Turl changed the topic of #linux-sunxi to: Allwinner/sunxi /development discussion - did you try looking at our wiki? https://linux-sunxi.org - Don't ask to ask. Just ask and wait! - https://github.com/linux-sunxi/ - Logs at http://irclog.whitequark.org/linux-sunxi
<apritzel> MoeIcenowy: b3d2861eb20a795b99292b823c923935df26dfc6
<apritzel> MoeIcenowy: if you revert this, it works
<apritzel> MoeIcenowy: do you have a working simple-framebuffer setup?
<MoeIcenowy> apritzel: in my own branch
<apritzel> I see kernel crashes
<apritzel> [ 0.000000] BUG: Bad page state in process swapper pfn:4257e
<MoeIcenowy> oh mysterious
BenG83 has quit [Ping timeout: 260 seconds]
wzyy2 has joined #linux-sunxi
BenG83_PB has joined #linux-sunxi
wzyy2 has quit [Remote host closed the connection]
BenG83_ has quit [Quit: Leaving]
Mr__Anderson has quit [Ping timeout: 240 seconds]
<MoeIcenowy> apritzel: it now seems to be a problem on Pinebook if PL bank is fully occupied by ATF...
<MoeIcenowy> the ANX6345 is controlled via the I2C on PL bank
<MoeIcenowy> but its reset and enable pins are at PD and PH banks
<apritzel> MoeIcenowy: where is PL occupied by the ATF?
<MoeIcenowy> I think you want to do so ;-)
<apritzel> we _could_ do so if it would make sense
<apritzel> oh wait, the RSB is there, right
<apritzel> ?
<BenG83_PB> yes
<MoeIcenowy> now it do not make many sense ;-)
<apritzel> so what do you need from the bridge?
<MoeIcenowy> send some commands to it
<apritzel> the whole idea is that all those R_ peripherals should belong to the firmware
<apritzel> be that ATF or the arisc
<apritzel> that's why they are switchable
<apritzel> to secure, I mean
<BenG83_PB> PL is used for Wifi/BT, RSB and PWM on most A64 designs I saw
<apritzel> connecting a non-system critical peripheral like the bridge to that is not the smartest choice
<apritzel> BenG83_PB: is that for actual GPIO or just of power switching
<apritzel> ?
<BenG83_PB> Wifi/BT uses PL for wakeup and reset
<apritzel> I have a nice solution for that: SCPI provides a "device power state" interface
<apritzel> which allows you to switch peripheral devices on and off
<MoeIcenowy> In fact I still think we should drop SCPI now
<MoeIcenowy> we may leave SCPI thoughts for further boards
<MoeIcenowy> if they want to fully seperate the secure side and the non-secure side from its design
theone has quit [Quit: PanicBNC account has been auto selected for removal due to login inactivity.]
<apritzel> MoeIcenowy: to some degree it's optional
<apritzel> but I still want to give it a go
<MoeIcenowy> maybe you should implement some device tree fixup support in ATF...
<MoeIcenowy> as we may not supply the info about the SCPI interface in a mainline device tree
<apritzel> MoeIcenowy: indeed I was thinking about exactly that
<apritzel> this concept of having the DT in Linux is broken IMHO anyway
<MoeIcenowy> and the stock devicetree will mostly possible to also manage R_ peripherals
<MoeIcenowy> s/stock/mainline/
<apritzel> that was my plan: to allow ATF to fix up the DT to match what it implements
ErwinH has joined #linux-sunxi
<apritzel> removing R_ peripherals, for instance, and adding SCPI nodes
<BenG83_PB> MoeIcenowy, btw
<apritzel> upstream ATF has DT support, though I haven't checked what it actually covers
<BenG83_PB> how many of those questions to AW got answered?
ErwinH has quit [Ping timeout: 240 seconds]
<MoeIcenowy> BenG83_PB: not too much...
<MoeIcenowy> and I still didn't receive the mail from AW
<MoeIcenowy> it do contains some infomation about Secure Boot
<MoeIcenowy> apritzel: AW guys said that they want some demo code to prove TZPC do not work without SB bit burned
<apritzel> MoeIcenowy: they don't believe it?
<MoeIcenowy> yes they think it shouldn't happen ;-)
<apritzel> it's dead simple: read the SID or access SRAM A2 from non-secure
<apritzel> you can do this from U-Boot
<apritzel> both those peripherals are always secure
<apritzel> and with the SB bit burned access is indeed not possible
<apritzel> but the best prove is upstream U-Boot
<apritzel> it runs in EL2, which is architecturally non-secure
<MoeIcenowy> read SID means directly read from 0x01c14200?
<apritzel> both ways
<MoeIcenowy> directly read is more simpler ;-)
<apritzel> in secure boot both reading from memory and doing the register "dance" produce all 0s
<apritzel> in non-secure it works
<MoeIcenowy> have you tried to read it from BSP U-Boot?
<apritzel> yes, I think so (though I haven't booted this for a year or so)
<MoeIcenowy> P.S. The DRAM chip on Pinebook sample is from "ISDT"...
<MoeIcenowy> an unknown vendor...
<apritzel> great :-(
<apritzel> let's hope the production batch fixes that
<MoeIcenowy> the production batch even uses different memory design, did you forget?
<apritzel> MoeIcenowy: so to summarise this: prove is very simple: reading the SID works from U-Boot, which shouldn't in their security model
<MoeIcenowy> yes
<apritzel> that's why they have a "read SID" callback in their ATF
<BenG83_PB> MoeIcenowy, mine are from Micron
<MoeIcenowy> BenG83_PB: so things changed?!
<BenG83_PB> though I had some trouble finding that out since there was only stampcodes on the chips
<MoeIcenowy> P.S. The ISDT chip seems to have high relation with SKhynix
<BenG83_PB> let me find the post
<MoeIcenowy> just change the first letter of chip model number from I to H it will be a valid chip model number from SKhynix
<MoeIcenowy> oh ISDT is now also a company of SK C&C
<MoeIcenowy> and Pinebook sample batch seems to use dual-rank DRAM, this is why I firstly got only 1GiB on it
wzyy2 has joined #linux-sunxi
<apritzel> MoeIcenowy: re ranks> shouldn't the U-Boot DRAM code detect the rank automatically?
<MoeIcenowy> apritzel: currently not
<MoeIcenowy> set the default value of dual_rank in the dram_para to 1 seems to be able to trigger the detect
<MoeIcenowy> I tested it on Pinebook and Pine64
<MoeIcenowy> the former have dual rank and latter single rank
<MoeIcenowy> but I want regression tests to be done on H3 boards ;-)
<apritzel> MoeIcenowy: ah, OK, that makes sense
<apritzel> I just remembered this comment about "detecting rank"
<apritzel> it seems like it needs to fail first, then it will try with the other setting
leio has quit [Remote host closed the connection]
premoboss has quit [Ping timeout: 240 seconds]
vishnup has joined #linux-sunxi
gzamboni has quit [Ping timeout: 240 seconds]
premoboss has joined #linux-sunxi
ganbold has quit [Quit: This computer has gone to sleep]
GrimKriegor_ is now known as GrimKriegor
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
dave0x6d has joined #linux-sunxi
sgteem has joined #linux-sunxi
sgteem_ has quit [Ping timeout: 260 seconds]
ninolein_ has joined #linux-sunxi
ninolein has quit [Ping timeout: 240 seconds]
apritzel has quit [Ping timeout: 260 seconds]
ganbold has joined #linux-sunxi
lurchi_ has joined #linux-sunxi
lurchi__ has quit [Ping timeout: 240 seconds]
Andy-D has quit [Ping timeout: 240 seconds]
cnxsoft has joined #linux-sunxi
lkcl has joined #linux-sunxi
wzyy2 has quit [Ping timeout: 260 seconds]
wzyy2 has joined #linux-sunxi
lkcl has quit [Ping timeout: 240 seconds]
vishnup has quit [Ping timeout: 255 seconds]
vishnup has joined #linux-sunxi
pg12 has quit [Ping timeout: 260 seconds]
pg12 has joined #linux-sunxi
beeble_ has joined #linux-sunxi
beeble has quit [Ping timeout: 252 seconds]
medvid_ has joined #linux-sunxi
plaes_ has joined #linux-sunxi
mfa298_ has joined #linux-sunxi
sW`` has joined #linux-sunxi
topi`_ has joined #linux-sunxi
rtp has joined #linux-sunxi
medvid has quit [Ping timeout: 252 seconds]
bobryan has quit [Ping timeout: 252 seconds]
plaes has quit [Ping timeout: 252 seconds]
mfa298 has quit [Ping timeout: 252 seconds]
sW` has quit [Ping timeout: 252 seconds]
topi` has quit [Ping timeout: 252 seconds]
rtp_ has quit [Remote host closed the connection]
sW`` is now known as sW`
JohnDoe_71Rus has joined #linux-sunxi
bobryan has joined #linux-sunxi
wzyy2 has quit [Ping timeout: 240 seconds]
TheSeven has quit [Ping timeout: 255 seconds]
TheSeven has joined #linux-sunxi
premoboss has quit [Ping timeout: 260 seconds]
wzyy2 has joined #linux-sunxi
buZz has quit [Ping timeout: 240 seconds]
cyrozap has quit [Read error: Connection reset by peer]
cyrozap has joined #linux-sunxi
cyrozap has quit [Remote host closed the connection]
cyrozap has joined #linux-sunxi
buZz has joined #linux-sunxi
buZz is now known as Guest33235
Net147 has quit [Quit: Quit]
<montjoie_> MoeIcenowy: I pushed my v2 which should fix your issue on V3S (early init PHY)
montjoie_ is now known as montjoie
wzyy2 has quit [Ping timeout: 268 seconds]
wzyy2 has joined #linux-sunxi
Jan has joined #linux-sunxi
Jan is now known as Guest67308
Guest67308 is now known as DiscoStewDeluxe
Net147 has joined #linux-sunxi
DiscoStewDeluxe has quit [Ping timeout: 260 seconds]
terra854 has joined #linux-sunxi
JohnDoe_71Rus has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/]
Nacho_ has quit [Ping timeout: 255 seconds]
Nacho_ has joined #linux-sunxi
leviathan has joined #linux-sunxi
premoboss has joined #linux-sunxi
f0xx has joined #linux-sunxi
lkcl has joined #linux-sunxi
Nacho_ has quit [Ping timeout: 240 seconds]
Nacho_ has joined #linux-sunxi
|Jeroen| has joined #linux-sunxi
gzamboni has joined #linux-sunxi
IgorPec has joined #linux-sunxi
ErwinH_ has joined #linux-sunxi
Bridgeman has joined #linux-sunxi
apritzel has joined #linux-sunxi
Bridgeman has quit [Ping timeout: 260 seconds]
MiladJ40 has joined #linux-sunxi
<MiladJ40> i'm trying to design a allwinner a64 based module, i want to know if it supports android and if it is easy to port android on the device
jstein_ has joined #linux-sunxi
wzyy2 has quit [Ping timeout: 260 seconds]
jstein_ is now known as jstein
jstein has quit [Read error: Connection reset by peer]
<MiladJ40> is there anybody to help?
komunista has joined #linux-sunxi
apritzel has quit [Ping timeout: 255 seconds]
sgteem has quit [Ping timeout: 260 seconds]
sgteem has joined #linux-sunxi
Bridgeman has joined #linux-sunxi
wzyy2 has joined #linux-sunxi
MiladJ40 has quit [Quit: Page closed]
<Bridgeman> How would it be possible to upgrade cubian with wheezy to jessy and newer kernel without reinstalling current services?
<Bridgeman> Is that something you can help with?
Bridgeman has quit [Ping timeout: 260 seconds]
ErwinH_ has quit [Remote host closed the connection]
ErwinH has joined #linux-sunxi
<KotCzarny> MiladJ40: see topic
<KotCzarny> in short, this channel is about mainline linux
reinforce has joined #linux-sunxi
ErwinH has quit [Ping timeout: 255 seconds]
IgorPec has quit [Ping timeout: 260 seconds]
ErwinH has joined #linux-sunxi
cnxsoft has quit [Read error: Connection reset by peer]
cnxsoft has joined #linux-sunxi
BenG83 has joined #linux-sunxi
ErwinH has quit [Remote host closed the connection]
ErwinH has joined #linux-sunxi
ErwinH has quit [Ping timeout: 260 seconds]
LargePrime has quit [Ping timeout: 260 seconds]
JohnDoe_71Rus has joined #linux-sunxi
lkcl has quit [Ping timeout: 255 seconds]
netlynx has joined #linux-sunxi
netlynx has joined #linux-sunxi
netlynx has quit [Changing host]
bonbons has joined #linux-sunxi
LargePrime has joined #linux-sunxi
ErwinH has joined #linux-sunxi
ErwinH has quit [Ping timeout: 255 seconds]
Ntemis has joined #linux-sunxi
premoboss has quit [Ping timeout: 260 seconds]
tucker has joined #linux-sunxi
ErwinH has joined #linux-sunxi
IgorPec has joined #linux-sunxi
ErwinH has quit [Remote host closed the connection]
r1mikey has joined #linux-sunxi
<tucker> Hey. Just wanted to report that I'm running 4.10 with dwmac ethernet (extracted from montjoies stmmac-sun8i-wip tree, internal phy) on my silly little H3 box. The dts is (for now) more or less a copy of the orangepi-pc dts.
ErwinH has joined #linux-sunxi
<tucker> Now, what would be needed to have USB0 (the type-A OTG port) fully hooked up as O-/EHCI?
<tucker> "The original driver wired it to OHCI/EHCI controller; however, as the code to use PHY0 as OHCI/EHCI is missing, it makes the PHY fully unusable." -MoeIcenowy on the list
lkcl has joined #linux-sunxi
ErwinH has quit [Ping timeout: 240 seconds]
r1mikey_ has joined #linux-sunxi
r1mikey has quit [Read error: Connection reset by peer]
ErwinH has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
mzki has joined #linux-sunxi
r1mikey_ has quit [Ping timeout: 255 seconds]
Mr__Anderson has joined #linux-sunxi
ErwinH has quit [Ping timeout: 255 seconds]
<plaes_> tucker: your own board?
BenG83_PB has quit [Quit: Leaving]
<tucker> HYH-TBH3, documented in wiki (small anonymous box with only the words "Smart Android Box" printed on the underside)
<plaes_> ah
plaes_ is now known as plaes
plaes has quit [Changing host]
plaes has joined #linux-sunxi
|Jeroen| has quit [Quit: dada]
<montjoie> tucker: thanks for the report
<tucker> montjoie: np.
premoboss has joined #linux-sunxi
<tucker> The latest work to incorporate DE2 into preexisting DE code looks promising, just hoping for H3+HDMI support there soonish
scream has joined #linux-sunxi
premoboss has quit [Quit: Sto andando via]
Gerwin_J has quit [Ping timeout: 260 seconds]
IgorPec has quit [Quit: Nettalk6 - www.ntalk.de]
r1mikey has joined #linux-sunxi
r1mikey has quit [Ping timeout: 255 seconds]
lkcl has quit [Ping timeout: 255 seconds]
leio has joined #linux-sunxi
r1mikey has joined #linux-sunxi
r1mikey has quit [Ping timeout: 255 seconds]
techping has joined #linux-sunxi
techping has quit [Remote host closed the connection]
bparker_ is now known as bparker
medvid_ is now known as medvid
cnxsoft has quit [Remote host closed the connection]
cnxsoft has joined #linux-sunxi
<BenG83> MoeIcenowy, I managed to boot your kernel tree into an Armbian rootfs
<BenG83> with simplefb LCD output ;)
<BenG83> I also get
<BenG83> [ 50.893139] BUG: Bad page state in process 40-updates pfn:af607
Gerwin_J has joined #linux-sunxi
r1mikey has joined #linux-sunxi
vinimac has joined #linux-sunxi
TheSeven has quit [Ping timeout: 255 seconds]
r1mikey has quit [Ping timeout: 255 seconds]
TheSeven has joined #linux-sunxi
<MoeIcenowy> tucker: I still forgot to do it...
cptG has joined #linux-sunxi
jstein_ has joined #linux-sunxi
cptG_ has quit [Ping timeout: 255 seconds]
<tucker> MoeIcenowy: No rush. I gather that you started with V3 since it's a simpler version, and wens is currentlty changing things in the drm code as well.
jstein_ is now known as jstein
<Putti> is it possible to connect to u-boot via usb cable like it is possible with FEL mode?
<Putti> the u-boot command prompt, I mean.
popolon has joined #linux-sunxi
cnxsoft has quit [Quit: cnxsoft]
<MoeIcenowy> tucker: I mean the H3 usb things
<MoeIcenowy> for DE2 HDMI support some work on dw-hdmi is waited
<MoeIcenowy> they are being done by Renesas people
scream has quit [Remote host closed the connection]
chlorine has joined #linux-sunxi
Bridgeman has joined #linux-sunxi
vishnup has quit [Remote host closed the connection]
<Bridgeman> Hello. Sorry to bother you again. I know this channel is for mainline kernel development. But where do I get best help if I want to upgrade my cubian with old 3.4.79 kernel to latest mainline and also upgrading distro to Debian Jessy?
jernej has joined #linux-sunxi
Putti has quit [Ping timeout: 240 seconds]
<Bridgeman> The wiki is very confusing. I always get lost on the uboot part.
vishnup has joined #linux-sunxi
mozzwald_ is now known as mozzwald
LargePrime has quit [Ping timeout: 240 seconds]
<BenG83> MoeIcenowy, does your PB tree support USB?
jstein has quit [Remote host closed the connection]
<MoeIcenowy> BenG83: yes ;-)
<MoeIcenowy> Bridgeman: mainline currently do not support NAND, so first get a SD ;-)
<BenG83> then I screwed up the config maybe...
jstein_ has joined #linux-sunxi
<MoeIcenowy> probe of USB failed with error 38...
<MoeIcenowy> 38 is ENOSYS...
<MoeIcenowy> have you enabled sun4i usb phy support?
<Bridgeman> MoeIcenowy: My Cubian is installed on SD-Card only and I suppose it boots from there. At least fdisk -l doesn't show something useful on /dev/nand
jstein_ is now known as jstein
<Bridgeman> MoeIcenowy: The issue is, that I have installed Cubian like 2 years ago and now it is not maintained anymore and I need to switch to something with more future...
<MoeIcenowy> Bridgeman: why don't you just re-install a armbian?
lkcl has joined #linux-sunxi
<BenG83> sun4i?
<BenG83> maybe I should not have removed all non A64 options ;P
<MoeIcenowy> BenG83: usually we name drivers with the earliest SoC with a similar device
chlorine has quit [Remote host closed the connection]
<Bridgeman> MoeIcenowy: Because it's nicely installed and running already some services (custom internet connection monitoring, boinc, AirPrint via Cups). I would like to avoid doing this work again.
<MoeIcenowy> I think the Cubian system it self is not maintained
<MoeIcenowy> and try to do your own kernel setup there is a bigger mess
LargePrime has joined #linux-sunxi
<TheLinuxBug> Armbian is nice and they have most of the packages
<MoeIcenowy> you will finally make yourself a distro vendor ;-)(
<MoeIcenowy> s/(//
<Bridgeman> MoeIcenowy, TheLinuxBug: Will try to find an upgrade path to armbian. Just didn't saw any help on their forums and wondering if not many more people have the same problem.
<MoeIcenowy> yes most people just won't select Cubian
Putti has joined #linux-sunxi
<MoeIcenowy> we do not believe systems by SBC vendors
<Bridgeman> MoeIcenowy: It was a different case 2 years ago and if you start fresh you'll take the easiest and most obvious image. In my case it was Cubian.
<Bridgeman> MoeIcenowy: But thanks for your help. Will try to work through it.
<TheLinuxBug> um I am not sure what you mean by upgrade path
<TheLinuxBug> I wouldn't even consider trying to upgrade Cubian in place
<TheLinuxBug> I would simply burn the armbian image, install any services you need and copy over any specific configs and such you may need
<TheLinuxBug> you will just waste a bunch of time
<MoeIcenowy> yes
<BenG83> MoeIcenowy, my PHY section only has
<BenG83> CONFIG_PHY_SUN9I_USB=y
<MoeIcenowy> ok CONFIG_PHY_SUN4I_USB is not set?
<BenG83> no
<MoeIcenowy> it may be the point ;-)
<MoeIcenowy> there's even no CONFIG_PHY_SUN4I_USB ?!
<BenG83> and I cant find it in the PHY drivers
leio has quit [Remote host closed the connection]
<MoeIcenowy> maybe its some dependencies are not met
<BenG83> arm defconfig has it
<BenG83> arm64 defconfig not
<BenG83> yeah I'll look what it depends on
<MoeIcenowy> yes arm64 defconfig needs some update
lurchi_ is now known as lurchi__
|Jeroen| has joined #linux-sunxi
<BenG83> ok missing EXTCON
mossroy has joined #linux-sunxi
<BenG83> thanks MoeIcenowy, got USB
<BenG83> still crashes right after login though
<BenG83> [ 184.799925] 40-updates[2005]: unhandled level 2 translation fault (11) at 0x000ccd85, esr 0x92000006
<MoeIcenowy> what's 40-updates?
<MoeIcenowy> P.S. I think this linux-next may be not so stable -- after 4.11-rc is out I will do a version based on it
<MoeIcenowy> is it an armbian process?
<BenG83> yes it's the armbian update check that runs after login
<BenG83> basically just apt update
wzyy2 has quit [Ping timeout: 240 seconds]
<MoeIcenowy> oh mysterious
<BenG83> let me check if it happens anywhere else
<BenG83> usb ethernet dongle seems to work ok
<MoeIcenowy> https://pastebin.anthonos.org/view/2c0fdff6 here's my .config
<MoeIcenowy> https://pastebin.anthonos.org/view/2c0fdff6 is the raw form ;-)
<MoeIcenowy> oh the same url...
<BenG83> thanks
<BenG83> mmh so far anything else doesnt triggers BUG
* BenG83 goes to investigate
<MoeIcenowy> how about just run apt update?
<BenG83> that works
<BenG83> just upgraded a couple things
|Jeroen| has quit [Quit: dada]
<BenG83> can I set console=... bootarg to have output on simplefb?
<MoeIcenowy> console=tty0 ;-)
<BenG83> k
<Putti> is simplefb the one that is outputted on the device's lcd (in case of a allwinner tablet)?
wzyy2 has joined #linux-sunxi
Bridgeman has quit [Quit: Page closed]
ErwinH has joined #linux-sunxi
<MoeIcenowy> simplefb is in fact only a part of memory passed by U-Boot to Linux
<MoeIcenowy> but U-Boot configured it as the output of LCD
wzyy2 has quit [Ping timeout: 260 seconds]
<Putti> cool, so simplefb would output kernel's messages to lcd, right (as I have u-boot console showing on the lcd)?
<MoeIcenowy> yes
<BenG83> time for more coffee
<BenG83> brb
<MoeIcenowy> It's just like vesafb on x86 -- BIOS does the initialization, and the driver only knows that write something there can change the display content.
ErwinH has quit [Ping timeout: 260 seconds]
r1mikey has joined #linux-sunxi
<Putti> MoeIcenowy, that is awesome. Thanks!
<MoeIcenowy> In fact there's some solution that makes vesafb to use simplefb driver ;-)
r1mikey has quit [Ping timeout: 255 seconds]
wzyy2 has joined #linux-sunxi
deskwizard_ has quit [Remote host closed the connection]
deskwizard has joined #linux-sunxi
dave0x6d has quit [Quit: Connection closed for inactivity]
wzyy2 has quit [Ping timeout: 240 seconds]
lkcl has quit [Read error: Connection reset by peer]
<willmore> MoeIcenowy, wens Looks like you're Phoronix famous: http://www.phoronix.com/scan.php?page=news_item&px=Allwinner-Disp-Engine-2.0-DRM
<jelle> lol
<lvrp16> *clap clap*
wzyy2 has joined #linux-sunxi
vinimac has quit [Quit: Saindo]
wzyy2 has quit [Ping timeout: 255 seconds]
tucker has quit [Ping timeout: 260 seconds]
<BenG83> MoeIcenowy, I have now the same general config like you, but that didnt change things, I still get BUG: Bad page .... in various processes, could this be linked to the dual rank config for the DRAM?
lurchi__ is now known as lurchi_
<MoeIcenowy> can you try to revert 49eb19f93ce557a596cca4cc25b84aa46f1b107a in u-boot?
wzyy2 has joined #linux-sunxi
<BenG83> MoeIcenowy, ok
lurchi_ has quit [Ping timeout: 240 seconds]
chlorine has joined #linux-sunxi
chlorine has quit [Ping timeout: 260 seconds]
TheSeven has quit [Remote host closed the connection]
IgorPec has joined #linux-sunxi
wzyy2 has quit [Ping timeout: 255 seconds]
<BenG83> MoeIcenowy, turning off the second rank fixed the crashes
<MoeIcenowy> so maybe there's some problem in current rank detect code...
<BenG83> doesnt have the Pine64 the same memory layout?
<MoeIcenowy> jemk: do you have any thoughts about it?
<BenG83> *have
<MoeIcenowy> Pine64 uses 4*8bit chips
<MoeIcenowy> but Pinebook uses 2*2*16bit chips
wzyy2 has joined #linux-sunxi
<MoeIcenowy> oh yes the channels in DE2 mixer is just like the "pipe" in DE1 BE
beeble_ has quit [Remote host closed the connection]
<wens> willmore: you mean MoeIcenowy is famous
TheSeven has joined #linux-sunxi
<MoeIcenowy> in fact it's the first time for me to touch DRM/KMS driver development
TheSeven has quit [Remote host closed the connection]
swiftgeek has quit [Ping timeout: 268 seconds]
<jemk> MoeIcenowy: dual rank seemed buggy, and since i didn't have the hardware to test it, i disabled it
<MoeIcenowy> wens: will your cleanups in 4.12?
<MoeIcenowy> when will you push multiple pipeline patch?
Andy-D has joined #linux-sunxi
<wens> MoeIcenowy: still working on it, not much has been done yet
dave0x6d has joined #linux-sunxi
<wens> wish i were still i student sometimes
<willmore> Oh, sorry, wens. I thought you and MoeIcenowy were working on that.
<wens> would have more energy and free time
<wens> willmore: nope, i'm doing DE 1.0 stuff
<MoeIcenowy> willmore: we are independently doing new functions for DE driver
<willmore> Yay!
<MoeIcenowy> what I do is sun8i-mixer support
<MoeIcenowy> and wens do is multiple pipeline
<BenG83> mmh
<BenG83> I have to retract my statement
<MoeIcenowy> finally our works will meet in A83T/A64/H3, which have dual display pipelines (A83T/A64 is LCD/HDMI, H3 is HDMI/TV)
<wens> the multi pipeline stuff for DE 2.0 is different though, because the mux points are in different places
<BenG83> it doesnt crash with BUG: Bad page state anymore, so far so good, but it dies now on it's own with stalled CPUs
<wens> for DE 2.0, it's between the mixers and tcons
|Jeroen| has joined #linux-sunxi
<wens> for DE 1.0, it's between the tcons and trailing encoders (TV, HDMI, DSI)
<MoeIcenowy> wens: for DE 2.0 I think TCONs will still have mux
<MoeIcenowy> for example, TCON0 in A64 can output to RGB/LVDS/DSI
swiftgeek has joined #linux-sunxi
<MoeIcenowy> layer is set to layers[i], then layer is updated, but layers[i] is not
<MoeIcenowy> wens: in sun4i_layers.c, sun4i_layers_init, I doubt the code about layer and layers is wrong
sgteem has quit [Ping timeout: 260 seconds]
<MoeIcenowy> could you check it?
<wens> MoeIcenowy: i already sent a fix for it. it will be in 4.12
<MoeIcenowy> ok
<wens> MoeIcenowy: but TCON1 does not have channel 0, so it cannot do those
<wens> the tcons on the newer chips are single purpose
<wens> they have only 1 channel
<MoeIcenowy> and newer chips can have two full pipelines running together ;-)
<MoeIcenowy> not only changable TCON output
swiftgeek has quit [Ping timeout: 240 seconds]
<wens> old chips can as well
<wens> tcons are bound to bes, which form the pipeline
<MoeIcenowy> only if it have multiple BEs
<wens> you always have 1 of each, in pairs
<wens> so either you have 1 tcon + 1 be, or 2 tcons + 2 bes
<wens> they are not muxable
wzyy2 has quit [Ping timeout: 240 seconds]
<plaes> wens: I have ccu for A20, now need to figure out why it doesn't work :)
<wens> plaes: clk_ignore_unused on the cmdline is your friend
<plaes> thx
<wens> or add some debug messages to the clk core where unused clks are shutdown
<wens> if these show nothing, it's probably you have some offset or clk name wrong
<wens> boot with clk_ignore_unused, and check /sys/kernel/debug/clk/clk_summary
<wens> be back tomorrow
TheSeven has joined #linux-sunxi
r1mikey has joined #linux-sunxi
Guest33235 is now known as buZz
swiftgeek has joined #linux-sunxi
wzyy2 has joined #linux-sunxi
r1mikey has quit [Ping timeout: 255 seconds]
mfa298_ is now known as mfa298
wzyy2 has quit [Ping timeout: 268 seconds]
sgteem has joined #linux-sunxi
fdcx has quit [Remote host closed the connection]
fdcx has joined #linux-sunxi
ErwinH has joined #linux-sunxi
ErwinH has quit [Remote host closed the connection]
ErwinH has joined #linux-sunxi
ErwinH has quit [Remote host closed the connection]
ErwinH has joined #linux-sunxi
interrobangd has joined #linux-sunxi
terra854 has quit [Quit: Connection closed for inactivity]
<plaes> \o/
<plaes> first boot :)
ErwinH has quit [Remote host closed the connection]
BenG83_PB has joined #linux-sunxi
ErwinH has joined #linux-sunxi
chlorine has joined #linux-sunxi
chlorine has quit [Remote host closed the connection]
chlorine has joined #linux-sunxi
<MoeIcenowy> mripard: will you try to mainlineize the GEM allocation ioctl of sun4i-drm used by mali dmabuf?
ErwinH has quit [Ping timeout: 260 seconds]
interrobangd has quit [Read error: Connection reset by peer]
leviathan has quit [Remote host closed the connection]
<MoeIcenowy> jernej: I hope you can share what you know on DE2_Registers_Guide page ASAP ;-)
leviathan has joined #linux-sunxi
BenG83 has quit [Quit: Leaving]
<jernej> MoeIcenowy: Which part exactly do you have in mind?
scream has joined #linux-sunxi
<MoeIcenowy> now what I want most is things about alpha blending
Gerwin_J has quit [Quit: Gerwin_J]
Gerwin_J has joined #linux-sunxi
<jernej> I don't know that good enough to help you
<jernej> After all, I'm writing only U-Boot driver :)
<jernej> Did you check jfm driver?
<MoeIcenowy> some parts are confusing in jfm driver
<MoeIcenowy> e.g. he uses a VI channel's first layer as the primary layer
ErwinH has joined #linux-sunxi
<jernej> imo, it should be as easy as enabling wanted layers, set correct alpha and enable correct routes
<jernej> you should read also register descriptions in BSP source
<jernej> after all, they are GPL licensed :)
BenG83_PB has quit [Ping timeout: 240 seconds]
<MoeIcenowy> I wonder when will Allwinner release GPL headers for DE2 sources
<MoeIcenowy> he promised this when meeting with me
<MoeIcenowy> jernej: can layers in a single channel do alpha blending?
<TheLinuxBug> hehe, did he give you a date? If not that could mean 20 years from now..they still 'will' do it... just not on any realistic timeframe ;p
<jernej> hm, there is not much comments, but descriptive register names should help
<jernej> MoeIcenowy: There is not much blending, if you have only one channel?
<jernej> MoeIcenowy: Except maybe with background color
ErwinH has quit [Ping timeout: 260 seconds]
BenG83_PB has joined #linux-sunxi
<jernej> MoeIcenowy: I'm afraid you are on your own here, I have specialized more for a crtcs
<MoeIcenowy> ?
<jernej> HDMI, TV...
<jernej> well, more or less only on hdmi
<MoeIcenowy> I think HDMI and TV is only bridge over TCON's channel 1 ;-)
chlorine has quit [Remote host closed the connection]
<jernej> oh, sorry, hdmi is in this case called only connector in Linux DRM speak
<MoeIcenowy> P.S. you can do drivers for the bridges ;-)
<jernej> I would love to write HDMI bridge, if you don't mind :)
<jernej> I think that TV doesn't need a bridge
<jernej> it is exactly the same IP block, only on a different addrsss
<jernej> *address
<jernej> CSC, which is needed to display something, is part of DE2
<MoeIcenowy> oh I think I should buy a OPi Zero expansion board and a OPi's 3.5mm to TV cable
<MoeIcenowy> but only buy these two things cost too little money and will lead to a shippment fee around or even more than the boards' value
<jelle> oh even in china itself?
<MoeIcenowy> jelle: yes
<MoeIcenowy> the expansion board only costs ¥9.99
<MoeIcenowy> but the shippment costs ¥10
<jelle> ah same as here in europe
<MoeIcenowy> ¥10 is ~$1.6 US dollar
<jelle> well worse even extension card is ~ 2 euro, shipping is 3,37 euro >_>
<MoeIcenowy> or maybe I should buy the NAS expansion board? ;-)
<MoeIcenowy> (or just ask Steven for one?
ErwinH has joined #linux-sunxi
<MoeIcenowy> or maybe I can try to temporarily use jfm's bridge driver
<MoeIcenowy> (for a final bridge driver we should do a modified dw-hdmi.c
mossroy has quit [Quit: Leaving]
<jernej> I follow DRM maillist and it is interesting that there were two attempts already to add custom HDMI PHY
<jernej> one from Renesas and one from Amlogic
<jernej> it still didn't land
<jernej> MoeIcenowy: If you use jfm driver for hdmi, then you are limited with resolutions
<MoeIcenowy> he didn't fix this?!
<MoeIcenowy> sorry to hear that.
lynxis_ is now known as lynxis
<jernej> no, he didn't want to reuse dw-hdmi driver
<MoeIcenowy> maybe it's the reason that he abandoned.
<MoeIcenowy> but I think he should have fixed resolution issues in his own hdmi driver
<jernej> he also argued something about reusing sun4i code, iirc
<MoeIcenowy> yes his driver is difficult to co-exist with sun4i code
leviathan_ has joined #linux-sunxi
<jernej> so, I did a bit of a research and wrote U-Boot HDMI driver, which seems the easiest approach
<MoeIcenowy> and being denied by the DM wanters
<jernej> I'm not against DM, I actually support it
<MoeIcenowy> but it's currently different to do DM
<jernej> I'm half way in my effort to support it
<jernej> imo DM is very good attempt to avoid the mess we already have
leviathan has quit [Ping timeout: 240 seconds]
atsampson has quit [Ping timeout: 240 seconds]
atsampson has joined #linux-sunxi
<MoeIcenowy> P.S. seems that TV code called codepath exported by TCON code...
<MoeIcenowy> which means the TV code is binded with TCON code...
<MoeIcenowy> jernej: do H3 have clocks for tcon1?
<jernej> no
<MoeIcenowy> so it have only bus gate clock, no lcd channel clock?
<jernej> if you mean tcon, then yes
ErwinH has quit [Remote host closed the connection]
<jernej> also, TVE block requires fixed clock set to 216 MHz
<jernej> any other value doesn't produce any image
<jernej> That's why default DE clock value on H3 and H5 is 432 MHZ
vagrantc has joined #linux-sunxi
r1mikey has joined #linux-sunxi
<MoeIcenowy> what fixed clock?
<jernej> I mean, TVE block requires clock source to be set to exactly 216 MHz
ErwinH has joined #linux-sunxi
<jernej> and you have two possibilities, DE or PERIPH1 with divider
<MoeIcenowy> how to set it?
<MoeIcenowy> DE module clock?
<jernej> BSP and my driver use PLL_DE set to 432MHz with divider 2, which is set in CCU TVE
r1mikey has quit [Ping timeout: 255 seconds]
<MoeIcenowy> so you mean assigned-clocks = <&ccu CLK_TVE>; assigned-clock-rates = <216000000>; ?
<jernej> what assigned-clocks mean?
netlynx has quit [Quit: Ex-Chat]
<MoeIcenowy> assign a certain clock to a certain frequency
<MoeIcenowy> but as CLK_TVE is not CLK_SET_PARENT_RATE, we should also assign <&de2_clocks CLK_MIXER1> to 432000000
<jernej> then yes, I guess
ErwinH has quit [Ping timeout: 260 seconds]
chrishell has joined #linux-sunxi
<jernej> wouldn't be better to set that flag?
<jernej> but maybe 216 MHz is too low for de2
<MoeIcenowy> yes
<MoeIcenowy> so assign different clocks is the best way ;-)
komunista has quit [Quit: Leaving.]
<chrishell> hello, I have some issues building yocto on pine64 https://nopaste.me/view/13646bea. I followed this recipe: https://github.com/mont3z/meta-pine64. Build on a Debian unstable with AMD64 CPU
<chrishell> does that mean that the compiler does not support the right instruction set on sunxi ARM?
<chrishell> No Thrumb instruction set?
<vagrantc> you probably need a cross-compiler toolchain
<chrishell> I thougt this comes with gcc-multilib?
<vagrantc> chrishell: you got the toolchain from linaro, suggested in the README.md ?
<chrishell> yes I did
<vagrantc> i don't think gcc-multilib provides cross-toolchains
<Putti> is there a way to run boot.cmd in u-boot manually in the u-boot console?
<vagrantc> chrishell: you might need a toolchain for aarch64-linux-gnu instead of arm-linux-gnueabihf
<vagrantc> chrishell: since the pine64 is aarch64 (arm64)
<chrishell> vagrantc: neither do I, the info about gcc-multilib is not clear to me, how can I see which compiler is used by the Bitbake?
<vagrantc> i'm just speculating here, never worked with yocto before
<chrishell> all right that means another midnight oil for a long RTFM session, I want to learn Yocto anyway
<jernej> MoeIcenowy: Can your current DE2 DRM driver handle anything other than tcon0?
<jernej> MoeIcenowy: Or how do you plan to use TV? TV is always on TCON1
<MoeIcenowy> I made some patches locally which can make use of channel1-only TCONs
<BenG83_PB> MoeIcenowy, the crashes are not rank switch related, they just happen in other places/times if I switch off one rank
<MoeIcenowy> try to lower DRAM frequency?
<MoeIcenowy> I think you have different DRAM chip with me ;-)
<BenG83_PB> mmh
<MoeIcenowy> I remember you said you have Micron chips
<BenG83_PB> yes I do
<MoeIcenowy> I derived the memory rate from Pine64
<BenG83_PB> but wouldnt the crashes be more random with too high frequency?
<MoeIcenowy> stock DRAM freq of PB is 552
<jernej> MoeIcenowy: That's ok, but what about DT?
<MoeIcenowy> jernej: I haven't did the DT part
<jernej> MoeIcenowy: You probably have to have endpoints in DT?
<MoeIcenowy> yes
<MoeIcenowy> I have also no color correction code done yet
JohnDoe_71Rus has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/]
<MoeIcenowy> BenG83_PB: updated pinebook_defconfig
<MoeIcenowy> now you can use the new pinebook_defconfig to rebuild, then reflash the SPL
<BenG83_PB> thanks
premoboss has joined #linux-sunxi
BenG83 has joined #linux-sunxi
BenG83_PB has quit [Quit: Leaving]
chrishell has quit [Quit: Leaving.]
Andy-D has quit [Ping timeout: 240 seconds]
<jernej> MoeIcenowy: If I'm not mistaken, your driver supports cropping?
<jernej> MoeIcenowy: What about scaling?
<jernej> MoeIcenowy: I managed to make it work with U-Boot driver to have cloned output
<jernej> you have to set only two size fields differently, e.g. source size, output size
|Jeroen| has quit [Quit: dada]
f0xx has quit [Ping timeout: 240 seconds]
<BenG83> MoeIcenowy, this looks a lot better
<BenG83> no crashes so far
<BenG83> just built the wifi driver that works too
reinforce has quit [Quit: Leaving.]
<plaes> o/
<plaes> anyone aware how to use tables with SUNXI_CCU_MP_WITH_MUX_GATE ?
<plaes> CLK_OUTA_REG and CLK_OUTB_REG on A20
Andy-D has joined #linux-sunxi
IgorPec has quit [Ping timeout: 268 seconds]
premoboss has quit [Ping timeout: 260 seconds]
premoboss has joined #linux-sunxi
BenG83_PB has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
premoboss has quit [Ping timeout: 268 seconds]
paulk-collins has quit [Quit: Leaving]
scream has quit [Remote host closed the connection]
bonbons has quit [Quit: Leaving]
apritzel has joined #linux-sunxi
jernej has quit [Ping timeout: 260 seconds]
fl_0 has quit [Quit: STRG + Q]
Ntemis has quit [Remote host closed the connection]
lurchi_ has joined #linux-sunxi
fl_0 has joined #linux-sunxi
beeble has joined #linux-sunxi
msev- has quit [Quit: PanicBNC - http://PanicBNC.net]
BenG83 has quit [Ping timeout: 268 seconds]