I just ran the update to Linux 7.0.0-1 and pamac does not open:
pamac-manager:5013): GLib-ERROR **: 14:40:40.776: ../glib/glib/gmem.c:106: failed to allocate 6 bytes
zsh: IOT instruction (core dumped) pamac-manager
zsh was included in the update.
Are you trying it via CLI or via pamac-manager?
I can confirm both is running with that kernel version.
Both CLI and icon in panel (Xfce) are not working. I also just built via AUR and it doesn’t work. I may just start using octopi.
journal shows a very large systemd-coredump
I just stumbled over the issue, that mkinitcpio throws a warning because all the /usr/lib/initcpio/hooks referencing to #!/usr/bin/ash.
shouldn’t it be bash?
1 Like
I have them too:
/usr/lib/initcpio/hooks/dmraid:#!/usr/bin/ash
/usr/lib/initcpio/hooks/plymouth:#!/usr/bin/ash
/usr/lib/initcpio/hooks/grub-btrfs-overlayfs:#!/usr/bin/ash
/usr/lib/initcpio/hooks/consolefont:#!/usr/bin/ash
/usr/lib/initcpio/hooks/encrypt:#!/usr/bin/ash
/usr/lib/initcpio/hooks/keymap:#!/usr/bin/ash
/usr/lib/initcpio/hooks/memdisk:#!/usr/bin/ash
/usr/lib/initcpio/hooks/resume:#!/usr/bin/ash
/usr/lib/initcpio/hooks/shutdown:#!/usr/bin/ash
/usr/lib/initcpio/hooks/sleep:#!/usr/bin/ash
/usr/lib/initcpio/hooks/udev:#!/usr/bin/ash
/usr/lib/initcpio/hooks/usr:#!/usr/bin/ash
/usr/lib/initcpio/hooks/btrfs:#!/usr/bin/ash
/usr/lib/initcpio/init:#!/usr/bin/ash
/usr/lib/initcpio/init_functions:#!/usr/bin/ash
/usr/lib/initcpio/shutdown:#!/usr/bin/ash
But sudo mkinitcpio -P works as expected.
Maybe you have ash (the Almquist Shell) installed? 
No, I checked that already. I didn’t even know, that ash still exists.
1 Like
i don’t have ash installed.
because it’s a new warning i’m wondering if maybe my configuration is changed/broken in an other place. (according to gitlab the hooks have not changed since 2004)
Well, do either of you have dash installed? It’s Debian’s version of ash, and it’s the successor to it. If so, it may have /usr/bin/ash as a symlink for compatibility reasons.
No, from the shells with ash in the name there is only bash.
1 Like
No, nothing.
From what I’ve Googled so far, mkinitcpio uses BusyBox, which includes an ash-compatible shell.
Since I can’t say exactly when the warning started appearing, but everything seems to be working fine for now (except for occasional extreme lags with the 7.01 kernel, but that’s a different issue), I’ll leave it alone for now. Sorry for causing a fuss.
Yes, but only if you’re still using the base and udev hooks in /etc/mkinitcpio.conf, instead of the newer (and now recommended) systemd hooks.
The base hook does indeed use busybox, with an ash emergency shell — with the systemd hooks, the emergency shell is disabled by default for security reasons.
I forgot to mention that I am on unstable and systemd hooks:
HOOKS=(systemd autodetect microcode modconf kms keyboard sd-vconsole block sd-encrypt filesystems fsck)
Well, that makes it even more bizarre, because the systemd hook doesn’t normally include an emergency shell. You normally need the base hook as well for that, even though systemd locks it down by default.
The lock-down can be overridden by way of a boot parameter, though, but then you’d still need the base hook to provide the busybox shell. And given that you don’t have the base hook, I’m stumped.
since start version linux7.0 , no more build for nvidia-580xx-dkms
( 7/14) Install DKMS modules
==> dkms install --no-depmod nvidia/580.142 -k 7.0.0-1-MANJARO
Error! Bad return status for module build on kernel: 7.0.0-1-MANJARO (x86_64)
Consult /var/lib/dkms/nvidia/580.142/build/make.log for more information.
==> WARNING: `dkms install --no-depmod nvidia/580.142 -k 7.0.0-1-MANJARO' exited 10
8/14) Updating linux initcpios...
==> Building image from preset: /etc/mkinitcpio.d/linux70.preset: 'default'
==> Using default configuration file: '/etc/mkinitcpio.conf'
-> -k /boot/vmlinuz-7.0-x86_64 -g /boot/initramfs-7.0-x86_64.img
==> Starting build: '7.0.0-1-MANJARO'
-> Running build hook: [base]
-> Running build hook: [systemd]
-> Running build hook: [autodetect]
-> Running build hook: [microcode]
-> Running build hook: [modconf]
-> Running build hook: [kms]
-> Running build hook: [keyboard]
-> Running build hook: [keymap]
-> Running build hook: [block]
-> Running build hook: [filesystems]
-> Running build hook: [fsck]
==> ERROR: module not found: 'nvidia'
==> ERROR: module not found: 'nvidia_uvm'
==> ERROR: module not found: 'nvidia_drm'
==> Generating module dependencies
==> Creating zstd-compressed initcpio image: '/boot/initramfs-7.0-x86_64.img'
==> WARNING: errors were encountered during the build. The image may not be complete.
erreur : la commande n’a pas pu être exécutée correctementfrom /var/lib/dkms/nvidia/580.142/build/make.log
DKMS (dkms-3.4.0) make.log for nvidia/580.142 for kernel 7.0.0-1-MANJARO (x86_64)
mer. 15 avril 2026 07:19:44 CEST
Building module(s)
# command: 'make' -j12 IGNORE_PREEMPT_RT_PRESENCE=1 NV_EXCLUDE_BUILD_MODULES='' KERNEL_UNAME=7.0.0-1-MANJARO modules
make[1] : on entre dans le répertoire « /usr/lib/modules/7.0.0-1-MANJARO/build »
make[2] : on entre dans le répertoire « /var/lib/dkms/nvidia/580.142/build »
SYMLINK nvidia/nv-kernel.o
SYMLINK nvidia-modeset/nv-modeset-kernel.o
MODPOST Module.symvers
WARNING: modpost: missing MODULE_DESCRIPTION() in nvidia.o
WARNING: modpost: missing MODULE_DESCRIPTION() in nvidia-uvm.o
WARNING: modpost: missing MODULE_DESCRIPTION() in nvidia-modeset.o
WARNING: modpost: missing MODULE_DESCRIPTION() in nvidia-drm.o
CC [M] nvidia.mod.o
CC [M] .module-common.o
CC [M] nvidia-uvm.mod.o
CC [M] nvidia-modeset.mod.o
CC [M] nvidia-drm.mod.o
CC [M] nvidia-peermem.mod.o
LD [M] nvidia-modeset.ko
LD [M] nvidia-peermem.ko
LD [M] nvidia.ko
LD [M] nvidia-uvm.ko
BTF [M] nvidia-peermem.ko
LD [M] nvidia-drm.ko
BTF [M] nvidia-modeset.ko
awk: ligne de commande:1: 'BEGIN
awk: ligne de commande:1: ^ caractère incorrect « ' » dans l'expression
awk: ligne de commande:1: 'BEGIN
awk: ligne de commande:1: ^ caractère incorrect « ' » dans l'expression
make[4]: *** [/usr/lib/modules/7.0.0-1-MANJARO/build/scripts/Makefile.modfinal:60: nvidia-peermem.ko] Error 1
make[4]: *** Suppression du fichier « nvidia-peermem.ko »
make[4]: *** Attente des tâches non terminées....
make[4]: *** [/usr/lib/modules/7.0.0-1-MANJARO/build/scripts/Makefile.modfinal:60: nvidia-modeset.ko] Error 1
make[4]: *** Suppression du fichier « nvidia-modeset.ko »
BTF [M] nvidia-drm.ko
awk: ligne de commande:1: 'BEGIN
awk: ligne de commande:1: ^ caractère incorrect « ' » dans l'expression
make[4]: *** [/usr/lib/modules/7.0.0-1-MANJARO/build/scripts/Makefile.modfinal:60: nvidia-drm.ko] Error 1
make[4]: *** Suppression du fichier « nvidia-drm.ko »
BTF [M] nvidia-uvm.ko
awk: ligne de commande:1: 'BEGIN
awk: ligne de commande:1: ^ caractère incorrect « ' » dans l'expression
make[4]: *** [/usr/lib/modules/7.0.0-1-MANJARO/build/scripts/Makefile.modfinal:60: nvidia-uvm.ko] Error 1
make[4]: *** Suppression du fichier « nvidia-uvm.ko »I had the same issue and could track down the issue to jemalloc 1:5.3.1.
When downgrading to jemallac 1:5.3.0-7 it was working fine again
1 Like
Still need to add the 7.0 patches to 580xx series …
Confirmed on an Arch Linux Xfce VM that jemalloc is indeed, the culprit that prevented pamac GUI from launching. Downgrade to 1:5.3.0-7 fixed the issue.
In the meantime, I had installed octopi/yay and removed pamac-gtk.
Update: I re-installed Pamac, removed jemalloc, and now pamac works as expected.
Jemalloc is not required for anything on this particular machine, but on the Arch VM, jemmalloc is needed for at least bind and reflector-simple. I have no idea why jemalloc was installed on this machine. It has never shown up as an orphan. Explicitly installed, maybe?
pamac-gtk 11.7.4-4 is rebuilt without jemalloc for now. It was added as an alternative memory allocator.
1 Like
nvidia-580xx-dkms / nvidia-580xx-open-dkms 580.142-3 will be along shortly with kernel 7.0 support along with the precompiled linux70-580xx-nvidia / linux70-580xx-nvidia-open 580.142-1 packages.