<mnemoc>
can you boot the stock system and echo something to /dev/ttyS0 ?
<mnemoc>
and see if you can see it on your uart?
itamarjp has joined #arm-netbook
itamarjp has quit [Changing host]
itamarjp has joined #arm-netbook
<mnemoc>
s/it/if/
<techn>
I dont have uart :D
<ibot>
mnemoc meant: and see if you can see if on your uart?
<mnemoc>
doh
<mnemoc>
then don't console=ttyS0
<mnemoc>
only console=tty1
<mnemoc>
and init=/bin/sh
<techn>
yes. yes. :)
<mnemoc>
and get a keyboard
<techn>
and I'm having bin/sh.. but init.d is not somewhy progressing
<mnemoc>
init=/bin/sh bypasses all the normal init process
<techn>
ah :)
<mnemoc>
but you should be able to mount and edit files
<mnemoc>
to get the init of your distro working
DEAT_ is now known as DEAT
eh2scqw has left #arm-netbook [#arm-netbook]
<techn>
ok.. I removed that init=/bin/sh, and got the real errors.. "mount: wrong fs type, bad option, bad superblock on none, missing codepage or helper program, or other error " ..
<techn>
compared to a10 device which has same userspace, EXT4-fs (mmcblk0p2): re-mounted. Opts: (null)
<techn>
hmm.. I'm propably missing some debian/ubuntu kernel settings :/
cat_n9 has quit [Ping timeout: 264 seconds]
<mnemoc>
running the same version on both?
itamarjp has quit [Ping timeout: 272 seconds]
<techn>
yep
pawel5870 has joined #arm-netbook
<techn>
hmm.. I enabled couple fs types and now it boots to X :)
<techn>
propably tmpfs was the problem? :/
pawel5870 has quit [Client Quit]
<techn>
Now.. I'll need to findout how to enable goodix_touch_811 module
<mnemoc>
techn: eh? defconfig lacked of tmpfs?
<techn>
-# CONFIG_TMPFS is not set
<techn>
+CONFIG_TMPFS=y
<techn>
+CONFIG_TMPFS_POSIX_ACL=y
<techn>
+CONFIG_TMPFS_XATTR=y
<mnemoc>
ouch
<mnemoc>
can you make a commit to make a13_defconfig closer to sun4i_defconfig (and functional)?
<techn>
I'll evening latest.. currently trying findout what's working and not :p
<techn>
also should we have separate crane config?
cat_n9 has joined #arm-netbook
pawel5870 has joined #arm-netbook
ssalevan has quit [Quit: Leaving.]
cat_n9 has quit [Ping timeout: 260 seconds]
<mnemoc>
techn: yes, but I think it's called "nuclear" for A13
<mnemoc>
(nickname of the devkit)
pawel5870 has quit [Quit: Leaving.]
cat_n9 has joined #arm-netbook
ibrah has joined #arm-netbook
pawel5870 has joined #arm-netbook
ibrah has quit [Ping timeout: 244 seconds]
ssalevan has joined #arm-netbook
merbzt has joined #arm-netbook
<rz2k>
I confirm that, my a13 tablet has nuclear or nuclear-pfdq88 in places where sun4i crane where
<merbzt>
is there some faq where I can ready about the issues with the cedarx chip and it's support libs, right now I don't understand why things are at a standstill
<merbzt>
the only real issue I heard is that the decoder isn't fast enough and that people get memleaks when using the decoder
eflatun has quit [Ping timeout: 272 seconds]
<rz2k>
merbzt: please write to arm-netbook mailing list, there was a guy who worked on h264 decode with cedarx and he had same issues.
<merbzt>
well maybe I have to
<merbzt>
I was hoping someone here could enlighten me
<rz2k>
also couple days ago hipnoi said that he passed info about mem leaks to guy who worked on this libs.
<rz2k>
maybe fixes are coming, maybe not.
<merbzt>
I saw someone working on RE'ing the libcedarxalloc.a lib
eflatun has quit [Remote host closed the connection]
<hno>
I think a large contributing factor to the cedarx issues is that Allwinner never released sources to aworking reference player, only a binary player, partial example sources, libs and some documentation.
ceo16 has quit [Ping timeout: 245 seconds]
<L84Supper>
Any idea why the player for A10 has only been released as a binary? Does the VPU violate patents and keeping the register spec private helps keep them out of the courts or is it only shared with paying customers to give those manufacturers an advantage?
<rz2k>
omg
<rz2k>
[ 3458.633] drmOpenDevice: node name is /dev/dri/card0 [ 3458.633] drmOpenDevice: open result is 7, (OK)
<rz2k>
mali400 dri2 working
<RaYmAn>
ooh
<rz2k>
tons of UMP<1>: No handler for IOCTL. cmd: 0x400c9007, arg: 0xbeb95e28
<mnemoc>
L84Supper: usually fear for competition
<mnemoc>
rz2k: a10 or a13?
<rz2k>
and [ 3458.838] GET_UMP_SECURE_ID_BUF2 returned 0xffffffff offset: 3145728 virt address: (nil) fb_virt: 0x40648000 is clearly not OK
<rz2k>
A10
<rz2k>
mele a2000
<mnemoc>
rz2k: a13 doesn't have ump integration yet. A10 needs the disp drivers to be built as 'm'
<rz2k>
I know
<rz2k>
its 'm'
<mnemoc>
so we miss an ioctl
<mnemoc>
:)
<rz2k>
I think I need to integrate something to mali_exa about this ioctl, readme says that
<rz2k>
but mali_drm actually works
<mnemoc>
rz2k: please on the mali_next branch
<rz2k>
ok
ceo16 has joined #arm-netbook
<rz2k>
I've stolen driver from igloocommunity and it has two fb's, this needs to be fixed to one.
<rz2k>
or a10 can do two?
<RaYmAn>
It should theoretically be able to, I'd think
<rz2k>
I'm keeping it, so.
* rz2k
ping techn
<techn>
rz2k:
<rz2k>
you said that dri2 with igloocommunity drivers didnt work for you
<techn>
yes.. I didn't get gles2 stuff working
<rz2k>
I thought it never did boot up, lol.
<rz2k>
sorry for confusing everyone. but it is one step closer now.
<mnemoc>
mali_next is waiting for you :)
hp__ has quit [Ping timeout: 246 seconds]
<techn>
rz2k: X startup was complaining about missing dri2_mali.so or something :/
<mnemoc>
that code comes from the SDK... but we can fairly assume the same license of the drivers in 3.0-v2 where Tom fixed the lack of header by instruction of Allwinner
<mnemoc>
and so we just copy their standard gpl header
eflatun has quit [Ping timeout: 272 seconds]
eflatun has joined #arm-netbook
<techn>
mnemoc: is someone merging this commit 200a6a9b00900b880726a1b4c69a3afc5a5ebcfb
<mnemoc>
techn: not as bulk at least
eflatun has quit [Ping timeout: 248 seconds]
mysteryname has joined #arm-netbook
eflatun has joined #arm-netbook
<mnemoc>
techn: I made those based on what I found in the A13 SDK and the rest of the branch has fixes to let it compile
<mnemoc>
eventually
<mnemoc>
it should be merged into 3.0-v2
<mnemoc>
but we need to implement a real sw_get_chip_id()
<hno>
only A10 needs it.
<mnemoc>
there are sun5i drivers in that branch using it
<mnemoc>
but I guess that's the best tool we will have to unifity sun4i/sun5i
<techn>
maybe it should be patched to main baseline
<mnemoc>
but we can take parts of it
<mnemoc>
that git diff gives you the diff against the base we used to construct 3.0-v2
<mnemoc>
already rebased, reencoded and sanitized
<mnemoc>
also removing the destructive parts
<mnemoc>
it was a pain to make :|
eflatun_ has joined #arm-netbook
eflatun has quit [Ping timeout: 272 seconds]
eflatun_ has quit [Ping timeout: 272 seconds]
eflatun_ has joined #arm-netbook
Turl has quit [Ping timeout: 250 seconds]
<merbzt>
oh neat, the CedarX sdk contains code that I wrote
<mnemoc>
o.o
eflatun_ has quit [Ping timeout: 264 seconds]
eflatun_ has joined #arm-netbook
<mysteryname>
The Ax are almost un-brickable right? I've recently got a Mele A2000 and about to dive into all of this for the first time, I read somewhere that the rhombus site does not have the latest information, Is there somewhere that has all the information?
<hno>
Yes the Mele is quite unbrickable.
eflatun_ has quit [Read error: Connection reset by peer]
<hno>
rhombus-tech is reasonably up to date in most areas. But there is no single place that have all information.
<lundman>
its more of a National Treasure of clues and quests
<mysteryname>
Ah brilliant, Clues and quests.. This could be quite interesting.
<mnemoc>
linux-sunxi.org tries to be more structured, but still lacks of tons of info which is available in other sites
<hno>
merbzt, you worked on CedarX, or they used code you wrote for something else?
qubits has joined #arm-netbook
qubits has quit [Changing host]
qubits has joined #arm-netbook
<mnemoc>
ffmpeg code I guess
<mysteryname>
mnemoc: Ah cheers.
<merbzt>
hno: A10-MEDIA-SDK-2012-04.tar.bz2 contains lots of ffmpeg/libav code
<merbzt>
I think it is the code that forms the Android player
eflatun_ has joined #arm-netbook
<hno>
merbzt, ok. kind of expected.
<lundman>
andnow I have a copy of it too
<hno>
lundman?
gimli has quit [Ping timeout: 272 seconds]
ssalevan has quit [Quit: Leaving.]
<techn>
is allwinner on ffmpeg's blacklist now? :p
<lundman>
hno?
<techn>
I noticed also that there is very familiar object in libav