| Summary: | sys-boot/plymouth-openrc-plugin: doesn't work and causes plymouth with intel drivers does not show splash animations in the shutdown process, though it does at boot | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Kete Tefid <ketetefid> |
| Component: | Current packages | Assignee: | No maintainer - Look at https://wiki.gentoo.org/wiki/Project:Proxy_Maintainers if you want to take care of it <maintainer-needed> |
| Status: | RESOLVED OBSOLETE | ||
| Severity: | normal | CC: | ketetefid, treecleaner |
| Priority: | Normal | Keywords: | PMASKED |
| Version: | 10.1 | ||
| Hardware: | AMD64 | ||
| OS: | Linux | ||
| Whiteboard: | Pending removal: 2016-09-21 | ||
| Package list: | Runtime testing required: | --- | |
| Attachments: |
Kernel configuration for gentoo-sources-3.8.1
Plymouth shutdown log Output of emerge --info (on working system) Kernel configuration for gentoo-sources-4.4.6 |
||
|
Description
Kete Tefid
2013-04-23 08:57:54 UTC
Created attachment 346360 [details]
Kernel configuration for gentoo-sources-3.8.1
Created attachment 346362 [details]
Plymouth shutdown log
That's weird. It works for me™, and I have intel HD 4000. Which model is yours? And do you use any initramfs? If so, which tool do you use to generate it? (In reply to comment #3) > That's weird. It works for me™, and I have intel HD 4000. Which model is > yours? It's an Arrandale cpu combined with an Ironlake gpu (aka intel hd) The cpu: http://ark.intel.com/products/50178 The gpu: http://en.wikipedia.org/wiki/Comparison_of_Intel_graphics_processing_units#Fifth_generation It's really weird. In your idea is it more related to the kernel itself or to the driver, i.e. xf86-vidoe-intel? I should mention that I have not tested different intel drivers. Thus, I think I should give different verison of the intel driver a try. I'll inform you of the result. (In reply to comment #4) > And do you use any initramfs? If so, which tool do you use to generate it? I used dracut for producing the initramfs. The only module was plymouth. emerge -pv dracut [ebuild R ~] sys-kernel/dracut-027-r1 USE="optimization -debug -device-mapper -net (-selinux)" DRACUT_MODULES="plymouth -biosdevname -bootchart -btrfs -caps -cifs -crypt -crypt-gpg -crypt-loop -dmraid -dmsquash-live -gensplash -iscsi -livenet -lvm -mdraid -multipath -nbd -nfs -ssh-client -syslog -systemd" 0 kB (In reply to comment #3) > That's weird. It works for me™, and I have intel HD 4000. Which model is > yours? Also, could you please kindly let me know your configuration (kernel, intel driver and other related information) so that I can try to check the issue with your configuration as well? the video-intel driver does not matter at all in my understanding. You are not under X11, so the X DDX driver is not involved at all. Only KMS and drm are to my understanding. And I think they are not the problem at all. I use dracut too. Do you use openrc? With the sys-boot/plymouth-openrc-plugin ? (In reply to comment #8) > the video-intel driver does not matter at all in my understanding. You are > not under X11, so the X DDX driver is not involved at all. Only KMS and drm > are to my understanding. And I think they are not the problem at all. > > I use dracut too. Do you use openrc? With the > sys-boot/plymouth-openrc-plugin ? Yes, you are right because I just tested different versions (stable and keyworded) with the same results. Furthermore, I now remember that I did a lot of tricks to test the issue and one of them were in the time when I had disabled xdm. So, this is only kernel related. I have tested (3.7.1/3.7.4/3.8.1/3.8.2 pf/gentoo) and now I am going to test vanilla 3.9.rc8 and 3.6.x and 3.5.x. This is going to be an epic testing! What a stubborn bug to kill... I don't think it is kernel either. That's why I asked about openrc and the plugin. It might just be a wrong communicatin between the init system and plymouth. But since I don't use openrc it is a bit hard for me to understand if this can be the case or not. And I just realized that I'm listed as the maintainer of the openrc plugins.... well since proxy-maint is in CC: guys can you please remove me from that? I maintain plymouth and that's ok but I have only systemd on my only gentoo system, can't test the openrc plugin. Sorry. (In reply to comment #9) > > sys-boot/plymouth-openrc-plugin ? Yes, [ebuild R ~] sys-boot/plymouth-openrc-plugin-0.1.2 0 kB While diggin every website and downloading irc logs to find something useful, I got this: https://github.com/aidecoe/plymouth-openrc-plugin/issues/2 A gentoo user explained a problem, but I don't think it is related. What's your idea? Furthermore, the plugin is two years old. Also, did you have a chance to have a look at the plymouth debug log? (In reply to comment #10) > I don't think it is kernel either. That's why I asked about openrc and the > plugin. It might just be a wrong communicatin between the init system and > plymouth. > > But since I don't use openrc it is a bit hard for me to understand if this > can be the case or not. > > And I just realized that I'm listed as the maintainer of the openrc > plugins.... well since proxy-maint is in CC: guys can you please remove me > from that? I maintain plymouth and that's ok but I have only systemd on my > only gentoo system, can't test the openrc plugin. Sorry. Hmmmm. I changed my idea accordingly. I am going to test systemd instead of the kernel testing. Just because gentoo was udev friendly, I stayed with udev to avoid bugs. But now it seems that it might be worth to change to systemd. I will install systemd and will report the results. (In reply to comment #11) > A gentoo user explained a problem, but I don't think it is related. What's > your idea? > Furthermore, the plugin is two years old. > Also, did you have a chance to have a look at the plymouth debug log? My idea is that plymouth-openrc-plugin is no more maintained [I don't maintain it, again I was not even aware I was the maintainer of that package in gentoo O_O], so it should just be removed. I looked at the log, but I don't really understand what's going on. This is pretty sad I know, sorry about that, but I can't work on plymouth-openrc-plugin. Let's hope someone will step up to maintain it. If you want to try other kernels feel free to do so, but I think the problem is the openrc plugin. It is 2 years old, no surprise it is going to break. So dont' want to waste your time. The intel KMS and drm code are pretty solid, I really doubt the problem is the kernel. About systemd: why not just dropping plymouth? I mean having plymouth doesn't sound as a good reason to switch init system, really. You are going to hit bigger problems with systemd. Also systemd in gentoo is in an experimental stage. It is far from being well supported. This is not to change your mind, this is just to let you know what your are going to find. Indeed your help is welcome if you want to contribute :) (In reply to comment #13) > (In reply to comment #11) > > A gentoo user explained a problem, but I don't think it is related. What's > > your idea? > > Furthermore, the plugin is two years old. > > Also, did you have a chance to have a look at the plymouth debug log? > > My idea is that plymouth-openrc-plugin is no more maintained [I don't > maintain it, again I was not even aware I was the maintainer of that package > in gentoo O_O], so it should just be removed. Thank you anyway for your nice and quick replies. > > I looked at the log, but I don't really understand what's going on. > :( > This is pretty sad I know, sorry about that, but I can't work on > plymouth-openrc-plugin. Let's hope someone will step up to maintain it. > It is completely OK. The gentoo users would be grateful if you could offer a person to maintain it. > If you want to try other kernels feel free to do so, but I think the problem > is the openrc plugin. It is 2 years old, no surprise it is going to break. > So dont' want to waste your time. The intel KMS and drm code are pretty > solid, I really doubt the problem is the kernel. > You are right. > About systemd: why not just dropping plymouth? I mean having plymouth > doesn't sound as a good reason to switch init system, really. You are going > to hit bigger problems with systemd. Also systemd in gentoo is in an > experimental stage. It is far from being well supported. This is not to > change your mind, this is just to let you know what your are going to find. > I actually came over again to say that, after a brief searching, I found the new init system experimental in gentoo (exactly as you stated) and there were some complaints about it with some bugs unanswered in the forum (eeeeeie...). Simply I became too afriad to switch to it since it might break many things while I do rely on openrc and its way of init scripting and have my own custom stuff. Thus, it's not wise to switch to it unless Gentoo officials officially support it with a clear change of the default init system. > Indeed your help is welcome if you want to contribute :) I will try to have a look on the openrc plugin for plymouth. I think it will be more beneficial than changing the whole layout just to have plymouth and then become trapped in its bugs (And it seems that after switching to systemd, people have had problems switching back to openrc). Thus, you can leave this bug either "opened" or just "won't fix until a maintainer for the plugin appears." Thank you very much for your support and company. I'll leave it open, at least for now. What now puzzle me is that on you AMD hardware it is working. It is still a weak argument given the plugin is 2 year old and not maintained, but still.... (In reply to comment #10) > And I just realized that I'm listed as the maintainer of the openrc > plugins.... well since proxy-maint is in CC: guys can you please remove me > from that? I maintain plymouth and that's ok but I have only systemd on my > only gentoo system, can't test the openrc plugin. Sorry. + 25 Apr 2013; Sergey Popov <pinkbyte@gentoo.org> metadata.xml: + Drop maintainer by his request, remove proxy maintainers herd, reassigning to + maintainer-needed All done (In reply to comment #16) > + 25 Apr 2013; Sergey Popov <pinkbyte@gentoo.org> metadata.xml: > + Drop maintainer by his request, remove proxy maintainers herd, > reassigning to > + maintainer-needed > > All done Thank you Sergey commit 7f805bae938b58b86f697da02258e2ebe0372981 Author: Sam Jorna <wraeth@gentoo.org> Date: Tue Apr 26 20:21:36 2016 +1000 sys-boot/plymouth: remove proxy maintainer Proxy maintainer has requested to drop the package per mail to the project. Removing both maintainer and project from metadata.xml. FWIW, I installed plymouth-openrc-plugin today on a Lenovo T61 w/ Intel graphics following the instructions available at https://wiki.gentoo.org/wiki/Plymouth . Everything works as expected, showing the splash at boot, reboot and shutdown. I tested both with and without an initramfs (generated through genkernel-next-64) with the same results (except that w/o the iniramfs the splash starts after around 20 lines of OpenRC loading services etc). I tested once available themes, but extensively performed my tests with "spinfinity" one. Incidentally, I misread the Wiki and masked >plymouth-0.9.0 (instead of >=) and everything worked fine w/ 0.9.0 too. I then tried 0.9.2 but got some artifacts during boot. Now I'm back to plymouth-0.8.8-r4 in order to fill this information. Since the original submitter was using gentoo-sources-3.8.1 and current stable (the one I'm using) is now gentoo-sources-4.4.6, I'd ask to test again, as the plymouth-openrc-plugin seems to work fine here (and, by reading the available source code, it doesn't look obsolete in regards to openrc code). I'm available to provide further information if needed. Pasting emerge --info produces an error (message too long), so I will attach it as a file. The currently (relevant) installed packages: danae ~ # emerge genkernel-next plymouth plymouth-openrc-plugin -pv These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild R ] sys-boot/plymouth-0.8.8-r4::gentoo USE="gtk libkms pango -debug -gdm -static-libs" VIDEO_CARDS="intel -radeon" 0 KiB [ebuild R ] sys-kernel/genkernel-next-64::gentoo USE="plymouth -cryptsetup -dmraid -gpg -iscsi -mdadm (-selinux)" 0 KiB [ebuild R #] sys-boot/plymouth-openrc-plugin-0.1.2::gentoo 0 KiB Created attachment 444776 [details]
Output of emerge --info (on working system)
Created attachment 444778 [details]
Kernel configuration for gentoo-sources-4.4.6
thanks for testing, unmasked Incidentally, I believe this has nothing directly to do with Intel graphics, but rather the fact that the plugin orders Plymouth to terminate almost immediately after starting up, at which point it may or may not have had time to actually show the splash screen, depending on graphics hardware. https://github.com/tlhonmey/plymouth-openrc-plugin has a fixed version that waits until mount-ro (near the end of the shutdown) to terminate Plymouth instead of doing it at localmount (which is near the beginning in most cases.) Since aidecoe is no longer maintaining the plugin, how do you want to handle this? Specifically, should I create a new bug since this one was marked resolved, even though the root cause really wasn't? |
