[RESOLVED] gcc9 / gcc10 version and partprobe problem
barturblits
Status: New User - Welcome
Joined: 13 Sep 2020
Posts: 2
Reply Quote
Hi there,

Lately the default gcc version was update to gcc10 in debian testing, the liquorix kernel still uses gcc9. Nvidia dkms fails at making the module (vbox dkms works properly btw). I have to manually rebuild it with each kernel update with export IGNORE_CC_MISMATCH=1. Any reason why gcc9 is still used?

Another bug which I encountered since the 5.8 release is that partprobe gives an error:

Partition(s) 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64 on /dev/sdc have been written, but we have been unable to inform the kernel of the change, probably because it/they are in use. As a result, the old partition(s) will remain in use. You should reboot now before making further changes.

I thought it was 5.8 related, 5.7.0-19.1-liquorix-amd64 and the debian 5.7 release did not give me those errors. But today 5.8 came in with testing and it also does not give this error. Seems to be specific for 5.8 liquorix.
Back to top
damentz
Status: Assistant
Joined: 09 Sep 2008
Posts: 1142
Reply Quote
Sorry about that, I pushed out a new build that should be using latest gcc. The last build used a docker image cache that I incrementally update, but sometimes it won't take the meta package updates. Remaking the cache with a Dockerfile fixes it but is expensive bandwidth and time-wise.

As for the partprobe issues, I'm not sure why you're getting that. When I run partprobe myself, I get similar output, but it looks like something you can get otherwise on RHEL 8 based systems: www.linuxquestions.org/questions/red-hat-31/rhel8-udev-and-partprob-error-partition-s-1-2-3-dev-sdc1-have-been-written-but-we-have-been-unable-to-inform-the-kernel-of-the-change-4175663204/

All I can say is that it's probably not an issue, more likely it's a bug in how partprobe interacts with the kernel.
Back to top
damentz
Status: Assistant
Joined: 09 Sep 2008
Posts: 1142
Reply Quote
Ok and some good news, it looks like this was reported for upstream Arch Linux:

FS#67891 - [linux] mdadm and gparted broken by kernel change - bugs.archlinux.org/task/67891

I applied the fix to upstream zen-kernel [1] and new builds for Liquorix are on the way. You should get an update in the next hour or two that solves the partprobe regression.

[1] github.com/zen-kernel/zen-kernel/commit/460a32f85
Back to top
barturblits
Status: New User - Welcome
Joined: 13 Sep 2020
Posts: 2
Reply Quote
That was fast, thanks!
Works properly now, both errors are gone.
Back to top
Display posts from previous:   

All times are GMT - 8 Hours