nighty-- has quit [Remote host closed the connection]
nighty-- has joined #linux-rockchip
bbelos has quit [Ping timeout: 260 seconds]
ganbold has joined #linux-rockchip
nashpa has quit [Quit: Going away]
nashpa has joined #linux-rockchip
nashpa has quit [Quit: Going away]
nashpa has joined #linux-rockchip
lkcl has quit [Ping timeout: 272 seconds]
lkcl has joined #linux-rockchip
cnxsoft has joined #linux-rockchip
techping has joined #linux-rockchip
techping has quit [Remote host closed the connection]
lkcl has quit [Ping timeout: 240 seconds]
wzyy2 has joined #linux-rockchip
_whitelogger has joined #linux-rockchip
zzarr_ has quit [Quit: Lämnar]
wadim_ has joined #linux-rockchip
<micken>
ayaka: I have my HDMI driver now, but it seems like every write to hdmi ends up in void.. , right now I am monitoring the phy slave address, and it isn't set. I also tried to set it with our memory tool, but it refuses the write (no error) just doesn't get there.
<micken>
ayaka: I assume that I write to the write address ff98c080
<micken>
ie x4 the offset
matthias_bgg has quit [Remote host closed the connection]
matthias_bgg has joined #linux-rockchip
matthias_bgg has quit [Remote host closed the connection]
<micken>
ayaka: ok was'nt in svc when I did the physical write, tried at the top of our kernel before MMU and it worked. So it is a problem with mmu mapping on our side. sorry .
matthias_bgg has joined #linux-rockchip
techping has joined #linux-rockchip
techping has quit [Ping timeout: 260 seconds]
<stdint>
he may be here at work time
ganbold has quit [Ping timeout: 260 seconds]
ganbold has joined #linux-rockchip
stdint has quit [Remote host closed the connection]
stdint has joined #linux-rockchip
nighty-- has quit [Quit: Disappears in a puff of smoke]
wzyy2 has quit [Ping timeout: 246 seconds]
nighty-- has joined #linux-rockchip
cyteen has quit [Quit: Leaving]
cyteen has joined #linux-rockchip
wzyy2 has joined #linux-rockchip
mrjay has joined #linux-rockchip
wzyy2 has quit [Quit: Konversation terminated!]
wzyy2 has joined #linux-rockchip
afaerber has joined #linux-rockchip
lkcl has joined #linux-rockchip
tx0h has joined #linux-rockchip
<tx0h>
hi
<tx0h>
my new toy: yundoo y8 (a rk3399 tv-box). i have a few questions and hope for some answers and advice. the hardware is more or less equal to the firefly rk3399, can i use the firefly .dts for the kernel configuration? when i build the distribution with rockships yocto, do i use the android recovery to flash the linux image from sd card or via usb adb?
cyteen_ has joined #linux-rockchip
cyteen has quit [Read error: Connection reset by peer]
LongWork has quit [Ping timeout: 246 seconds]
nighty-- has quit [Read error: Connection reset by peer]
<wzyy2>
; ） You need find how to enter maskrom to avoid brick.
<phh>
meh, just shortcut some random pins near eMMC, always worked fine for me :P
LongWork has joined #linux-rockchip
<tx0h>
oh hell, thats some answer for me, right?
<tx0h>
phh: isn;t it to lower the clk pin? ie connect the clk pin to ground?
<phh>
well if you have the clk pin easy to access, then yes, just that
<tx0h>
i guess i have no clue where i can find the pin.
nighty-- has joined #linux-rockchip
<tx0h>
emmc are so much more complicated compared to a microsd
<tx0h>
but before it comes to flashing: is it a good way to go to derive the kernel config from the firefly rk3399 to get a good kernel for the yundoo y8? it seems that just the ethernet interface is different
<tx0h>
i mean, can i use the firefly device tree source and modify just the ethernet port?
<phh>
yes firefly sounds good
<phh>
you'd need to check the PMIC to be sure
<phh>
wzyy2: in rk tvbox reference design, there is only one pmic, or multiple pmic supported?
<phh>
tx0h: like kernel source from the manufacturer?
<phh>
tx0h: you know where the emmc is at least?
<tx0h>
phh: no, i just decompiled a device tree from the running android box. which is a bit different because it is all things in one dts (no dtsi's or so)
<phh>
tx0h: right, decompiled is a way
<tx0h>
sure, on the images it is on the top
<phh>
a bit more complicated though
<tx0h>
south from the sd card slot
<phh>
yup yup seen
<phh>
I'm seeing no pin screaming clk test pin :s
<tx0h>
:-)
wzyy2 has quit [Ping timeout: 260 seconds]
<tx0h>
but shorten the clk pin is required for a success, right?
<tx0h>
or i have this miniloader from rockchip, right?
<phh>
it's required to have a way back when you flash something that doesn't boot :P
<tx0h>
the problem with the emmc is, that the contacts are below the chip, right? so if there is no clk pad designed there is no way to shorten it?
<phh>
oh actually underneath there are many test pins
<phh>
there are two test pins at the height of in-between usb and hdmi connector
<phh>
which could be under emmc if my brain geometry works
<tx0h>
thats true, on the back side there are some pins. but i require, according to the specs, to shorten pin M6 :-)
<phh>
there is nothing written on the test pins?
<tx0h>
thats the current images
<tx0h>
security by obscurity
<phh>
I suggest putting a dc-meter, put heavy I/O charge, the clk pin will be ~ 1V
<tx0h>
i have a oscilloscope
<phh>
though I think that myself I'd be courageous enough to just shortcut them without even checking :P
<tx0h>
i'm not so brave, more a chicken
LongWork1 has joined #linux-rockchip
LongWork has quit [Ping timeout: 268 seconds]
<tx0h>
but according to Miniloader Rockusb mode described on http://opensource.rock-chips.com/wiki_Rockusb i don't need to maskrom (shorten the pin), when the current system runs with the rockusb miniloader.
LongWork1 has quit [Ping timeout: 246 seconds]
<tx0h>
the pitfall here is, to figure out the right mmc offset
<phh>
again, you need to shorten the pin only when you brick your device
LongWork has joined #linux-rockchip
hardik_ has joined #linux-rockchip
<hardik_>
hi all, how access partition 8 an 9 created using mtdpartitions CMDLINE:earlyprintk=ttyFIQ0,115200n8 console=ttyFIQ0 init=/sbin/init initrd=0x62000000,0x00800000 root=/dev/mmcblk0p7 rw rootwait rootfstype=ext4 quiet loglevel=0 lpj=8157341 mtdparts=rk29xxnand:0x00002000@0x00002000(uboot),0x00002000@0x00004000(misc),0x00006000@0x00006000(kernel),0x00006000@0x0000c000(boot),0x00010000@0x00012000(recovery),0x0001E000@0x00022000(s
<tx0h>
hardik_: do you made a filesystem on the partition? mkfs.ext4 or something?
<hardik_>
tx0h: yes /mnt is there
<tx0h>
but invalid argument sounds really a bit strange
<hardik_>
no i don't have mkfs command
<hardik_>
* on buildroot
<hardik_>
tx0h: u have any idea ...how to do it in buildroot rootfs
<tx0h>
hardik_: but you need to have a filesystem on the partition and the kernel needs to support the filesystem type (vfat, ext4, ntfs...)
<hardik_>
my rootfs file system is ext4 type
<tx0h>
hardik_: sorry, my last buildroot experience is about 15 years ago
<hardik_>
tx0h: buildroots was not exist at that time.......i think............
<tx0h>
or was it 10 years.. about
<tx0h>
forgot
<hardik_>
so can we give any option at kernel command line ....like we are giving for rootfstype=ext4
<tx0h>
yeah, 12 years old now..
<hardik_>
tx0h: Ok....:) .......i wonder how all rockchip people are using data and backup partitions
<tx0h>
hardik_: if the partition didn't contain a filesystem, you can't mount it
<hardik_>
mostly in android ....i haven't tried but............
<tx0h>
google does to linux what apple does to bsd - obfuscating simple operating systems
<phh>
sorry, what?
<hardik_>
tx0h: yes...that's right........but i need a way to format and access this memory area
cnxsoft has quit [Quit: cnxsoft]
<tx0h>
hardik_: can't you build e2fsprogs or mount / copy the binaries from another source?
<tx0h>
phh: well i think osx/ios and android are bad versions of linux and bsd
<phh>
tx0h: android is much ahead of most linux distros in term of security for instance
<phh>
and saying "google does to linux..." while they do chromium os which is one of the biggest effort of mainlining embedded systems is harsh
<tx0h>
phh: yes, it disables the owners of the boxes of using and maintaining it. but it's just my point of view. make the os'es a matrioska didn't mean it's safe, it's obfuscated and hard to maintain.
<phh>
tx0h: yes, Google devices are locked, and the users can't build the firmwares themselves...
<phh>
oh wait.
<tx0h>
phh: i'm happy that this yundoo devices is rooted.
<tx0h>
whatever root means in android
<hardik_>
tx0h: e2fsprogs is available in buildroots ...let me enable this ..
<tx0h>
hardik_: this would give you mkfs.ext4
<tx0h>
phh: ok, call me old fashioned, but i like plain unix/linux and i like system v init.d :-)
<tx0h>
ok, i'm afl
<tx0h>
afk
<hardik_>
tx0h: now it's "# mount /dev/mmcblk0p8" "mount: can't find /dev/mmcblk0p8 in /etc/fstab"
<hardik_>
i have formatted this successfully with command " mkfs.ext4 /dev/mmcblk0p8"
<hardik_>
tx0h: u were asking me why i create "system" partition twice ...... so wheever i boot from recovery image..it's creating one extra partition named as "parameter" so i had to change root=/dev/mmcblk0pX
tx0h has quit [Quit: ChatZilla 0.9.93 [SeaMonkey 2.46/20161213175921]]
<micken>
ayaka: OK. So I have the driver now, and it updtaes the registers correctly I think. The only thing I am worried about is the clocks. U-boot doesn't set them in the HDMI code at all, it only does a reset via CRU. The CRU lists two clocksettings related to HDMI. What do I need?
<ayaka>
micken, let me check it later
<ayaka>
it is very late here
<micken>
ayaka: alright np
<ayaka>
micken, you mean clk_hdmi_hdcp and hdmi_cec_clk_gate_en ?
<micken>
ayaka: yes
<ayaka>
micken, none of them looks mandatory
<micken>
ayaka: on the i.mx 6 you need to configure system clocks to output to hdmi
<ayaka>
most of clocks is controller by vop
<micken>
ah
<micken>
so are you saying that putting data on VOP against HDMI (enabling hdmi output) will clock hdmi?
<micken>
I don't get sync , so thats why I wonders
<micken>
But I don't put anything on VOP yet, just doing FC_DBGFORCE
<micken>
with blue color
amstan has quit [Ping timeout: 240 seconds]
cyteen_ has quit [Remote host closed the connection]