Package Details: plymouth-git 22.02.122.r48.g9e72df3d-1

Git Clone URL: (read-only, click to copy)
Package Base: plymouth-git
Description: A graphical boot splash screen with kernel mode-setting support (Development version)
Upstream URL:
Licenses: GPL
Conflicts: plymouth
Provides: plymouth
Submitter: PirateJonno
Maintainer: Taijian
Last Packager: Taijian
Votes: 266
Popularity: 1.70
First Submitted: 2009-05-02 09:53 (UTC)
Last Updated: 2022-07-19 19:43 (UTC)

Required by (140)

Sources (14)

Latest Comments

Taijian commented on 2022-08-10 08:50 (UTC)

@Lutrim: Thanks for your report, I'll look into it when I get back from vacationnext week.

Lutrim commented on 2022-08-10 08:21 (UTC) (edited on 2022-08-10 08:24 (UTC) by Lutrim)

~/» sudo mkinitcpio -P              
==> Building image from preset: /etc/mkinitcpio.d/linux-libre.preset: 'default'
  -> -k /boot/vmlinuz-linux-libre -c /etc/mkinitcpio.conf -g /boot/initramfs-linux-libre.img
==> Starting build: 5.18.14-gnu-1
  -> Running build hook: [base]
  -> Running build hook: [udev]
  -> Running build hook: [plymouth]
  -> Running build hook: [autodetect]
  -> Running build hook: [modconf]
  -> Running build hook: [block]
  -> Running build hook: [filesystems]
  -> Running build hook: [keyboard]
  -> Running build hook: [plymouth-encrypt]
==> ERROR: file not found: `/usr/share/fonts/cantarell/Cantarell-Thin.otf'
==> ERROR: file not found: `/usr/share/fonts/cantarell/Cantarell-Regular.otf'
  -> Running build hook: [lvm2]
  -> Running build hook: [fsck]
==> Generating module dependencies
==> Creating zstd-compressed initcpio image: /boot/initramfs-linux-libre.img
==> WARNING: errors were encountered during the build. The image may not be complete.
==> Building image from preset: /etc/mkinitcpio.d/linux-libre.preset: 'fallback'
  -> -k /boot/vmlinuz-linux-libre -c /etc/mkinitcpio.conf -g /boot/initramfs-linux-libre-fallback.img -S autodetect
==> Starting build: 5.18.14-gnu-1
  -> Running build hook: [base]
  -> Running build hook: [udev]
  -> Running build hook: [plymouth]
  -> Running build hook: [modconf]
  -> Running build hook: [block]
==> WARNING: Possibly missing firmware for module: isci
  -> Running build hook: [filesystems]
  -> Running build hook: [keyboard]
  -> Running build hook: [plymouth-encrypt]
==> ERROR: file not found: `/usr/share/fonts/cantarell/Cantarell-Thin.otf'
==> ERROR: file not found: `/usr/share/fonts/cantarell/Cantarell-Regular.otf'
  -> Running build hook: [lvm2]
  -> Running build hook: [fsck]
==> Generating module dependencies
==> Creating zstd-compressed initcpio image: /boot/initramfs-linux-libre-fallback.img
==> WARNING: errors were encountered during the build. The image may not be complete.

Installing cantarell-fonts does not change anything...

daenney commented on 2022-07-29 14:47 (UTC)

Whatever it was, it seems to have fixed itself. I wonder if I ended up forgetting to rebuild something, but I'm positive I rebuilt plymouth-git first and then gdm-plymouth before rebuilding the initramfs. The only real change between then and now is a slightly newer kernel.

For reference, the hardware in this case was a Lenovo IdeaPad 5 Pro 14ACN6. Initcpio is a systemd based stack (HOOKS=(base systemd keyboard sd-plymouth autodetect sd-vconsole modconf sd-encrypt fsck filesystems)) with amdgpu as the first entry in MODULES.

Taijian commented on 2022-07-22 20:56 (UTC)

@daenney: Also helpful would be some information about your hardware setup (NVIDIA?) and your mkinitcpio.conf.

whynothugo commented on 2022-07-22 18:56 (UTC)

@daenney Can you post logs from journalctl? You might still have the ones for last time you tried; you'll need to use journalctl --since XXX --until YYY.

That should point out what's failing.

daenney commented on 2022-07-22 17:21 (UTC)

Since 22.02.122.r48.g9e72df3d I can't boot with plymouth-git anymore. It gets past the sd-encrypt prompt, but at some point something locks, gdm doesn't launch and nothing responds to input anymore. I've rebuilt gdm-plymouth too and rebuilt the initramfs, but no dice. I get a kernel oops when booting with it. If I disable plymouth things boot just fine.

Reverting to 22.02.122.r2.g716b5bd solved the issue for me, but I'm entirely unclear as to why.

dacioromero commented on 2022-07-19 19:17 (UTC)

There was an update to plymouth making --with-runtimedir obsolete.

Switching it to --runstatedir seems to do the trick, but I'm not sure if they're strictly equivalent.

diff --git a/PKGBUILD b/PKGBUILD
index ecaa0a2..82eb13e 100644
@@ -4,7 +4,7 @@
 #               Lara Maia, Padfoot, Jorge Barroso, carstene1ns, Sebastian Lau

 pkgdesc="A graphical boot splash screen with kernel mode-setting support (Development version)"
@@ -88,7 +88,7 @@ build() {
        --with-background-end-color-stop=0x4D4D4D \
        --without-rhgb-compat-link \
        --without-system-root-install \
-       --with-runtimedir=/run
+       --runstatedir=/run


michaldybczak commented on 2022-03-15 19:21 (UTC)

@Taijian, the problem with documentation is that it's never 100% clear for regular users as me. Arch wiki is also often very badly written, in a dry, technical, manner, assuming some greater, wider knowledge.

From my point of view it looks like that: the package worked well for years, and now it stopped. I don't know what changed and where to look.

I have two options. Either I try to find some answers, so I could fix the issue, or give up on the package, since it became worthless to me. The fact that it worked well, and I was even able once to find and report a bug that was later fixed, shows that, even regular users as me, could provide something valuable. However, we can't do that if we have no support. I'm used to troubleshooting myself, but in this case, this is too obscure to know where to start. Maybe if I searched and red for hours, I would figure it out. I don't have that kind of time, whereas for someone familiar, it's quicker to process information from me and point me toward the solution. I know, I know, everybody is busy and that is why, I don't expect quick answers. As you said, there is little incentive to help individual users (which is a pity, but this is how it is), so possibly I won't get much help or at all, but I can at least try.

sjnims commented on 2022-03-15 02:21 (UTC) (edited on 2022-03-15 02:25 (UTC) by sjnims)

@Taijian, I switched to using a similar boot configuration as you, so instead of using udev and GRUB, I'm now using systemd-style init, and booting from efistub. However, still not getting the Fedora-style unencrypt screen. My efistub parameters are as follows:

efibootmgr --disk /dev/nvme0n1 --part 1 --create --label "Arch Linux" \
--loader /vmlinuz-linux-zen --unicode 'initrd=\intel-ucode.img \
initrd=\initramfs-linux-zen.img<UUID of 2nd nvme partition>=cryptroot \
root=UUID=<UUID of cryptroot> rootflags=rw,subvol=@' --verbose

Boots much faster than w/ GRUB, and I don't see any console messages at all now which is a plus, but as I mentioned, still no Fedora-style. Look forward to hearing what you find out!

sjnims commented on 2022-03-14 21:19 (UTC) (edited on 2022-03-14 21:19 (UTC) by sjnims)

@Taijian, it also doesn't help that Fedora uses dracut vs mkinitcpio to generate their initramfs...makes it harder to peek at what they're doing! :P

Taijian commented on 2022-03-14 20:45 (UTC)

@michaldybczak: Since the info I asked you for is very specific to Arch (and derivatives), I do not think that upstream will find your bug report very interesting/helpful. I does however suggest to me that you have not yet fully read through the wiki pages for mkinitcpio.conf or GRUB.

Please do so and then come back with the info I asked for.

Taijian commented on 2022-03-14 20:42 (UTC)

@sjnims: Uups, yeah, my bad - that what I get for typing from memory...

Other than that: Thank you for trying. I guess upstream (dominated by Red Hat) is mostly on all-systemd systems, so they probably do not test for udev compatability as much - or the issue is with our encrypt_hook... I'll have another look when I find some hours to do so!

michaldybczak commented on 2022-03-14 19:31 (UTC)

@Taijan, I submitted a bug:

sjnims commented on 2022-03-14 18:07 (UTC)

@Taijian, neither worked, and I think it was consolefont (vconsole threw an error when generating the initramfs) - see, sd-vconsole is used when using systemd init vs busybox init (I think that means replacing udev w/ systemd, along with a few others depending on if using encrypt, plymouth, etc...).

Taijian commented on 2022-03-14 16:09 (UTC)

@sjnims: Thank you for your mail!

Could you try the following: Add 'keymap' to your HOOKS array (between 'keyboard' and 'modconf')? If that doesn't work: also add 'vconsole' after that?

Just to see if either of these are the culprits, because from a cursory reading of the code, the 'Fedora' splash might be looking for a keymap...

sjnims commented on 2022-03-14 11:20 (UTC)

@Taijian, interesting!

My setup is:

  • mkinicpio generates the initramfs, same as you
  • I use udev vs systemd
  • Same module loaded early (i915), also loading btrfs early as that is the file system I'm using
  • cantarell-fonts is installed
  • booting from grub
  • kernel command line = 'quiet loglevel=3 rd.udev.log_level=3 splash cryptdevice=UUID=<device UUID>:cryptroot root=/dev/mapper/cryptroot'

Taijian commented on 2022-03-14 10:13 (UTC)

@michaldybczak: Please do provide more information in order for us to help you debug.

  • What hardware do you use?
  • Which kernel?
  • initramfs generation method?
  • boot loader?
  • initramfs generation config?
  • kernel command line?
  • have you installed any opt-depends?
  • any other information that might be helpful here...

Taijian commented on 2022-03-14 10:09 (UTC)

@sjnims: OK, this is fascinating! I also use BGRT and I DO get the 'Fedora' password prompt and zero messages displayed... So, let's debug, shall we?

My setup is as follows:

  • mkinitcpio generates the initramfs
  • I use the systemd-style init system (as opposed to udev)
  • Intel iGPU with i915 loaded early in MODULES
  • extra/cantarell-fonts is installed
  • booted from efistub
  • 'quiet splash' on the kernel command line

What does your setup look like?

sjnims commented on 2022-03-14 02:01 (UTC) (edited on 2022-03-14 04:42 (UTC) by sjnims)

@Taijian, I’m using the BGRT theme, the output I get is like the Fedora theme in that it displays the laptop manufacturer’s logo, the spinner below that, and then the Arch logo below that, up until it requests my password in order to unencrypt my drive. At that point it looks more like the Ubuntu theme with the text “Please unlock disk xyz…” vs like Fedora 35 where it’s a picture of a keyboard with the keyboard layout next to that (in my case US) below the password prompt. I don’t know what F36 beta looks like or Fedora rawhide for that matter, so maybe they didn’t merge Hans’ branch in yet, or if they do it’s not in the main branch yet? Also, even after following the wiki on how to get a silent boot, I still see some messages displayed, e.g. "Loading Linux 5.16..." whereas on Fedora 35 there are no messages at all. "Gee, why don't you just use Fedora?" I know, I've used it before and while I do like it, I don't like dnf and I love pacman.

michaldybczak commented on 2022-03-13 17:07 (UTC)

I noticed, that the plymouth doesn't work from some time. I tried to recompile it, but it didn't help. When I boot, in place of a plymouth, I just see a black screen with "Plymouth Terminate" notification.

Taijian commented on 2022-03-12 15:33 (UTC) (edited on 2022-03-12 15:34 (UTC) by Taijian)

@sjnims: Maybe I am getting old, but I was under the impression that the changes you reference (regarding the difference between Ubuntu & Fedora login flow) are already part of upstream plymouth and therefore well integrated into this package?

If yours looks different: Which theme are you using?

EDIT: me no typy well...

sjnims commented on 2022-03-11 17:32 (UTC)

Thanks for maintaining this for us! Question: is there a way to grab the changes that Hans de Goede has in his fork for Fedora 35 ( He made some changes to how the Plymouth themes display the text box for entering the password to unencrypt an encrypted drive (see for sample pictures). I guess it stems from changes Hans made as he details in a blog post here: Fedora actually uses Hans' fork for the F35 release, see the rpm spec here:

cthulhaire commented on 2022-01-30 21:57 (UTC)

I made a small change to the plymouth-encrypt hook which allows it to skip the password prompt if the root device has already been decrypted earlier.

diff --git a/plymouth.encrypt_hook b/plymouth.encrypt_hook
index 8acb9af..97b1c45 100644
--- a/plymouth.encrypt_hook
+++ b/plymouth.encrypt_hook
@@ -69,29 +69,33 @@ EOF
     if resolved=$(resolve_device "${cryptdev}" ${rootdelay}); then
         if cryptsetup isLuks ${resolved} >/dev/null 2>&1; then
             [ ${DEPRECATED_CRYPT} -eq 1 ] && warn_deprecated
-            dopassphrase=1
-            # If keyfile exists, try to use that
-            if [ -f ${ckeyfile} ]; then
-                if eval cryptsetup --key-file ${ckeyfile} luksOpen ${resolved} ${cryptname} ${cryptargs} ${CSQUIET}; then
-                    dopassphrase=0
-                else
-                    echo "Invalid keyfile. Reverting to passphrase."
-                fi
-            fi
-            # Ask for a passphrase
-            if [ ${dopassphrase} -gt 0 ]; then
-                echo
-                echo "A password is required to access the ${cryptname} volume:"
-                plymouth ask-for-password --prompt="Password for ${cryptname} volume" --dont-pause-progress --number-of-tries=5 --command="/sbin/cryptsetup luksOpen --key-file=- ${resolved} ${cryptname} ${cryptargs} ${CSQUIET}"
-                sleep 2
-            fi
+            # Check whether device has already been decrypted
             if [ -e "/dev/mapper/${cryptname}" ]; then
-                if [ ${DEPRECATED_CRYPT} -eq 1 ]; then
-                    export root="/dev/mapper/root"
-                fi
+                echo "${cryptname} already unlocked, continuing..."
-                err "Password succeeded, but ${cryptname} creation failed, aborting..."
-                exit 1
+                dopassphrase=1
+                # If keyfile exists, try to use that
+                if [ -f ${ckeyfile} ]; then
+                    if eval cryptsetup --key-file ${ckeyfile} luksOpen ${resolved} ${cryptname} ${cryptargs} ${CSQUIET}; then
+                        dopassphrase=0
+                    else
+                        echo "Invalid keyfile. Reverting to passphrase."
+                    fi
+                fi
+                # Ask for a passphrase
+                if [ ${dopassphrase} -gt 0 ]; then
+                    echo "A password is required to access the ${cryptname} volume:"
+                    plymouth ask-for-password --prompt="Password for ${cryptname} volume" --dont-pause-progress --number-of-tries=5 --command="/sbin/cryptsetup luksOpen --key-file=- ${resolved} ${cryptname} ${cryptargs} ${CSQUIET}"
+                    sleep 2
+                fi
+                if [ -e "/dev/mapper/${cryptname}" ]; then
+                    if [ ${DEPRECATED_CRYPT} -eq 1 ]; then
+                        export root="/dev/mapper/root"
+                    fi
+                else
+                    err "Password succeeded, but ${cryptname} creation failed, aborting..."
+                    exit 1
+                fi
         elif [ -n "${crypto}" ]; then
             [ ${DEPRECATED_CRYPT} -eq 1 ] && warn_deprecated

whynothugo commented on 2021-12-12 02:03 (UTC)

Can gdm-plymouth.service be dropped?

The link is broken and systemd reports over and over this when applying presets.

It seems that it's no longer recommended or necessary anyway:

hoban commented on 2021-11-09 16:31 (UTC)

I ran into the same issue as @medoix which was fixed on my end by manually deleting the directory ~/.cache/paru/clone/plymouth-git. Is there any easier way to clean build? Perhaps it was awkward on my end because of an issue on the Paru side? I did not have luck building this package after running paru -Sc, for what it's worth.

medoix commented on 2021-10-29 19:44 (UTC)

@Taijan A cleanBuild fixed it thanks! Always the simple things you forget to check.

Taijian commented on 2021-10-29 11:10 (UTC)

@medoix: Have you tried a clean build directory? Because it sure works for me...

medoix commented on 2021-10-29 07:35 (UTC) (edited on 2021-10-29 07:36 (UTC) by medoix)

Trying to update to 0.9.5.r84 and getting the following, any ideas?

==> Starting prepare()...
can't find file to patch at input line 4
Perhaps you used the wrong -p or --strip option?
The text leading up to this was:
|diff -aur plymouth.src/scripts/plymouth-update-initrd plymouth.patch/scripts/plymouth-update-initrd
|--- plymouth.src/scripts/plymouth-update-initrd        2014-02-20 19:38:13.319427393 +1100
|+++ plymouth.patch/scripts/plymouth-update-initrd      2014-02-20 19:54:13.035553244 +1100
File to patch:

Taijian commented on 2021-07-09 11:26 (UTC) (edited on 2021-07-09 11:27 (UTC) by Taijian)

@whynothugo: Could you post your complete HOOKS entry from mkinitcpio.conf?

Also: What does your disc layout look like and are you sure that your initramfs is not on an encrypted partition?

whynothugo commented on 2021-07-08 16:06 (UTC)

I'm using the systemd hook, and adding the sd-encrypt hook did not work for me.

When the system starts, I get prompted for the disk password and plymouth starts after that.

d_fajardo commented on 2021-07-02 06:12 (UTC)

Any update on this infinite loop / trailing 0's?

invidian commented on 2021-06-28 22:42 (UTC)

Once again the latest update is flagging an infinite update loop. Is there a fix?

Yep, evil trailing 0 strikes again :(

d_fajardo commented on 2021-06-28 17:11 (UTC)

Once again the latest update is flagging an infinite update loop. Is there a fix?

bdeshi commented on 2021-06-28 05:30 (UTC) (edited on 2021-06-28 07:53 (UTC) by bdeshi)

@Taijian I'm sorry but could you double-check that please? it works with plain encrypt hook too, without systemd-boot or systemd init hook.

how I checked:

  • # cryptsetup --persistent refresh /dev/sdXY -- removed any persistent flags
  • removed those flags from cryptdevice kernel command parameter too, and rebooted.
  • # dmsetup table -- confirmed no flags are present
  • added flags back, eg ... cryptdevice=UUID=<UUID>:<mapped_name>:allow-discards,no-read-workqueue,no-write-workqueue ...
  • regenerated bootloader config (grub) & initramfs, using plymouth-encrypt patched as my last comment.
  • rebooted and checked with # dmsetup table that all three flags were present.

Taijian commented on 2021-06-27 21:54 (UTC)

@bdeshi: The kernel command line parameters you are trying to use only work in a systemd-based boot setup. You need to switch out plymouth for sd-plymouth and encrypt for sd-encrypt (and a bunch of other hooks too - best have a look at the wiki).

bdeshi commented on 2021-06-27 20:28 (UTC)

hi, I've been using the no-*-workqueue options in cryptdevice param with cryptsetup's encrypt initcpio hook, but these echo warnings at boot with plymouth-encrypt. could you patch please? (copied from the default encrypt hook file)

diff --git a/usr/lib/initcpio/hooks/plymouth-encrypt b/-
index a623deb..0000000 100644
--- a/usr/lib/initcpio/hooks/plymouth-encrypt
+++ b/-
@@ -51,9 +51,15 @@ EOF

     for cryptopt in ${cryptoptions//,/ }; do
         case ${cryptopt} in
-            allow-discards)
+            allow-discards|discard)
                 cryptargs="${cryptargs} --allow-discards"
+            no-read-workqueue|perf-no_read_workqueue)
+                cryptargs="${cryptargs} --perf-no_read_workqueue"
+                ;;
+            no-write-workqueue|perf-no_read_workqueue)
+                cryptargs="${cryptargs} --perf-no_write_workqueue"
+                ;;
                 echo "Encryption option '${cryptopt}' not known, ignoring." >&2

KenFromBarbie commented on 2021-06-06 08:13 (UTC) (edited on 2021-06-06 09:55 (UTC) by KenFromBarbie)

I have an infinite update loop where with every check it wants to update to the same version over and over again:

:: Searching AUR for updates...

:: 1 Packages to upgrade.

1 aur/plymouth-git 0.9.5.r84.ge554475-1 -> 0.9.5.r84.ge554475-3

x70b1 commented on 2021-05-08 14:06 (UTC)

@Taijian That works perfect. Thanks!

Taijian commented on 2021-05-08 12:48 (UTC)

@x70b1: should be fixed in the latest commit. Please test.

x70b1 commented on 2021-05-07 22:12 (UTC) (edited on 2021-05-07 22:13 (UTC) by x70b1)

mkinitcpio fails to create my initramfs.

==> ERROR: file not found: `/etc/fonts/conf.d/60-latin.conf'

The error seems to bei here:

I have Cantarell installed, but in the package cantarell-fonts is no 60-latin.conf.

Taijian commented on 2021-04-26 08:12 (UTC) (edited on 2021-04-26 08:15 (UTC) by Taijian)

@p00h: Are you sure you have installed base-devel?

p00h commented on 2021-04-26 03:21 (UTC) (edited on 2021-04-26 03:22 (UTC) by p00h)

Trying to build the package, but it fails this way (several last lines): error: required file './ABOUT-NLS' not found
docs/ warning: '%'-style pattern rules are a GNU make extension
docs/ warning: '%'-style pattern rules are a GNU make extension
src/ installing 'build-tools/depcomp'
src/libply/ warning: subst /usr/lib,/lib,$(libdir: non-POSIX variable name
src/libply/ (probably a GNU make extension)
src/plugins/splash/script/ warning: '%'-style pattern rules are a GNU make extension
autoreconf: error: automake failed with exit status: 1
configure: error: cannot find required auxiliary files: config.rpath
==> ERROR: A failure occurred in build().

Full log:

asasione commented on 2021-04-21 22:53 (UTC)

Latest update fixed the issue, thanks

d_fajardo commented on 2021-04-21 19:42 (UTC)

I am also getting an infinite loop to update because of the trailing zero. It's harmless but slightly annoying.

invidian commented on 2021-04-20 09:10 (UTC)

Commit changes the length of the version and I'm pretty sure it shouldn't. Once removed, I think everything should be back to normal.

asasione commented on 2021-04-19 09:09 (UTC)

I cloned the repo and tried again and same issue. Some snippets of the PKGBUILD and this is without using an AUR helper.


Here is the checksum


This is clearly the same as what you have for your version with 0 in it, it matches but the versions are not the same hence the infinite loop for a newer version

Taijian commented on 2021-04-19 07:30 (UTC)

Yeah, as I said: I have no idea what is going on.

When I clone the upstream repo and do the git describe thingy, I get the tag without the trailing zero.

When I build the package, I get the pkgver as in this PKGBUILD here. Beats me.

AsamK commented on 2021-04-19 07:11 (UTC) (edited on 2021-04-19 07:11 (UTC) by AsamK)

Same for me. When I run the pkgver command in the plymouth git repo I get the version without 0 in the end

$ git describe --long | sed 's/-/.r/;s/-/./'

$ git --version
git version 2.31.1

michaldybczak commented on 2021-04-18 18:21 (UTC) (edited on 2021-04-18 18:23 (UTC) by michaldybczak)

Outputs on my end are exactly the same as on @asasione's. I tried to uninstall the package and install again, but the result is the same :(. All AUR helpers (pamac, trizen, yay) and pacman show the same version 0.9.5.r84.ge554475-1, so I doubt this is an error on AUR helper part. I have also been using this package for years.

Taijian commented on 2021-04-18 16:26 (UTC)

@asasione: Sorry about that. But yeah, I'm afraid I really do not know what might be causing the problem on your end.

asasione commented on 2021-04-18 16:21 (UTC)

I am running Arch not some random derivative, I have been using your package for years FYI.

Pacman v5.2.2 - libalpm v12.0.2

Taijian commented on 2021-04-18 16:10 (UTC)

@asasione: Well, then my next question(s) would be:

a) are you running Arch or some derivative, and

b) what does pacman -V tell you?

asasione commented on 2021-04-18 16:03 (UTC) (edited on 2021-04-18 16:04 (UTC) by asasione)

pacman -Qs plymouth-git
local/plymouth-git 0.9.5.r84.ge554475-1
    A graphical boot splash screen with kernel mode-setting support (Development version)
yay -Qs plymouth-git   
local/plymouth-git 0.9.5.r84.ge554475-1
    A graphical boot splash screen with kernel mode-setting support (Development version)

These are the results I'm getting even after having just installed the latest ge5544750-1 version. The update works correctly on your device but doesn't on ours it seems

Taijian commented on 2021-04-18 14:54 (UTC) (edited on 2021-04-18 14:57 (UTC) by Taijian)

$ pacman -Qs plymouth-git
local/plymouth-git 0.9.5.r84.ge5544750-1
    A graphical boot splash screen with kernel mode-setting support (Development version)

So if your output is different from mine, you are not running the same version of pacman/makepkg as I am, and I really cannot help you with that.

Taijian commented on 2021-04-18 14:53 (UTC)

@michaldybczak: OK, please do the following:

  1. pacman -Qs plymouth-git

  2. $(your_aur_helper) -Qs plymouth-git

  3. Compare the output. If there is a difference, the problem lies in the way your AUR helper parses the pkgver. Then fiel a bug report with them and/or use a different AUR helper in the future, one that does not misreport package updates.

The problem is not the way you install the package. The problem is the way you check for available updates. And that is an issue with your helper program, not with pacman (which does not care for AUR packages) nor with this package.

michaldybczak commented on 2021-04-18 14:47 (UTC) (edited on 2021-04-18 14:49 (UTC) by michaldybczak)

But I told you that the same happens when installed manually via makepekg -i. And -Qs returns: local/plymouth-git 0.9.5.r84.ge554475-1, so no, pacman is not aware of the extra 0 in the version. So what should I do to avoid that issue?

Taijian commented on 2021-04-18 12:21 (UTC)

@michaldybczak & Depau:

Again, the problem is neither with this PKGBUILD, nor with pacman. Go ahead, do a proper pacman -Qs plymouth-git and you will see that pacman is well aware of the trailing 0 in the pkgver. The issue is with whatever AUR helper you are using. Complain there.

michaldybczak commented on 2021-04-18 11:54 (UTC) (edited on 2021-04-18 11:56 (UTC) by michaldybczak)

I'm not acquainted with technicalities. I'm just an average user. This may be some minor issue but it causes the update to be displayed endlessly, and it disturbs the whole system of showing if there is an update or not. Plus, the said package is repeatedly updated/recompiled with every update, so it prolongs the process. In overall, small issue but irritating as hell. That's said, what non-technical users as me should do about it? All I could do is to put plymouth-git on the ignore list, but there is surely a better solution to this. I already tried installing it manually (with makepkg -i) and the issue is the same, so this is not the aur-helper issue but how pacman reads the PKGBUILD.The additional 0 is omitted and not written to the pacman database, so it looks like there was no update, because on the other hand pacman sees the versioning difference (hence the update need) and then fails to write the whole version number. Isn't it a pacman issue that should be reported somewhere?

Depau commented on 2021-04-18 09:48 (UTC) (edited on 2021-04-18 09:48 (UTC) by Depau)

There's an extra 0 in the PKGBUILD version that triggers the AUR helpers. git describe --long uses the first 8 chars of the SHA1 but pkgver (the variable) contains 9.

Ultimately I think AUR helpers should be smarter than this, or users should just not really care. It's not a real issue.

michaldybczak commented on 2021-04-18 08:46 (UTC) (edited on 2021-04-18 08:59 (UTC) by michaldybczak)

I did the update with pamac. On your suggestion, I did the update via yay and the issue is still the same. Update is successful, but item is showed on the list to update. So the last resource would be to install it manually, So I copied the PKGBUILD and the rest of the files from AUR, and did makepkg -i. Again, that did not solve this problem. Here is the info of versions: The version on my system: 0.9.5.r84.ge554475-1 The version to update: 0.9.5.r84.ge5544750-1 The weird thing is, during update I get info: warning: plymouth-git-0.9.5.r84.ge554475-1 is in the newest version - reinstallation So maybe pacman's database is reading the info on the installed version incorrectly? Yay shows: aur/plymouth-git 0.9.5.r84.ge554475-1 -> 0.9.5.r84.ge5544750-1

Taijian commented on 2021-04-17 14:31 (UTC)

@michaldybczak: I'm sorry, but it seems to me that this is likely an issue with whatever AUR helper you are using.

michaldybczak commented on 2021-04-17 11:50 (UTC)

No matter how many times I update/build it, it is always shown as a package to update. Probably versioning is broken on PKGBUILD since yesterday's update.

Daniel071 commented on 2020-08-16 10:01 (UTC)

@zynex: I had the same error as you had, I re-installed the base-devel package and now the build worked! Thanks!

zynex commented on 2020-07-17 20:48 (UTC) (edited on 2020-07-24 13:54 (UTC) by zynex)

Tried downloading the package and build it with pkgbuild, got the same error. Might be something I have installed that collide with it. Worked with previous release though.

[EDIT] Found the problem, I had somehow removed some packages from the base-devel group. Saw it them Octopi also failed to compile :)

Taijian commented on 2020-07-16 13:45 (UTC)

@zynex: Sorry to have to say this, but it 'works for me'.

Have your tried building in a clean chroot? Or at least with plain makepkg, i.e. without an AUR-helper?

zynex commented on 2020-07-16 09:42 (UTC) (edited on 2020-07-16 09:46 (UTC) by zynex)

/tmp/pikaur/build/plymouth-git/src/plymouth/configure: line 11825: syntax error near unexpected token IMAGE,' /tmp/pikaur/build/plymouth-git/src/plymouth/configure: line 11825:PKG_CHECK_MODULES(IMAGE, libpng >= 1.2.16 )' ==> FEL: Ett fel uppstod i build(). Avbryter...

Getting error when trying to compile it. Any idea what it could be. libpng is at version 1.6.37-3.

Taijian commented on 2020-07-01 20:57 (UTC)

@stertingen OK, that should be possible - do you have a proposal for modifying the install hook?

stertingen commented on 2020-07-01 17:28 (UTC)

Right now, the initcpio install hook copies the spinner theme files if the bgrt themes is selected.

I'd like to propose that the install hook should look into the ImageDir option of the plymouth theme file and copy this. This way, I can customize a plymouth theme by copying the bgrt theme without modifying the install hook.

commented on 2019-09-08 19:54 (UTC)

The Cantarell fonts (Regular and Thin) need to be added to the plymouth-encrypt install script. The default BGRT theme uses them for the password dialog and update messages. A optional dependency to 'cantarell-fonts' would also make sense.

commented on 2019-09-08 19:52 (UTC)

The Cantarell font needs to be added to the install script. The default BGRT theme is using it for the dialogs and update messages. A optional dependency to 'cantarell-fonts' would also make sense.

commented on 2019-09-08 19:48 (UTC)

The Cantarell font needs to be added to the install script, since the default BGRT theme uses it. Defining a optional dependency to the font would also be nice.

Depau commented on 2019-06-01 01:09 (UTC) (edited on 2019-06-01 14:09 (UTC) by Depau)

When selecting the BGRT theme, the Spinner theme should also be selected as it's a variation and images are recycled.

Here you can find my commit with patched mkinitcpio install hooks, if you'd like to merge:

PedroHLC commented on 2019-03-04 12:44 (UTC) (edited on 2019-03-04 12:45 (UTC) by PedroHLC)

checking for intltool-extract... no 
configure: error: intltool-extract is required to build plymouth 
==> ERROR: A failure occurred in build(). 

intltool seems to be required for boot. (New dep as it wasn't required yesterday)

Gigadoc2 commented on 2019-02-25 13:55 (UTC)

Could you change the git url to https? The sources are not verified, so adding transport encryption and CA trust might increase security a little bit here.

firedrake commented on 2019-01-17 00:51 (UTC)

When using theme bgrt, it doesn't display properly. Because upstream delete all images from the dir of theme bgrt recently. I think we should pick up image dir from the property "ImageDir", and add files from there in plymouth.install.

nullptr_t commented on 2017-10-14 13:37 (UTC) (edited on 2017-10-14 13:40 (UTC) by nullptr_t)

I've adopted the package since I think it is easy for me to port development to the stable version, which I maintain, too. This development version got a few patches ported from stable and is up-to-date now.

siavashs commented on 2017-08-26 14:46 (UTC)

I currently don't have the required resources to continue contributing to this package. Feel free to take ownership.

sovanyio commented on 2017-08-15 17:57 (UTC) (edited on 2017-08-15 18:05 (UTC) by sovanyio)

For anyone struggling to get themes to work during boot--this is the fix alluded to by Aetf: /lib/initcpio/install/plymouth

DoogyRev commented on 2017-06-18 18:04 (UTC)

I can't get sd-plymouth to take a graphical password input, I have to esc and input into the text console to proceed. I have used sd-plymouth and sd-encrypt (with and without plymouth-encrypt) mkinitcpio hooks with graphics driver modules included and I have used luks.uuid=, and luks.options= on the kernel command line. It works fine with the udev equivalent hooks (plymouth, plymouth-encrypt) and command line options (cryptdevice=).

sylveon commented on 2017-06-18 14:56 (UTC)

The gdm-plymouth service is complety borked on my machine. I had to chroot and enable the regular gdm service instead

prMoriarty commented on 2017-06-03 03:19 (UTC)

siavashs, please add sddm-plymouth.service

siavashs commented on 2017-04-27 00:45 (UTC)

Fixed :)

babel commented on 2017-04-25 14:17 (UTC)

New error compiling: patching file scripts/plymouth-update-initrd patching file systemd-units/ patching file src/main.c Hunk #1 FAILED at 1529. 1 out of 1 hunk FAILED -- saving rejects to file src/main.c.rej @acz13 is it related?

acz13 commented on 2017-04-25 01:35 (UTC)

patch no longer needed and will fail build:

babel commented on 2017-04-18 23:04 (UTC)

I can confirm it works now. Thank you siavashs.

siavashs commented on 2017-04-17 13:39 (UTC) (edited on 2017-04-17 13:47 (UTC) by siavashs)

It makes me sad when I see a C developer who commits his/her code before trying to compile it! Anyway I fixed the compilation error, and as I tested it works fine(Not very useful on my MBP though as it boots very fast).

ryester19 commented on 2017-04-10 18:51 (UTC)

I got the same error as babel, but this time in english :) main.c: In function ‘on_escape_pressed’: main.c:1532:34: error: ‘toggle_between_splash_details’ undeclared (first use in this function) toggle_between_splash_details, ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ main.c:1532:34: note: each undeclared identifier is reported only once for each function it appears in main.c:1540:43: error: expected ‘)’ before ‘state’ state->details_trigger); ^~~~~ main.c:1536:9: error: too few arguments to function ‘ply_event_loop_watch_for_timeout’ ply_event_loop_watch_for_timeout (state->loop, ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In file included from ./libply/ply-command-parser.h:27:0, from main.c:45: ./libply/ply-event-loop.h:74:6: note: declared here void ply_event_loop_watch_for_timeout (ply_event_loop_t *loop, ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ make[3]: *** [Makefile:613: plymouthd-main.o] Error 1 make[3]: Leaving directory '/home/kurbster/.cache/pacaur/plymouth-git/src/plymouth/src' make[2]: *** [Makefile:702: all-recursive] Error 1 make[2]: Leaving directory '/home/kurbster/.cache/pacaur/plymouth-git/src/plymouth/src' make[1]: *** [Makefile:457: all-recursive] Error 1 make[1]: Leaving directory '/home/kurbster/.cache/pacaur/plymouth-git/src/plymouth' make: *** [Makefile:389: all] Error 2 ==> ERROR: A failure occurred in build().

babel commented on 2017-04-09 14:17 (UTC) (edited on 2017-04-09 14:18 (UTC) by babel)

Hello. I get this error trying to install it: main.c:1532:34: error: ‘toggle_between_splash_details’ no se declaró aquí (primer uso en esta función) toggle_between_splash_details, ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ main.c:1532:34: nota: cada identificador sin declarar se reporta sólo una vez para cada función en el que aparece main.c:1540:43: error: expected ‘)’ before ‘state’ state->details_trigger); ^~~~~ main.c:1536:9: error: faltan argumentos para la función ‘ply_event_loop_watch_for_timeout’ ply_event_loop_watch_for_timeout (state->loop, ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In file included from ./libply/ply-command-parser.h:27:0, from main.c:45: ./libply/ply-event-loop.h:74:6: nota: se declara aquí void ply_event_loop_watch_for_timeout (ply_event_loop_t *loop, ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ make[3]: *** [Makefile:613: plymouthd-main.o] Error 1 make[3]: se sale del directorio '/tmp/yaourt-tmp-babel/aur-plymouth-git/src/plymouth/src' make[2]: *** [Makefile:702: all-recursive] Error 1 make[2]: se sale del directorio '/tmp/yaourt-tmp-babel/aur-plymouth-git/src/plymouth/src' make[1]: *** [Makefile:457: all-recursive] Error 1 make[1]: se sale del directorio '/tmp/yaourt-tmp-babel/aur-plymouth-git/src/plymouth' make: *** [Makefile:389: all] Error 2 ==> ERROR: Se produjo un fallo en build(). Cancelando… I'm sorry it is in Spanish. I won't translate it because I'm not sure what words to use exactly, but I hope you get the idea.

Aetf commented on 2016-11-04 23:18 (UTC)

Hi, the plymouth.initcpio_install script is incompatible to breeze-plymouth[1] theme. In addition to normal files under /usr/share/plymouth/themes/breeze/, this theme also has a images folder at /usr/share/plymouth/themes/breeze/images/. In the plymouth.initcpio_install script, around line 37, it checks and skips all non-file children under the folder. I'm not sure I understand the reason why not just add the whole themes/${PLYMOUTH_THEME_NAME} folder to initramfs, maybe there are other themes that puts non-related files in the theme folder? Missing images folder is causing problems like blank screen when ask password input should be shown, as reported in the breeze-plymouth package. [1]:

javashin commented on 2016-05-22 14:54 (UTC)

Something Weird Just Happen And After upgrade systemd to version 230 the hook sd-plymouth is no longer compatible with systemd so please check the sd-plymouth hook and fix it to use it with systemd well plymouth works with HOOKS="base udev plymouth resume autodetect modconf block filesystems keyboard" but the combo HOOKS="systemd sd-plymouth" dont boot

commented on 2015-11-05 07:32 (UTC)

It compiles fine, but splash screen doesnt actually show. I have to use -legacy version. Is that the case for everyone else?

giorgianb commented on 2015-09-29 01:54 (UTC)

For the *-git packages, how does one sign them? Most *-git packages I've tried installing fail with 'package missing required signature'.

skeggse commented on 2015-09-17 17:26 (UTC)

The plymouth-encrypt hook differs from the encrypt hook in that it does not used the resolved device for the cryptsetup luksOpen command following the ask-for-password prompt on line 79 of plymouth.encrypt_hook (/usr/lib/initcpio/hooks/plymouth-encrypt).

Gryffyn commented on 2015-05-23 07:46 (UTC)

I have adopted this package, and in the newest version it compiles perfectly on my 3 testbeds.

Linux_Yang commented on 2015-05-03 05:29 (UTC)

I have some troubles when I install "plymouth-git" from AUR. /usr/share/aclocal-1.15/silent.m4:52: warning: macro 'AM_SUBST_NOTMAKE' not found in library /usr/share/aclocal-1.15/silent.m4:54: warning: macro 'AM_SUBST_NOTMAKE' not found in library libtoolize: putting auxiliary files in AC_CONFIG_AUX_DIR, 'build-tools'. libtoolize: linking file 'build-tools/' libtoolize: Consider adding 'AC_CONFIG_MACRO_DIRS([m4])' to, libtoolize: and rerunning libtoolize and aclocal. libtoolize: Consider adding '-I m4' to ACLOCAL_AMFLAGS in /usr/share/aclocal-1.15/silent.m4:52: warning: macro 'AM_SUBST_NOTMAKE' not found in library /usr/share/aclocal-1.15/silent.m4:54: warning: macro 'AM_SUBST_NOTMAKE' not found in library configure:4877: error: possibly undefined macro: _AM_SUBST_NOTMAKE If this token and others are legitimate, please use m4_pattern_allow. See the Autoconf documentation. configure:4923: error: possibly undefined macro: AM_SUBST_NOTMAKE autoreconf: /usr/bin/autoconf failed with exit status: 1 configure: error: cannot find install-sh,, or shtool in build-tools "."/build-tools Online

commented on 2015-04-26 02:38 (UTC)

and the sddm-plymouth for ussing sddm with plymouth?

commented on 2015-03-16 08:49 (UTC)

Brought this version mostly into line with the non-git version, and added the sd-plymouth hook for using plymouth with systemd initcpios. Among other things, this should work seamlessly with any necessary password asking (i.e., for encryption) in the initcpio without configuration — but I haven't yet tested it.

commented on 2015-03-16 08:47 (UTC)

Brought this version mostly into line with the non-git version, and added the sd-plymouth hook for using plymouth with systemd initcpios.

carstene1ns commented on 2015-03-08 14:01 (UTC)

I have cleaned up this package a bit and copied the initrd hook from the stable plymouth package. With this, I got a working bootsplash. However, as the splash is only shown for a few seconds anyway, I do not intend to use/maintain this any further.

Kubax commented on 2015-01-28 10:11 (UTC)

Just for the ones who tried to edit the file, without success (as i did), here is the path from the manually edited files. --------------------------- --- plymouth.src/systemd-units/ 2015-01-28 11:01:26.267298987 +0100 +++ plymouth.patch/systemd-units/ 2015-01-28 11:05:27.439604785 +0100 @@ -1,8 +1,9 @@ [Unit] Description=Terminate Plymouth Boot Screen After=rc-local.service plymouth-start.service systemd-user-sessions.service +Before=getty@tty1.service [Service] -ExecStart=-@PLYMOUTH_CLIENT_DIR@/plymouth quit +ExecStart=-@PLYMOUTH_CLIENT_DIR@/plymouth quit --retain-splash Type=oneshot TimeoutSec=20

wchouser3 commented on 2014-11-08 20:30 (UTC)

I'm getting the same build errors as shihjay2.---package won't install

shihjay2 commented on 2014-08-04 12:13 (UTC)

For others interested, you'll also need to change this in the file in addition to what @techryda mentioned... @@ -1,8 +1,9 @@ instead of @@ -1,10 +1,12 @@ Was able to compile successfully afterwards, however, plymouth-git still does work. Instead of %[] %[] %[] now only showing [] [] [] no matter what theme you use

techryda commented on 2014-08-04 02:04 (UTC)

Per this commit: Remove the last three lines (below) from --- + [Install] ---

shihjay2 commented on 2014-07-30 22:06 (UTC)

Build error since new commits to plymouth git repository yesterday. patching file scripts/ patching file scripts/plymouth-update-initrd patching file systemd-units/ Hunk #1 FAILED at 1. 1 out of 1 hunk FAILED -- saving rejects to file systemd-units/

commented on 2014-06-17 21:17 (UTC)

Fixed encrypt-hook where cryptargs were not being passed while unlocking volume.

commented on 2014-04-03 06:40 (UTC)

@ randomer679 Yes git is currently broken. While it builds and installs all ok, in operation, it fails to find any of the graphical renderers, and ends up falling back to a text splash. Use plymouth for the time being. It is the most recent stable and is working.

randomer679 commented on 2014-04-02 22:40 (UTC)

@padfoot I'm not sure if it matters now, but I use radeon with early KMS. When my laptop boots the screen will sort of flash when it starts plymouth but then falls back to text.

commented on 2014-02-24 08:43 (UTC)

I have confirmed plymouth is failing to find any renderer on boot hence falling back to a text only splash. Now to find the reason why.

commented on 2014-02-22 09:40 (UTC)

After a lot of messing around, I found the default for a new config option may be the cause of bootsplashes not showing. There is now a showdelay config option, which defaults to 5 seconds. If your system boots before this you will only have a blank screen between EFT/bootloader and DM/WM. To get your splash showing again, edit /etc/plymouth/plymouthd.conf and add a showdelay config option: [Daemon] Theme=<your theme name> ShowDelay=<seconds> I used ShowDelay=1 and my splash returned. However, with the latest git, for some reason, plymouth is failing to find the frame-buffer renderer (ati or nvidia), so I am falling back to a text theme. I am interested to see if anyone using the drm renderer (intel, nouveau or radeon with early kms) falls back to text. Please let me know, then I have something solid to work on getting corrected. In the meantime, I am back on plymouth-release.

commented on 2014-02-22 05:23 (UTC)

@aricat Thanks for that. I'll look into it and see what I can figure out.

aricart commented on 2014-02-21 22:46 (UTC)

Not working here. I can get debug output by passing "plymouth.debug" on the kernel command line though. Here's an interesting bit of output, though I don't know how useful it'll be: ./plugin.c:698 show_splash_screen:couldn't load views ply-boot-splash.c:519 can't show splash: Success

commented on 2014-02-20 10:36 (UTC)

I have picked up this package. While I have updated the package, I am yet to upload it as I don't want to break what may be working for others. Please let me know if this package is working for you so I can decide if to upload the updated package or not. Git is currently not working for me. No bootsplash is displayed, yet with debugging turned on, no errors are logged. I just get a blank screen on boot. After boot, if I change tty and start the daemon, the bootsplash is displayed, so I am uncertain as to what the problem is (I am suspecting an issue when building the initrd).

mytbk commented on 2014-02-19 07:50 (UTC)

I haven't used plymouth for a long time, and I think I'm not capable to maintain it, so I disown this package.

kaede commented on 2014-02-17 12:16 (UTC)

==> ERROR: file not found: `/etc/system-release' when mkinitcpio-ing the standard linux.

commented on 2014-02-15 21:41 (UTC)

The AUR mdm-display-manager package already includes mdm-plymouth.service

nachoig commented on 2014-02-13 19:59 (UTC)

@malinas pacman -Qo /etc/os-release /etc/os-release is owned by filesystem 2013.05-2 We are on Arch, default init is systemd and Plymouth here depends on systemd anyway. According to file, Plymouth supports custom release files, so, it's possible to use something like this in the LDFLAGS. --with-release-file=<path_to_release_file> In this case, it would be: --with-release-file=/etc/os-release About plymouth-quit.service, if upstream provides it, I don't know why we need to use a custom file (also, the file from upstream was identical to the file that was provided here). If it's broken, the better way to fix it is reporting at upstream. This version is from Git, so first of all, it's not expected to work fine like a release or production version. ________ About MDM, if it provides a systemd file, you can do the following: cp /usr/lib/systemd/system/mdm.service /etc/systemd/system/ Edit /etc/systemd/system/mdm.service and add plymouth-quit.service in After. systemctl daemon-reload Reboot the system and test.

malinas commented on 2014-02-10 10:15 (UTC)

I use catalyst (uvesafb) and mdm with this, there is no mdm-plymouth, might wanna add that... I am not sure if needed but to make sure, I just created a phony system-release. but with a <2 sec boot, plymouth is pointless anyhow I realise.

malinas commented on 2014-02-10 10:05 (UTC)

@nachoig: well, not sure about plymouth-quit.. but I certainly remember in the past, I had to manually patch it to make it work on arch... so instead of doing a lot of pattern matching, I guess they are just providing a working file. Now, for os-release... that's a systemd decision to use, breaking the old lsb dependency and so on... since plymouth is not system-centric, I can understand why we would have logic to add it.

commented on 2014-02-09 08:57 (UTC)

Requires gtk2 to compile, so it should probably be added to makedepends.

nachoig commented on 2014-02-02 03:32 (UTC)

I'd like to understand what these files are doing here: plymouth-quit.service Upstream already provides the same file. system-release filesystem provides /etc/os-release, which contains the details about Arch (and it's much more complete).

commented on 2014-01-25 23:53 (UTC)

I had a good mess around with plymouth today, and I have the current release version 0.8.8 working for me. I have uploaded the package here for those having issues with git to try: If it proves itself, I welcome the changes being incorporated into the other plymouth packages and we can then remove the plymouth-release package. Cheers.

pezz commented on 2014-01-14 10:53 (UTC)

@iron8core Your version compiles, thanks mate. But this whole thing seems pretty broken. Doesn't work well with v86d (as it used to) and plymouth-encrypt seems busted. Disabled it all until some new good version comes out.

commented on 2014-01-14 09:51 (UTC)

PKGBUILD and updated patch for latest git version: The latest git version, for me and many others, won't display splash on boot, but sometimes on shutdown. works fine though, so unless the latest works, git it a try. PKGBUILD for version (just a small fix on line 40 compared this one): Note: The latest kernel I've tried with is 3.12.7-1

commented on 2014-01-14 07:26 (UTC)

Unfortunately for me, plymouth is no longer working with the update to kernel 3.12.7-2. System hangs on boot unless plymouth is disabled in your initramfs.

nachoig commented on 2014-01-12 05:40 (UTC)

@catlover2 You are right, this commit changed plymouth-update-initrd. Unfortunately, the original script is Fedora-specific (uses Dracut), so it needs to be patched to work on Arch. Maybe the Plymouth mantainers will drop this someday, as described in this commit (what they are waiting to do this? They should kill it with fire!!!! =). The right way to do this is updating the initrd manually (in the case of Arch, mkinitcpio -p linux),

catlover2 commented on 2014-01-10 23:57 (UTC)

This is relevant to It appears the file that "plymouth-update-initrd.patch" patches has changed, so the patching now fails. Please update the patch file. After I updated the patch and installed the package, following the instructions on the wiki, plymouth still didn't work. It just hung at boot after running the [plymouth] hook, and nothing happened, and the instructions for previwing themes didn't do anything either. My GPU is an Intel HD Graphics 3000.

nihonsuki commented on 2014-01-07 05:51 (UTC)

Further to this: Occasionally get brief splash on shutdown...

nihonsuki commented on 2014-01-05 02:32 (UTC)

@mytbk: thanks for your efforts. I've rebuilt the package and installed. No joy I'm afraid. On boot the syslinux menu comes up then quiet boot as disks are mounted, until gdm-plymouth.service takes over. When I look for plymouth in htop I see root has /usr/bin/plymouth --wait dawdling. BTW, the package built purports to be Thanks again.

mytbk commented on 2014-01-04 15:40 (UTC)

@nihonsuki: you can change the url in source as 'git://' And I've written the commit in the PKGBUILD already.

nihonsuki commented on 2014-01-04 12:48 (UTC)

@taxfusion: Perhaps I've misunderstood. I've changed the pkgver to but when I makepkg -s it seems to pull down the most recent source and overwrite the changes I've made to PKGBUILD.

tuxfusion commented on 2014-01-03 09:19 (UTC)

@padfoot, since this is the git version, you may edit the PKGBUILD when promped and just change the version string to "". Git will pull that codebase or via the wiki:

nihonsuki commented on 2014-01-01 13:33 (UTC)

padfoot: Looking to put plymouth on a new install. Where would I acquire the tarball for (Arch newbie)

commented on 2013-12-17 05:38 (UTC)

Downgraded to Splash now works again.

aricart commented on 2013-12-16 09:16 (UTC)

Same bug here, no splash on boot with radeon. Looks like it effects everyone.

commented on 2013-12-16 05:19 (UTC)

Actually, there seems to be something with the commits since that stop the splash working. I have updated on 3 machines, testing with nvidia and intel drivers (nvidia machine uses efifb pre X) and does not display splash at all whereas does.

commented on 2013-12-14 21:19 (UTC)

Package now builds. With the latest kernel update, I no longer get splash on boot. And on shutdown, I end up with console messages and an aretfact of the splash (spinner is displayed). Checking my systemd journal (for boot at least) plumouth is no longer started as early as it previously was.

Det commented on 2013-12-14 15:13 (UTC)

The deprecated comment about the "kill Plymouth" lines in the .install is completely and utterly unnecessary. Someone even went ahead and added it to the Wiki.

mytbk commented on 2013-12-14 00:06 (UTC)

Add LDFLAGS, and change --enable-drm-renderer to --enable-drm(but ./configure --help still says --enable-drm-renderer, it needs fixing.)

alucryd commented on 2013-12-13 23:19 (UTC)

padfoot: A recent commit must have a missing linker flag. Add this before make: export LDFLAGS="$LDFLAGS -ludev" BTW, it seems --enable-drm-renderer does not exist anymore, it's --enable-drm now.

commented on 2013-12-13 22:13 (UTC)

Build is failing for me: make[3]: Entering directory '/tmp/yaourt-tmp-padfoot/aur-plymouth-git/src/plymouth/src' CC plymouthd-ply-boot-server.o CC plymouthd-plugin.o CC plymouthd-main.o CCLD plymouthd libply-splash-core/.libs/ undefined reference to `udev_enumerate_add_match_property' libply-splash-core/.libs/ undefined reference to `udev_device_get_syspath' libply-splash-core/.libs/ undefined reference to `udev_new' libply-splash-core/.libs/ undefined reference to `udev_enumerate_scan_devices' libply-splash-core/.libs/ undefined reference to `udev_monitor_receive_device' libply-splash-core/.libs/ undefined reference to `udev_monitor_unref' libply-splash-core/.libs/ undefined reference to `udev_unref' libply-splash-core/.libs/ undefined reference to `udev_enumerate_add_match_parent' libply-splash-core/.libs/ undefined reference to `udev_enumerate_new' libply-splash-core/.libs/ undefined reference to `udev_queue_unref' libply-splash-core/.libs/ undefined reference to `udev_device_get_property_value' libply-splash-core/.libs/ undefined reference to `udev_device_get_action' libply-splash-core/.libs/ undefined reference to `udev_enumerate_add_match_subsystem' libply-splash-core/.libs/ undefined reference to `udev_list_entry_get_next' libply-splash-core/.libs/ undefined reference to `udev_device_new_from_syspath' libply-splash-core/.libs/ undefined reference to `udev_device_has_tag' libply-splash-core/.libs/ undefined reference to `udev_device_get_parent' libply-splash-core/.libs/ undefined reference to `udev_monitor_get_fd' libply-splash-core/.libs/ undefined reference to `udev_enumerate_get_list_entry' libply-splash-core/.libs/ undefined reference to `udev_device_unref' libply-splash-core/.libs/ undefined reference to `udev_monitor_new_from_netlink' libply-splash-core/.libs/ undefined reference to `udev_device_get_is_initialized' libply-splash-core/.libs/ undefined reference to `udev_enumerate_add_match_is_initialized' libply-splash-core/.libs/ undefined reference to `udev_enumerate_unref' libply-splash-core/.libs/ undefined reference to `udev_device_get_sysname' libply-splash-core/.libs/ undefined reference to `udev_monitor_filter_add_match_tag' libply-splash-core/.libs/ undefined reference to `udev_monitor_filter_add_match_subsystem_devtype' libply-splash-core/.libs/ undefined reference to `udev_queue_new' libply-splash-core/.libs/ undefined reference to `udev_monitor_enable_receiving' libply-splash-core/.libs/ undefined reference to `udev_device_get_sysattr_value' libply-splash-core/.libs/ undefined reference to `udev_list_entry_get_name' libply-splash-core/.libs/ undefined reference to `udev_device_get_devnode' libply-splash-core/.libs/ undefined reference to `udev_queue_get_queue_is_empty' collect2: error: ld returned 1 exit status Makefile:537: recipe for target 'plymouthd' failed make[3]: *** [plymouthd] Error 1 make[3]: Leaving directory '/tmp/yaourt-tmp-padfoot/aur-plymouth-git/src/plymouth/src' Makefile:689: recipe for target 'all-recursive' failed make[2]: *** [all-recursive] Error 1 make[2]: Leaving directory '/tmp/yaourt-tmp-padfoot/aur-plymouth-git/src/plymouth/src' Makefile:445: recipe for target 'all-recursive' failed make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory '/tmp/yaourt-tmp-padfoot/aur-plymouth-git/src/plymouth' Makefile:376: recipe for target 'all' failed make: *** [all] Error 2

mytbk commented on 2013-12-04 10:46 (UTC)

@gormux: I have no problem to build it.

gormux commented on 2013-12-03 18:13 (UTC)

Hi, I have a problem during configure : /tmp/yaourt-tmp-julien/aur-plymouth-git/src/plymouth/configure: line 5221: syntax error near unexpected token `dlopen' /tmp/yaourt-tmp-julien/aur-plymouth-git/src/plymouth/configure: line 5221: `LT_INIT(dlopen disable-static pic-only)' Am I the only one ?

grufo commented on 2013-11-19 19:08 (UTC)

Hello everyone! Congratulations for this very very good aur package! :) I have just a question. Are you able to terminate the gnome session using the current systemd gdm-plymouth.service? When I used gdm.service I had no problems with "terminate session". But now, when I try to terminate, it freezes on linux console and to solve I have to launch: # systemd stop gdm-plymouth.service && systemd start gdm.service This problem occurs only on terminating current gnome session. At system boot there are not problems and everything works fine.

ylecuyer commented on 2013-11-05 01:33 (UTC)

I have an error while trying to install

mytbk commented on 2013-10-31 07:33 (UTC)

updated plymouth-update-initrd.patch

aricart commented on 2013-10-27 08:09 (UTC)

Needs an update to plymouth-update-initrd.patch: --- scripts/plymouth-update-initrd.old 2013-10-27 17:02:15.200368142 +0900 +++ scripts/plymouth-update-initrd 2013-10-27 17:02:57.686768138 +0900 @@ -1,2 +1,4 @@ #!/bin/bash -dracut -f /boot/initramfs-$(uname -r).img $(uname -r) +find /etc/mkinitcpio.d/ -name \*.preset -a \! -name example.preset | while read p; do + mkinitcpio -p $p +done

mytbk commented on 2013-09-24 04:29 (UTC)

plymouth.initcpio_install is updated. By the way, I saw the aur/plymouth package also use the git sources.

intelfx commented on 2013-09-23 16:41 (UTC)

As said above, the install hook shall be rewritten (at the very least). Also, I'd recommend that you update your whole PKGBUILD from plymouth package.

duht commented on 2013-09-03 08:35 (UTC)

Please replace the line SCRIPT='plymouth' with add_runscript in plymouth.initcpio_install script to make plymouth mkinitcpio HOOK work again and start plymouth at early boot stage (right after GRUB). More info in this thread:

aricart commented on 2013-09-02 00:01 (UTC)

For this to work I had to modify both plymouth-quit.service and plymouth-quit-wait.service so they read: ExecStart=/usr/bin/plymouth instead of ExecStart=-/usr/bin/plymouth Also, one of them had DOS newline characters for some reason...

Dreamkey commented on 2013-06-06 23:46 (UTC)

I think it would be cleaner if you add "--sbindir=/usr/bin" on the configure line instead of moving everything in the package function. Thanks.

commented on 2013-06-04 07:01 (UTC)

Please update package to move binaries in /bin /sbin and /usr/sbin to /usr/bin Thanks.

mytbk commented on 2013-06-02 01:53 (UTC)

@mikeroll: Sorry. PKGBUILD updated.

mikeroll commented on 2013-06-01 16:43 (UTC)

@mytbk: service files are added, but they are not mentioned in package(), therefore not installed.

mytbk commented on 2013-05-19 11:57 (UTC)

{lightdm,slim}-plymouth.service are added. I saw many -git packages still use all dots in their pkgver, so I think there's no need to change.

commented on 2013-05-15 00:19 (UTC)

no lightdm-plymouth.service?? and not use . in git, replace those . for .r or ~ because minor updates to the four digit can not be detected due to vercam isues... with this schema

mytbk commented on 2013-05-14 11:44 (UTC)

@jarryson: Well,are you using gdm-plymouth.service or gdm.service?

jarryson commented on 2013-05-14 09:38 (UTC)

Seems can not get plymouth work in early start(before mounting /). so add plymouth hook is not necessary. PS: please add one line to gdm.service or plymouth won't stop (I don't understand why plymouth-quit.service did not called by gdm.service). "ExecStartPre=-/usr/bin/plymouth quit"

mytbk commented on 2013-05-09 14:40 (UTC)

@Boohbah: updated

commented on 2013-05-08 20:03 (UTC)

Fix pkgver() function

commented on 2013-05-08 19:59 (UTC)

Fix pkgver() function

commented on 2013-04-28 17:54 (UTC)

@mytbk: Thanks! :D

mytbk commented on 2013-04-28 16:33 (UTC)

@AStrangeEnigma: I saw that the patch of systemd-unit-dir.patch is no longer in need, I'll update it later.

commented on 2013-04-28 15:37 (UTC)

Hi, I'm still having issues with patching :/ ==> Starting prepare()... patching file scripts/ patching file scripts/plymouth-update-initrd patching file Hunk #1 FAILED at 254. 1 out of 1 hunk FAILED -- saving rejects to file ==> ERROR: A failure occurred in prepare(). Aborting... Not sure if this is my fault, or if the package needs to be updated

mytbk commented on 2013-04-26 00:26 (UTC)


AlD commented on 2013-04-21 19:02 (UTC)

plymouth-pid.patch doesn't apply anymore due to

francoism commented on 2013-02-26 22:39 (UTC)

To get this to work, I need to use the --force command: # systemctl enable kdm-plymouth.service --force

lmello commented on 2013-02-08 23:09 (UTC)

Ok, the latest git pulls works... but gdm doesn't start. Neither by enabling gdm.service nor gdm-plymouth.service.

lmello commented on 2013-01-22 14:59 (UTC)

@ricardofunke: using gdm.service solves the sluggishness but makes the tty's look weird (with partial splash images permanently on the screen) and does not transition well to gdm...

orumin commented on 2013-01-21 12:39 (UTC)

I can't install plymouth run script into initramfs when running mkinitcpio. But, plymouth's modules correctly installed. So, I modified plymouth.initcpio_install --- 2012-07-14 08:04:03.000000000 +0900 +++ plymouth.initcpio_install 2013-01-21 20:22:34.357898902 +0900 @@ -47,7 +47,7 @@ add_binary "$(readlink -e /lib/" add_file /lib/ - SCRIPT='plymouth' + add_runscript } help() {

ricardofunke commented on 2013-01-17 13:42 (UTC)

@lmello: I wasn't usind gdm-plymouth.service, I was using gdm.service instead.

lmello commented on 2013-01-17 08:20 (UTC)

@ricardofunke: yes, but the gdm-plymouth.service in the stable version makes my desktop very slow.

ricardofunke commented on 2012-12-13 13:42 (UTC)

@lmello and @stibbard, are you running "mkinitcpio -p linux" after install the stable version?

lmello commented on 2012-12-13 01:34 (UTC)

Same issue observed by ricardofunke and stibbard. Using an Intel integrated graphic card. Unfortunatelly for me the stable version doesn't work either.

commented on 2012-11-13 22:57 (UTC)

Thanks for the tip. Unfortunately, the stable version returns the same error. I continue to dig and if I come up with an answer I'll post it.

ricardofunke commented on 2012-11-13 20:42 (UTC)

Well @stibbard, It's the mainstream git version that causes that. So, as this package tends to be unstable for use git last pulls, I moved to "plymouth" aur package, it's uses the last stable revision. The "plymouth" package has now a mantainer, so I think this is the recomended for you.

commented on 2012-11-13 18:43 (UTC)

Please excuse the typo in my earlier post, I meant to say "I have"... For what it's worth, I have tried the plymouth-git pkg using both the Nvidia and nouveau graphics drivers with the same results. Thanks.

commented on 2012-11-13 17:49 (UTC)

I'm have the same issue as ricardofunke. I had to reinstall my Arch system this morning and after installing the plymouth-git package I get the same error: "failed to read status from child immediately after starting to daemonize". Previous installations were trouble free. I am using lightdm and Mate. Thanks for any comments/assistance.

ricardofunke commented on 2012-11-09 02:05 (UTC)

Today I installed plymouth-git in my wife's laptop but I'm getting this error when running plymouthd: "failed to read status from child immediately after starting to daemonize" Can you help me find out what is wrong?

ricardofunke commented on 2012-11-03 00:45 (UTC)

I'm not having the smooth transition to GDM, am I the only one? I already disable gdm.service and enable gdm-plymouth.service. Am I missing something?

ShyPixie commented on 2012-11-02 21:17 (UTC)

@webmeister Fixed!

popsUlfr commented on 2012-11-02 11:15 (UTC)

@webmeister: It looks like pangox has been removed from pango: commenting the line in '/usr/lib/initcpio/install/plymouth-encrypt': #add_file "/usr/lib/pango/1.6.0/modules/" removes the error. And I don't see any differences in functionality after recreating my initramfs.

webmeister commented on 2012-11-02 08:49 (UTC)

-> Running build hook: [plymouth-encrypt] ==> ERROR: file not found: `/usr/lib/pango/1.6.0/modules/'

ShyPixie commented on 2012-11-02 06:57 (UTC)

@fzerorubigd Thank you, fixed.

fzerorubigd commented on 2012-11-02 06:30 (UTC)

in plymouth-start.service the /bin/udevadm is wrong. udevadm is inside /usr/bin and also linked in /sbin

Det commented on 2012-10-28 20:54 (UTC)

Nonono: make DESTDIR="$pkgdir" install || return 1

Det commented on 2012-10-26 10:52 (UTC)

Be my guest.

commented on 2012-10-26 02:27 (UTC)

Quick fix for the issues with the latest version: sed -i 's/^\(\s*patch \)/\1--follow-symlinks /;s/^\(\s*curl \)-sO/\1-LsO/' PKGBUILD Or, for those who prefer patch files: --- PKGBUILD.orig 2012-10-25 00:00:00.000000000 +0000 +++ PKGBUILD 2012-10-25 00:00:00.000000000 +0000 @@ -74,24 +74,24 @@ msg "Applying Patches..." rm -f encrypt_hook encrypt_install - curl -sO \ - -O + curl -LsO \ + -O msg2 "Fixing encrypt HOOK" - patch -i encrypt_hook.patch - patch -i encrypt_install.patch + patch --follow-symlinks -i encrypt_hook.patch + patch --follow-symlinks -i encrypt_install.patch cd $_gitname-build msg2 "Setting a human-readable error for '# plymouth-set-default-theme -R'" - patch -p0 -i "$srcdir/" + patch --follow-symlinks -p0 -i "$srcdir/" msg2 "Fixing updating initrd" - patch -p0 -i "$srcdir/plymouth-update-initrd.patch" + patch --follow-symlinks -p0 -i "$srcdir/plymouth-update-initrd.patch" msg2 "Changing PID file to /run/" - patch -p0 -i "$srcdir/plymouth-pid.patch" + patch --follow-symlinks -p0 -i "$srcdir/plymouth-pid.patch" msg2 "Changing SYSTEMD_UNIT_DIR to /usr/lib/systemd/system/" - patch -p0 -i "$srcdir/systemd-unit-dir.patch" + patch --follow-symlinks -p0 -i "$srcdir/systemd-unit-dir.patch" sed -e 's:png_set_gray_1_2_4_to_8:png_set_expand_gray_1_2_4_to_8:g' \ -i src/libply-splash-graphics/ply-image.c

Det commented on 2012-10-04 18:19 (UTC)


Zell_89 commented on 2012-10-04 15:52 (UTC)

Anyone can provide a service for lightdm?

Det commented on 2012-09-15 09:19 (UTC)

I'm failing in the first one already, since the GNU folks apparently had this brilliant idea about symlinks:

liquidsky commented on 2012-09-15 00:45 (UTC)

There is a "bug" while I want to compile plymouth. I use a pure systemd Installation. patching file scripts/plymouth-update-initrd -> Changing PID file to /run/ patching file systemd-units/ Hunk #1 FAILED at 5. 1 out of 1 hunk FAILED -- saving rejects to file systemd-units/ patching file systemd-units/ patching file systemd-units/

Dreamkey commented on 2012-08-18 14:54 (UTC)

Hello, I have the same problem as GNRS: after the splash screen and before my login manager (slim), the screen is turned off (no backlight) for 2 seconds. I didn't have this problem with an old build: systemd was not a dependency at this time, but I needed to rebuilt it for the /lib directory becoming a symlink. I built it without systemd, removing lines 91-92, 100 and 126-129.

commented on 2012-08-18 12:30 (UTC)

I've made a nice theme for Plymouth, based on [url=""]this one[/url] but with fixed logo and colors (accodring to official site colors\logo) here it is [url][/url]

xzy3186 commented on 2012-08-18 06:52 (UTC)

With gdm 3.4.1-3 in testing repo installed, the gdm-plymouth.service can not be loaded anymore (every time returns as failed in systemctl). This is not a critical issue because gdm.service can work anyway but do you know how to get rid of this error? Thank you.

Det commented on 2012-08-17 22:21 (UTC)

It was reverted as you can see here:

commented on 2012-08-17 18:00 (UTC)

works great for me (testing system), plymouth-git installed today I have just the "fsck-not-quiet" issue mentioned below in comment of July 20th from Det

Det commented on 2012-08-09 20:17 (UTC)

'plymouth' seems to work fine. I assume it was this commit that did it:

lynix commented on 2012-08-09 19:53 (UTC)

Currently broken for me with: ply-text-display.c:198:1: warning: no previous prototype for ‘ply_text_display_reset_background_color’ [-Wmissing-prototypes] ply-text-display.c: In function ‘ply_text_display_reset_background_color’: ply-text-display.c:205:31: error: ‘color’ undeclared (first use in this function)

Det commented on 2012-07-25 07:57 (UTC)

Yeah, it has nothing to _do_ with libtool. It's just the file that was updated in git.

commented on 2012-07-25 06:16 (UTC)

I've (libtool instaled and got) error in build: -> Changing PID file to /run/ patching file systemd-units/ Hunk #1 FAILED at 5. 1 out of 1 hunk FAILED -- saving rejects to file systemd-units/ patching file systemd-units/ patching file systemd-units/ ==> BŁĄD: Wystąpił błąd w build(). Przerywam... The build failed.

commented on 2012-07-23 19:20 (UTC)


Samsagax commented on 2012-07-23 19:15 (UTC)

@lupylucke: should I be the RTFM guy here?

commented on 2012-07-23 19:10 (UTC)

Why? Plymouth fails to build (at least for me) without 'libtool' installed. Looks like a build-time dependency to me...

Det commented on 2012-07-23 12:14 (UTC)

I don't(...)

commented on 2012-07-22 21:49 (UTC)

I think 'libtool' should be listed in 'makedepends'...

Det commented on 2012-07-21 09:48 (UTC)

Well, you can _install_ it but you can't actually _use_ it. Not unless somebody makes it work again.

sadika commented on 2012-07-21 06:15 (UTC)

how can i install plymouth without systemd?

Det commented on 2012-07-20 14:58 (UTC)

Reverted the pid path but the [systemd-fsck] message is still there - even with the stable one. How did we hide this in the first place?

Det commented on 2012-07-20 13:18 (UTC)

Lol. I wonder what did I just do to make the boot messages appear _twice_ with both systemd and initscripts. Took like half an hour to figure out why was it still happening even after uninstalling plymouth (forgot the hook). E: Ok, I don't exactly know why but the line: /usr/sbin/plymouthd --mode=boot --pid-file=/run/plymouth/pid in "/usr/lib/initcpio/hooks/plymouth" is somehow causing the double-printing. Reverting the pid path to '/run/' and rebuilding the initrd fixes it. I still need to figure out why I'm getting the systemd-fsck message, though, which minimizes plymouth to the right lower corner of my screen.

Det commented on 2012-07-20 10:54 (UTC)

K, some of the changes: * Make the cryptsetup links clickable * Install our own quit service (ricardofunke) * Make the dm services neither conflict or run after getty@tty1 (ricardofunke) (I actually didn't even try abbradar's solution - this one just sounded like the right way) * Create the 'plymouth' directory in /run instead of patching all the service files (abbradar) * Use '--depth 1' with 'git pull' too * Simplify the patch messages * More efficient install for systemd services

abbradar commented on 2012-07-20 09:40 (UTC)

I've tried some modifications to gdm-plymouth.service, but I cannot achieve desired behaviour (smooth transition from plymouth to gdm). I just fixed killing plymouth with some kind of workaround (ricardfunke made same or more complete modifications, is my version). Here is my patch for restoring "/run/plymouth/pid" path:

ricardofunke commented on 2012-07-19 15:34 (UTC)

plymouth-quit.service must run before getty@tty1.service, in this way display-manager.service (yourdm-plymouth.service) doesn't need to conflict with getty@tty1.service neither run after that. My plymouth-quit.service: My gdm-plymouth.service: Doing that, tty1 is available again without showing you password when typing and maybe will not show tty1 in transition to display manager.

abbradar commented on 2012-07-18 17:07 (UTC)

A lot of places in systemd have /run/plymouth/pid path, so won't it be simpler to just do "mkdir -p /run/plymouth" before daemon starting in mkinitcpio hook? (just do search for "run/plymouth/pid" in /usr/lib/systemd, for example) Just built package with this implemented, but I see no improvements - plymouth really never gets killed and gdm starts on another VT (I use gdm-plymouth.service)

dbuch commented on 2012-07-18 15:08 (UTC)

How do you fix the PID error on non-systemd setup?

BasT commented on 2012-07-15 16:52 (UTC)

All .service files need to be updated: The [Install] section should be changed from to Alias=display-manager.service

Det commented on 2012-07-15 10:38 (UTC)

Surprising how painful these changes can be. The 'systemd-ask-password-console.path' and 'systemd-ask-password-console.service' files still hold the old '/run/plymouth/pid' path, though.

Samsagax commented on 2012-07-15 03:30 (UTC)

I can confirm @ricardofunke's solution works. But now, plymouth never gets killed :S

ricardofunke commented on 2012-07-15 00:43 (UTC)

To fix the '/run/plymouth/pid not found' error, you must change the systemd and mkinitcpio hook files from '/var/run/plymouth/pid' to '/run/'. Since the /run directory is volactile (all it's contents is removed in shutdown or reboot), the plymouth isn't able to create the plymouth directory inside it for itself, but it can create a file, so /run/ turns to be the solution.

Samsagax commented on 2012-07-14 22:25 (UTC)

Removing and reinstalling plymouth to see what's wrong. The filesystem thing was long ago, and worked fine, the first time I got plymouth working, but then I got this error over and over again.

commented on 2012-07-14 20:19 (UTC)

@Samsagax The pid file is located in /var/run/plymouth/pid. Maybe there is a problem with that directory's permissions(?). Please use the files uploaded by Det in this page. Also have you done this?: "ls -lah /var/run" should show it as a symlink to /run although I don't see why that could be a problem.

Samsagax commented on 2012-07-14 17:08 (UTC)

@Zephyr: Plymouth won't start in my config: It says it can't write pidfile. I'm using grub2 + systemd. How can I fix that? Thank you :)

commented on 2012-07-14 14:37 (UTC)

@Det Thanks for fixing the hooks and units :D About the '--depth 1', my local PKGBUILD never had that so I didn't actually remove it.

deadcode commented on 2012-07-14 08:11 (UTC)

The splash screen still shows in tty1

Det commented on 2012-07-13 23:11 (UTC)


Samsagax commented on 2012-07-13 22:51 (UTC)

I'm getting errors in mkinitcpio hooks: -> Running build hook: [plymouth] ==> ERROR: file not found: `/sbin/plymouthd' ==> ERROR: file not found: `/bin/plymouth'

Det commented on 2012-07-13 19:23 (UTC)

Nice. But why did you remove '--depth 1' from 'git clone'?

commented on 2012-07-13 19:02 (UTC)

Sorry guys, I had old files in my build dir. I have modified the encrypt patches so that they apply cleanly and corrected the sha1sums.

commented on 2012-07-13 17:13 (UTC)

@qifei9 sha1sums are also not right

qifei9 commented on 2012-07-13 13:09 (UTC)

@Zephyr I used your PKGBUILD and tarball. And... "patching file encrypt_hook Hunk #1 FAILED at 69. 1 out of 1 hunk FAILED -- saving rejects to file encrypt_hook.rej"

commented on 2012-07-12 21:21 (UTC)

@ImNtReal You are correct, I have uploaded a src tarball: Please notice that I am using/testing this with systemd only. I do not know if plymouth is working with initscripts but it should.

ImNtReal commented on 2012-07-12 20:37 (UTC)

@Zephyr, it looks like your PKGBUILD is trying to use some patches that aren't included in the old tarball.

commented on 2012-07-12 20:19 (UTC)

Hello, you can try this PKGBUILD for the glibc/usr move update. Note that the systemd units are now included in the plymouth package and the ones inside the tarball should be deleted. (They are also "compiled" so that they follow the /usr move) (the maintainer has my name because I used to maintain plymouth-systemd-git, a new maintainer should update the package and remove that line)

gog commented on 2012-07-08 21:55 (UTC)

Hi guys. I tried to get things to work, but other matters require more of my attention. I apologize for the inconvenience.

dieghen89 commented on 2012-07-08 12:55 (UTC)

hi! in few days the rebuild of glibc will hit core. Please read [1] to get your package ready. [1]

commented on 2012-07-07 20:04 (UTC)

do i have to change all the paths in the service files for systemd ? o.O

fs4000 commented on 2012-07-07 16:35 (UTC)

Please change to --without-system-root-install. Arch is just removing /lib, plymouth makes the update to glibc 2.16.0-2 impossible. And you need to strip off the paths in the initcpio files. Path resolution works in initcpio so just use the program name. To fix the systemd service files you can use "sed -i 's:/lib/systemd/system:/usr&:' configure" and --enable-systemd-integration. The symlinks still have to be created. Some other improvements: configure with --without-rhgb-compat-link, remove --enable-gdm-transition since it is deprecated and --with-gdm-autostart-file=yes as it doesn't make sense without log-viewer

commented on 2012-07-07 11:27 (UTC)

Prompoting for encrypted drive password not working yet. From what I can see the encrypt and plymouth-encrypt hooks on my system are identical. Does anyone have this working?

ImNtReal commented on 2012-06-20 20:37 (UTC)

I think that the PKGBUILD should be updated to, the plymouth-buildfixes.patch to, and add plymouth-no-gtk.patch in case someone wants to fix it separately since it keeps changing. It should look like this:

BasT commented on 2012-06-18 08:59 (UTC)

Any news on how to get it to work with the patch?

ImNtReal commented on 2012-06-12 16:31 (UTC)

Maybe it would be a good idea to move it to a separate patch. That could make it easier to maintain, or at least allow users to exclude the patch instead of modifying it when it breaks.

gog commented on 2012-06-12 16:14 (UTC)

That hunk is necessary to remove a dependency on GTK+, but it keeps changing upstream.

solsticedhiver commented on 2012-06-10 18:13 (UTC)

One need to remove the first hunk of the patch to make it work i.e. Remove those lines in plymouth-buildfixes.patch @@ -53,10 +53,6 @@ AC_SUBST(PANGO_LIBS) fi -PKG_CHECK_MODULES(GTK, [gtk+-2.0 >= 2.12.0 ]) -AC_SUBST(GTK_CFLAGS) -AC_SUBST(GTK_LIBS) - AC_ARG_ENABLE(libdrm_intel, AS_HELP_STRING([--enable-libdrm_intel],[enable building with libdrm_intel support]),enable_libdrm_intel=$enableval,enable_libdrm_intel=no) AM_CONDITIONAL(ENABLE_LIBDRM_INTEL, [test "$enable_libdrm_intel" = yes])

commented on 2012-06-10 08:11 (UTC)

Hi! Trying to compile the package plymouth-git 20120610-1 and I get an error with patches: ==> Applying Patches... patching file ./ Reversed (or previously applied) patch detected! Assume -R? [n] Building is aborted whatever I answer, yes or no... How can I fix it?

gog commented on 2012-06-08 19:42 (UTC)

I'm setting up a testbed to get to the bottom of the matter.

LeonidasXIV commented on 2012-06-08 14:48 (UTC)

Thanks for taking over but I just tried and it still does not ask for passphrase while booting like it used to.

gog commented on 2012-06-07 01:03 (UTC)

Hi, plymouth-git fans. You should have pretty boot screens again.

commented on 2012-06-02 15:25 (UTC)

I have the encrypt_hook ... FAILED encrypt_install ... FAILED error on on archlinux x86_64. Please, fix the sha1 summs in the PKGBUILD. The sha1summs is: f39177bcfe423d03b362888025d48903f6dea75c encrypt_hook 26fb380d56e64cc777ac32e2c124c41ac0769ff5 encrypt_install

ricardofunke commented on 2012-06-02 15:19 (UTC)

Sorry people! I'm disowning this package. No time to maintain...

commented on 2012-05-30 01:16 (UTC)

@ricardofunke Alright, next time include some sort of warning haha! For anybody else with this issue, simply take the "plymouth" hook out of /etc/mkinitcpio.conf and change the "plymouth-encrypt" hook back to "encrypt", then rebuild the image with mkinitcpio -p linux (stock, core kernel is assumed, if your running anything else you should be able to figure it out on your own)

ricardofunke commented on 2012-05-28 13:34 (UTC)

Hello @liquidsky and @Nictrasavios, I have to remove some patches related to that in order to compile because they wasn't working anymore. At this moment I don't know how to fix these patches yet, so I'm apologizing for this problems and I hope I can fix this soon. Regards.

commented on 2012-05-27 20:02 (UTC)

I have the same issue as liquidsky. Did a complete reinstall before I figured out what it was....

liquidsky commented on 2012-05-27 08:55 (UTC)

Since the last update I can't use Plymouth with my crypted LVM. The Text field don't appears. And if I switch with F2 to the Text mode I can't type in the Pass code. After two keystrokes the cursors moves in a new line and I get the message that the passcose is wrong.

ricardofunke commented on 2012-05-24 16:16 (UTC)

@KriK, it's fixed, thanks.

KriK commented on 2012-05-24 15:23 (UTC)

==> Validating source files with sha1sums... arch-logo.png ... Passed encrypt_hook ... FAILED encrypt_install ... FAILED

LeonidasXIV commented on 2012-05-14 15:04 (UTC)

@christian.d, you're right. For some reason, I uninstalled libtool although it is of course in base-devel. My fault, sorry for the false alarm and thank you!

commented on 2012-05-12 09:22 (UTC)

LeonidasXIV You must install libtool

LeonidasXIV commented on 2012-05-09 12:00 (UTC)

@ricardofunke: Thanks for looking into it, but I don't see the file in git: I suppose it is generated by the autotools, because I see this error, before it continues with configure (and fails): [snip many more errors of this type] src/tests/ warning: variable 'ply_boot_splash_test_LDADD' is defined but no program or src/tests/ library has 'ply_boot_splash_test' as canonical name (possible typo) src/tests/ 'src/tests/' included from here autoreconf: automake failed with exit status: 1

ricardofunke commented on 2012-05-08 14:08 (UTC)

Well @LeonidasXIV, It's working for me... This file is pulled by git, so I think it's not related to the package. Have you tried to install again? Check if git is retrieving all the files without errors.

LeonidasXIV commented on 2012-05-08 02:10 (UTC)

./configure fails with: config.status: error: cannot find input file: `src/libply/'

ricardofunke commented on 2012-05-06 15:27 (UTC)

Done @Damon, thanks

daimonion commented on 2012-05-06 11:33 (UTC)

Initcpio hooks need to be installed in /usr/lib/initcpio/hooks

ricardofunke commented on 2012-05-01 13:45 (UTC)

Thanks @Shaeffer

commented on 2012-05-01 13:21 (UTC)

systemd unit files have changed directory. They are stored into '/usr/lib/systemd' instead of '/lib/system'. Please update your PKGBUILD.

ricardofunke commented on 2012-04-29 16:28 (UTC)

Fixed, thanks @Maxr.

Maxr commented on 2012-04-29 15:27 (UTC)

encrypt_install got a wrong checksum

ricardofunke commented on 2012-04-04 00:19 (UTC)

systemd isn't build by this package, it's installed as a dependency

h31 commented on 2012-04-03 22:30 (UTC)

Did you tried to use --enable-plymouth option while building systemd? Are there any difference with it?

ricardofunke commented on 2012-04-03 19:32 (UTC)

Thankyou for you comment GNRS. I must understand a lot more about systemd, I already doing some efforts to solve this problem. If someone already knows how to fix it or want to make a try, feel free to send me your suggestions

GNRS commented on 2012-04-03 13:15 (UTC)

I get a black screen between the plymouth splash screen and GDM. I think it is due to GDM loading (it is not a black screen with a blinking cursor, it's just an empty screen). I wonder if it is normal and whether it is possible to tell plymouth to wait until GDM is completeley loaded so there is no black screen between them. Any ideas?

ricardofunke commented on 2012-02-17 13:14 (UTC)

Ok, done! Thank you guys!

Det commented on 2012-02-17 02:11 (UTC)

I'm pretty sure people got initscripts installed. The ones who don't obviously use systemd (otherwise they couldn't even boot). Just put both to optdeps.

ImNtReal commented on 2012-02-16 20:51 (UTC)

It seems like there should be a common provides between the two. It's not that big of a deal for me to remove the dependency when I update, and it's probably safe to assume anyone willing to go to native systemd can probably do the same.

ricardofunke commented on 2012-02-16 20:39 (UTC)

But, on reflection, if I do so, both systemd and initscripts will be optdepends and this package require that one of them is installed on the system in order to work. So, I think I prefer to keep it with what is working better, what do you think?

ricardofunke commented on 2012-02-16 20:34 (UTC)

Oh, you're right! I'll do that, thanks!

ImNtReal commented on 2012-02-16 20:25 (UTC)

@ricardofunke, I don't blame you for wanting to leave initscripts on there. I just think it should be moved to optdepends, so it can be built without modifying the PKGBUILD if one is running with native systemd. I just switched to native, myself, thanks to the new systemd-sysvcompat package.

ricardofunke commented on 2012-02-16 20:22 (UTC)

Well @ImNtReal, I think it still works better with initscripts. For me, systemd option still have some problems that I'm working on... For example: 1. The plymouth-start.service stay with failed status in systemctl. 2. Sometimes, actually most of them, plymouth don't work on shutdown or reboot. Are these problems happening with you too?

ImNtReal commented on 2012-02-16 20:00 (UTC)

It's working for me this way. I would recommend moving initscripts to optdepends, and maybe add systemd there, so people can tell they have a choice. Thanks.

ImNtReal commented on 2012-02-16 15:43 (UTC)

Does this really require initscripts, or will it run from native systemd, also?

ricardofunke commented on 2012-02-10 16:22 (UTC)

Well @Maxr, I've tested the checksum and it's still correct. Maybe you had some temporary trouble with your connection. So try again

Maxr commented on 2012-02-10 15:45 (UTC)

encrypt_hook ... FAILED wrong checksum?

ricardofunke commented on 2012-02-08 13:03 (UTC)

Agree @Det, done

Det commented on 2012-02-07 23:09 (UTC)

It should rather be an optdeb, since this one now includes systemd support.

ricardofunke commented on 2012-02-07 18:55 (UTC)

@webmeister, I agree

webmeister commented on 2012-02-07 18:33 (UTC)

Is the systemd dependency really necessary? The package seems to build and run fine without it.

ricardofunke commented on 2012-02-07 17:13 (UTC)

Fixed last problems! Sorry for the confusion! Thanks @brcha, @Det and @Shaeffer! Sorry my bad english too, I'm Brazilian...

Det commented on 2012-02-07 16:08 (UTC)

I'd assume ricardo is either using an older PKGBUILD of this thing or he's just verifying it somehow incorrect. If you're expecting those "Removing X11-Gtk+ renderer" messages, you'd need to add them yourself with 'msg' or 'msg2'. Otherwise it's just a couple of "patching file" lines with no clear indication which patch is being applied.

brcha commented on 2012-02-07 11:54 (UTC)

Hi Ricardo, I've just updated to latest plymouth-git and noticed that rebuilding initcpio gives an (non fatal) error that x11 renderer is missing. Of course it is missing, since you've included the no-x11-no-gtk patch (btw, shouldn't you remove gtk2 from deps?), but, it seams, you forgot to remove the x11-renderer line from initcpio install script. :) Anyway, thank you for maintaining this package & keep up the good work :) PS: The no-x11-no-gtk patch is being applied on my computer, and honestly I cannot see a reason why makepkg would skip over that patch...

commented on 2012-02-07 04:21 (UTC)

The patch was applied. I used --verbose option with the patch command. Sorry for my bad English, I'm French.

ricardofunke commented on 2012-02-06 17:23 (UTC)

The changes was done. But @Det and @Shaeffer, could you help me to find out why no-x11-no-gtk.patch is not being applied?

Det commented on 2012-02-06 15:56 (UTC)

Well you of course also need to install/symlink them as in: Afterwards 'plymouth-systemd-git' should be removed.

ricardofunke commented on 2012-02-06 15:53 (UTC)

@Shaeffer, so it's just add this files and it works? Nothing else?

commented on 2012-02-06 11:44 (UTC)

@ricardofunke Thanks ! You can take files from this PKGBUILD : . I use them with yours and it works perfectly.

ricardofunke commented on 2012-02-03 19:58 (UTC)

Done @Det fixes! #Todo: 1. Aply the settings "StandartInput=null" to without package conflicts 2. Add systemD units files if possible @Shaeffer, I don't forgot you! ;)

ricardofunke commented on 2012-02-03 19:38 (UTC)

Ok @Det, I'll do that Thank you

Det commented on 2012-02-03 19:20 (UTC)

So could you please also use '--depth 1' with 'git clone' (to not fetch the complete history) and fix the encrypt_hook patch (you can grab it from the tarball I linked, remember to update the sha1sum)? Thank-you. E: nevermind, didn't see your reply there.

ricardofunke commented on 2012-02-03 19:09 (UTC)

Ok @Det, done. As soon as possible I'll verify your package and the @Shaeffer request. Thanks

Det commented on 2012-02-03 19:00 (UTC)

I think it's the same as with plymouth, meaning: # Conbtributor: Det <nimetonmaili at gmail dot com> # Contributor: Biginoz <biginoz _at_ free _point_ fr> # Contributor: PirateJonno <>

ricardofunke commented on 2012-02-03 18:16 (UTC)

Well @Det, I think you're right about the list of contributors. But I don't have the list anymore, if you have it, please, post here and I'll put back again.

ricardofunke commented on 2012-02-03 18:08 (UTC)

@Det, this package was orphan for a long time. This means that there was no mainteiners, much less contributors! Man, I know you want to contribute and that you have skills, but I cannot understand why you have to be so rude...

Det commented on 2012-02-03 16:58 (UTC)

@ricardofunke, that's not how it goes. The "contributors" list means that those people have _at some point_ been maintainers of this package (or helped in some other major way). You can't just remove them just because you don't know them and I don't see how this qualifies as a psychiatric problem on my side.

divansantana commented on 2012-02-03 16:52 (UTC)

@Det There's no need to disrespect. One can still get their point across in a more helpful way. Thanks @ricardofunke for your work and patience.

ricardofunke commented on 2012-02-03 16:47 (UTC)

@Det Man I think you're having some kind of psychiatric problem! I removed so because this package was orphan, there's just a few days that I became manteiner and I don't know those people. I'll put their name again once they return to contribute. Why don't you'll find some medical help?

Det commented on 2012-02-03 16:17 (UTC)

That's funny. You actually removed all the contributors from the beginning of the PKGBUILD? You want this package to be force orphaned from you?

ricardofunke commented on 2012-02-03 16:07 (UTC)

Hi @Shaeffer, I'll take a look at this as soon as possible. Thanks

Det commented on 2012-02-03 13:20 (UTC)

@ricardofunke, don't pretend it's anybody else's problem than your own. Do you think people _care_ that you're going to stand for principle here? You think they'll come to congratulate you for how stubborn you have been? Well they won't. They care about this package. You care about teaching me a lesson (good luck with that). The simple fact is, you're fighting against the _users_ - not me. Why is that? Only one of us needs to grow up. Can't take criticism? Too bad, you're gonna face it in your life one way or another. So you got a choice: keep on fighting or don't and update your package, _show_ that it's me who needs to grow up. Pick one.

commented on 2012-02-03 09:48 (UTC)

Can you add systemD units files in this PKGBUILD ? You can use files from plymouth-systemd-git PKGBUILD.

ricardofunke commented on 2012-02-02 19:51 (UTC)

No @Det, I think you should grow up and learn to respect people. I will not give any more attention to you.

Det commented on 2012-02-02 17:07 (UTC)

I had to laugh. First of all, _you_ are the maintainer of this package. _You_ have the responsibility of making the changes. This means _you_ also claim to be able to do this. I'm telling you that being the maintainer means you should also know how to do diffs (you could even do that with a graphical app like KDE's Kompare) and if that actually makes you so ridiculously angry then I can only assume you don't. And that's very bad for the package - _and_ its users. If you can't take criticism, then disown this package. If you can, then deal with it with a little more grown-up fashion and learn to put the package before yourself.

ricardofunke commented on 2012-02-02 16:14 (UTC)

Well @Det, this package is being orphan for a long time, why don't you take such a "responsability" if you think you are that might power script kiddie? I tried to be polite with you, but now I have to warn you that I won't give any attention to displeasing people.

Det commented on 2012-02-02 08:47 (UTC)

Being the maintainer you should be able to figure that out yourself. Just some simplifications, '--depth 1' for git clone and fixed the encrypt_hook patch (forgot a space in the beginning of the line where it's being used, though. You should add that).

ricardofunke commented on 2012-02-01 20:49 (UTC)

@Det Why would I? What does it provides?

Det commented on 2012-02-01 17:53 (UTC)

Ok. Could you still please pull my stuff?

ricardofunke commented on 2012-02-01 14:19 (UTC)

Oh it's right @Det! Now I get ya... =) Well, I don't removed the file, I just commented the lines, but makepkg doesn't packaged the file because of that. I just commented because I don't know yet if it will be necessary in the near future... Was just a quick solution, I haven't given a good attention to this yet...

Det commented on 2012-02-01 12:45 (UTC)

Pull my stuff:

Det commented on 2012-02-01 12:30 (UTC)

I meant that why did you leave the rc-local.service lines there? If you already removed the file itself then why not the lines too?

ricardofunke commented on 2012-01-30 22:59 (UTC)

Well @Det, I mean, if someone has a better way to set "StandartInput=null" in rc-local.service, to feel free to post to me, because I can't realize now how to do it.

ricardofunke commented on 2012-01-30 22:57 (UTC)

Hi @Det, I don't understand what you mean...

Det commented on 2012-01-30 21:54 (UTC)

So you removed the actual file, but just commented everything in the PKGBUILD? How about just like removing the stuff in the PKGBUILD too?

ricardofunke commented on 2012-01-30 21:42 (UTC)

I commented the lines that installs rc-local.service. If someone has a better way to set this, please feel free to post here. ;)

Feanor12 commented on 2012-01-30 21:31 (UTC)

file conflict : /lib/systemd/system/rc-local.service plymouth-git & initscripts-systemd

StS commented on 2011-10-09 06:28 (UTC)

smooth plymouth -> gdm yaourt -S plymouth-systemd-git # systemctl disable gdm.service # systemctl enable gdm-plymouth.service

StS commented on 2011-10-09 05:45 (UTC)

Hi, is it possible to keep plymouth screen till gdm is loading ? . I mean like it was in ubuntu , after plymouth u see desktop right away, there wasn't black screen between them. And if some1 still have a bug. just edit /lib/systemd/system/rc-local.service change: -StandartInput=tty +Standartinput=null

mrbit commented on 2011-10-01 13:50 (UTC)

Password properly only in the terminal tty1 clear, odd keyboard input pressing arrow keys

commented on 2011-09-21 00:23 (UTC)

Anyone have a viable fix for the "Small black squares" glitch? I use inittab to start X and GDM; the only way I can avoid this is by killing Plymouth partway through boot.

Samsagax commented on 2011-09-08 23:49 (UTC)

Lol > Look & Feel If you get it working again that would be great.

Det commented on 2011-09-08 21:30 (UTC)

L&F? Lose and forget?

Samsagax commented on 2011-09-08 19:06 (UTC)

@Det: In fact it is. And if the package is well maintained, then the time spent is returned in good feedback. I'm looking forward on this one. Really completes the Arch L&F

Det commented on 2011-09-07 16:31 (UTC)

@Samsagax, it probably _is_ related to plymouth.functions. I wonder if I should start maintaining this again. I wouldn't even need to really come up with anything myself - it would be more like for appearances to get people give more input (which clearly decreased after I disowned this thing). I guess people are just used to only packages with a maintainer(s) being maintained - so when a package doesn't have any they are more reluctant to contribute.

Samsagax commented on 2011-09-07 16:15 (UTC)

Found a problem related to plymouth + ufw: If you have plymouth installed, when you try to start ufw daemon, it will fail and print no messages. I've managed to print a trace (using set -x in the script) and seems to be the "/bin/plymouth --ping" call that is failing the entire script. (Somewhere at start ufw-init-functions enables set -e so it will fail entirely if any command fails). Here is the trace: After uninstalling plymouth it works normally. I'll dig it more to have a better conclusion. As this maybe related to modified functions for Arch, should I report this upstream?

Maxr commented on 2011-09-06 20:27 (UTC)

working here.

commented on 2011-09-04 06:22 (UTC)

Will not install, here are my errors ==> Applying Patches... patching file encrypt_hook patching file encrypt_install patching file scripts/plymouth-update-initrd ==> Starting make... /tmp/yaourt-tmp-xaer0/aur-plymouth-git/./PKGBUILD: line 77: ./ Permission denied ==> ERROR: A failure occurred in build(). Aborting... ==> ERROR: Makepkg was unable to build plymouth-git.

Det commented on 2011-08-14 13:09 (UTC)

Yeah, that's the one.

Cilyan commented on 2011-08-14 12:07 (UTC)

Old plymouth.functions (if I'm not mistaken)

Det commented on 2011-08-13 22:32 (UTC)

Soooo... does anybody have it?

ricardofunke commented on 2011-08-10 23:23 (UTC)

I think the /etc/rc.d/functions.d/plymouth.functions file is running all your stuff both in the boot and the shutdown, because my theme is running in shutdown mode on the boot process... /o\ Well, think the old plymouth.functions were more correct than this one... =P

Det commented on 2011-08-02 22:30 (UTC)

Disowning. Clearly there's people more capable of maintaining this thing than I am. Kinda busy with other stuff too.

ricardofunke commented on 2011-08-02 21:48 (UTC)

Hi again, About the "can't shutdown" and the "not showing splash on shutdown" problem... I've tried all the solutions proposed here, but no success, so I undo all the changes and put back the original file. Then I decided to try one more thing, modifying the ply_shutdown_start() function in plymouth.functions, like this: __ply_shutdown_start(){ ____XPID=`pidof X` ____if [ "$XPID" ]; then ______if [ "`runlevel | cut -c 3`" != '5' ]; then ________local DM ________for DM in slim gdm kdm xdm entrance; do __________ck_daemon "$DM" || stop_daemon "$DM" ________done ______fi ______kill -9 $XPID &> /dev/null ____fi ____ply_daemon --mode=shutdown --pid-file=/tmp/plymouthd ____ply_client --show-splash ____while [[ ( $(ps h $(cat /tmp/plymouthd)) ) && ( $ply_wait -lt 3 ) ]]; do ______$(( ply_wait++ )) ______sleep 1 ____done ____[[ $(ps h $(cat /tmp/plymouthd)) ]] && kill -9 $(cat /tmp/plymouthd) ____# don't get killed by kill_all ____add_omit_pids `cat /tmp/plymouthd` __} Though, I think just sleeping 3 seconds and then kill plymouthd will be enough. But, anyway, I put all that "while" thing to verify if plymouthd will stop correctly before the kill... So, with that, my machine can shutdown showing the splash correctly... I have submitted a bug to mainstream about the hang of "plymouthd --mode=shutdown", just waiting for response... []'s

Det commented on 2011-08-02 10:31 (UTC)

Humm.. just stopped working completely for me. Hangs on _both_ startup _and_ shutdown (with and without Nouveau). Just wasted like 3 hours of my life trying to figure out what was wrong.. without any luck. The stable one ('plymouth') didn't work either so I can only assume it was some [testing] update that did it.

ricardofunke commented on 2011-08-01 17:09 (UTC)

Hi guys, I have submited a bug to mainstream yesterday about the problem with tty1: Here is what they said: "It sounds like you're quitting plymouth after starting your getty, make sure you quit plymouth before starting your getty. plymouth will restore tty settings to the way it found them when it quits. That could overwrite any settings your getty tries to change."

ricardofunke commented on 2011-08-01 16:32 (UTC)

@auti, My theme is displayed on boot, but not on shutdown But, I'll remove vga=773 as you said, thanks! ;) @Samsagax, I'll try commenting this too and see if I'll can shutdown

autermann commented on 2011-08-01 00:27 (UTC)

'ply_client quit --retain-splash' kills plymouthd but does not clean the framebuffer. Who can guess that it would prevent init from terminating...

Samsagax commented on 2011-07-31 23:59 (UTC)

I Can't shutdown or reboot system with last update. Fixed this by commenting line 40 in function ply_quit_shutdown(). Now everything works as expected. Should look like this: ... # stop plymouth before shutdown ply_quit_shutdown() { #ply_quit_boot [[ $(ps h $(cat /tmp/plymouthd)) ]] && kill -9 $(cat /tmp/plymouthd) } add_hook shutdown_poweroff ply_quit_shutdown ... Why is "ply_quit_boot()" function call there in the first place? This is what it does: ... # stop plymouth after rc.multi ply_quit_boot() { ply_client quit --retain-splash } ... Why quit retaining splash if we are going to turn the system completely off? ply_quit_shutdown() is called when the system turns to the proper runlevel and just need to kill plymouth so INIT can terminate.

autermann commented on 2011-07-31 23:16 (UTC)

@ricardo: i have the same problem with my nvidia based box. I didn't set up a framebuffer and plymouth defaults to the text theme. That's why everything is echoed. I think the vga=773 in your kernel command line conflicts with KMS, such that your theme can't be displayed. @Det: thanks :)

ricardofunke commented on 2011-07-31 21:03 (UTC)

@Cilian You misunderstood me, I mean: - tty1 is still echoing everything that is typed after the prompt, including the password

ricardofunke commented on 2011-07-31 21:00 (UTC)

@Cilyan Yes, I did! What's the relation? $ cat /proc/cmdline root=/dev/mapper/vg00-root ro vga=773 quiet splash i915.modeset=1

Cilyan commented on 2011-07-31 20:38 (UTC)

"- tty1 is still echoing everything" Just a check: did you add the option "quiet" to the boot line in grub/...?

ricardofunke commented on 2011-07-31 19:24 (UTC)

Well guys, I've upgraded the package and it still have some problems: - tty1 is still echoing everything - plymouthd --mode=shutdown is still hanging, so I still have to force kill [1] - when shuting down or rebooting, the splash is not showed I have tested it in two notebooks with the same results, both are running Gnome3 with GDM [1] Here is the code to force kill in plymouth.function: __# stop plymouth after rc.multi __ply_quit_boot() { ____ply_client quit --retain-splash & ____if [[ -f /tmp/plymouthd ]]; then ______while [[ ( $(ps h $(cat /tmp/plymouthd)) ) && ( $ply_wait -le 3 ) ]]; do ________$(( ply_wait++ )) ________sleep 1 ______done ______[[ $(ps h $(cat /tmp/plymouthd)) ]] && kill -9 $(cat /tmp/plymouthd) ____fi __}

Det commented on 2011-07-31 17:04 (UTC)

And even the ~/.xinitrc trick doesn't work?

commented on 2011-07-31 14:54 (UTC)

It works I can now see the plymouth animation, still having problems with the small black squares using nvidia with proprietary drivers ¿Any ideas? Great job Thanks.

Det commented on 2011-07-31 08:29 (UTC)

@auti, I don't get to say this too often but: a brilliant package! There were a couple of minor "carelessnesses" but I mean: wow!

autermann commented on 2011-07-31 00:16 (UTC)

This would require patching of X[1], KDM[2], GDM[3] or whatever display manager you use. More information at [4] ;) [1];a=blob;f=xserver-1.5.0-bg-none-root.patch [2];a=blob;f=kdebase-workspace-4.4.92-kdm_plymouth081.patch [3];a=blob;f=plymouth.patch [4]

commented on 2011-07-30 23:02 (UTC)

Would it be easy to make it like fedora and ubuntu and make plymouth fade out with the display manager?

ricardofunke commented on 2011-07-30 21:03 (UTC)

@auti, no solves, same problems and the old problem of "doesn't shut down" is back. Sorry for the bad news, but thanks anyway ;)

autermann commented on 2011-07-30 19:50 (UTC)

You may try my version. The current package won't start plymouth at shutdown as the wrong hook is registered. And no problems with echoed commands on tty1 here...

ricardofunke commented on 2011-07-30 18:49 (UTC)

Just one more issue... The splash is not showing in shutdown anymore, what can I do?

ricardofunke commented on 2011-07-30 18:41 (UTC)

I'm having an issue with my tty1, it's echoing every command I type there, even the password! It's just in tty1, the other tty works ok. Have you facing this trouble too?

autermann commented on 2011-07-30 17:01 (UTC)

@Det: the shutdown_start hook is only called if /etc/rc.shutdown is executed and the multi_end hook is called everytime you switch to runlevel 5. I've updated the package [1] and made some changes.. feel free to adopt those: * partly rewritten plymouth.functions. registers now lots of hooks and works like a charm * killing the daemon in rc.local is not needed * the plymouth-encrypt hook is generated by patching the original cryptsetup hook files. * plymouth-update-initrd.patch is updated to use *.preset instead of *.kver in /etc/mkinitcpio.d [1]

Det commented on 2011-07-30 16:43 (UTC)

Started maintaining this thing (for now). It has so many sweet little votes.

Det commented on 2011-07-30 09:17 (UTC)

Btw. I updated the Wiki. It's up-to-date again :p.

Det commented on 2011-07-30 07:24 (UTC)

Guys, there's no maintainer with this thing. You are free to do that yourselves. E: @ricardofunke, I did those. @auti, I don't think if it should register either one at the end of the file. Doesn't this mean it does it every time the file is called (on startup _and_ shutdown (though, this is not currently implemented)).

autermann commented on 2011-07-30 01:00 (UTC)

shouldn't /etc/rc.d/functions.d/plymouth.functions register a shutdown_start hook instead of a multi_end hook? from /etc/rc.d/functions: # shutdown_start: at the beginning of rc.shutdown # multi_end: at the end of rc.multi

ricardofunke commented on 2011-07-30 00:25 (UTC)

please change /etc/rc.d/functions.d/plymouth.functions line from: while [[ ( $(ps h $(cat /tmp/plymouthd)) ) && ( $ply_wait -lt 5 ) ]]; do to while [[ ( $(ps h $(cat /tmp/plymouthd)) ) && ( $ply_wait -lt 3 ) ]]; do 5 seconds is such a long time, 3 is better...

ricardofunke commented on 2011-07-30 00:21 (UTC)

Every update still overwrites our /etc/plymouth/plymouthd.conf. This should not occur in order do keep our theme selection.

commented on 2011-07-18 21:00 (UTC)

Please rename the "install" function in "/lib/initcpio/install/plymouth" and "/lib/initcpio/install/plymouth-encrypt" to "build"

hellnest commented on 2011-07-10 18:15 (UTC)

Guys don't really have time to updated and maintenance it anymore, also will away from my linux box for a while. Here a recent tarball for this package i don't have time to re-create this on my Arch machine. Thanks i will drop this package and feel free to maintenance :)

divansantana commented on 2011-07-10 17:24 (UTC)

with latest mkinitcpio 0.7.2-1 below warning. ==> WARNING: Hook 'plymouth-encrypt' uses a deprecated 'install' function. This should be renamed 'build'

commented on 2011-07-10 13:48 (UTC)

I'm running with the kernel26-lqx and also keep the kernel26 around for good measure. It builds and runs fine - so long as I first change 'install' to 'build' in the '/lib/initcpio/install/plymouth' file after installation. Since I'm also running with systemd, I only see the splash animation on bootup and not on shutdown as I was without systemd.

hellnest commented on 2011-07-06 07:08 (UTC)

@sidneyk i done that on saturday. :) but just now have time to upload anyway thanks for your information. Plymouth should build now against new mkinitcpio, but i still got an error "disconnected from boot daemon". I'm running BTRFS system. Let me know if this package run fine on others :)

commented on 2011-07-06 05:08 (UTC)

Seems that all that is required to fix plymouth-git is to change a couple items in /lib/initcpio/install/plymouth. On line #2, change 'install' to 'build' and remove the 'add_device' line, line #3. # vim: set ft=sh: build() { add_dir /dev/pts add_dir /usr/share/plymouth/themes It now builds and works just fine with minitcpio again.

hellnest commented on 2011-07-02 06:15 (UTC)

Ok i will take a look at it later, i will put the fix on monday probably. :) if there any suggestion for new mkinitcpio just let me know, i will be off for holiday this weekend.

commented on 2011-07-01 17:33 (UTC)

The mkinitcpio-0.7-1 update from 6-30-11 in testing breaks plymouth whenever the first images are built with it. It says that the 'install' function is deprecated in '/lib/initcpio/install/plymouth', that the function should be renamed to 'build' and also complains about the command 'add_device' not being found. If I change the 'install' function name to 'build' that error disappears, but the other error is still present. I tried replacing 'add_device' with 'mknod', as suggested in another post, but this does not work either. Reading the announcement in one of the mailing lists says that the 'install' function has been deprecated and the 'add_device' command removed. I don't know if there were any other things that are now deprecated or changed as far as the hooks go, but I couldnt get past these and build any images that had plymouth incorporated.

hellnest commented on 2011-06-29 04:30 (UTC)

@mond Thanks for your suggestion, i already added it to plymouth-update-initrd

commented on 2011-06-28 17:15 (UTC)

plymouth-update-initrd doesn't work with custom kernelversions e.g. kernel26-ck. When you change it to #!/bin/bash for i in /etc/mkinitcpio.d/*.kver; do i=${i##*/} i=${i%.kver} /sbin/mkinitcpio -p $i done it should rebuild the initrds for every kernel installed on the system.

hellnest commented on 2011-06-25 19:27 (UTC)

@ricardo done :), thanks for the correction :). PKGBUILD updated and fix plymouth.functions

ricardofunke commented on 2011-06-20 17:29 (UTC)

Another thing... The plymouth.conf file shouldn't be overriden at the updates... ;) Thanks!

hellnest commented on 2011-06-20 00:34 (UTC)

Nicely done :), actually i just realize i had a same problem when shutting down using 'shutdown -h' command. But using 'halt' is run fine ^^. Plymouth.functions updated

ricardofunke commented on 2011-06-19 15:33 (UTC)

Sorry, I forgot to thank you for the modifications... ;)

ricardofunke commented on 2011-06-17 19:00 (UTC)

Sorry, but I think you should reduce the time from 5 to 3, I saw later that 5 is too much... Could you put it indented?

ricardofunke commented on 2011-06-16 03:54 (UTC)

My plymouth don't let my notebook shutdown, then I found that "plymouthd --mode=shutdown" hangs and it stops only with kill -9. So I wrote that code above on plymouth.functions to fix it: #### (...) if [ "$0" == "/etc/rc.shutdown" ]; then XPID=`pidof X` if [ "$XPID" ]; then RUNLEVEL=`runlevel` if [ "${RUNLEVEL:2:4}" != '5' ]; then [ -e /var/run/daemons/slim ] && /etc/rc.d/slim stop [ -e /var/run/daemons/gdm ] && /etc/rc.d/gdm stop [ -e /var/run/daemons/kdm ] && /etc/rc.d/kdm stop [ -e /var/run/daemons/xdm ] && /etc/rc.d/xdm stop [ -e /var/run/daemons/entrance ] && /etc/rc.d/entrance stop fi kill -9 $XPID &> /dev/null fi ply_daemon --mode=shutdown --pid-file=/tmp/plymouthd ply_client --show-splash while [[ ( $(ps h $(cat /tmp/plymouthd)) ) && ( $ply_wait -lt 5 ) ]]; do ply_wait=$(( $ply_wait + 1 )) sleep 1 done [[ $(ps h $(cat /tmp/plymouthd)) ]] && kill -9 $(cat /tmp/plymouthd) fi ###

Cilyan commented on 2011-06-11 15:09 (UTC)

pkg-config (once again) is part of the base-devel, which is a prerequisite to build ANY package. That's why it should NOT appear in the [make]depends array.

hellnest commented on 2011-06-11 12:39 (UTC)

The PKGBUILD should running fine, just try it and give any info if it failed to build,

ricardofunke commented on 2011-06-11 11:40 (UTC)

So @dimitar, should pkg_config be as "makedepends" in PKGBUILD?

hellnest commented on 2011-05-22 00:31 (UTC)

Since i'm using it, i will try to maintenance the package

Cilyan commented on 2011-05-21 10:33 (UTC)

Update of PKGBUILD: - Quotes in PKGBUILD to protect paths with spaces - Separate package function - Get package version from autotools (for provides option) - Include fix for getpwid_r with a tweak to cope with different version of nss (should work till version 3 is released)

commented on 2011-03-17 15:48 (UTC)

about that getpwuid_r(): did that for me (ok, i had a different libnss version so i had to change that)

dimitar commented on 2011-03-16 21:21 (UTC)

@antiwinhacker: I also got the same error when I tried to makepkg. The problem is that the macro PKG_CHECK_MODULES cannot be found, i. e. you need to install pkg-config: $ sudo pacman -S pkg-config You'll also need patch: $ sudo pacman -S patch

toropisco commented on 2011-02-18 16:38 (UTC)

Make that getpwuid_r() :-)

toropisco commented on 2011-02-18 16:36 (UTC)

When booting finishes and just before the login manager starts, slim in my case, there is a message that says that getpwid_r() failed because there is no user with system id 0 (that would be root!). It seems that plymouth needs some sort of basic passwd file with basic accounts defined present in the initrd. I haven't actually checked that up myself, though.

commented on 2010-12-28 08:35 (UTC)

I just recompiled and it worked fine for me

commented on 2010-12-28 07:42 (UTC)

Current plymouth does not support libpng>= 1.4. I get this during configure: ./configure: line 12270: syntax error near unexpected token `IMAGE,' ./configure: line 12270: `PKG_CHECK_MODULES(IMAGE, libpng >= 1.2.16 )' I installed libpng12, but the configure script did not found :(

jorgemota commented on 2010-10-22 05:17 (UTC)

I got a unknow user id error after boot ends and before slim get loaded. but this dissappears so quick that I can't read entire msg, but sometimes the laptop freeze at it and reboot :S

jarryson commented on 2010-10-05 18:52 (UTC)

Sorry, i don't have much time right now. i will drop this package, plz adopt it.

daenyth commented on 2010-10-05 17:53 (UTC)

Make sure to make a massive warning about that.

commented on 2010-10-05 17:31 (UTC)

plymouth-encrypt is not a good hook name, because it cannot be skipped using the 'disablehooks=' boot option. The reason for this is that Arch's init script pushes kernel cmdline args into the environment using export, which doesn't like variable names with hyphens in them. Renaming it to plymouth_encrypt would fix this problem.

commented on 2010-10-04 22:57 (UTC)

this produces an ERROR during mkinitcpio without the ttf-dejavu font package installed. should this be listed as a dependency? I assume the font is configurable (haven't looked into it), so maybe just adding to the message after installation to mention that a font needs to be installed/configured?

commented on 2010-09-09 13:06 (UTC)

have you tried rebuilding the package? it downloads everytime you built it the newest git sources, you can change the last changes here:

fettouhi commented on 2010-07-10 07:08 (UTC)

Currently plymouth is causing graphical corruptions on my KDE desktop if I use it. Does anybody else experience that?

MuPuF commented on 2010-07-03 18:43 (UTC)

jarryson: Any update ?

jarryson commented on 2010-06-21 08:46 (UTC)

mkinitcpio 0.6.6 still can not mount /dev/pts, so have to remove this part.

jarryson commented on 2010-06-16 07:46 (UTC)

see this bug i reported:

jarryson commented on 2010-06-16 07:34 (UTC)

i did not recompiled 2.6.34 kernel. but i still have this problem. i think this is a bug or something.

MuPuF commented on 2010-06-15 21:38 (UTC)

Jarryson: I have the same bug. Only a few people must have it as only a few of us actually recompile their kernels ;) The bug happens on both my computers.

PirateJonno commented on 2010-06-15 07:02 (UTC)

to check your image run zcat /boot/kernel26.img | cpio -t 2>&1 | grep '/dev' and replace zcat with xzcat or bzcat if you're using lzma or bzip2 compression

jarryson commented on 2010-06-15 04:20 (UTC)

i know why now. i downgrade mkinitcpio to 0.6.4-1 and everything is fine now. i am not sure if this is a bug of mkinitcpio package, if anyone else met the same problem, i will try to report a bug.

jarryson commented on 2010-06-15 04:09 (UTC)

i am not sure, and even i remove plymouth, i still can see the "can not open /dev/null" warning. i don't know how to check if /dev/null and /dev/pts exist in my initcpio image, but i checked the scripts /lib/initrd, /dev/null and /dev/pts will be created when booting, so i really don't know how this issue happen.

PirateJonno commented on 2010-06-15 00:37 (UTC)

make sure /dev/null and /dev/pts are in your initcpio image?

jarryson commented on 2010-06-14 17:28 (UTC)

i met some problem when booting. these messages are too early to log, so i write it down from my memory. cannot find /dev/null mount: can not mount none on /dev/pts, file or directory doesnot exist. can not create session these all shown about plymouth started. but i don't think it's caused by plymouth. do anyone else met the same issue?

jarryson commented on 2010-06-14 16:37 (UTC)

thanks! Fixed.

fettouhi commented on 2010-06-14 16:31 (UTC)

plymouth.initcpio_hook has the wrong md5sum.

commented on 2010-05-24 12:03 (UTC)

Forgot to say: I get that error (which after googling seems to have nothing to do with plymouth itself) using an intel card 4500 I think

smahs commented on 2010-05-23 17:39 (UTC)

Have you removed temp.patch ? This pkg runs fine with vesa fb on nvidia. But the shutdown is screwed up. Compiling with --enable-gdm-transition will work for KDM ?

commented on 2010-05-15 19:10 (UTC)

GLib-WARNING **: getpwuid_r(): failed due to unknown user id (0) Also got this ;)

kolos commented on 2010-05-15 06:23 (UTC)

i got the same warning message @antigerme

daenyth commented on 2010-05-04 01:03 (UTC)

Yes, I've seen that also.

commented on 2010-05-04 00:20 (UTC)

Hook plymouth erro: GLib-WARNING **: getpwuid_r(): failed due to unknown user id (0) After the warning message closes plymouth. Someone is going through the same problem?

jarryson commented on 2010-04-30 14:25 (UTC)

if there is any daemons failed, plymouth will quit, so plz check if all the services run normally. if your shutdown all messedup, please try --enable-gdm-transition, or add the tmp.patch i provided.

Huulivoide commented on 2010-04-30 14:15 (UTC)

Now I do have right theme on start up, but the plymouth gets ?killed? just before the udev starts do its things. also i only have all messed up on shutdown

fettouhi commented on 2010-04-25 07:39 (UTC)

Yep. Many thanks!

jarryson commented on 2010-04-20 12:53 (UTC)

Fixed. need to build /etc/plymouth/plymouthd.conf into initrd image. PS: tmp.patch is for the ones who only see all messed when shutdown.

jarryson commented on 2010-04-19 14:15 (UTC)

i think i know what's the problem now, need time to test. plz wait for a while

Huulivoide commented on 2010-04-19 13:23 (UTC)

I get just simple blue bar moving to right on boot instead of the solar/any other theme. Also the Plymouth closes when the pc inters init 3/5. when shutting down I get picture of the solar theme but it closes imidiatly

fettouhi commented on 2010-04-13 17:37 (UTC)

Same problem here also.

commented on 2010-04-12 21:26 (UTC)

i ve got the same problem with my ati card

jarryson commented on 2010-04-10 11:21 (UTC)

do u guys have any problem with this recently? i can get boot splash work. the boot splash looks just like shutdown mode splash.

jarryson commented on 2010-03-30 16:10 (UTC)

i have already explained many times. you must compile libdrm with intel radeon nouveau support. acutally no need to use libdrm-latest at all, libdrm in testing is pretty new already.

commented on 2010-03-30 14:09 (UTC)

I got error compiling with libdrm_radeon:

jarryson commented on 2010-03-25 08:16 (UTC)

recently plymouth-git won't work properly again, shutdown mode won't show splash. but good news: version 0.8 is out, hope it can enter community