ChanServ changed the topic of #linux-sunxi to: Allwinner/sunxi development - Did you try looking at our wiki? https://linux-sunxi.org - Don't ask to ask. Just ask and wait for an answer! - This channel is logged at https://oftc.irclog.whitequark.org/linux-sunxi
vagrantc has quit [Quit: leaving]
cnxsoft has joined #linux-sunxi
ftg has quit [Read error: Connection reset by peer]
cnxsoft has quit [Remote host closed the connection]
junari has joined #linux-sunxi
junari_ has joined #linux-sunxi
bauen1 has quit [Ping timeout: 480 seconds]
junari is now known as Guest9862
junari_ is now known as junari
Guest9862 has quit [Ping timeout: 480 seconds]
apritzel has quit [Ping timeout: 480 seconds]
radxanaoki has joined #linux-sunxi
diego71_ has joined #linux-sunxi
diego71 has quit [Ping timeout: 480 seconds]
radxanaoki has quit [Quit: radxanaoki]
hexdump0815 has joined #linux-sunxi
hexdump01 has quit [Ping timeout: 480 seconds]
junari has quit [Remote host closed the connection]
hallyn has quit [Quit: Lost terminal]
hallyn has joined #linux-sunxi
hallyn has left #linux-sunxi [#linux-sunxi]
junari has joined #linux-sunxi
cnxsoft has joined #linux-sunxi
cnxsoft has quit [Ping timeout: 480 seconds]
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
junari has quit [Remote host closed the connection]
JohnDoe_71Rus has joined #linux-sunxi
bauen1 has joined #linux-sunxi
apritzel has joined #linux-sunxi
cnxsoft has joined #linux-sunxi
apritzel has quit [Ping timeout: 480 seconds]
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
bauen1 has quit [Ping timeout: 480 seconds]
cnxsoft has quit [Ping timeout: 480 seconds]
bauen1 has joined #linux-sunxi
ungeskriptet_ has joined #linux-sunxi
ungeskriptet has quit [Ping timeout: 480 seconds]
Robot_ has joined #linux-sunxi
bauen1 has quit [Ping timeout: 480 seconds]
hazardchem has quit [Read error: Connection reset by peer]
hazardchem has joined #linux-sunxi
evgeny_boger has joined #linux-sunxi
apritzel has joined #linux-sunxi
cnxsoft has joined #linux-sunxi
radxanaoki has joined #linux-sunxi
radxanaoki has quit []
bauen1 has joined #linux-sunxi
bauen1 has quit [Ping timeout: 480 seconds]
Schimsalabim has quit [Read error: Connection reset by peer]
Schimsalabim has joined #linux-sunxi
<Robot_> apritzel: last june we talked about the t113-s4 vs s3 dram init code, did you ever get to create patch to support both soc ?
<apritzel> Robot_: sorry, no. Can you confirm that the fuse reads 10 for both, but we need table 0 instead of 5 for the T113-s4?
<apritzel> I wonder if we should change that code to always use table 5 for the -s3 and table 0 for the -s4, ignoring the fuse
<apritzel> we already read some SID data in the driver, so should be an easy change
<apritzel> Robot_: if you could make a patch, I am more than happy to review and apply it
<Robot_> no worries :) I just didnt want to duplicate things
<Robot_> I can confirm that the patch to remap is needed for S4
<Robot_> This project on github is dealing with the same thing, but in bare metal: https://github.com/search?q=repo%3Aua1arn/hftrx%20allwnrt113_get_chipid&type=code
freemangordon1 has joined #linux-sunxi
freemangordon has quit [Read error: Connection reset by peer]
<Robot_> looking at the chipid of different chips:
<Robot_> f133 - 93005c00
<Robot_> t113-S3 - 93406000
<Robot_> t113-S4(M4020) - 93407200
<Robot_> when compared to SUN6I implementation, it seems that the 0xf000 mask on these is the bonding id ?
<apritzel> looking at other SoCs it's just like the whole lower 16 bits
<apritzel> so it's 0x6000 for the -s3 and 0x7200 for the -s4
<apritzel> unfortunately I don't know if there are -s3 chips out there which don't have the fuse set to 10, and which use a different pin mapping, so I guess it's best to leave this alone
<apritzel> and just check for 0x7200 and use map 0 then
<Robot_> sure
mripard has joined #linux-sunxi
szemzoa_ has quit []
JohnDoe_71Rus has quit [Quit: KVIrc 5.2.6 Quasar http://www.kvirc.net/]
JohnDoe_71Rus has joined #linux-sunxi
<kuba2k2> jernej: in sun8i_tcon_top, when it registers the clock gates, do you know what else is needed to change the "tcon-top-tv0" parent to "tve0" instead of "tcon-tv0"?
<kuba2k2> if I just change the argument to sun8i_tcon_top_register_gate() and switch TV0_CLK_SRC, I get " [CRTC:42:crtc-0] vblank wait timed out"
<kuba2k2> and the kernel locks up for over 60 sec during bootup
<kuba2k2> I assume that "tcon-top-tv0" should use the TVE clock as parent, along with changing the mux, but that alone doesn't work
bauen1 has joined #linux-sunxi
ity has quit [Remote host closed the connection]
ity has joined #linux-sunxi
cnxsoft has quit [Ping timeout: 480 seconds]
hexdump0815 has quit [Quit: WeeChat 3.8]
hexdump0815 has joined #linux-sunxi
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
<jernej> kuba2k2: vblank timeout means that TCON isn't configured properly and doesn't produce vblank interrupt
<kuba2k2> this part I understand
<kuba2k2> it means that its clock isn't running, right?
<jernej> most likely
<jakllsch> well, functional clock
<jakllsch> (it's not like the registers aren't readable)
<kuba2k2> yeah
<jernej> I'm just thinking if you actually need to change TCON clock. Maybe this clock is still needed for some functionality
<kuba2k2> I don't know if I need to change it
<kuba2k2> but if I don't, then simultaneous LVDS and TVE output is not possible (one will change the other's clock rate, they're both VIDEO0)
<jernej> you can try with assigned clocks trick in DT for test
<kuba2k2> I can force TVE_CLK to VIDEO1 using assigned-clocks in DT, but the TCON TV clock is still VIDEO0
<kuba2k2> ...which wouldn't be a problem, normally - except *nothing* seems to explicitly set the TVE clock rate - the only call with crtc_clock is for the TCON TV clock
<kuba2k2> it seems to just rely on the fact that TCON and TVE will *usually* use the same clock parent
<jernej> TVE clock rate should be set by TVE driver
<kuba2k2> it isn't, not on A10, not on the WIP H3/H5 patch
<jernej> well, you can alway modify the driver, right?
<kuba2k2> I can, but I assume that it's not there because it wasn't supposed to :)
<jernej> fyi, TVE clock is also the reason why H3/H5 driver isn't mainlined
<jernej> there was no good understanding how clock is connected internally, so it wasn't clear how to properly describe it in DT
<kuba2k2> oh? I thought the patch was working, wasn't it?
<jernej> making things work is easy, make them work the right way is harder :)
<kuba2k2> hmm, okay. So was there any particular reason not to add the CLK_TVE to DT then?
<kuba2k2> at least on T113, it seems to be pretty straightforward how it's connected
<jernej> I forgot actual H3 conversation, but you can dig out patches and discussion if you want
<jernej> yeah, T113 documentation is decent regarding this, but it has different problem
<jernej> there is no driver yet with TVE_TOP
<jernej> which should be represented as its own node
<kuba2k2> yep, it seems I'm writing one now
<kuba2k2> though as a single node
<kuba2k2> the TVE part itself is pretty much different from the one on A10, let alone the TOP
<jernej> if you check R40, TVE_TOP actually represents 4 analog outputs, where they 3 can be combined as VGA output
<jernej> so binding must be universal enough to support such cases too
<kuba2k2> ah, so it has 4 TVEs then?
<kuba2k2> hm no just two
hazardch1m has joined #linux-sunxi
<jernej> but one could be CVBS and other VGA
<jernej> or two CVBS and then two DACs are unused
<kuba2k2> I see
<kuba2k2> when making the TVE TOP as a separate node, do you think it's enough to add a reference to it in the &tve node?
<kuba2k2> or should the TOP also be connected using endpoints and stuff
hazardchem has quit [Ping timeout: 480 seconds]
<jernej> good question, usually display related stuff use endpoints, like tcon_top
<kuba2k2> the &dphy doesn't, but I guess there's another reason for that
<kuba2k2> so.. ideally, the TCON TV should output to TVE TOP, and that should output to TVE(s)?
<jernej> yeah, phys are weird exception, but it is what it is
<jernej> kuba2k2: this needs some brainstorming. TCON outputs signal to TV encoder and encoder outputs signal via DACs to connector
<kuba2k2> drm connectors aren't in dt, right?
<kuba2k2> also, I can't test anything else than CVBS because I don't have the hardware
<jernej> and for CVBS that is actually a no brainer, since every node has one input and one output. However, VGA case isn't so straightforward, but it must be covered.
<kuba2k2> is that even used for the A10 TVE?
<kuba2k2> I haven't seen it before
<jernej> no, it should be, but some mistakes have been made
pmp-p has quit [Ping timeout: 480 seconds]
<jernej> but if you check hdmi, there is always hdmi-connector at the end
pmp-p has joined #linux-sunxi
<jernej> I have one R40 board, but it only has test pad for one DAC, so it's not useful for any serious testing
<kuba2k2> Hmm okay. So if I was to write a new TVE driver, it should configure the outputs based on the connectors that are connected to it in DT?
<jernej> yes
hazardchem has joined #linux-sunxi
<kuba2k2> Okay, see you in a few months then :D
<jernej> as I said, proper solution takes much more time :D
<kuba2k2> I'll let you know once I have something that works
<kuba2k2> Thanks for the help :)
hazardch1m has quit [Ping timeout: 480 seconds]
dsimic is now known as Guest9904
dsimic has joined #linux-sunxi
Guest9904 has quit [Ping timeout: 480 seconds]
ftg has joined #linux-sunxi
apritzel has quit [Ping timeout: 480 seconds]
JohnDoe_71Rus has quit [Quit: KVIrc 5.2.6 Quasar http://www.kvirc.net/]
mnemoc has quit [Remote host closed the connection]
mnemoc has joined #linux-sunxi
apritzel has joined #linux-sunxi
mnemoc has quit [Quit: leaving]
Robot_ has quit [Ping timeout: 480 seconds]
mnemoc has joined #linux-sunxi
evgeny_boger has quit [Ping timeout: 480 seconds]
bauen1 has quit [Ping timeout: 480 seconds]