01:10 <macromorgan> so what all do I need to do to get USB OTG working on the H616? I've correctly identified the usb_id_det pin (when I plug in a host device I can see an interrupt, when I plug in nothing or not a host device I get no interrupt). I also have a working driver for the USB charger that the otg controller can see when there's a voltage or not (usb0_vbus_power-supply). And I've set the dr_mode to OTG.

01:10 <macromorgan> How can I check then if the controller is correctly switching between OTG and Host?

02:52 <ItsKaitlyn03> Hey, it's been awhile! So, I'm starting to think DRAM initialization for A133 is closer to A523

04:03 <jernej> tokyovigilante: having both outputs active at the same time is a bit tricky and to fully utilize that, further changes are needed

04:03 <jernej> I suggest to test with LCD only first

04:04 <jernej> there is no need to have h700 variant of lcd tcon

05:02 <Jookia> ItsKaitlyn03: yeah? what's the news?

05:03 <ItsKaitlyn03> I got it to somehow "recognize" 128 MiB of DRAM, also going to have to modify H616 further, as there is more DRAM parameters

05:03 <Jookia> interesting!

05:03 <Jookia> that's some really cool progress

05:03 <Jookia> DRAM is magic to me :)

05:04 <ItsKaitlyn03> unfortunately I don't think its initialized correctly, but yeah

05:05 <ItsKaitlyn03> all i know is that h616 has some similarities, but it needs more modifications. so I already have decided to branch off from H616 dram code

05:08 <Jookia> yeah that makes sense

06:46 <loki666> tokyovigilante: I'm already using your rg35xx-upstream-extras branch with anbernic_rg35h_defconfig for u-boot

10:09 <Jookia> does dts pinctrl work on the d1 in u-boot?

11:27 <Jookia> i got my pwm driver working. kind of. the number of active cycles seem to 'stick' when updating the register

11:36 <Jookia> it's probably because i'm not using writel or readl but instead clrsetbits. writel seems to cause compiler warning and data abort though. not sure

11:39 <Jookia> yeah, two writes are interfering :(

11:40 <Jookia> i'll fix it tomorrow but otherwise it works fine!

13:01 <ItsKaitlyn03> okay doing more comparisons between h616 there is more differences

13:01 <ItsKaitlyn03> so branching out into my own dram c file was probably a good thing

13:01 <ItsKaitlyn03> PRCM on a133 doesn't hold ohms240 and res_cal_ctrl configuration vals

13:01 <ItsKaitlyn03> instead its in SYS_CFG

13:02 <ItsKaitlyn03> conveniently, the addresses are also in the same spot as in the D1

16:15 <loki666> tokoyvigilante: here is my dmesg log https://pastebin.com/1wdJ8rTq

16:17 <loki666> don't know if it's the issue, but the driver "sun50i-iommu.c" doesn't compile

16:17 <loki666> https://git.sr.ht/~tokyovigilante/linux/tree/anbernic-hdmi/item/drivers/iommu/sun50i-iommu.c#L1007 label "err_free_cache" doesn't exists

16:18 <loki666> should I patch it to use label "err_free_group"

16:29 <loki666> mmh when I plug the hdmi back in, I get

16:29 <loki666> [ 1972.798881] Console: switching to colour frame buffer device 128x48

16:29 <loki666> [ 1972.832268] sun4i-drm display-engine: [drm] fb0: sun4i-drmdrmfb frame buffer device

16:29 <loki666> but no output

17:02 <loki666> according to dts, mixer0 is referencing iommus ...

17:10 <loki666> clearly, something went wrong when applying jernejsk patches see https://github.com/jernejsk/linux-1/blob/h616-full/drivers/iommu/sun50i-iommu.c

18:48 <macromorgan> so it looks like (on the Anbernic H700 based stuff) boost is connected to the OTG port through another regulator. The regulator has 5 pins and one of them is connected to a GPIO. The problem is when I drive the GPIO high the regulator only outputs about half of what is input, so with boost running at full blast I'm getting about 2.6v to the OTG port. I haven't yet figured out what more I need to do to get it outputting 5v

18:48 <macromorgan> to the OTG port (or even what the regulator's part number is... the text is just too damn tiny).

18:49 <macromorgan> if boost is off it looks like the VSYS is what's connected to the mystery regulator

19:35 <loki666> tokyovigilante: bah I'm lost, I think my DTB wasn't properly loaded by u-boot, now it seems to be ok, but I don't have the fb0 log when I plug the hdmi...

19:35 <loki666> https://pastebin.com/0zA08ykt