For the fallback I use command:
booster build --universal --force /boot/booster-linux-fallback.img
I haven't checked the docs on this but it seems to work similar when I
used it.
artist
On 8/19/25 07:10, Javier wrote:
> On 2025-08-18 00:56 PM, artist wrote:
>> Another advantage of booster is that it is a lot faster.
>>
>> artist
>>
>> On 8/18/25 20:29, Javier wrote:
>>> On 2025-08-18 00:13 PM, artist wrote:
>>>> To preserve space you might also have a look at switching from
>>>> mkinitcpio too booster.
>>>> From a test box:
>>>> % ls -lh /boot/*.img
>>>> -rw-r--r-- 1 root root 24M Aug 8 14:34
>>>> /boot/booster-linux-fallback.img
>>>> -rw-r--r-- 1 root root 25M Jul 27 17:13
>>>> /boot/booster-linux-lts-fallback.img
>>>> -rw-r--r-- 1 root root 4.8M Jul 27 17:13 /boot/booster-linux-lts.img
>>>> -rw-r--r-- 1 root root 6.0M Aug 8 14:34 /boot/booster-linux.img
>>>> -rw------- 1 root root 48M Aug 8 14:34
>>>> /boot/initramfs-linux-fallback.img
>>>> -rw------- 1 root root 49M Aug 8 14:34
>>>> /boot/initramfs-linux-lts-fallback.img
>>>> -rw------- 1 root root 16M Aug 8 14:33 /boot/initramfs-linux-lts.img
>>>> -rw------- 1 root root 16M Aug 8 14:34 /boot/initramfs-linux.img
>>>>
>>>> artist
>>>
>>> Thanks @artist, btw, by changing:
>>>
>>>> #COMPRESSION_OPTIONS=()
>>>
>>> to:
>>>
>>>> COMPRESSION_OPTIONS=(--long --ultra -22)
>>>
>>> and by changing:
>>>
>>>> #MODULES_DECOMPRESS="no"
>>>
>>> to:
>>>
>>>> MODULES_DECOMPRESS="yes"
>>>
>>> doesn't really gain much, but at least gets me out of the 90+% usage:
>>>
>>>> % df -h /boot/
>>>> Filesystem Size Used Avail Use% Mounted on
>>>> /dev/nvme0n1p2 290M 216M 55M 80% /boot
>>>
>>> And now:
>>>
>>>> % du -hs /boot/*
>>>> 14M /boot/grub
>>>> 902K /boot/grub-common
>>>> 159M /boot/initramfs-linux-fallback.img
>>>> 14M /boot/initramfs-linux.img
>>>> 13M /boot/intel-ucode.img
>>>> du: cannot read directory '/boot/lost+found': Permission denied
>>>> 12K /boot/lost+found
>>>> 154K /boot/memtest86+
>>>> 16M /boot/vmlinuz-linux
>>>
>>> The initrd backup is still big... So perhaps time to look for
>>> booster or dracut, whatever gives smaller images, :) I see the
>>> linux image at a reasonable size, 16M Roy, not sure why you
>>> mentioned 67M, perhaps you're using something different than the
>>> "linux" package?
>>>
>>>> % uname -a
>>>> Linux m1 6.16.1-artix1-1 #1 SMP PREEMPT_DYNAMIC Fri, 15 Aug 2025
>>>> 23:14:52 +0000 x86_64 GNU/Linux
>>>
>>>
>>> Thanks a lot @artis !
>
> Besides my prior fallback question (no idea how you got it with
> booster) I can report that it's impressively smaller, and it generates
> in no time:
>
>> % ls -l /boot/*.img
>> -rw-r--r-- 1 root root 6.2M Aug 18 21:02 /boot/booster-linux.img
>> -rw------- 1 root root 159M Aug 18 19:02
>> /boot/initramfs-linux-fallback.img
>> -rw------- 1 root root 14M Aug 18 18:59 /boot/initramfs-linux.img
>> -rw-r--r-- 1 root root 13M Aug 12 15:24 /boot/intel-ucode.img
>
> I'm a bit hesitant given the lack of fallback. Actually mkinicpio is
> bad given the fallback. Without it (it can be removed from the
> mkinitcpio presets) the default image more then doubles the booster
> one (14M vs. 6.2M) but it's still reasonable, the current fallback
> size is not.
>
> Unfortunately grub-mkconfig pushes the default mkinitcpio preset image
> as the item not requiring a sub-menu (I guess that can be changed on
> /etc/grub.d/10_linux). For now I can just copy the one created under
> the submenu out of it...
>
> Now I found a problem, as I'm using a "la-latin1" keyboard I need the
> booster "vconsole" flag set to true to make my life a bit easier given
> special chars, however I can't include that flag since if I do when
> loading the image I just get a note to hit "enter" to reboot, :( I
> guess I'm hitting this:
>
> https://github.com/anatol/booster/issues/234 (vconsole: true blocks
> booting w/ booster exit status 71)
>
> I use terminus on console/tty (I used it on terminal emulator before
> until finding ttf-iosevkaterm-nerd)... Not that I couldn't live
> without vconsole, but it's sort of handy to refer to the keyboard on
> its own layout, :)
>
> But I'm happy to see the size decrease, and also how fast it is...
>
More information about the artix-general
mailing list