[12:51] <tseliot> ricotz: hey, just FYI, I've just uploaded a new libvdpau, which I merged from debian
[18:24] <alkisg> tjaalton: hello! Regarding LP: #1815172, all schools close for the summer, so I only have the i915 PC in my office to test with, but I'm not sure if it's affected or not (I was testing with 64bit at the time and it wasn't):
[18:24] <alkisg> 00:02.0 VGA compatible controller [0300]: Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor Integrated Graphics Controller [8086:0412] (rev 06)
[18:24] <alkisg> 	Subsystem: Gigabyte Technology Co., Ltd Xeon E3-1200 v3/4th Gen Core Processor Integrated Graphics Controller [1458:d000]
[18:24] <alkisg> Would it make sense to test there?
[18:26] <tjaalton> alkisg: I was able to repro after all
[18:26] <tjaalton> am writing a comment now
[18:26] <alkisg> Ah great, thank you
[18:28] <tjaalton> was kinda hard to find hw for it.. had to install to an external ssd and use a t470s, everything else I tried before that was too old
[18:28] <tjaalton> or didn't support legacy boot at all (hello intel)
[18:29] <tjaalton> but, I missed the bionic kernel respin by some hours
[18:29] <alkisg> I arranged to have remote access in many school just for things like these, where bugs only happen in certain hardware... but I'd have to wait for September when they open again :/
[18:32] <tjaalton> no worries
[18:32] <tjaalton> I should have a setup for this :P
[18:33] <tjaalton> an external disk is handy
[18:34] <tjaalton> the internal one was partitioned for gpt and whatnot, couldn't install a bootloader on it, and booting via the installer ended up in initramfs
[18:39] <tjaalton> still, I don't get it why folks run 32bit on skylake and up :P
[18:40] <tjaalton> at least that will be a thing in the past come next year..
[18:45] <alkisg> (09:39:36 PM) tjaalton: still, I don't get it why folks run 32bit on skylake and up :P ==> that's my "fault" for developing ltsp, and allowing a single installation to be used as a template to netboot any number of clients
[18:45] <alkisg> So since the clients are 32bit, the server needed a 32bit installation
[18:45] <alkisg> The plus side is that it's so easy, that 1000+ schools managed to do it; the negative side... you know ^ :)
[18:46] <alkisg> So now I'm implementing an easy way to offer both 64bit (using the server as a template) and 32 bit using a virtualbox vm as the template
[18:47] <alkisg> It's possible to install grub-pc (bios) in gpt, but it requires a special grub-boot partition
[18:47] <alkisg> I have my own PCs like that, bootable in both uefi and bios modes, for easy testing
[18:49] <tjaalton> ah
[18:50] <tjaalton> I need to test that
[18:50] <tjaalton> the installer did complain about that
[18:55] <alkisg> https://termbin.com/xomb => these are the grub packages needed to have a dual efi/bios boot
[18:55] <alkisg> grub-pc conflicts, so I remove it, but keep grub-pc-bin
[18:56] <alkisg> Other than that, it makes no difference if you start with uefi install or bios install, the end result is the same
[18:57] <tjaalton> ok, cool
[18:57] <alkisg> /dev/sda1 : start=        2048, size=     1048576, type=C12A7328-F81F-11D2-BA4B-00A0C93EC93B, uuid=5BA773EE-D197-4237-9BCB-3FF0AB9B4522, name="Efi"
[18:57] <alkisg> /dev/sda2 : start=     1050624, size=        2048, type=21686148-6449-6E6F-744E-656564454649, uuid=568C6765-64E6-4240-A66C-D2FB3C17D992, name="Grub"
[18:57] <alkisg> And these are the boot partitions involved, i.e. efi and unformatted grub-boot
[19:12] <tjaalton> got it