theborger has quit [Quit: The only mystery in life is why the kamikaze pilots wore helmets]
y0gurt has joined #linux-sunxi
rav0__ is now known as rav0
hramrach has quit [Remote host closed the connection]
hramrach has joined #linux-sunxi
bhoj has joined #linux-sunxi
<bhoj>
hi !
<bhoj>
I have a Allwinner A31 tablet which I tried updating using phoenixsuite. Now the device is not booting and Its not being detected by phoenixsuite.
<bhoj>
It looks like android kernel is booting and then partitions are not getting mounted. if I try adb it says " adb shell "exec '/ system / bin / sh' failed: No such file or directory (2)""
<bhoj>
Is there anyway I can unbrick this device ?
<bhoj>
Can I boot it from sdcard ?
<ZaEarl>
bhoj, do you have a livesuit img?
<bhoj>
ZaEarl, no. I am new to allwinner platform. Any details about livesuit ?
<ZaEarl>
it
<ZaEarl>
it's a flashing tool that most manufacturers use. with a stock rom img, any allwinner device is unbrickable
<ZaEarl>
you'll need to find a rom for your device
<bhoj>
ZaEarl, are you referring to a working android image here ?
<ZaEarl>
yes, whatever the manufacturer created
bhoj has quit [Ping timeout: 248 seconds]
bhoj has joined #linux-sunxi
ZaEarl has quit [Ping timeout: 245 seconds]
rav0_ has joined #linux-sunxi
rav0 has quit [Ping timeout: 245 seconds]
eebrah has joined #linux-sunxi
<bhoj>
Does Allwinner A31 tablet boot from SDcard ?
<xenoxaos>
probably
n01|work|afk is now known as n01
<bhoj>
xenoxaos, I have a tablet which I was flashing with phoenixsuite and now it doesn't boot. looks like partitions are not mounted... I get adb shell "exec '/ system / bin / sh' failed: No such file or directory (2)" error ..
<bhoj>
phoenix suite is no more detecting it... I don't know how I can recover from this.
<bhoj>
and to boot from sdcard I would need u-boot sources for a31 .. which I am not able to find ..
<xenoxaos>
well....have you tried reflashing it over usb?
<xenoxaos>
power off completely, then hold, probably the home button, while plugging in usb otg?
<bhoj>
yes I have tried that ... It doesn't seem to go into flash mode.. It boots the kernel and fails to mount system partition ... adb at that point is working ...
<xenoxaos>
probably try all the buttons then?
<bhoj>
I don't even see a splash screen
<xenoxaos>
you wont....
<xenoxaos>
the recovery button/buttons might be different on different models
<xenoxaos>
then use livesuite or whatever to reflash
<bhoj>
how does livesuite work ? is there any documentation anywhere ?
<xenoxaos>
that's a good question for google
<bhoj>
xenoxaos, :) . googling currently ..
<oliv3r>
bhoj: i wouldn't be supprised if the A31 has the exact same boot probing as the A1*
<xenoxaos>
probably using the exact same bootrom
eebrah has quit [Ping timeout: 246 seconds]
rav0__ has joined #linux-sunxi
rav0_ has quit [Ping timeout: 245 seconds]
Quarx has joined #linux-sunxi
rellla has joined #linux-sunxi
bhoj has quit [Ping timeout: 264 seconds]
bsdfox_ has quit [Ping timeout: 245 seconds]
dolapevich has joined #linux-sunxi
bsdfox_ has joined #linux-sunxi
<dolapevich>
Hello all, I have an A10 media PC running debian
<dolapevich>
I am trying to compile a kernel for it
<mripard_>
and did you unify the other clocks as well?
<mripard_>
or just the osc for now?
<Turl>
mripard_: I could also unify apb1_mux and apb1, but I think that's it as regards unification
<mripard_>
and the gates?
<Turl>
mripard_: they have a different register to control the gates
<oliv3r>
Undertasker: yep :) the partition does not have a FS, a 'boot.img' is dd'ed into place. but yeah
<Turl>
besides I wouldn't know how to solve the clock interdependencies
<Turl>
mripard_: ahb depends on axi, but not any gates, and if I unify them I'd need a fake number on there to take a reference to the proper clock or something
<Undertasker>
Can I do this from a running system, or is it only possible from a pc with the tab in recovery mode?
<Turl>
Undertasker: you should be able to do it from a running system if you have root
<Undertasker>
I do
hansg has joined #linux-sunxi
<mripard_>
Turl: what fake number?
<Turl>
mripard_: say I unify axi and axi_gates
<Turl>
mripard_: then I can no longer do <&axi> from ahb to form the parent relationship
<Turl>
I need to do something like <&axi fakenumber>
<mripard_>
ahhh
<mripard_>
axi has a gate
<mripard_>
but ahb just passes through
<mripard_>
ok
<Turl>
mripard_: btw, what's the standard for internal usage function naming? double underscore?
<Undertasker>
oliv3r, what partition is the right one? /dev/block/nandd ?
<mripard_>
I don't know if that's possible, but maybe you could just try to get the cell size, and see from there if you need a gate or none
<mripard_>
but I don't know how dtc would handle that
<mripard_>
1 _, then 2, then 3, etc. :)
<mripard_>
btw, I still have that problem when earlyprintk and debug_ll are disabled
<mripard_>
on both the olinuxino and the hackberry
<Turl>
mripard_: are you by any chance using the same power source for the board and uart?
<mripard_>
I'm more and more enclined to think that I screwed the map_io up
<fra79Wii>
from some time I have the insane fixation to try to boot android from nand… with a 3.4 kernel.. I've tried replace u-boot to the lichee-dev branch, with no luck...
<mripard_>
well, the UART is on the SoC, so I hope the board and the SoC have the samep power source :)
Tartarus has quit [Ping timeout: 245 seconds]
<Turl>
mripard_: I meant the usb part :)
<Turl>
mripard_: if I plug both my board and the uart on a hub, the uart stalls all the time
<Turl>
it might be my crappy hub though
<fra79Wii>
Now I saw that there is the 3.3. branch for the A20… I suspect the nand related stuff are the same… could I cherry pick the commit I should apply to make a 3.4 bootable
<Turl>
3.4 should be bootable as is
<mripard_>
Turl: nope, it's a "real" power source, for the board at least
<mripard_>
plugged directly in an outlet
<fra79Wii>
Turl: hmm.. what I'm doing wrong then… it's a config problem.. I can boot from sd…
<mripard_>
Turl: and it works fine until the kernel boots :)
<oliv3r>
Undertasker: i think it was /dev/black/nandc
<oliv3r>
Undertasker: but i'm compiling your kernel now, so should be able to build the module for you ;)
<Turl>
mripard_: maybe I should move the gate and fixed kzalloc and setup inside composite, what do you think?
Slash__ has quit [Quit: Page closed]
<oliv3r>
Undertasker: what was that rtl chip you have again?
<fra79Wii>
Turl: Have you tried to boot android on nandd with a kernel 3.4?
<Turl>
fra79Wii: no, I haven't, but I've booted 3.4 and I know nand works
<fra79Wii>
I've seen booting linux by doing a repartition of the nand… I'd like to make a flashable kernel image with the 3.4 branch, and just swap the current 3.0.x version
<oliv3r>
Undertasker: there was no support for the rtl back in that version.
<fra79Wii>
I think the problem is some allwinner boot hack.. are missing in the sunxi-3.4 branch. I'd like to take those hacks from the A20 3.3 lichee branch :P… but I have no clue where to look (mach-sun7i.. or where else...)
<oliv3r>
fra79Wii: lichee-3.3/sun7i i belive
<oliv3r>
otherwise lichee-3.3/a20-dev
<oliv3r>
those should be the latest sources
<fra79Wii>
..hu, yes, I've seen that.. but the eventual boot option, in wich subfolders reside?
<oliv3r>
fra79Wii: for now, my a10 won't buut anything but stock kernel either; i have to see whats going on via uart
<oliv3r>
fra79Wii: doubt you'll find it ;)
<oliv3r>
3.0 even refuses to boot on my kernel
<oliv3r>
erm, tablet
<oliv3r>
it runs 3.0.23+ right now; but can't get 3.0.62+ to boot
<fra79Wii>
wow.. that's odd. I usually can boot any sunxi-3.0. Have you tried to change the u-boot binary?
Tartarus has joined #linux-sunxi
<oliv3r>
fra79Wii: yep using mnemoc's aw precompiled nand supporting one
<oliv3r>
on mmc it works fine btw
<oliv3r>
just stupid stuff broken and i can't 'see' whats wrong
<Undertasker>
oliv3r sorry was away. Rtl2832u is the chip
<Undertasker>
The driver was backported down to 2.6.x afaik
<Undertasker>
I found sources that were said to compile from 3.0 to 3.2, but unfortunately I can't try them now
ZaEarl has quit [Read error: Connection reset by peer]
<drachensun>
gzamboni: I sent you a PM about that driver
Undertasker has joined #linux-sunxi
hansg has quit [Quit: Leaving]
n01 has joined #linux-sunxi
dolence has quit [Read error: Connection reset by peer]
vinifm has quit [Quit: Saindo]
Undertasker has quit [Read error: Connection reset by peer]
Undertasker has joined #linux-sunxi
sankasan has joined #linux-sunxi
dolence has joined #linux-sunxi
sankasan has quit [Quit: Leaving]
sankasan has joined #linux-sunxi
FergusL is now known as Kush
Kush is now known as FergusL
vinifm has joined #linux-sunxi
ZaEarl has joined #linux-sunxi
shineworld has joined #linux-sunxi
sankasan has quit [Quit: Leaving]
rav0_ has joined #linux-sunxi
shineworld has quit [Client Quit]
rav0__ has quit [Ping timeout: 245 seconds]
shineworld has joined #linux-sunxi
tinti has quit [Read error: Connection reset by peer]
paulk-desktop has quit [Quit: Ex-Chat]
<Turl>
hramrach: ping
<hramrach>
Turl: pong
<Turl>
hramrach: I just saw your patchset
<hramrach>
and?
<Turl>
you're supposed to write a subject and a description when launching email with cover letter :)
<hramrach>
I did. It just sent the backup too :)
Undertasker has quit [Quit: Bye]
<Turl>
heh
<hramrach>
if that's the only objection that's cool :)
<Turl>
hramrach: your emails also seem to lack the references to the other emails that create the 'subtree' view thingy
<Turl>
I didn't look at the actual code, fwiw :P
<hramrach>
you have to manually add references with git so next to nobody has them
<Turl>
manually?
<Turl>
git adds them for me when I send all my commits :P
<hramrach>
and there is nothing to reference either, logically
<Turl>
do you format-patch and then send the files?
<hramrach>
yes
<hramrach>
yes, it should make a list or 1-level tree if you specify --thread
<hramrach>
I wonder why google threads some together and not others without the refs
FergusL has left #linux-sunxi ["Quit"]
eebrah has quit [Ping timeout: 264 seconds]
<Turl>
gmail uses the title
<Turl>
I use thunderbird for my email though, so I get proper threads
<techn__>
hramrach: did you compare sizes of hwpack before and after that kitchensink stuff?
<hramrach>
no.
<hramrach>
I expect tehm grow a lot ;-)
<Turl>
hramrach: if you use git "send-email --annotate --cover-letter starthash^..endhash" you get the threading for free, stuff to write your cover letter and hassle free patch review :)
<Turl>
*"git send email.."
<Turl>
and with the dash too, I'm faily today
<hramrach>
format-patch does not have annotate
<Turl>
hramrach: "Unify usb peripherial controller choice." looks.. wrong; why switch everything to bool?
<hramrach>
send-emails is not going to work
<hramrach>
Turl: because the damn kconfig can't get that a choice can have both bool and tristate options
<hramrach>
did you read the patch message?
<Turl>
annotate opens every patch on your editor so you can review them and add a comment or something before sending
<Turl>
hramrach: yeah but why is there mixed options in the first place?
<hramrach>
because sunxi usb cannot do module
<Turl>
make it tristate anyway and leave it broken until someone fixes it? :)
<hramrach>
the patchset is not meant to break the kernel
<hramrach>
if you think it's the right way suggest it in a reply and if everyone agrees it's easy to make a different patch ;-)
<Turl>
hramrach: how does it break btw? I don't recall kconfig exploding or anything when I configured usb
<hramrach>
did you configure it as module?
<hramrach>
I expect that to explode
<Turl>
not really, I'm an =y/n guy
<Turl>
I disable all modules except the wifi one
<hramrach>
it's not possible even but it's possible for all the other controllers
<Turl>
hramrach: re. "Make OTG the default since it cannot be selected by defconfig."
<Turl>
hramrach: why are you moving unrelated stuff around? or is that diff being funny?
<hramrach>
what is unrelated stuff?
<Turl>
USB_FUSB300, other renesas controller stuff, resorting includes on another Kconfig file
<hramrach>
that's not unrelated
<hramrach>
that makes the sunxi controller the default
<hramrach>
that thing is sorted among platform controllers and has no platform dependency which is bogus
<hramrach>
and changing the include order is not unrelated either
dolence has quit [Quit: Saindo]
<hramrach>
it seems you cannot select something that's not been included yet
<hramrach>
well, I cannot make it work as tristate
<hramrach>
possible because it has cyclic dependency with also bogus choice of USB_SW_SUN4I_USB0_*
<hramrach>
if you have better solution for that sunxi usb mess I would like to see but since the driver is bogus I don't care about the configuration reflecting that too much
shineworld has quit [Quit: Leaving]
n01 has quit [Ping timeout: 248 seconds]
vinifm has quit [Quit: Saindo]
<oliv3r>
Turl: --annotate launches your editor to write the cover-letter?
<oliv3r>
hramrach: we desperatly need a new usb driver
<oliv3r>
:)
Tartarus has quit [Excess Flood]
Tartarus has joined #linux-sunxi
<rm>
Subject: [linux-sunxi] [PATCH 3.4 0/4] *** SUBJECT HERE ***
<rm>
aw forget it
<rm>
I give up on making a witty yet not too insulting comment
<rm>
maybe this will do: *** JOKE HERE ***
<rm>
:))
<dolapevich>
:P
<dolapevich>
rm, after some testing it looks like if you don´t compile sunxi modules static into kernel they fail
<dolapevich>
Yesterday I came here asking for an unresolved module on SND_SUN4I_SOC_SPDIF
<rm>
yes
<rm>
quality allwinner code
<rm>
afk
<dolapevich>
ok
<hramrach>
dolapevich: that shoould work
<hramrach>
I have that option with m too
dolapevich has quit [Read error: Connection reset by peer]
<hramrach>
and htere is check for unresolved symbols on build so would fail