Hi Diederik, Great news! I'm glad I did it correctly.
On 10/20/22 11:32 AM, Diederik de Haas wrote: > Hi Dan, > > On Thursday, 20 October 2022 17:26:42 CEST Dan Coleman wrote: >> Yeah, I tried to apply it, but I had that same problem where I managed to >> compile the .debs, but got a bunch of errors installing them with a black >> screen at the end of the test. I use DKMS for nvidia and openrazer, and >> that's been posing problems, which you'll see below. In addition, there are >> dependency issues that dpkg and apt both complain about. > I see that you, Salvatore and the upstream maintainer are in discussion in the > upstream repo, so I'll leave that part in their capable hands :-) > > As for your build+install attempts, it looks like you did everything correct! > > Apparently "test-patches" does not build a 'headers-common' package and > therefor installing the headers failed which in turn made DKMS fail. That seems like less-than-ideal behavior for test-patches. Is that a bug that would benefit from being reported? > I understand that you'd want that for a complete functioning system, but for > testing this particular issue/patch it is sufficient. Once the proper fix is > validated, all the kernel (related) packages will be rebuild and then the > other things like DKMS should work again too. > Unfortunately it turned out that just this patch wasn't sufficient to fix the > issue as Alex Deucher indicated. How to proceed with that, I'll leave that up > to Alex/Salvatore :-) > > You also removed the normal (i.e. Debian's) kernel meta package (linux-image- > amd64) which prevented the installation of your patched kernel. After that the > installation of your patched kernel succeeded. And you rebooted into that. > Earlier you indicated you had doubts whether you did it correctly and therefor > whether you were actually testing the patch and I'm quite sure you did (at > least this time based on the output you shared). > > HTH, > Diederik
