08:41 <apalos> sjg1: is there a way to define a default sysinfo driver that would run without matching any specific compatible?

08:53 <apalos> because the entire lib/smbios.c assumes that sysinfo is used and it also assumes that special DT entry about chassis etc is there

08:53 <apalos> but the device tree of the majority if the boards lacks that, however we have model, compatible and chassis-type from the DT spec

08:53 <apalos> that we can reuse and fill that information

08:54 <apalos> instead of system, baseboard and chassis that the u-boot,sysinfo-smbios expects

12:18 <Domon> Hi, maybe somebody have experience with broadcom arm soc bootloader?

13:14 <Tartarus> xypron, sjg1: Should I take https://patchwork.ozlabs.org/project/uboot/list/?series=312114 or https://patchwork.ozlabs.org/project/uboot/patch/20220731182707.921722-1-sjg@chromium.org/ or is something else still going to be posted to address the rtc test fails in CI sometimes issue?

13:29 <sjg1> Tartarus: yes I think you should take that series...the GMT issue is not something I plan to take on at present and isn't really necessary

13:29 <Tartarus> sjg1: Thanks, xypron that OK with y ou?

13:29 <Tartarus> sjg1: I'm reviewing the LTO series,btw

13:30 <sjg1> Tartarus: yay!

13:32 <sjg1> apalos: No there isn't but I have been thinking we could create a global-u-boot.dtsi that every board includes. But from the discussion yesterday it seems that a) UEFI is used on these server boards, and b) where U-Boot is used there are only a few of them anyway so we are better to update the DT for each one

19:32 <xypron> Tartarus those RTC patches are good step forward eve if some further refinement remains possible.

19:36 <Tartarus> ok, thanks

20:19 <apalos> sjg1: I think it's the other way around

20:19 <apalos> Boards are starting to run distros, because they got lots of RAM now

20:20 <apalos> but because a small number of boards has the smbios .dtsi stuff, they end up showing "unkwown product"

20:20 <apalos> and lot's of them are defining chassis-type for example from the DT spec

20:21 <apalos> I'll send a hacky RFC at some point next week, hopefully it will explain some stuff