narmstrong changed the topic of #linux-amlogic to: Amlogic mainline kernel development discussion - our wiki http://linux-meson.com/ - ml linux-amlogic@lists.infradead.org - official channel moved from Freenode - publicly logged on https://libera.irclog.whitequark.org/linux-amlogic
vagrantc has quit [Quit: leaving]
Daanct12 has joined #linux-amlogic
Daanct12 has quit [Client Quit]
kilobyte_ch has quit [Ping timeout: 244 seconds]
kilobyte_ch has joined #linux-amlogic
<steev> Cheaterman: oh, thanks for that patch, gonna try it here with my waveshares which seem to be the only displays i have with broken edids
<steev> i used to have a SOYO monitor when i was at Genesi that we kept around specifically because the edid was so crazy and incorrect
<steev> and yeah, gentoo isn't nearly as insane or crazy as it used to be. i definitely miss it
buzzmarshall has quit [Quit: Konversation terminated!]
<steev> Cheaterman: also have you considered submitting it upstream so that everyone gets it, not just people using tobetter's fork? :D
tortoisedoc has quit [Ping timeout: 252 seconds]
Daanct12 has joined #linux-amlogic
tortoisedoc has joined #linux-amlogic
Ballerburg9005 has joined #linux-amlogic
tortoisedoc has quit [Ping timeout: 252 seconds]
Daanct12 has quit [Remote host closed the connection]
<steev> hm, doesn't seem to help here, at least, not with the 1080x1920 version
<steev> or i screwed up the patch
<steev> hm, okay so with Cheaterman's patch, and my patch to lower the bounds of what is a bad mode, the 400x1280 waveshare works, though its top pixels are shifted down about 10 or so
<steev> now to figure out why the 1080x1920 doesn't
<steev> oh
<steev> i bet
<steev> the upper vertical size is 1600 (with the 2K display support patch), so make that 1920, since the waveshare 5.5" amoled is 1080x1920...
<steev> Now just this minor issue of the jiggle? Not sure how to describe it https://usercontent.irccloud-cdn.com/file/sXNpGQNF/IMG_0094.MP4
<steev> oh, and also the fact that it has the wrong margins somehow
Ballerburg9005 has quit [Ping timeout: 244 seconds]
camus has quit [Ping timeout: 268 seconds]
camus has joined #linux-amlogic
<Cheaterman> steev: It's such a trivial patch and the issue has been around for so long that I'm very surprised it wouldn't be in patchwork somewhere
<Cheaterman> The jiggles you're getting are pretty weird ; does the _exact same_ mode work on the same screen when driven by other devices?
tortoisedoc has joined #linux-amlogic
<steev> As far as I know, yes - I can plug it in to a windows box with no issue, likewise the thinkpad x13s via a usb-c to hdmi adapter, additionally, on the others, there’s no issue with the margins
tortoisedoc has quit [Ping timeout: 252 seconds]
<xdarklight> steev: RE "the jiggle" - I had some screen tearing issue on Meson8b (32-bit SoC, so much older than what you're using). there the fix was https://github.com/xdarklight/linux/commit/67ba46ecad0e89084a83ae36432cad2f9961f0d7 (not upstream yet)
<xdarklight> not saying that it's the fix for your issue - but maybe it can give you a starting point
gab has quit [Ping timeout: 255 seconds]
GNUtoo has quit [Ping timeout: 258 seconds]
tortoisedoc has joined #linux-amlogic
nashpa has joined #linux-amlogic
GNUtoo has joined #linux-amlogic
kilobyte_ch has quit [*.net *.split]
repk_ has quit [*.net *.split]
adema has quit [*.net *.split]
dliviu has quit [*.net *.split]
GNUtoo has quit [Remote host closed the connection]
naoki has quit [*.net *.split]
camus has quit [*.net *.split]
warpme___ has quit [*.net *.split]
Consolatis has quit [*.net *.split]
BlueMatt[m] has quit [*.net *.split]
jelly has quit [*.net *.split]
nashpa has quit [*.net *.split]
BlueMatt has quit [*.net *.split]
Terry13732293 has quit [*.net *.split]
montjoie has quit [*.net *.split]
fiii has quit [*.net *.split]
orkid has quit [*.net *.split]
steev has quit [*.net *.split]
Pinchiukas_ has quit [*.net *.split]
rtp has quit [*.net *.split]
mrec has quit [*.net *.split]
ldevulder has quit [*.net *.split]
zkrx has quit [*.net *.split]
dlan has quit [*.net *.split]
cyrozap has quit [*.net *.split]
elastic_dog has quit [*.net *.split]
psydroid has quit [*.net *.split]
xdarklight has quit [*.net *.split]
Tartarus has quit [*.net *.split]
ccaione has quit [*.net *.split]
dagmcr has quit [*.net *.split]
mturquette has quit [*.net *.split]
khilman has quit [*.net *.split]
narmstrong has quit [*.net *.split]
mkorpershoek has quit [*.net *.split]
shoragan has quit [*.net *.split]
gabes has quit [*.net *.split]
cottsay has quit [*.net *.split]
filt3r has quit [*.net *.split]
doppo has quit [*.net *.split]
Danct12 has quit [*.net *.split]
wens has quit [*.net *.split]
Stricted- has quit [*.net *.split]
ndufresne has quit [*.net *.split]
tomeu has quit [*.net *.split]
kbingham has quit [*.net *.split]
tortoisedoc has quit [*.net *.split]
ChanServ has quit [*.net *.split]
montjoie has joined #linux-amlogic
gab has joined #linux-amlogic
nashpa has joined #linux-amlogic
tortoisedoc has joined #linux-amlogic
camus has joined #linux-amlogic
naoki has joined #linux-amlogic
elastic_dog has joined #linux-amlogic
ldevulder has joined #linux-amlogic
Danct12 has joined #linux-amlogic
warpme___ has joined #linux-amlogic
Consolatis has joined #linux-amlogic
doppo has joined #linux-amlogic
zkrx has joined #linux-amlogic
psydroid has joined #linux-amlogic
Tartarus has joined #linux-amlogic
BlueMatt[m] has joined #linux-amlogic
mrec has joined #linux-amlogic
tomeu has joined #linux-amlogic
kbingham has joined #linux-amlogic
ChanServ has joined #linux-amlogic
Stricted- has joined #linux-amlogic
wens has joined #linux-amlogic
BlueMatt has joined #linux-amlogic
jelly has joined #linux-amlogic
khilman has joined #linux-amlogic
xdarklight has joined #linux-amlogic
cyrozap has joined #linux-amlogic
dlan has joined #linux-amlogic
steev has joined #linux-amlogic
rtp has joined #linux-amlogic
Pinchiukas_ has joined #linux-amlogic
orkid has joined #linux-amlogic
fiii has joined #linux-amlogic
mturquette has joined #linux-amlogic
dagmcr has joined #linux-amlogic
narmstrong has joined #linux-amlogic
ccaione has joined #linux-amlogic
mkorpershoek has joined #linux-amlogic
Terry13732293 has joined #linux-amlogic
shoragan has joined #linux-amlogic
gabes has joined #linux-amlogic
filt3r has joined #linux-amlogic
ndufresne has joined #linux-amlogic
cottsay has joined #linux-amlogic
anessen977 has joined #linux-amlogic
anessen97 has quit [Quit: Ping timeout (120 seconds)]
JerryXiao has quit [Max SendQ exceeded]
anessen977 is now known as anessen97
adema has joined #linux-amlogic
kilobyte_ch has joined #linux-amlogic
repk_ has joined #linux-amlogic
Cheaterman has quit [*.net *.split]
GNUtoo has joined #linux-amlogic
JerryXiao has joined #linux-amlogic
Tartarus has quit [Ping timeout: 268 seconds]
BlueMatt[m] has quit [Ping timeout: 250 seconds]
psydroid has quit [Ping timeout: 248 seconds]
Cheaterman has joined #linux-amlogic
Tartarus has joined #linux-amlogic
tortoisedoc has quit [Ping timeout: 252 seconds]
ldevulder has quit [Remote host closed the connection]
BlueMatt[m] has joined #linux-amlogic
ldevulder has joined #linux-amlogic
gab has quit [Changing host]
gab has joined #linux-amlogic
<Cheaterman> I see that people are sadly sharing old fixes that have never been mainlined so I figured I'd ask
psydroid has joined #linux-amlogic
<Cheaterman> I'm getting freezes on N2+, oopses with vblank timeouts, or axg-tdm-iface audio-controller-1: ASoC: error at snd_soc_component_set_bias_level on audio-controller-1:
<Cheaterman> I'm doing UI stuff so I need the display to be updated at reliable interval, sadly these freezes are in the way
<Cheaterman> The one that's obviously sound-related can be worked around by removing audio sources from the application I run
<Cheaterman> but I don't think the vblank timeouts are audio related (they occured at a point where no audio was playing)
<Cheaterman> If you search for the exact error message I gave above about audio, you'll find literally ONE Google result
<Cheaterman> Pretty unusual, haha
<Cheaterman> lands me on this, user reporting the same issue (and UI freezes) but no fixes
<Cheaterman> So, thanks in advance if anyone has clues about this :)
<Cheaterman> Oh also panfrost gpu sched timeout
<Cheaterman> again no clue whether/how all of these are related, i just know they cause display freezes and I'd like to fix that
<Cheaterman> Dunno if it's a real thing but it almost feels like there's some power management going on, and whatever handles hdmi audio encoding (or something like that) takes a while to wake up
<Cheaterman> but once it's awake it responds quickly until it becomes inactive for a while
<Cheaterman> or something to that effect
<Cheaterman> possibly an "auto" i could change to "on" in sysfs
naoki has quit [Quit: naoki]
naoki has joined #linux-amlogic
Ballerburg9005 has joined #linux-amlogic
<steev> Cheaterman: old fixes that were never mainlined because... people assume it's trivial and/or already in a patchwork somewhere :P
<Cheaterman> could be, but I don't see anything like that in tobetter fork's issues
<Cheaterman> or PRs
<steev> oh i was being a bit sarcastic, like when i suggested you submit your patch for mainline and you said it's probably already in a patchwork somewhere - others could be doing the same thing, or not have the time to chase down/fix up any mainline code review requests, and just wanted things to work for them. the amlogic stuff isn't particularly heavy traffic, and you don't even need to be subscribed to the mailing list
<steev> xdarklight: hm, okay, i see, i'll give something along those lines a whirl
<steev> hm, 12 was still really jumpy on the 1080x1920, but taking it back to the original 4 pre AFBC patch works, may need to bump it a bit higher... i don't know what AFBC is, or if i'm using it here so...
<steev> added bonus is that the pixels don't seem to be off anymore
<steev> that's running bottom in alacritty on sway on the 1080x1920
<steev> And 2K monitor (was too lazy to reload the confit and unrotate display) https://usercontent.irccloud-cdn.com/file/VOLXgoPI/68538934673__34A8D808-10B2-4C50-9AC3-F5B6216128D8.mp4
<xdarklight> steev: AFBC is "Arm Frame Buffer Compression". IIRC it's used when importing a decoded video stream onto the display - for high resolutions this can consume a lot of memory bandwith. AFBC is supposed to save memory bandwidth by compressing the data between the video decoder and display IPs
<steev> ah, hm, okay so play a video and see if it makes things go very bad
<xdarklight> more like "decode a video through the VDEC (video decoder IP) and pipe it using zero-copy to the display output"
<steev> that is above my skillset/knowledge level
<xdarklight> I also haven't tried the video decoder myself. if Neil doesn't reply here you can try sending the patch as RFC (request for comments) and get some feedback from him on the mailing list
<steev> that's a good idea as well
<steev> xdarklight: an interesting side effect is that the pixels aren't pushed all off anymore with that change?
<xdarklight> steev: I've no idea why that would be but I'm happy for you if that's a welcome side-effect :)
<Cheaterman> steev: yeah but at least I 1/ tested Neil's fix instead of staying on mine, 2/ came here to report results, 3/ updated my issue on tobetter's fork and 4/ included the patch
<steev> :)
<Cheaterman> Not perfect for sure but I think I'm being a good steward of the ecosystem
<steev> oh please don't take it as a criticism at all, i'm 100% guilty of things, and was just teasing, you have no idea how much i appreciate that patch
<Cheaterman> There's another fun issue BTW that could be tangentially related to what you're experiencing
<Cheaterman> Some sort of off-by-one error in the framebuffer or something
<steev> very possible
<Cheaterman> one pixel row that should be at top is at the bottom, and one pixel column that should be on the left is on the right
<Cheaterman> Nobody noticed so I didn't investigate too much but it probably will need fixing
<steev> with the afbc related patch xdarklight mentions, i don't see that here (you can check the last 2 videos, i zoom in on the corners
<steev> or i suppose really, it's with
<Cheaterman> From what I read above of what AFBC should do, it should be unrelated, I'm observing this both on X11 desktop and in my apps, it's also worth noting I often use display rotation (eg portrait) for work stuff
<steev> though, that's the 5.10 branch, 5.19 isn't wanting to boot here for me, and that's my main testing device so i haven't looked into it
<Cheaterman> so perhaps it's less obvious or not exhibited at all on normal rotation
<steev> the fwiw, both of the videos are with the monitors being rotated via "fbcon=rotate=1"
<Cheaterman> but either way I sadly can't pipe decoded video straight to display because I do GLES stuff ; it would be neat if that could be piped directly to GPU though, as a GLES texture
<Cheaterman> oh, I rotate in X11
<steev> oh, actually, i do both
<Cheaterman> It would be a bit inconvenient to use a kernel cmdline for that in my case
<steev> #output HDMI-A-1 resolution 1280x400 transform 270
<steev> #output HDMI-A-1 resolution 1920x1080 transform 90
<steev> that's in my sway config, for when i switch to those
<Cheaterman> In my case it's just old school X11 stuff, Section "Monitor", Option "Rotate" "left"
<Cheaterman> or something along those lines
<steev> sway is wayland :(
<Cheaterman> ah, right, i should have guessed from the name haha
<Cheaterman> i know, weston is it? but not sway ; might have heard of it but that's it
<steev> fwiw, here is unrotated
<steev> sway is i3 for wayland
<Cheaterman> oh sway is a WM not a compositor
<Cheaterman> yeah I see
<Cheaterman> for me it's pretty old school setup hahaha... gentoo, openrc, x11, openbox (just because some apps might need a WM that respects eg full screen hints and whatnot)
<Cheaterman> a tweaked inittab for autologin in tty, a tweaked bashrc that starts X
<Cheaterman> steev: considering what you're displaying it would actually be pretty hard to tell!
<Cheaterman> what i advise is fullscreen video
<steev> yeah, i need to find something decent to check
<steev> something that isn't a *cough* home video *cough*
<Cheaterman> what's your machine BTW, N2+ like me?
<steev> radxa zero
<Cheaterman> lol home video haha
<Cheaterman> S905Y2
<steev> yeah
<Cheaterman> G31 instead of my G52
<Cheaterman> it may or may not have the same bugs, I take it it's not quite the same module
<steev> mali and bugs, name a more iconic duo
<Cheaterman> lol
<Cheaterman> IKR.... but see I'll take something fast that has a few bugs anyday
<Cheaterman> As I said earlier I'm really tired of VideoCore haha
<Cheaterman> and generally Broadcom stuff
<Cheaterman> Reminds me
<steev> but seriously, i really do appreciate all the workings, i'm able to now plug in either the 1080x1920 or 400x1280 and have the display work (althuogh i do need to reboot between them)
<Cheaterman> on an app we developed for XU4 we had refresh issues
<Cheaterman> I fixed it by having a permanently "blinking" transparent pixel (0-0.0001 alpha) in a corner
<Cheaterman> I don't mind these hacks, honestly, if it means the whole thing performs twice as fast as a Pi
<Cheaterman> Yeah that's pretty good news! Although the reboot is less than ideal
<Cheaterman> OTOH I suppose your G31 is covered by panfrost all the same as my G52
<steev> it's also a bit odd (or maybe not, who knows) that one waveshare has to be 270 to be "bottom" opposite of hdmi, and the other 90
<Cheaterman> So a large part of the bugs should be shared.... assuming they're not in silicon... hehe
<Cheaterman> That's interesting indeed
<Cheaterman> are you sure they're the exact same revision of the same model
<Cheaterman> if so, that's really curious for sure
<steev> no, they're not at all
<Cheaterman> Oh, well, then anything's possible
<steev> the one is 7.9" 400x1280, the other is 5.5" 1080x1920 :)
<Cheaterman> Hahaha literally a different panel
<Cheaterman> Then yeah it's not necessarily surprising
<steev> yeah, i'd jus thave expected the manuf to realize that rotation bit but... i should know better from my time at genesi
buzzmarshall has joined #linux-amlogic
vagrantc has joined #linux-amlogic