You are not logged in.
Hello pals
Is there some way that I could force pacman install first kernel headers before kernel "itself"?
At this point I should install kernel, kernel-headers, and reinstall kernel to dkms compile it with nvidia driver
Last edited by ilkyest (2014-05-15 02:19:03)
Offline
I've all ready said this in the last 24 hours to someone else: When I asked Allan McCrae regarding the return of the SyncFirst option in /etc/pacman.conf, he said that it will not be happening.
Personally, I think it's the user's responsibility to make sure it's updated anyways. Either that or use a kernel package that includes the headers in the main package.
Last edited by clfarron4 (2014-05-02 10:06:48)
Claire is fine.
Problems? I have dysgraphia, so clear and concise please.
My public GPG key for package signing
My x86_64 package repository
Offline
I'm not sure what do you want to do and why.
Is your issue / idea the same as https://bbs.archlinux.org/viewtopic.php?id=171568 or https://bbs.archlinux.org/viewtopic.php?id=180821 ?
Offline
This is a very clear "Manual intervention required" situation. You run pacman -Syu, you see a kernel upgrade, you stop, you run pacman -S linux-headers, you run pacman -Su. Really, how hard is that?
Offline
I also saw the thread clfarron4's referring to, and I'm a little confused. Can't you just rebuild the kernel image after the update is finished? Neither of the threads I've seen explicitly state that it makes updating impossible, and unless you're blindly running updates you can see there's a kernel (and therefore headers) update available. These threads are all from long-time users too, which just adds to my confusion.
Last edited by ANOKNUSA (2014-05-02 12:59:08)
Offline
This is a very clear "Manual intervention required" situation. You run pacman -Syu, you see a kernel upgrade, you stop, you run pacman -S linux-headers, you run pacman -Su. Really, how hard is that?
I'd say this is a very clear 'write a script to automate this process' situation.
But whether the Constitution really be one thing, or another, this much is certain - that it has either authorized such a government as we have had, or has been powerless to prevent it. In either case, it is unfit to exist.
-Lysander Spooner
Offline
alphaniner,
If you mean the USER should write that script, i agree with you.
Imo the problem is that most posters that complain about the dkms/kernel / kernel-headers issue seem to think the arch devs need to write that script FOR them.
Disliking systemd intensely, but not satisfied with alternatives so focusing on taming systemd.
clean chroot building not flexible enough ?
Try clean chroot manager by graysky
Offline
Oaky pals, I've read about script but, I couldn't understand it, in so less time I've, to it...
But, it's solved... with pacman -Sdd linux-headers-xxxxxx and after pacman -S linux-xxxxxx it makes dmks works, without headers fault, and everything it's okay
I compile, and would compile yet, linux-ck, linux-ice, linux-lqx, and linux-uksm-ck... both of them I need headers cause nvidia driver, that dmks/hook compiles it....
At first time I should install image, headers, and after, reinstall image. With -Sdd, just headers, and after, image.....
thanks.
Offline
When I asked Allan McCrae regarding the return of the SyncFirst option in /etc/pacman.conf, he said that it will not be happening.
Hmmmm. Maybe one could easily do this with the future pacman hooks? A "SyncFirst"-like hook wouldn't do any harm...but I'm not sure about the difficulties of maintaining this.
Last edited by thiagowfx (2014-05-03 18:53:52)
Offline
clfarron4 wrote:When I asked Allan McCrae regarding the return of the SyncFirst option in /etc/pacman.conf, he said that it will not be happening.
Hmmmm. Maybe one could easily do this with the future pacman hooks? A "SyncFirst"-like hook wouldn't do any harm...but I'm not sure about the difficulties of maintaining this.
I think the dificulty is "analyse" each package to see what needs "be before" another
Offline