05:57 <loki666> I was thinking, I don't think bisecting will help me, as the order of the patch matters, let say I start with mailine 6.10rc3 which is a good commit, and I know that the top of the branch is bad

05:58 <loki666> if the last patch set is cpufreq dvfs pathches, bisecting will tell me that cpufreq is the issue

05:58 <loki666> but apply cpufreq dvfs first is good

06:28 <tokyovigilante> loki666: can I just confirm if you rebase my display and GPU patches on top of 6.10rc3 they’re ok?

06:29 <tokyovigilante> or is that not working? as before probably not huge value in trying to get my trees that are WIP working, better to focus on testing what is coming through the ML. I’m back from holiday and going to send a v2 of the display patches, probably rebased on apritzel’s IOMMU v2

06:30 <tokyovigilante> I haven’t had a chance to test -rc3 however, only -rc2

06:35 <loki666> well your branches are pretty stable (if I disable latest opp speeds), I have quake running right now

06:36 <loki666> I'll try your v2 patches, when published

06:44 <loki666> it's missing the sounds, and analogue sticks dont' seems to works, but it's close to being usable for a batocera build

06:56 <Raqbit> tokyovigilante: You know what, you're right! I discarded that idea as I only saw it a few weeks after last opening it, but I guess I just never realized. *facepalm* Thanks!

07:01 <tokyovigilante> whew, so the battery is ok?

07:02 <tokyovigilante> in theory it should be safe to plug a PD charger in, the device has to request anything above USB2 charging (as I understand it) but that relies on both the device and charger supporting the various standards correctly.

07:05 <tokyovigilante> ah yeah, a quick read suggests the 2A -> 3A current increase between USB PD v1 and v2/3 may be an issue. but again current is pulled not pushed so maybe bot

07:05 <tokyovigilante> *not

07:05 <Raqbit> This could be a good excuse to get one of those PD debug display cables

08:46 <tokyovigilante> are they a thing?

09:32 <apritzel> I think that the battery is supplied (and drained) through the AXP chip only, so I guess any funny input would be filtered out by its circuitry?

09:33 <apritzel> the AXP PMIC does a lot about the battery charging on its own (monitoring voltage and temperature), though it can be *tuned* via software

11:43 <Raqbit> tokyovigilante: Yeah, you can get little adapters that are C-C but have a little display showing voltage, current draw and PD negotiation details.

23:30 <tokyovigilante> I'm preparing a V2 of the DE patchset, and confess I absolutely do not understand Conor's review of the v1 DT bindings, do I just need to bite the bullet and actually learn YAML, or os there a reference beyond the kernel.org DT binding docs?

23:39 <apritzel> I find it often more helpful to browse through existing examples

23:42 <apritzel> tokyovigilante: what's the problem, exactly? Is it about the first hunk in RFC patch 1/8?

23:45 <apritzel> and you should indeed make the clock driver in 8/8 a separate patch, but I guess that's not a real problem?