nighty-_ has quit [Quit: Disappears in a puff of smoke]
levd has joined #linux-rockchip
levd1 has joined #linux-rockchip
levd has quit [Ping timeout: 244 seconds]
levd has joined #linux-rockchip
levd1 has quit [Ping timeout: 255 seconds]
prop395 has joined #linux-rockchip
naobsd has joined #linux-rockchip
prop395 has quit [Ping timeout: 246 seconds]
levd1 has joined #linux-rockchip
levd has quit [Ping timeout: 244 seconds]
akaizen_ has joined #linux-rockchip
akaizen_ is now known as akaizen
markm has joined #linux-rockchip
markm has quit [Quit: Leaving]
levd has joined #linux-rockchip
levd1 has quit [Ping timeout: 252 seconds]
levd1 has joined #linux-rockchip
levd has quit [Read error: Connection reset by peer]
levd has joined #linux-rockchip
levd1 has quit [Ping timeout: 244 seconds]
levd1 has joined #linux-rockchip
levd has quit [Ping timeout: 252 seconds]
naobsd has quit [Ping timeout: 240 seconds]
levd has joined #linux-rockchip
levd1 has quit [Ping timeout: 260 seconds]
naobsd has joined #linux-rockchip
Zafar has joined #linux-rockchip
* bashintosh
is working on Android 5.1 for the Firefly..
GriefNorth has joined #linux-rockchip
Zafar has quit [Ping timeout: 246 seconds]
levd1 has joined #linux-rockchip
levd has quit [Ping timeout: 252 seconds]
akaizen has quit [Remote host closed the connection]
levd has joined #linux-rockchip
levd1 has quit [Ping timeout: 244 seconds]
levd has quit [Remote host closed the connection]
levd has joined #linux-rockchip
GriefNorth has quit [Read error: Connection reset by peer]
lautriv__ has joined #linux-rockchip
lautriv_ has quit [Ping timeout: 246 seconds]
levd has quit [Remote host closed the connection]
levd has joined #linux-rockchip
nighty-_ has joined #linux-rockchip
apritzel has joined #linux-rockchip
levd1 has joined #linux-rockchip
levd has quit [Ping timeout: 240 seconds]
levd1 has quit [Ping timeout: 240 seconds]
_whitelogger has joined #linux-rockchip
<sjoerd>
sanjay_: Don't ask to ask, just ask
<sanjay_>
lol sorry
<sanjay_>
i have en MemoPad 10 (ME102A) bricked
<sanjay_>
with RK3188
<sanjay_>
the NAND is bad
<sanjay_>
rkandroidtool, rkflashkit, rkflashtool not working
<sanjay_>
fastboot is not unlockable, only stock recovery
<sanjay_>
is RK3188 sdcard bootable ?
<sjoerd>
yes
<sanjay_>
can't apply any update.zip officiel, assert file error, or apk file check error (system is empty i think)
<sanjay_>
because on androtab, i saw that ME102a is RK3188 but it's particular
<sanjay_>
but for sdcard boot
<sanjay_>
i must put parameter / kernel.img and boot.img from my official firmware ?
<sanjay_>
or sdboot_rk3188_miniroot.zip file ?
* sjoerd
doesn't know anything about the official firmwards
<naobsd>
sanjay_: how did you bricked it? you flashed firmware image for different device?
<sanjay_>
my wife has apply an update with no battery
<sanjay_>
so ME102a has shutdown on update
<naobsd>
sanjay_: btw if you have a lot of experiences with non-rockchip Android devices, please forgot it at first e.g. "fastboot is unlockable" is useless for rockchip devices
<sanjay_>
rkandroidtools return "download file failed" on system.img, boot.img all file. I can flash only parameter
<naobsd>
sanjay_: then you can enter fastboot mode with VOL-, and rkusb mode with fastboot reboot-bootloader, right?
<sanjay_>
rkflashkit return "flash memory at xxxx is differnt from file"
<sanjay_>
yes
<sanjay_>
fastboot unlock return
<naobsd>
no need to do anything for fastboot unlock
<naobsd>
forgot it
<sanjay_>
"oem admmunlock not requested"
<naobsd>
not required at all
<sanjay_>
so i can't write clockworkmod
<naobsd>
please forgot it
<naobsd>
otherwise I cannot talk anymore
<sanjay_>
i can boot in bootload rockchip
<naobsd>
cwm can be flashed without fastboot unlock
<sanjay_>
rkflashkit, rkandroidtool and rkflashtool reconize my device
<naobsd>
please forgot about fastboot unlock at all
<sanjay_>
ok
<naobsd>
then
<sanjay_>
rkandroidtool can't flash cwm
<sanjay_>
same error
<naobsd>
there is no official system.img for memo pad 10
<sanjay_>
yes
<sanjay_>
there is one
<sanjay_>
on asus website
<sanjay_>
zip contain
<sanjay_>
system.img
<sanjay_>
parameter
<sanjay_>
boot.img
<naobsd>
ah,
<naobsd>
sure
<sanjay_>
kernel.img
<sanjay_>
recovery.img
<sanjay_>
but
<sanjay_>
all fail
<sanjay_>
"Download file failed"
<sanjay_>
only parameter can be flash, then reboot ME102a but no change
<sanjay_>
ERASE IDB return successfull but reboot tablet in normal mode after flash
<sanjay_>
and no change
<naobsd>
eMMC might be damaged...
<sanjay_>
i think that partition is bad
<sanjay_>
or bootloader has be update, but not other partition ?
<sanjay_>
i have try to boot into MASKROM
<sanjay_>
but can't find the pin
<naobsd>
if you know something about non-rockchip android device, please forgot it too for now
<sanjay_>
MASKROM is for rockchip ? no ?
<naobsd>
working loader and parameter should enough for flashing other partitions
<sanjay_>
bootloader.img return an IDB error
<sanjay_>
IDB download failed or something like this
<naobsd>
2nd command need to be run just after 1st one
<naobsd>
mmmm
<naobsd>
oen -> oem
<naobsd>
anyway
<naobsd>
I'm not sure about loader in latest zip
<naobsd>
sorry
<naobsd>
let's return to home now
<naobsd>
later
<naobsd>
there is small possibility that asus locked loader mode... I have interest to try me102a, but it's not available in JP
<naobsd>
bye
naobsd has quit [Quit: naobsd]
naobsd has joined #linux-rockchip
naobsd has quit [Quit: naobsd]
<rperier>
just in case, do you know a container in the linux kernel which allows to share the same object between different containers ? it is not possible with a struct list_head. I would like to send a message from a process to another without copy the data each time, just the structure inside the container. I plan to use reference counting to free the object transparently
ssvb has quit [Ping timeout: 264 seconds]
<sanjay_>
noabsd => the rklashtool b 3 is not working
GriefNorth has joined #linux-rockchip
ssvb has joined #linux-rockchip
<sjoerd>
mmind00: btw did you ever get the usb uart stuff to work ?
<mmind00>
sjoerd: it did and does work fine ... just the form was contested - where it should live
<mmind00>
sjoerd: I need to dig my way out of uboot first though :-) ... but I hope to give it another go in time for 4.4
<sjoerd>
mmind00: what board was that on ooi?
<mmind00>
I'm using this regularly on my jerry chromebook
* sjoerd
relays to Kevin who was asking on #kernelci :)
<sjoerd>
mmind00: cool
<sjoerd>
mmind00: does that work over all usb host ports or only the dwc ones?
<mmind00>
sjoerd: that is a property of the usb phy (routing the uart signals to d+ and d- pins)
<sjoerd>
aha
<mmind00>
sjoerd: on jerry it is the port next to sd slot btw
<sjoerd>
I should try it on my rock2 at some point
<sjoerd>
would be intersting once it's in a case as that makes the serial pins less accessible
<mmind00>
:-) ... also a nice feature when you need to fight for uart pins, like on the rk3368 where the debug uart shares pins with an mmc controller
<sjoerd>
heh
<maz>
mmind00: ah, that's exactly what I need to start playing with u-boot on speedy.
<maz>
mmind00: Yes, I found it. I need to do something similar for u-boot so that I can trace things as early as possible (well, just after leaving coreboot).
<rperier>
so uboot mainline seems to work fine on veyron... nice
khilman has joined #linux-rockchip
Nanosonde has joined #linux-rockchip
<Nanosonde>
Hi! How do I deactivate the dwmmc or the slot the eMMC is connected to? It is a RK3188 based Tablet (Odys Gate). I would like to use SD boot.
<mmind00>
Nanosonde: for nands it's always shorting two pins (8 and 9), but not sure if this also works for emmcs ... but as the emmc/nand reading is happening in the maskrom I don't think there is a permanent solution
<Nanosonde>
The memory is a Samsung KLM eMMC type (16GB). The problem is that it is a BGA package and not a TSOP package as it is commonly used for NAND.
<Nanosonde>
"If there is onboard eMMC available it will take priority over SD Card when booting a precompiled kernel preventing a boot from SD Card. Similarly, bad eMMC can prevent the device from booting from SD Card any more. To avoid/solve these issues, deactivate the dwmmc or the slot the eMMC is connected to. "
<Nanosonde>
The question is: what is meant by saying "deactivate the dwmmc"? Is it the peripherial that is inside the RK3188?
<Nanosonde>
Or is it a driver?
<sjoerd>
there is no driver you can disable at the point where you boot
<Nanosonde>
exactly.
<sjoerd>
I thought the soc rom looked at SD first though on 3188 but i'm probably wrong
<sjoerd>
unless there is say a test pad for the emmc clk line you could short to ground there isn't much else you can do then to overwrite the loader in emmc
<sjoerd>
which will make it try sd
JohnDoe_71Rus has joined #linux-rockchip
<Nanosonde>
sjoerd, yes you are right. The PCB layout is very compact. I hope I can find some signal which I can "disturb" somehow so that eMMC is not recognized anymore.
<lautriv__>
sjoerd, 3188 is used to look on SD first, maybe it could not read/use what you had there ?
<sjoerd>
I know it looks at SD before NAND, does it do SD before eMMC as well ?
<Nanosonde>
However, the SD card was not recognized. Instead the normal Android boot took place.
<sjoerd>
Nanosonde: erasing the first 4 megabytes or so is probalby a good bet
<Nanosonde>
Can you give a hint on how to do a complete(!) backup of the eMMC content first (incl. bootloader, idb, etc.)?
<Nanosonde>
I have used to backup rkflashtool to do a backup of every partition. Is this enough? I am unsure on how to create a full IDB backup.
* sjoerd
does not know, never done that
<Nanosonde>
I have an "update.img" file from the tablet manufacturer for use with the rkbatchtool. Maybe this enough.
<Nanosonde>
It is supped to contain the factory image.
<Nanosonde>
Is it correct that the rkbatchtool shows connected devices which are in MASKROM mode in blue color and normal bootloader mode in green color?
<lautriv__>
maybe someone changed things on this box but mine looks on SD even before emmc ...
<Nanosonde>
mhmm. Are there any pins on the RK3188 that are read in during reset/boot and configure the boot behaviour?
<Nanosonde>
I am just reading the RK3188 Technical Refernce Manual. It says this order: RESET->romcode->check ID block from NAND->check ID block from SPI nor flash->check ID block from external eMMC.
<Nanosonde>
It does not even mention SD. |confused|
Nanosonde has left #linux-rockchip [#linux-rockchip]
dlezcano has quit [Ping timeout: 250 seconds]
dlezcano has joined #linux-rockchip
akaizen has joined #linux-rockchip
gb_master has joined #linux-rockchip
GriefNorth has joined #linux-rockchip
else- has joined #linux-rockchip
<akaizen>
bashintosh: awesome, I just got the repo last night
<lautriv__>
akaizen, what repo ?
VargaD has quit [Ping timeout: 260 seconds]
VargaD has joined #linux-rockchip
<rperier>
mmind00: using your driver armsoc-rockchip from an archlinuxarm, xorg try to load "armsoc_dri.so" from /usr/lib/xorg/modules/dri/
<rperier>
I have a graphical session but even with veyron-libgl and mali userspace library , AIGLX complains about this missing module
<mmind00>
rperier: of course ... glx stuff wants GL ... but the mali-driver of course only brings GLES ... but it should drop you into software-rendered GL
<rperier>
exactly, this is what happens
<rperier>
but... you did not say earlier that gl rendering works just fine with a chromeos kernel ?
<mmind00>
rperier: nope
<rperier>
:D
<mmind00>
rperier: with the mali drivers you only get GLES ... there is glshim, which does GL->GLES (for a GL subset)
<rperier>
okay okay, my bad...
* lautriv__
raises an eyebrow ...
<lautriv__>
mmind00, i assume that mentioned driver is userspace , how does it apply ?
<mmind00>
lautriv__: downloading a .run file, extracting proprietary libGLES*.so and moving them into position ... or on a debian userspace, building a package from https://github.com/mmind/mali-driver and install that
<mmind00>
lautriv__: of course the whole thing is rk3288-only at the moment
<lautriv__>
mmind00, that is what i wondered about because i checked for 3188 a few days ago and there was not even a framebuffer ...
<mmind00>
lautriv__: fbdev is deprecated, so I don't think anybody will work in this area ... we have a drm/kms driver for Rockchip ... which would need to be extended for the older chips without iommus
akaizen has quit [Remote host closed the connection]
<amstan>
mmind00: glshim? that's interesting
<amstan>
i wonder how that compares to mesa
<mmind00>
amstan: bad ... GL1.2 mainly and a little bit of GL2 at the moment
<mmind00>
amstan: enough for extremetuxracer, but not to much more ;-)
<amstan>
aren't we at 4?
<mmind00>
yep :-)
<rperier>
seriously... ^^
<lautriv__>
mmind00, in other words it would not make sense to run a 3188 on vanilla the next few years ...
<mmind00>
lautriv__: it all depends on someone implementing the cma-based memory handling
<mmind00>
lautriv__: but sadly you're probably right for the near future if visual output is needed [I remember some people using radxa rock as server-thingy though]
<rperier>
did you test with a GLES vm ? like enlightenment or kwin ?
<rperier>
s/vm/wm/
<lautriv__>
mmind00, i'm used to screw on thingies for wondermedia where we have not even a datasheet (NDA) but a framebuffer. This 3188 was choosen because i found more info and had the impression it is better supported. Unfortunately it appears the box in question is not even rootable :(
gb_master has quit [Quit: Leaving]
<mmind00>
rperier: plasma5 does use it a bit [couldn't get the old kwin to use it] ... although it fails often with strange EGL errors
<akaizen>
mmind00: cool, and it looks like they are using DRM/KMS in that driver, is that right? What's the user space rendering library? Freon / Aurora?
<mmind00>
akaizen: that is for the chain drm/kms -> x11