<naobsd>
btw, probably recovery can boot directly from power on, press & hold recovery but without connection USB OTG
<tonikasch>
wow! I'll test
<naobsd>
if PC is connected via OTG, it goes bootloader mode, but PC is not connected, it goes recovery after some seconds
<naobsd>
I'm not sure, but I saw it some times when I forget to connect OTG ;)
<naobsd>
timeout is not so long, I guess
<tonikasch>
ahm
<naobsd>
(I'll leave here soon...)
<tonikasch>
me too
naobsd has quit [Quit: Page closed]
arokux2 has quit [Remote host closed the connection]
tonikasch has quit [Quit: Bye!]
akaizen has quit [Read error: Connection reset by peer]
akaizen has joined #linux-rockchip
Astralix1 has joined #linux-rockchip
Astralix has quit [Ping timeout: 264 seconds]
atiti has quit [Ping timeout: 252 seconds]
mcbrick has quit [Quit: Leaving]
atiti has joined #linux-rockchip
markm has joined #linux-rockchip
markm has quit [Client Quit]
markm has joined #linux-rockchip
rellla has joined #linux-rockchip
markm has quit [Quit: Leaving]
markm has joined #linux-rockchip
akaizen has quit [Remote host closed the connection]
akaizen has joined #linux-rockchip
akaizen has quit [Remote host closed the connection]
rellla2 has joined #linux-rockchip
rellla has quit [Ping timeout: 246 seconds]
Theueip has joined #linux-rockchip
AstralixNB has joined #linux-rockchip
akaizen has joined #linux-rockchip
eebrah|noPC has joined #linux-rockchip
deviker has joined #linux-rockchip
massi_ has joined #linux-rockchip
tonikasch has joined #linux-rockchip
<ganbold>
AstralixNB: ping
<AstralixNB>
hey
<AstralixNB>
Do you have a question or did you only want to check my existence?
* lunra
pokes AstralixNB
<lunra>
yep, definitely exists
<ganbold>
AstralixNB: had question, in rk30xx doc pl310 l2 cache controller doesn't have irq assigned, is it by design?
<AstralixNB>
Needed two cups of coffee but after that I felt existing too
<AstralixNB>
I just did a quick check, but it loks like PL310 IRQ is just thrown together with system and timer interrupts. You find L2C interrupts at offset 0x0204 (Page 276) and other places near this one.
<AstralixNB>
But could be that the manual is just incomplete at this point as it is in other topics.
<ganbold>
yeah, it could be manual incomplete
<AstralixNB>
you just could check the sources.
eebrah|noPC has quit [Ping timeout: 245 seconds]
ganbold_ has joined #linux-rockchip
<tonikasch>
a perhaps stupid question, how to do a sleep statement?
<tonikasch>
ah
<tonikasch>
confimed it is a cr1225
<lunra>
Sleep statement in what language/environment? I often wish I knew how to do a sleep statement in Human/RealLife
<tonikasch>
haha
<tonikasch>
in C, on kernel code
<tonikasch>
I've just mounted debugfs and checked that /sys/kernel/debug/gpio has no gpios requested
<tonikasch>
so something is failing with this kernel on radxa rock
Galland has joined #linux-rockchip
<mnemoc>
tonikasch: wiki wiki
Galland has quit [Remote host closed the connection]
<tonikasch>
ok
<tonikasch>
I'll put a link to linux-sunxi.org about making your own rootfs with debootstrap
pfalcon has joined #linux-rockchip
<tonikasch>
mmmm
<tonikasch>
on radxa something is wrong with gpios... RK30_PIN3_PD0 binds to 14848 instead of 280
tonikasch has quit [Remote host closed the connection]
naobsd has joined #linux-rockchip
Galland has joined #linux-rockchip
pfalcon has quit [Ping timeout: 248 seconds]
deviker has quit [Ping timeout: 264 seconds]
pfalcon has joined #linux-rockchip
deviker has joined #linux-rockchip
pfalcon has quit [Ping timeout: 256 seconds]
deviker has left #linux-rockchip [#linux-rockchip]
<deviker>
Hi, I'm having packet loss issues with the GK908. Using ping -f -l 250 someip everything is ok. With 275 I got 2% dropped packets. with 300 9%, with 400 29%. I tried changing the MTU to 250 and I still have the same problems.
<Galland>
you mean internal Wifi?
<deviker>
yes
<deviker>
it is the V3
<Galland>
GK908?
<Galland>
or MK908
<deviker>
mk908
<Galland>
if it's MK908V3, it has a **** antenna
<Galland>
so no wonder it has packet losses
<deviker>
does someone else have similar problems with the mk908?
<Galland>
me
<deviker>
Galland the problem is that it is exactly reproducible. It always losses 9% with 300 29% with 400 and 2% with 250
<Galland>
of course, the larger the message, the larger the possibilities of losing data
<deviker>
yes but always losses the same amount of data
<Galland>
so it's always the same **** antenna :)
<Galland>
wifi problems have been happening forever with chinese arm devices
<Galland>
that's why people like external antennas
<deviker>
:P ok Galland I'll try to put the AP nearer and put them inside a box covered with aluminium foil to test it better.
<Galland>
you don't believe me?
<Galland>
do it, repeat your measurements with the mk908 right by the AP
<Galland>
and then again at 10 meters
<hramrach>
the aluminium foil is probably not worth the effort
<Galland>
if the losses are greater, then you may start believing me
<Galland>
if the losses remain constant then you may start thinking conspiracy
<hramrach>
there are always wifi problems. that's why fiber > wire > WiFi
<hramrach>
also there is sime minimum distance at which stuff works
<Galland>
aluminium foil is a terrible idea, all the signal bounces would become echos and seriously affect the signal
<hramrach>
so right at the AP might not be the best position unless the antenna is really bad
<hramrach>
yeah, sounds like that but not really an expert at wireless signal
<hramrach>
the best wireless advice I have is to avoid it if possible without much extra cost :p
<Galland>
agreed :D
<pfalcon>
naobsd, Galland: Hi, if you care about that case I reported last week of not being able to boot kernel after flashing it to a recovery. I was able to overcome that and described my investigation in http://linux-rockchip.info/mw/index.php?title=Troubleshooting
<naobsd>
I guess deviker want to try both good condition (near AP) and bad condition (inside aluminium foil), regardless of "believe me" thing
<naobsd>
I'm sorry if I'm wrong
<Galland>
I'm not sure what his thought was with the reproducible losses :)
<Galland>
pfalcon: you should include the mtdparts as read from the stick: ./rkflashtool r 0x0 0x1 | head -n 11
<naobsd>
I also have some interest about "reproducible"
<pfalcon>
Galland: you mean, add to that page? well, I'm not sure it's too much relevant, that section is intended to be a bit higher-level than specific partition layout, but I can add it
<naobsd>
about constant percentage, not about reproducible lossing
<deviker>
Galland and naobsd (I meant putting both inside a case and cover the case with aluminium foil so that no *other* signal intereres)
<pfalcon>
Btw, I also had big issues with WiFi connection from my stick (mk809). An exactly after finishing "recovery doesn't boot" investigation, it's much improved
<Galland>
hehe in that page you say you overwrote recovery, but the behavior after that seems like you flashed somewhere else, so it's very relevant, if only to double-check your steps and be of help :P
<pfalcon>
2 things changes: 1) I reflashed/did factory reset on the device (rom image is the same though). 2) it is currebtly without (plastic) case
<naobsd>
pfalcon: you flashed CWM?
<pfalcon>
Galland: we already went over "somewhere else" last time ;-). I did not make mistakes and used the same command (from bash history for "before" and "after")
<pfalcon>
naobsd: yep, from your (apparently) site
<Galland>
pfalcon: really? :S bad memory hehe
<pfalcon>
all in all, that's FYI, if someone else will have such issues, please point them to that page ;-)
<Galland>
pfalcon: however isn
<Galland>
t that a 3.0.8 kernel?
<Galland>
not that it matters to your prob, just curious about why use such an old one, if it's the case
<pfalcon>
Galland: which one: android's or one I flashed. the latter is from picuntu original release (Jan)
<Galland>
pheeew then it's 3.0.8
deviker has quit [Ping timeout: 240 seconds]
<pfalcon>
Galland: I just tried to make any 3rd-party kernel boot on my device. though would take 5 min and I'll proceed with building mine, took 3 days ;-)
<naobsd>
pfalcon: sorry, I have no idea why flashing recovery with rkflashtool doesn't work
<pfalcon>
naobsd: well, now you know that it works, after CWM was flashed first from an SD card. just fyi and hope you'll be bale to hint someone having that issue later (I probably won't be around often ;-) )
<Galland>
pfalcon: so if I understand correctly, the solution is to flash CWM to the recovery partition?
<Galland>
or to which?
<pfalcon>
Galland: yes, flashing CWM to recovery partition *via sd card update.zip* somehow "unlocks" the process, and after that, kernel image can be written to recovery part. (thus overwriting CWM) and will boot w/o issues
<Galland>
have you tried diffing the full flash before and after recovery stops being overwritten?
<Galland>
that is, if you flash CWM to recovery and that fixes it, it must have changed something somewhere else (so it keeps working after you overwrite CWM)
<Galland>
maybe the misc or the parameters parts of flash
<pfalcon>
Galland: yeah, I tried, but I cannot be sure that I did it 100% cleanroom, and didn't find any changes the behavior could have been atttibuted too. Ia ctually anted to do clean-room diff, but apparently forgot ;-(
deviker has joined #linux-rockchip
pfalcon_ has joined #linux-rockchip
<pfalcon_>
fscken connection
<pfalcon_>
<pfalcon> Galland: yeah, I tried, but I cannot be sure that I did it 100% cleanroom, and didn't find any changes the behavior could have been atttibuted too. Ia ctually anted to do clean-room diff, but apparently forgot ;-(
<pfalcon_>
<pfalcon> Galland: misc part is what had some changes. but with working recovery boot, I tried to revert misc part. to previosu content, and it still worked. reflashing all 4gb to original state is what broke it again, until CWM was flashed anew
<naobsd>
recovery-from-boot.p is very common thing among Android devices
<naobsd>
if you google it, you can find a lot of information
<mihir>
I am getting information. But I want reference board with SDK for development. I have mailed rockchip also.
<hramrach>
some people got a board from rockchip, yes
<hramrach>
you can probably find a SDK somewhere on the internet
<pfalcon_>
naobsd: I personally didn't know about it (wonder how to keep up with all this google mumbo-jumbo). Neither I got suggestion about it from anyone, even though the behavior 100% fits.
<pfalcon_>
naobsd: So, the problem should happen to 100% of the owners of Android devices. That's why I don't like "custom roms" - they obfuscate real situation with Android, and they not Android themselves.
<pfalcon_>
naobsd: All in all, what do we do so everyone knew about that recovery-from-boot.p curse? ;-)
<hramrach>
I don't like the need to falsh custom roms to get a working device
<hramrach>
*flash
<naobsd>
I guess "custom roms" kills it by default
<naobsd>
it's common for stock roms
<pfalcon_>
naobsd: So, let me put question straight: why didn't you point me to it last week? ;-) No preaching obviously, my point is that it's not high enough in people's mind cache, and ideally, there should be straight reaction of: Q: I flash, but id doens't boot: A: did you check /system/recovery-from-boot.p ? ;-)
<pfalcon_>
naobsd: btw, http://androtab.info/clockworkmod/rockchip/install/ talks about /system/etc/install-recovery.sh . It's probably "more correct", but even more correct is to mention about both files. If you could add that, it would be appreciated.
<pfalcon_>
naobsd: also, it should be mentioned in "from rkflash, rkflashtool, or RKAndroidTool.exe" on that page...
<naobsd>
I thought you're trying to boot your own kernel and Linux userland ;p
<naobsd>
probably I wrote "from root shell" first
<naobsd>
well
<naobsd>
I can't remember. probably I was lazy.
<pfalcon_>
naobsd: well, all's cool - live and learn. and http://androtab.info/clockworkmod/rockchip is cool resource (only 2 missing things is "getting source" and "building" pages ;-) ). I'm just caring that other people didn't go same trap like I did...
<naobsd>
3. only few people has interest about source
<naobsd>
4. CWM is not my product. I just add small modification. there are a lot of information about CWM on the net.
<naobsd>
I'm not only person who can build CWM
<pfalcon_>
naobsd: yeah, that's true. but you know the spirit of open source - you just give it to people, to use whenever they need, and hope that everyone would need it sooner or later (or whole FOSS paradigm won't work if people won't contribute)
<pfalcon_>
anyway, just friendly suggestions ;-)
akaizen has quit [Remote host closed the connection]
<pfalcon_>
naobsd: one problem as you know s that there're *too many noise* around everything around android hacking. finding *correct* info is very time-consuming... ;-(
<naobsd>
sure, there are a lot of noises
<pfalcon_>
so, keeping sources close to binaries is good way to counter that ;-)
<naobsd>
I opened source as much as possible, but it was impossible to get time more than 24 hours in a day ;)
<naobsd>
most people just want "please give me working image for my device"
<naobsd>
or "please tell me how to root my device"
<naobsd>
most people ask me to write document, I'm sure it's very correct.
<naobsd>
but sometimes I wonder
<naobsd>
if someone summarize my message, I'll be happy ;)
<pfalcon_>
point taken. if you ever get time to update http://androtab.info/clockworkmod/ with all that info, people will appreciate that, and it can raise few high-quality android hackers ;-)
<naobsd>
you can see there is no update on my site for half year...
<pfalcon_>
I myself trying to contribute to making knowledge more available (see that link for example) - as much as time allows, which I have very little too (that's the point - I don't like to do same thing twice)
<naobsd>
I really want to do more things
<naobsd>
it's 3AM here. tomorrow, well, today is not holiday :(
<pfalcon_>
ok, have a nice rest then and thanks for all your efforts!
<naobsd>
my hacking work for RK device is not finished ;)
<naobsd>
but probably I'll sleep very soon
<naobsd>
I hope I can go to bed...
<hramrach>
good night :)
eebrah|noPC has quit [Read error: Connection reset by peer]
mihir has quit [Quit: Page closed]
akaizen has joined #linux-rockchip
pfalcon_ has quit [Quit: Leaving]
akaizen has quit [Ping timeout: 264 seconds]
tonikasch has joined #linux-rockchip
<tonikasch>
Galland!
<Galland>
hi
<tonikasch>
I'm trying to get RTL8188ETV working, but have no idea how to do it...
<tonikasch>
now with your Linux3188 kernel
<Galland>
for what device?
<tonikasch>
radxa rock
<Galland>
hmmmm You sure you got a board with no BT?
<Galland>
have you checked the marking on the Wifi chip under the board?
<Galland>
8188 or 8723?
<tonikasch>
8188ETV I think, let me check again
<tonikasch>
it's RTL8188ETV
mcbrick has joined #linux-rockchip
<tonikasch>
I've tried exporting gpio 256 and echoing 1 to value but only thing it does is get recognised in another usb port...
<tonikasch>
in fact, as I boot it's yet listed in lsusb
<tonikasch>
but I guess it has to be activated somehow, perhaps through GPIO or who knows
akaizen has joined #linux-rockchip
<Galland>
if it appears on lsusb then it's just a matter of insmodding the right module
<Galland>
since it's just a normal usb device (once powered on --> lsusb)
<Galland>
at most you may need some firmware, but you can get both from Androdi
<tonikasch>
aha
<tonikasch>
ok ;) thx
Galland has quit [Ping timeout: 248 seconds]
AstralixNB has joined #linux-rockchip
Galland has joined #linux-rockchip
AstralixNB has quit [Quit: Leaving.]
Galland has quit [Quit: Saliendo]
Galland has joined #linux-rockchip
Galland has left #linux-rockchip ["Saliendo"]
Galland has joined #linux-rockchip
hramrach has quit [Remote host closed the connection]
MakeMickey has joined #linux-rockchip
<MakeMickey>
hello
<MakeMickey>
i try to run qemu-system-i386 on my rk4
<MakeMickey>
rk3066
<tonikasch>
uf
<Galland>
that pretty much explains the result
<MakeMickey>
but i think it have a write to virtual disk problem
<Galland>
an RK device is not a i386 architecture
<Galland>
have you compiled it for arm?
<Galland>
and you want to emulate an i386 arch?
<MakeMickey>
it s a armhf debian system
<Galland>
ah ok, I thought the other way around :)
<MakeMickey>
openbsd boot and install, but after never boot on virtual disk
<MakeMickey>
freedos boot too
<MakeMickey>
all boot but i have a virtual disk proble i think
<MakeMickey>
yes it s not the best way to have a openbsd system
<MakeMickey>
but
<MakeMickey>
why not
<MakeMickey>
linux is good, but pf is more good
<Galland>
so what's the problem? any error msg?
<MakeMickey>
in fact all virtual machine boot, install, and have problem with virtual disk i think
<MakeMickey>
why ?
<MakeMickey>
Galland: no, openbsd install good, and error to boot
<MakeMickey>
no find logs errors on messages or syslog.....
<MakeMickey>
must find how to increase logs on qemu....
<MakeMickey>
galland speak about me kernel 3.0.35 vs 3.0.8 ?
<MakeMickey>
on rk 3066
<MakeMickey>
now i use 3.0.8
<Galland>
3.0.36 has many bugfixes everywhere in the kernel
<MakeMickey>
23:16 < Galland> and no, I don't have any olimex boards
<MakeMickey>
23:18 < MakeMickey> thk for this link
<MakeMickey>
23:18 < Galland> np :)
<MakeMickey>
23:18 < tonikasch> Galland
<MakeMickey>
pffff
<tonikasch>
Yes, there is wlan0 and after doing ifconfig wlan0 up I can iwlist wlan0 scan
<Galland>
then why do u think u need firmware?
<tonikasch>
Galland nothing, I forgot to do ifconfig wlan0 up before :p
<tonikasch>
And so, I couldn't iwlist...
Tsvetan2 has joined #linux-rockchip
<MakeMickey>
i not see my integrated wifi card, if i try lusb....
Tsvetan has quit [Ping timeout: 252 seconds]
<MakeMickey>
lsusb
<MakeMickey>
no see rk903.... or other
<MakeMickey>
is a good way to take linux source for rk3188 for RK3066 wifi ?
<tonikasch>
for wifi?
<MakeMickey>
yes
<tonikasch>
don't understand...
<MakeMickey>
kernel source for RK3188 run for 3066 no ?
<MakeMickey>
but have rk903 wifi driver ????
<Galland>
github.com/Galland/Linux3188 runs for some rk30
<Galland>
but no idea about rk903
markm has quit [Read error: Operation timed out]
sjoerd has quit [Ping timeout: 245 seconds]
sjoerd` has joined #linux-rockchip
<MakeMickey>
ok wifi is another problem, my great problem at this time is my qemu-system-i386
<tonikasch>
some people say that 3.0.36 kernel has correct gpio values for mk808b, but it's like a rumour
<MakeMickey>
work only on live cd or boot cd or install cd
Tsvetan has joined #linux-rockchip
<MakeMickey>
tonikasch: now i use a 3.0.8 kernel, good no problem, for a small server
<tonikasch>
:)
<MakeMickey>
think i get have a lot of uptime
<tonikasch>
:D uptime :D
<MakeMickey>
yes
Tsvetan2 has quit [Ping timeout: 246 seconds]
<tonikasch>
well, Galland, after installing wpasupplicant and configuring /etc/network/interfaces I can confirm radxa rock with native rtl8188eu module is working
<tonikasch>
so... I don't know what to do now
<MakeMickey>
tonikasch: yes small server, small money for energy
<tonikasch>
MakeMickey yes, I know, I have some webpages and other services serving in a mk808
Tsvetan2 has joined #linux-rockchip
<Galland>
tonikasch can u fork Linux3188 apply the patch for 8188 and pull request to me?
<MakeMickey>
don t now how this small server can stay ALIVE .......
<tonikasch>
Ok, let me see if I know how to do that :)