/srv/irclogs.ubuntu.com/2020/07/08/#ubuntu-kernel.txt

=== Eighth_Doctor is now known as Conan_Kudo
=== Conan_Kudo is now known as Eighth_Doctor
lynxisHow do you release kernel versions? Is there a written process about it? I've a regression in 16.04/xenial which was introduced in the upstream stable 4.4.x, but also fixed later in upstream and wonder how long I've to wait to be settled down into linux-image-4.4.16:15
sarnoldlynxis: the dates for the sru cycle are on https://kernel.ubuntu.com/ and you can see the progress per package on https://kernel.ubuntu.com/sru/dashboards/web/kernel-stable-board.html18:51
=== ben_r_ is now known as ben_r
mathnewbhi. i am new to kernel development. i am compiling focal using the master branch here: git.launchpad.net/~ubuntu-kernel/ubuntu/+source/linux/+git/focal and i notice that the master branch is at version 5.4.41, while a fresh install of the focal release image is at version 5.4.44. Why the discrepancy?22:34
mathnewbOr am I reading it wrong? Sorry. I am learning.22:35
cascardomathnewb: the master branch is outdated, possibly some tooling or process problem. but don't confuse corresponding upstream stable releases and our ABI versioning23:14
cascardoso, we release a package that is versioned like 5.4.0-40.4423:14
cascardothe important bit here is 5.4.0 and -40, ignore the .44 after -40, which we call upload number and increase for every upload, trying to couple with the ABI, which is -4023:15
cascardounfortunately, to add some confusion, that kernel has mostly all patches that are included in upstream 5.4.4423:16
cascardothe version that is included in the ISO would be 5.4.0-26, with upstream patches up to 5.4.3023:16
cascardomathnewb: does it make sense?23:16
mathnewbcascardo: I was just eating some dinner. Thank you for writing back. I will read it now.23:36
mathnewbOk. I think it makes sense, yes. In my VM (which I updated to latest after installing) a `uname -r` returns `5.4.0-40-generic`. the master branch is outdated, and building it gives me linux-*-5.4.0-39-generic*, which I believe squares with your explanation (master is slightly out of date).  23:41
mathnewband if I am understanding correctly, there is no exact "1-1" correspondence between an upstream ("mainline"?) kernel and an ubuntu release23:43
mathnewbthe upload number just happens to coincide at the moment23:43
mathnewb(which tricked me into thinking there was a relation there that doesn't exist)23:44
mathnewbDoes that seem correct?23:44
sarnoldit's certainly confused me in the past :)23:44
mathnewbHaha, I'm glad I am not the only one!23:47

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!