[10:31] <xnox> LocutusOfBorg:  I think it is time to simply drop virtualbox-guest-dkms and virtualbox-guest-source from the virtualbox & virtualbox-hwe packages. Since all of the kernel modules they provide, are available from the vanilla kernels, and have been for a long time. I.e. debian testing, ubuntu focal have them. Thus the ones from the virtualbox dkms package are not used.
[10:31] <xnox> i.e. something like: https://launchpadlibrarian.net/544986016/virtualbox_6.1.22-dfsg-1_6.1.22-dfsg-1ubuntu1.diff.gz which has been test built in my PPA
[10:32] <xnox> LocutusOfBorg:  would you be unboard with that? is there any value to still have dkms modules from the virtualbox package?
[10:32] <ogra> "unboard" ... thats like schroedingers onboard ?
[10:35] <xnox> LocutusOfBorg:  i.e. they have been available since v4.15 so Debian stable & Ubuntu bionic and up.
=== cking is now known as cking2
=== cking2 is now known as cking
=== doko_ is now known as doko
[12:12] <LocutusOfBorg> xnox, this is something I thought already lots of time
[12:12] <LocutusOfBorg> would be nice to do it, if we are sure the vanilla version is stable enough, something that didn't happen in the past
[12:12] <LocutusOfBorg> with Debian providing them, and Ubuntu too, this is a rational choice
[12:13] <xnox> LocutusOfBorg:  then i shall send the patch to debian bts for the maintainer to consider. vanilla version is more stable as far as i can tell. At least in the recent years.
[12:13] <LocutusOfBorg> will also probably allow us to drop virtualbox-hwe package
[12:13] <LocutusOfBorg> xnox, no need to open a bug, I can just apply the patch on git and upload
[12:13] <xnox> and the vanilla one is the one we have been using in all ubuntu provided virtualbox images
[12:14] <LocutusOfBorg> but we might consider keeping the binaries around until recent kernels are available on all lts
[12:14] <xnox> (vgrant boxes, etc)
[12:14] <LocutusOfBorg> this is good to hear
[12:14] <LocutusOfBorg> no need for guest-dkms at all then?
[12:14] <xnox> correct, not even in bionic.
[12:14] <LocutusOfBorg> I admit I have received feedbacks of resize screen issues that were fixed with guest-dkms tools
[12:14] <LocutusOfBorg> this is why I kept them around
[12:15] <xnox> imho we should drop guest-sources & guest-dkms from both virtualbox and virtualbox-hwe
[12:15] <xnox> we need the guest-tools package
[12:15] <xnox> it just needs not to depend on the guest-dkms anymore, as the kernel modules are implied.
[12:15] <xnox> and if i am wrong, i will eat my own words and fix all the things =)
[12:18] <LocutusOfBorg> virtualbox-hwe exists *just* because of guest tools
[12:18] <LocutusOfBorg> so virtualbox-hwe stuff can be just provided by main src:virtualbox for a couple of years, and then dropped
[12:19] <LocutusOfBorg> mmm virtualbox-guest-x11 still need to be around :/
[12:19] <LocutusOfBorg> sad
[12:19] <LocutusOfBorg> this is the main reason for virtualbox-hwe existance
[12:19] <LocutusOfBorg> because of newer xorg and runtime deps
[12:21] <xnox> LocutusOfBorg:  yeah, i think we still need to ship all the new userspace bits and bobs like guest-x11 etc.
[12:22] <LocutusOfBorg> :/
[12:22] <xnox> LocutusOfBorg:  so we are only dropping guest-dkms guest-source, but otherwise we are keeping src:virtualbox src:virtualbox-hwe etc.
[12:23] <LocutusOfBorg> sadly to know kernel folks can't provide udev rules for the guest-x11 startup things
[12:23] <LocutusOfBorg> :(
[12:23] <LocutusOfBorg> also Debian has such issue
=== sem2peie- is now known as sem2peie
[12:37] <xnox> src:linux is providing udev rules for a few things already.
[12:38] <xnox> for example see https://packages.debian.org/unstable/hyperv-daemons shipped by src:linux in debian
[12:39] <xnox> it has init.d scripts, systemd service, daemons, and a udev rule. They do simple small things, whenever special hyperv device is plugged in.
[12:39] <xnox> in ubuntu src:linux also ships it, but it is called different.
[14:38] <xnox> LocutusOfBorg:  i think we can kill udev rules by patching kernel driver to do that.
[14:38] <xnox> LocutusOfBorg:  but not other user space utils.
[14:57] <seb128> bdmurray, hey, could you review https://code.launchpad.net/~ubuntu-core-dev/update-notifier/+git/update-notifier/+merge/404404 when you have some free cycles?
[15:00] <bdmurray> seb128: I'm out this afternoon and tomorrow so may not get to it until next week
[15:00] <seb128> bdmurray, no hurry, next weeks is fine, enjoy the long weekend!
=== bdmurray_ is now known as bdmurray
[15:59] <laney> laney@raleigh> syncpackage -d experimental bijiben
[15:59] <laney> The source package 'bijiben' status Superseded not in (Pending,Published) in experimental for pockets Release, Security, Updates, Proposed with status in Pending, Published
[15:59] <laney> what a message
=== georgiag_ is now known as georgiag
=== tjaalton_ is now known as tjaalton
[21:58] <jawn-smith> doko: fwupd has a flashrom plugin that is enabled in Debian, but not enabled in Ubuntu because flashrom is in universe. In order to get fwupd back in sync with Debian we will need flashrom in main, and to make it a runtime dependency of fwupd
[22:00] <sarnold> jawn-smith: https://bugs.launchpad.net/ubuntu/+source/flashrom/+bug/1912371
[22:00] <ubottu> Launchpad bug 1912371 in libftdi1 (Ubuntu) "[MIR] flashrom + libftdi" [Undecided, New]
[22:00] <sarnold> oh heh, you're on that bug :) sorry