07:53 <memeka> mszyprow: ping

07:53 <mszyprow> memeka: pong

07:53 <memeka> Have you seen the patch I sent you yesterday?

07:54 <memeka> On HYP mode?

07:54 <mszyprow> memeka: nope...

07:55 <memeka> 8:03 AM <memeka> mszyprow|home: I saw in the tizen exynos list of todos that fixing the “inconsistent cpu state” is one of the things

07:55 <memeka> 8:03 AM <memeka> https://github.com/mihailescu2m/linux/commit/4aa48f68a6b6166049806b247efee6bbd84be4b5 - this revert fixes it, if this helps :)

07:56 <mszyprow> memeka: well, I know about that revert, but the goal is to have it fixed in mainline

07:57 <memeka> Ok

07:57 <mszyprow> memeka: and frankly I didn't check that issue yet

07:57 <memeka> Do you also remember the issue with 2 planes?

07:57 <memeka> Changing second plane to overlay => crashes?

07:58 <memeka> Your patch I think added 2 lines I think?

07:58 <memeka> The pointer was getting out of buffer space .. ?

07:59 <memeka> mszyprow: at the time I was running 4.18 I think, and now in 5.0 still happens

08:00 <mszyprow> memeka: there have been no changes related to plane switching in mixer driver, so it is not a surprise that the issue is still there

08:01 <memeka> From what I heard, it doesn’t happen on Hardkernel’s 4.14 kernel, although I think there are some (unrelated to this issue) drm commits on top of mainline 4.14

08:02 <memeka> But from what I remember, I didn’t test 4.14 properly because planes were also rendered incorrectly (e.g one plane was black)

08:02 <memeka> Anw, something to keep in mind if you need this info :)

08:04 <mszyprow> memeka: well, thanks for that info, although I don't remember any significant patches for mixer driver in hardkernel's branch...

08:06 <memeka> Yeah I think is just changes either in drm framework or exynos drm between 4.14-4.18

09:43 <mszyprow> memeka: I've reproduced vblank issue on Hardkernel's 4.14 :/

10:32 <memeka> mszyprow: ok that settles that... is it possible it’s a hw issue?

10:33 <mszyprow> memeka: I will not call it hw issue, it is rather a 'feature' ;P

10:33 <mszyprow> memeka: the driver does something in a wrong way

10:33 <mszyprow> memeka: the issue is on all exynos5 socs

10:34 <mszyprow> memeka: and exynos5 introduced so called 'hdmi stereoscopic' mode (not even supported in the current drivers)

10:35 <mszyprow> memeka: which requires something more to be done during page flip of 2 planes at once

10:35 <memeka> Which is not done?

10:35 <mszyprow> memeka: the same test code work fine on exynos4 without any issue, but that chip has no stereoscopic mode

10:35 <mszyprow> yep

10:36 <mszyprow> too bad the documentation is really not clear about how to handle that

10:36 <memeka> Maybe there will be a fix soon :P

10:36 <mszyprow> is a mix of exynos4 docs with just a few comments

10:36 <memeka> :(

10:36 <mszyprow> in a few places conflicting even... :/

10:37 <memeka> SA

10:37 <memeka> Same as the gscaler - mixer path :(

10:39 <memeka> Just got a mali binary with dmabuf import

10:49 <memeka> So the overlay plane is not as much critical as before :)

13:07 <PabloPL> mszyprow: ping

13:09 <PabloPL> mszyprow: Do You know any other devices using s6e63m0 lcd panel? I've wrote a panel driver for it (based on old video driver), but need's to know what timing should be used (not sure if they're identical across all devices)

13:15 <mszyprow> PabloPL: you can get some timigs from that ancient commit: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=3965fdee7723d01e97861cabf13c5cb8824cb4e6

13:15 <mszyprow> PabloPL: now I don't have access to Goni board to check with current kernel and your driver

13:16 <PabloPL> mszyprow: Thanks, i hope that they're the identical to one which i got for galaxys

13:34 <jackmitchell> Hi, I'm trying to debug an issue in the dwc2 otg driver, does anyone have an xu3 running a recent mainline uboot who would be willing to enable debug and capture the output of a test

14:02 <PabloPL> mszyprow: quick look shows that they're different (timings) than those used on galaxys :/. Difference is in vback-porch, refreshrate (didn't check clock yet)

14:03 <mszyprow> PabloPL: frankly, this doesn't mean anything

14:03 <mszyprow> PabloPL: I remember there way an issue with refresh rate

14:03 <mszyprow> jackmitchell: xu3 doesn't have dwc2 hw

14:11 <jackmitchell> mszyprow: ah my mistake, seems there was some confusion about this on the mailing list

14:13 <mszyprow> jackmitchell: dwc2 is only available in odroid x2/u3

18:51 <PabloPL> ahajda: Hi are You there?