read, for boot up having the core parts as a module won't work).
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


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

PostPosted: Sat Jan 10, 2015 6:24 pm    Post subject: Reply with quote

steveL wrote:
No s/he also in fact asserted that I am taking an "inflexible stance" in rejecting initramfs across the board which I never did. That is the strawman here, and you are merely repeating it, despite being told twice "no, that's not what I am saying".

khayyam wrote:
steve ... no, that conflation begins with you, you move from "boot failure when using initramfs" to the whole question of separate /usr, and what "your told on the list [...] initramfs is the new root", etc, so you come down as practically dismissive of any suggestion of an initramfs being in any way useful.

That is pure inference on your part, and I have told you several times that it is incorrect. When will you simply accept my word that that is not what I am saying?
Quote:
Additionally, when you return with your "I just don't see the point in pretending regressions are innovative progress", this is in reply to the following ... which you'd sniped:
tclover wrote:
There are valid reasons... but I don't see a _valid_ to reject an initramfs as de facto (inflexible like) stance. I do understand why those who did _not_ need an initramfs took such an opposition to such matter/sabotage.

To me that reads altogether different to an "assert[ion] that [you are] taking an "inflexible stance" in rejecting initramfs across the board"

Utter nonsense: that's exactly what s/he was getting at: you have reasons, but no reason "to reject an initramfs as de facto (inflexible like) stance". The clear implication is that I am taking an inflexible stance and it is unjustified.

Since I am not taking such a stance, the premise is flawed, so the conclusion is irrelevant.
steveL wrote:
Since we've all agreed that there was no technical reason for the drop of support, despite the pretext of several being presented, it was a bad move. Period.

Now by all means move on: just don't pretend that issue has anything whatsoever to do with how we discussed things here, now.

Quote:
... but so does bad, or unintelligible, advice ...

I'm sorry you're having such trouble with reading comprehension. ;) Seems perfectly clear to me: how we discuss things here has got absolutely zero to do with the flawed, politically-driven decision. And that was the issue we were discussing: no-one's really that interested in meta-discussion.
Back to top
View user's profile Send private message
greyspoke
Tux's lil' helper
Tux's lil' helper


Joined: 08 Jan 2010
Posts: 146

PostPosted: Sat Jan 10, 2015 7:37 pm    Post subject: Reply with quote

WWWW - roll your own initramfs. I am not a programmer, but I managed to do this (when I had a separate /usr) well enough to assemble arrays with mdadm, start lvm volumes and then mount and check /usr and /var with some basic shell scripting.

It is explained pretty well on the wiki: https://wiki.gentoo.org/wiki/Custom_Initramfs
Back to top
View user's profile Send private message
khayyam
Watchman
Watchman


Joined: 07 Jun 2012
Posts: 6228
Location: Room 101

PostPosted: Sat Jan 10, 2015 9:10 pm    Post subject: Reply with quote

steveL wrote:
khayyam wrote:
... no, that conflation begins with you, you move from "boot failure when using initramfs" to the whole question of separate /usr, and what "your told on the list [...] initramfs is the new root", etc, so you come down as practically dismissive of any suggestion of an initramfs being in any way useful.

That is pure inference on your part, and I have told you several times that it is incorrect. When will you simply accept my word that that is not what I am saying?

steve ... yes, inference ... derived from premises ... are you saying that without access to your inner thoughts that any such premises are false, and so I should simply accept your word on the matter?

steveL wrote:
khayyam wrote:
To me that reads altogether different to an "assert[ion] that [you are] taking an "inflexible stance" in rejecting initramfs across the board"

Utter nonsense: that's exactly what s/he was getting at: you have reasons, but no reason "to reject an initramfs as de facto (inflexible like) stance". The clear implication is that I am taking an inflexible stance and it is unjustified.

There is such a thing as generosity of interpretation, and no I simply do not see such a "clear implication", but then my focus is on the context, whereas you seem to focused entirely on the "implied" meaning of that one sentence.

steveL wrote:
Since I am not taking such a stance, the premise is flawed, so the conclusion is irrelevant.

Way to fail at logic ... your "stance" in relation to a request re "boot failure when using initramfs" is to offload on the questioner the grievances WRT separate /usr, "unpleasantly-enforced monoculture", etc, etc ... basically acting rather unpleasant ... when the poster then says (to paraphrase) "there are valid reasons ... I don't see any reason to reject it outright" you take this as 1). "falsely" accusing you of some position you don't hold 2). providing something you can justly respond to ... and so bring in another clang of "regression" a la separate /usr ... and 3). further your stance in regard to the whole question ... "as usual, everyone focuses on the people discussing the issue, rather than the actual issue" ... the "actual" issue being whatever you happen to want the issue to be. When I draw your attention to this you claim this is misrepresentation, spin, and I can't possibly know what you really meant so I should take your word on the matter. In short, I should ignore the context and any possibility of deriving premises from that context.

steveL wrote:
I'm sorry you're having such trouble with reading comprehension. ;) Seems perfectly clear to me: how we discuss things here has got absolutely zero to do with the flawed, politically-driven decision. And that was the issue we were discussing: no-one's really that interested in meta-discussion.

I still say that particular sentence is incomprehensible, anyhow, shame that the post in question has nothing whatsoever to do with the issue "we" are discussing ... but then it seems that you've decided what is, and what isn't, "meta" to your discussion.

best ... khay
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


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

PostPosted: Sun Jan 11, 2015 4:12 pm    Post subject: Reply with quote

@khayyam: I don't know why you're taking this so personally, nor indeed accusing me of all kinds of stuff. You seem to want to hold me to some artificial standard of conduct no-one else has to follow. My replies must be tailored to exactly what the other party has said, and I am not allowed to go off on a tangent, or if I do, I must be admonished about it over several posts, none of which acknowledge my right to do exactly that.

And it's not like you don't go off on tangents (sometimes massively so); but somehow if I state that the one thing we seem to get with consistency from the nub-skool is regression, that must be "aggression" on my part, and I must be dragged over the coals for it. Just for the record, I wasn't being aggressive; and you cannot possibly know what tone I am taking since this is non-verbal. So please, let it go.

Or don't; either way I'm out of this sub-thread as it is both tedious and futile. I'm sorry you're evidently so stressed-out atm.

All I'd say is: as usual people are trying to turn the spotlight on to the person discussing the flaw, rather than simply discussing the flaw, and moving on. Dunno if that's cognitive dissonance or what; all I know is it's both very unpleasant, and nothing I want to emulate. Ever.
Back to top
View user's profile Send private message
Naib
Watchman
Watchman


Joined: 21 May 2004
Posts: 5625
Location: Removed by Neddy

PostPosted: Fri Jan 16, 2015 8:24 pm    Post subject: Reply with quote

v3 is out

http://lkml.iu.edu/hypermail/linux/kernel/1501.2/00487.html
_________________
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
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware All times are GMT
Goto page Previous  1, 2, 3, 4, 5, 6 ... 25, 26, 27  Next
Page 5 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