akaizen has quit [Remote host closed the connection]
lkcl has joined #linux-sunxi
akaizen has joined #linux-sunxi
akaizen has quit [Ping timeout: 240 seconds]
<petrosagg>
How can the SD card be reliable since it's using a NAND flash itself? The hardware FTL it uses can't be that sophisticated, how can there not be a "good enough" FTL for linux?
<quitte>
there are good enough FTL for linux. however not yet in mainline for the sunxi nand controller
<quitte>
and the flash controller inside a sd-card needs to be able to handle exactly one kind of flash chip
<petrosagg>
can you point me to one such FTL for linux?
<petrosagg>
I'd expect that the FTL would be independent of the NAND controller
<petrosagg>
After all the controller is often used with multiple NAND chips and these are the onces that need the FTL, not the controller
<quitte>
okay. I cannot. I haven't looked into the abstraction enough to know those layers
<petrosagg>
okay. It's really counter intuitive to me :S
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
wingrime1 has joined #linux-sunxi
<quitte>
i got openwrt to spit out ubiniuzed images :) hopefully I can get it to do the same with MMC tomorrow
<petrosagg>
which device are you working on?
<quitte>
cubietruck
<petrosagg>
ah, nice
wingrime has quit [Ping timeout: 250 seconds]
<petrosagg>
does openwrt have special requirements from the kernel or is it just a different rootfs?
<quitte>
it's not like android. but there are a couple of patches that are necessary at least for building it.
<petrosagg>
I've never played with openwrt
<petrosagg>
Is it useful for applications other than routers?
<quitte>
I'm planning to use it as a fileserver/printserver and such on the cubietruck
wingrime has joined #linux-sunxi
wingrime1 has quit [Read error: Connection reset by peer]
<petrosagg>
ah nice
wingrime has quit [Read error: Connection reset by peer]
ricardocrudo has quit [Ping timeout: 255 seconds]
ricardocrudo has joined #linux-sunxi
akaizen has joined #linux-sunxi
lkcl has quit [Ping timeout: 264 seconds]
lkcl has joined #linux-sunxi
lkcl has quit [Ping timeout: 250 seconds]
ricardocrudo has quit [Remote host closed the connection]
schopen has quit [Quit: Page closed]
TheSeven has quit [Ping timeout: 272 seconds]
TheSeven has joined #linux-sunxi
bonbons has quit [Quit: Leaving]
akaizen has quit []
akaizen has joined #linux-sunxi
lkcl has joined #linux-sunxi
MY123 has quit [Ping timeout: 250 seconds]
JohnDoe_71Rus has joined #linux-sunxi
amitk has joined #linux-sunxi
rz2k has joined #linux-sunxi
lkcl has quit [Ping timeout: 260 seconds]
lkcl has joined #linux-sunxi
Andy-D has joined #linux-sunxi
ah has quit [Read error: Connection reset by peer]
PulkoMandy has joined #linux-sunxi
yann_s|AFK has joined #linux-sunxi
yann_s|AFK is now known as yann_s
Andy-D has quit [Ping timeout: 246 seconds]
lkcl has quit [Ping timeout: 246 seconds]
lkcl has joined #linux-sunxi
shineworld has joined #linux-sunxi
shineworld has quit [Changing host]
shineworld has joined #linux-sunxi
libcg has joined #linux-sunxi
HeavyMetal has quit [Ping timeout: 240 seconds]
anthony_emtrion has joined #linux-sunxi
_massi has joined #linux-sunxi
HeavyMetal has joined #linux-sunxi
Zboonet has joined #linux-sunxi
<oliv3r>
libv: i see you started to cleanup my ndh page :)
wingrime has joined #linux-sunxi
skoperst has joined #linux-sunxi
<oliv3r>
libv: did you push your tpr4 changes to meminfo yet?
MY123 has joined #linux-sunxi
<MY123>
libv: Hi.
_massi has quit [Remote host closed the connection]
lkcl has quit [Ping timeout: 264 seconds]
_massi has joined #linux-sunxi
Quarx has joined #linux-sunxi
Zboonet has quit [Quit: Leaving]
<wingrime>
Libv: banana pi is opensource, or pretent to be so?
orly_owl has quit [Quit: leaving]
<shineworld>
what is the widest display resistive touchscreen have you connected to A10/A20 TP device driver ? and what about noise ?
lkcl has joined #linux-sunxi
paulk-aldrin has joined #linux-sunxi
<bbrezillon>
petrosagg: sorry for the delay (I've lost my connection and then decided I should go to sleep :-))
<bbrezillon>
petrosagg: your wiki page is a good start
<bbrezillon>
petrosagg: some comments though
<bbrezillon>
petrosagg: in the ECC data section => ECC are characterized by their strength and step size, not just their strength (i.e. 40 bits / 1024 bytes, where the strength is 40, and the step size is 1024 bytes)
shineworld has quit [Quit: Leaving]
paulk-aldrin has quit [Ping timeout: 246 seconds]
weissschloss has quit [Remote host closed the connection]
weissschloss has joined #linux-sunxi
<wingrime>
shineworld: I think there is no physical limitation you can hit with resistive ts unless you want make ts wall
paulk-aldrin has joined #linux-sunxi
<longsleep>
bbrezillon: I just saw your reply from the weekend - thanks for checking - will try to figure out how i can merge better from your branch and still keep mergability with my othe branches :)
<longsleep>
bbrezillon: so i just fix that manually for now as i cannot rebase on your branch without loosing mergability with my other branches :/
lkcl has quit [Ping timeout: 245 seconds]
orly_owl has joined #linux-sunxi
Black_Horseman has quit [Quit: Zwi se logou mou!!!]
<libv>
oliv3r: yeah, i do that with every device page
<oliv3r>
libv: i ment you did it fast, before i got time to go over it :)
<libv>
wingrime: i'd say pretend
<paulk-aldrin>
oliv3r, seems like my tzx-q8-713b is not the same as yours
<paulk-aldrin>
oliv3r, can you check the last digit on the PCB?
<paulk-aldrin>
the one under the glue
<oliv3r>
paulk-aldrin: not right now, but at home, sure thing
popolon has joined #linux-sunxi
<paulk-aldrin>
ok
<paulk-aldrin>
mine has a different accelerometer
<oliv3r>
paulk-aldrin: the pcb probably can suit many differnt kinds?
<libv>
oliv3r: i wrote the book on these things, i did it on 90+ pages and made them all pretty conform, it does not require much effort anymore, so don't feel bad about it
<oliv3r>
libv: okay :)
<paulk-aldrin>
and probably a different ts controller
<libv>
mnemoc: ok, so now the thing got shipped. aliexpress is strange that way :)
<libv>
hrm, cons_index wasn't properly set for the pov mobii 703
* libv
goes and verifies that this is what stopped it from booting correctly
<longsleep>
quitte: yeah thanks i compile the i2c-mv64xxx driver, but it tells me mv64xxx_i2c: probe of 1c2ac00.i2c failed with error -22
skoperst has quit [Ping timeout: 246 seconds]
<longsleep>
quitte: are you saying that sun7i could be another driver?
<quitte>
longsleep: i just had a look. it's enabled in my kernel and it builds just fine
<quitte>
longsleep: I don't know. I was going by the commit message alone
<longsleep>
quitte: yeah building works, but the driver does not load on cubietruck with said error. So i thought it could be another driver as it is listed seperately in mainlining wiki.
<quitte>
it's not a module on mine, but it is in /proc/irq/[39,40,41/mv64xxx_i2c
<longsleep>
quitte: ok and you got an A20 based board?
<quitte>
yes. cubietruck
<longsleep>
quitte: ok thanks thats promising then
<libv>
btw, we have parts of the A80 SDK :)
<libv>
i will be uploading that later today
FreezingCold has quit [Ping timeout: 246 seconds]
<wingrime>
L
<wingrime>
libv: a31 manual still not uploaded
<wingrime>
libv: and, I don't get who are recived optimus board
<libv>
wingrime: ...
<libv>
who has the a31 manual?
<wingrime>
Olv3r
<wingrime>
For example
<wingrime>
Mripard
lkcl has quit [Ping timeout: 260 seconds]
<libv>
grmbl.
<wingrime>
I wonder you still not aware about
<libv>
anyway, a80 sdk has the same libnand issues
<libv>
and libisp
<libv>
oh, there's extra ones
<quitte>
they must be really proud or embarassed about libnand
kuldeepdhaka has quit [Ping timeout: 245 seconds]
<libv>
i somehow feel that it has little to do with that
wingrime has quit [Read error: Connection reset by peer]
FreezingCold has joined #linux-sunxi
wingrime has joined #linux-sunxi
<libv>
mdfs, ram frequency scaling comes with 2 blobs as well.
<libv>
but that's not the end of that story.
bengal has quit [Ping timeout: 264 seconds]
JohnDoe_71Rus has quit [Read error: Connection reset by peer]
JohnDoe_71Rus has joined #linux-sunxi
<wens>
wingrime: a31 manual was uploaded (by me)
wingrime1 has joined #linux-sunxi
wingrime has quit [Read error: Connection reset by peer]
<wens>
libv: huh, didn't they handle that with the openrisc core? (another blob)
<wens>
btw, where did you get the sdk?
wingrime1 has quit [Read error: Connection reset by peer]
wingrime has joined #linux-sunxi
<libv>
via-via
<libv>
like all of these things worked before
<libv>
also, did anyone pull the cubietech a80 kernel when it was alive?
<wens>
don't think so :(
wingrime1 has joined #linux-sunxi
wingrime has quit [Read error: Connection reset by peer]
<libv>
a80 kernel is riddled with binaries.
<libv>
not just libnand and libisp
wingrime1 has quit [Quit: Bye]
wingrime has joined #linux-sunxi
<libv>
since this working with the community and such is working out so well for allwinner, they decided to step up their efforts in violating the GPL.
<quitte>
wens: it's not that bad. it didn't actually do any conversion. However I'd like to know if that is because it was defective in the first place or some design error in the board
<wens>
libAW5306 and gsl_point_id_20131111 are in the a23 sdk as well
<libv>
wens: at least libAW is listed there iirc
<libv>
i will pass over them again
<JohnDoe_71Rus>
wingrime: yeah DNS shop
<wingrime>
Thay at least write down soc
<wens>
the mdfs blob looks like a small bunch of code to run from sram to deal with dram refresh, muxing, re-clocking and stuff
jemk has joined #linux-sunxi
<libv>
wens: it is
<libv>
wens: but it is still GPL
<libv>
it's like 4k only
<libv>
each
<libv>
so easy to RE
<wens>
libv: yeah, having source for this is crucial
<wens>
well, RE would be possible
<wens>
having commented source is better
<MY123>
libv: There is no powervr series6 Linux driver.
<MY123>
(A80
<MY123>
( REing PowerVR is complex, tried it with Series5)
<wens>
there's probably a blob for android somewhere in the sdk
<MY123>
wens: There is.
<libv>
MY123: you know my history on pvr
<libv>
i have been very actively discouraging people from working on that
<MY123>
libv: I tried that for the Samsung S5PC110.
<MY123>
Hard. Working on VC4 since the docs release ( and sometimes on etnaviv)
<wingrime>
My123 optocodes are known?
<wingrime>
I mean power vr instruction set
<MY123>
wingrime: Series5 (partially), Series6 (not an opcode)
<wingrime>
As i know power vr are tiled
<wingrime>
Renderer...
<MY123>
They are tiled since the first powervr
<wingrime>
Dreamcast has some
<wingrime>
How much difficult to upload code to it?
<wingrime>
And run
<MY123>
wingrime: Series5(not very easy), Series6(unknown)
<wingrime>
Heh
<MY123>
Series5 is in the A31 and Series6 is in the A80, wingrime.
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
wingrime1 has joined #linux-sunxi
wingrime has quit [Read error: Connection reset by peer]
wingrime1 has quit [Read error: Connection reset by peer]
<libv>
dreamcast is like pvr2
<libv>
or was it the same as the matrox m3d/dell card?
<libv>
there's little shared between those 1996 devices and the hw today
* libv
got himself an m3d 4ys ago, when working on the omap on the n9, just for owning one
Gerwin_J has joined #linux-sunxi
<MY123>
libv: I also have a 1,2MP feature phone camera , should I try?
<MY123>
(Kurio 7S)
enrico_ has quit [Quit: Bye]
<libv>
MY123: if it gets more stable pictures, then yes
<libv>
MY123: make sure the light is good, and that you take pictures on a white background to help it out
paulk-aldrin has quit [Ping timeout: 264 seconds]
paulk-aldrin has joined #linux-sunxi
<wens>
libv: have you looked at the lcd functions in u-boot?
<wens>
that part looks awfully messy :|
<libv>
yeah
<libv>
it will be pretty simple once i'm done with it
<libv>
ooh! ooh!
<libv>
new mali binaries in buildroot tree
<MY123>
libv: But they have a new backdoor.
<libv>
MY123: the mali binaries?
<MY123>
libv: Yes.
<libv>
MY123: please explain
<wens>
that doesn't sound good
paulk-aldrin has quit [Ping timeout: 240 seconds]
paulk-aldrin has joined #linux-sunxi
<MY123>
As the source code is not available and the Mali can access to all the system mem. It can contain shit as there is not a explanation of the changes.
<merbanan>
hmmm, isn't the SGX driver source available "somewhere" ?
<MY123>
merbanan: If you sign an NDA.
<libv>
MY123: oh, that. sure.
<merbanan>
hmmm, I think I actually found it online somewhere
lkcl has quit [Ping timeout: 264 seconds]
<MY123>
merbanan: If you find it, DON'T push it. It will destroy the clean RE rume.
wingrime has joined #linux-sunxi
<MY123>
*Rule
wingrime1 has joined #linux-sunxi
wingrime has quit [Read error: Connection reset by peer]
<merbanan>
MY123: well afaik no one is working on that
<oliv3r>
A33 is also called sun8i btw
<oliv3r>
in case anybody was wondering :)
<MY123>
libv: The another camera is much worse .
<libv>
MY123: ok
<libv>
oliv3r: let's call it sun10i
<libv>
since a80 is clearly sun9i
<libv>
if needs be, we need to start using aw16xx
<libv>
as this new naming scheme is pretty broken
<MY123>
libv: Should I push the script.bin on Dropbox?
<libv>
we need to add the new naming scheme to the wiki still though
<libv>
MY123: follow the ndh howto.
<MY123>
(Google Drive)
<libv>
1:12 until upload complete...
<wingrime1>
Aw16xx broken too for cedarx
<libv>
wingrime1: ?
<libv>
wingrime1: please explain
<wingrime1>
A20 report same id like a10
<libv>
16xx would be the version hardcoded in the register, the one i read in from meminfo
<wingrime1>
In cedarx regs
<libv>
wingrime1: which is ok, as the engine probably is exactly the same
<libv>
so we better document that this register should not be used as the chip wide version
<wingrime1>
Thats ,true,
<MY123>
Ok.will send my changes.
<oliv3r>
libv: the a33 tablet i got is blurped ufll of sun8i internally what i'm guessing, since sun8iw3 was suna23
<wingrime1>
I have notice in cedarx pages
<oliv3r>
is that sun8iw<something else> is A33
<libv>
oliv3r: yeah, you're right
<libv>
oliv3r: but this new naming scheme is pretty broken and confusing
<oliv3r>
i even saw in an asdk what the w stood for
<libv>
oliv3r: so we better document things nicely on the SoC pages
<libv>
and pick our own
<oliv3r>
with unified drivers, it won't be a big problem, as it's all sunxi :)
<oliv3r>
yeah probably
<libv>
either sun10i to continue our naming scheme
<libv>
our throw our naming scheme out entirely and go for aw16xx
<oliv3r>
well the internal aw16xx names are probably nicer, but we don't have all names do we?
<libv>
like i did for unichrome back in the day
<libv>
we have the names the second we get the hw
<oliv3r>
i think a33 really is sun8i with quad cores instead of dual cores
<oliv3r>
i think everything else is identical
<wingrime1>
More one schene will make confuse worse
<libv>
wingrime1: yes, and no
<libv>
wingrime1: having a completely different scheme, one which is very unlikely to be influenced by marketing bs, one that actually matches the hw, one that will last quite long, might be worth the initial pain
<wens>
oliv3r: sun8i-a33?
<oliv3r>
we should have done something for sun5i tbh
<wingrime1>
Sunxi-a20
<oliv3r>
like a13 and a10s
<libv>
oliv3r: take a look at my posted meminfo code
<libv>
oliv3r: no
<libv>
oliv3r: nononono
<libv>
oliv3r: a13 and a10s are one and the same
<oliv3r>
libv: have you posted the meminfo code yet?
<libv>
oliv3r: check our mailing list
<wingrime1>
sunxi-a32
<wens>
and what about a31 and a31s?
<libv>
oliv3r: a13 are one and the same, perhaps a few minor changes
<libv>
just the packaging is different
<libv>
turl pointed that out to me
<libv>
and the a13 really does have hdmi
<wingrime1>
I prefer sunxi prefix, as there much hw called a10
<libv>
as the engine is there
<libv>
it's just not available to the outside world
<wingrime1>
Libv: cedarx somehow different in a13
<libv>
wingrime1: but a13 vs a10s?
<libv>
wingrime1: that's what i am talking about now
<libv>
so both are sun5i, just in a different package
<libv>
i even think that we shouldn't have a different dtsi for both
<oliv3r>
so what if a23 and a33 are fully the same, except for the number of cores
<wingrime1>
Libv: I have no a10s so I have nothing to say
<oliv3r>
(e.g. a23 had maybe lower yield so 2 cores got binned for example)
<oliv3r>
wingrime1: i got a10s recently!
<wens>
mripard_: has your mmc phase clk series received any attention?
<libv>
oliv3r: check the hw id
<libv>
oliv3r: then we'll continue the discussion ;p
<wingrime1>
Oliv3r: you can read cedar version?
<oliv3r>
what's nove's IRL name/email?
<oliv3r>
wingrime1: if you tell me how, sure
<mripard_>
wens: some
<oliv3r>
libv: where can I find that?
<wens>
mripard_: i'd like to use some of the factors cleanup patches in my sun6i/sun8i pll6 series
<mripard_>
I need to repost a v2
<libv>
oliv3r: the meminfo code i posted!
<oliv3r>
libv: ok ok :)
<oliv3r>
i'll pull the patch
<wens>
oliv3r: just get a shell and use busybox devmem to poke a register :)
<wingrime1>
oliv3r i have uboot cedarx init sequnce
<libv>
oliv3r: it's safe to run the full meminfo rewrite
<oliv3r>
ok i can check it, but i can spend like 30 seconds on it :p
<libv>
oliv3r: it'll bail and state the id
<oliv3r>
what's the title of the mail?
<oliv3r>
found it
<oliv3r>
v2
<oliv3r>
libv: you don't happen to have a git tree of those patches do you? 'd have to pull it from googlegroups which is horrid
<libv>
oliv3r: let me mail you the binary
<petrosagg>
bbrezillon: wiki fixed
<libv>
oliv3r: it's been a while since you mailed the ml :p
<oliv3r>
libv: ok that's even better
<oliv3r>
libv: yeah it has been
<oliv3r>
i do have some patches for 3.4 ready :)
<mripard_>
iirc, sun8iw1 is A31, sun8iw2 is A31s, sun8iw3 is A20, and I guess w4 is A23, and w5 A33
<wens>
# busybox devmem 0x01c00024 32 0x8100
<wens>
# busybox devmem 0x01c00024
<wens>
oliv3r: ^
<oliv3r>
mripard_: i thought sun6i was a31 :)
<mripard_>
they changed the naming halfway
<oliv3r>
wens: ok running that
<mripard_>
so it's both :)
<oliv3r>
sunA10i :p
<wens>
oliv3r: they changed it in the A23 SDK... sun8iw1 is A31(s), sun8iw3 is A23
<oliv3r>
highly confusing for us
<oliv3r>
we should rename them all!
<wens>
and sun9iw1 is A80
<libv>
haha
<libv>
they don't even know themselves anymore!
<oliv3r>
i think the W stood for 'wafer' run or something
<mripard_>
the point was to change the sunNi for a generation of chip, and then have the wafer revision
<wens>
yup, wafer (for whatever reason)
<libv>
btw, i would still like to see someone give the ok for meminfo patches
<oliv3r>
paulk-aldrin: in the case of sunxi, replicant vs CM would be only mali?
<wingrime1>
Paulk, you can ask quarx about cm port
<paulk-aldrin>
oliv3r, Mali would be a big part, I'm also thinking about CedarX
<paulk-aldrin>
but I'm not sure they ever released proper android binaries for cedarx
<oliv3r>
good luck getting that to work at all anyway :p
<oliv3r>
to bad vdpau won't work on android
<paulk-aldrin>
Android uses OMX, AFAIK
<paulk-aldrin>
I haven't looked a lot into it
<paulk-aldrin>
maybe I could implement an OMX plugin based on the vdpau code
<paulk-aldrin>
looks simple enough
<paulk-aldrin>
oliv3r, I'm probably not going to do the CM port myself though :)
<paulk-aldrin>
even though my mom would probably like fast graphics on her sunxi tablet, I don't want to deal with blobs, it's such a pain
<oliv3r>
paulk-aldrin: i don't blame ya :)
<oliv3r>
paulk-aldrin: but we got the secret weapon called libv :p
<paulk-aldrin>
hell yeah :)
<oliv3r>
so maybe we can get him excited enough :)
<paulk-aldrin>
I did get mesa to run on Android in the past
<oliv3r>
paulk-aldrin: i think vdpau atm requires X, so va-api might be an interesting avenue too
<paulk-aldrin>
ah ok
<paulk-aldrin>
I'd have to look into details
<oliv3r>
i'm no vdpau expert, but last time i read aynthing about the matter was that vdpau only has X as possible output, with the rest 'open for impelmentation'
<MY123>
libv: Neat patch sent.
<MY123>
paulk-aldrin: Is the new patch good?
<paulk-aldrin>
MY123, The From line looks weird, but the rest seems normal
<paulk-aldrin>
however, it can be stripped manually, so I suppose it's fine
<MY123>
paulk-aldrin: It was using git send-email ,I can't do more
<wens>
the options line was wrapped
<paulk-aldrin>
did you do it as root?
<MY123>
paulk-aldrin: Yeah.
<paulk-aldrin>
do it as regular user next time
<MY123>
libv: Did you just disappear?
<paulk-aldrin>
libv, maybe don't harass people too much like this :p
<paulk-aldrin>
oops
<paulk-aldrin>
MY123, maybe don't harass people too much like this :p
<paulk-aldrin>
oliv3r, btw, the Build number on my tzx-q8-713b has: "713B7"
afaerber_ has quit [Quit: Verlassend]
<paulk-aldrin>
so the next digit is definitely relevant
<oliv3r>
yeah looks like it
link0_ has joined #linux-sunxi
<MY123>
paulk-aldrin :-)
<MY123>
:p
<libv>
MY123: the fex is needed as well
huehner_ has joined #linux-sunxi
<MY123>
libv: Where should the fex be transferred.
<MY123>
?
focus_it has joined #linux-sunxi
zumbi has joined #linux-sunxi
tm512` has joined #linux-sunxi
<paulk-aldrin>
MY123, sunxi-boards
<paulk-aldrin>
send the patch to the mailing list as well
<libv>
MY123: again, that's in the ndh
bsdfox_ has joined #linux-sunxi
<petrosagg>
quitte_: I know, I didn't mean to imply that. I couldn't link to the specific stage of the boot process
<libv>
paulk-aldrin: i need to make some further changes
massi_ has joined #linux-sunxi
atsampso1 has joined #linux-sunxi
keebler_ has joined #linux-sunxi
<libv>
like rename the dram file
<paulk-aldrin>
ok
andoma_ has joined #linux-sunxi
<paulk-aldrin>
yes indeed
<libv>
and alphabetical order
tm512 has quit [Ping timeout: 260 seconds]
bsdfox has quit [Ping timeout: 260 seconds]
vbmithr has quit [Ping timeout: 260 seconds]
huehner has quit [Ping timeout: 260 seconds]
link0 has quit [Ping timeout: 260 seconds]
_massi has quit [Ping timeout: 260 seconds]
keebler has quit [Ping timeout: 260 seconds]
atsampson has quit [Ping timeout: 260 seconds]
focus has quit [Ping timeout: 260 seconds]
bbrezillon has quit [Ping timeout: 260 seconds]
kelvan has quit [Ping timeout: 260 seconds]
lynxis has quit [Ping timeout: 260 seconds]
discopig has quit [Ping timeout: 260 seconds]
zumbi_ has quit [Ping timeout: 260 seconds]
andoma has quit [Ping timeout: 260 seconds]
jelly-home has quit [Ping timeout: 260 seconds]
bsdfox_ is now known as bsdfox
<libv>
i will do so later, first the SDK
<libv>
wow, a server went down it seems
jelly-home has joined #linux-sunxi
huehner_ is now known as huehner
lynxis has joined #linux-sunxi
<paulk-aldrin>
right
<paulk-aldrin>
lol
discopig has joined #linux-sunxi
kelvan has joined #linux-sunxi
<skoperst>
a noob question, why so many branches in the linux-sunxi repository in github? what branch should I start with if I want to compile it for my sunxi Android board?
<libv>
skoperst: that is explained on our kernel page
<wingrime1>
My123: power vr have conigurable amount of raster, texture units?
<MY123>
wingrime1: Seems to be.
xavia has joined #linux-sunxi
Nyuutwo has quit [Quit: No Ping reply in 180 seconds.]
Nyuutwo has joined #linux-sunxi
<MY123>
libv: Why meminfo does not show everything?
lkcl has quit [Ping timeout: 240 seconds]
<libv>
arete74: just patches for now
<wens>
hmm, i2c seems broken in 3.17-rc1
<wens>
s/3.17-rc1/sunxi-next
keebler_ is now known as keebler
keebler has quit [Changing host]
keebler has joined #linux-sunxi
keebler has left #linux-sunxi [#linux-sunxi]
techn has quit [Ping timeout: 255 seconds]
<quitte>
petrosagg: it is indeed confusing. I remember having a hard time understanding how booting works. With sunxi things are especially bad since there always are two ways. fex/dtb spl/boot0 boot1/nothing u-boot/u-boot libnand/mtd and probably more
techn has joined #linux-sunxi
<quitte>
and there are even two mtd implementations
ptitnuage has joined #linux-sunxi
lkcl has joined #linux-sunxi
<ptitnuage>
Hello everybody. I'm trying to follow the building guide, i would like to know if the part of defining the mac adresse in the fex file is mandatory or if it can be ommited
kuldeepdhaka has joined #linux-sunxi
<MY123>
ptitnuage: If you want to disable Ethernet, yes.
<ptitnuage>
MY123
* MY123
waits ptitnuage
<MY123>
(reponse)
<MY123>
*response
<ptitnuage>
MY123: thanks for answer, i was wondering if the board may refuse to boot if the mac defined is not the same as the board (as seen on android)
<ptitnuage>
MY123: typed to fast the first time
<MY123>
ptitnuage: It works when disabling the Ethernet module in the kernel compilation.
<ptitnuage>
MY123: so if ethernet not disabled (i didn't changed anything form the tutorial) it may refused to boot because i didn't set MAC adress?
<MY123>
ptitnuage: It MAY.
lkcl has quit [Ping timeout: 245 seconds]
kuldeepdhaka has quit [Max SendQ exceeded]
<ptitnuage>
MY123: thanks for answering
<ptitnuage>
i'm still trying to build my SD card
<ptitnuage>
hehe :-)
<ptitnuage>
see you later
<MY123>
ptitnuage: I also failed the SD card test
kuldeepdhaka has joined #linux-sunxi
Quarx has joined #linux-sunxi
libcg has quit [Ping timeout: 240 seconds]
bertrik has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
konradoo77 has joined #linux-sunxi
rz2k has quit []
<ptitnuage>
MY123: but did you succeed then?
techn_ has joined #linux-sunxi
<ptitnuage>
MY123: or do you still not booting linux at all?
<MY123>
ptitnuage: Still Android only
bonbons has joined #linux-sunxi
<ptitnuage>
MY123: is your board a Q8?
<MY123>
ptitnuage: The Kurio7S tablet.
techn has quit [Ping timeout: 245 seconds]
<ptitnuage>
MY123: but don't you know witch board is it?
<MY123>
ptitnuage: Will never open it.
<libv>
ptitnuage: i think i found the issue
<libv>
ptitnuage: the u-boot options should read: sun5i:POV_TAB_P703,SPL,CONS_INDEX=2
<libv>
i still need to go verify that it fails without that
<ptitnuage>
libv: did you spoke about compilation options or about runtime options?
<libv>
ptitnuage: boards.cfg
bengal has joined #linux-sunxi
astr has quit [Quit: Ex-Chat]
<ptitnuage>
libv: my boards.cfg states: Active arm armv7 sunxi - sunxi POV_TAB_P703 sun5i:POV_TAB_P703,SPL
<ptitnuage>
libv: or did i opened the wong file?
<ptitnuage>
libv:
<ptitnuage>
libv: sorry typed too fast
<MY123>
libv: Can you push u-boot Kurio7S support?(or there is a coding rule)
<ptitnuage>
libv: so i should add ,CONS_INDEX=2 at the end of the line, this is it?
konradoo77 has quit [Ping timeout: 240 seconds]
<ptitnuage>
libv: another question: does having a correct boot partition is enought to lead the tablet lightin the backlight, or will a correct root filsystem also be mandatory for this (beacouse it contains modules for example)
<MY123>
ptinuage: You can use a big initrd.
konradoo77 has joined #linux-sunxi
<ptitnuage>
MY123: my question was not "will the system boot without rootfs", but "will it ba able to light on the backlight without rootfs". I'm trying to localte where i'm doing wrong (witch stage) knowing that my backlight still drak...
<nedko>
libv: thanks, but i dont see anything that i don't know already. the problem is that this u-boot is different from the older one i used, it tries to boot from sata for example. i cleared almost all of the env because i want something simple
<nedko>
i have: bootargs=console=ttyS0,115200
<nedko>
i have the uImage loaded, i verify it with iminfo
<libv>
it cannot boot from sata iiirc
<nedko>
and when i do "bootm ${kernel_address}" i dont see the printks
<libv>
nedko: was this working without the sata stuff?
<nedko>
libv: it is a new board...
<libv>
if you don't know, start with sdcard, just don't give it a rootfs
<nedko>
i'm setting it for first time and i'm trying to reuse the uImage from my older board
<nedko>
the older board is a olinuxino-micro
<nedko>
a20
<libv>
nedko: you want rootfs on sata, and the rest on sd-card
<nedko>
the new one is also olinuxino a20, but a lime
<libv>
the boot stuff on sd-card really is the way to go today
<quitte>
doesn't that mean script.bin is the wrong one?
<nedko>
libv: this is how i have it in my older board, i know how to do it.
<libv>
nedko: you will need to change the bootloader as well
<nedko>
libv: what i don't know is how to pass the boot args...
<libv>
as you need to do with the script.bin as quitte says
<libv>
like in the example i gave you a link to
<nedko>
unfortunately the old bootloader didnt work on the new board
<libv>
just rootfs somewhere else
<nedko>
so i built a new one
<libv>
did you properly clear and then reinstall the boot area?
<nedko>
yes, with dd
<libv>
do you see any serial output from u-boot?
<nedko>
yes
<nedko>
it boots the kernel
<libv>
but...
<nedko>
i want to do setenv/saveenv commands to make it pass the bootargs
<nedko>
and it doesnt work
<nedko>
just setting bootargs in the env doesnt work
<libv>
nedko: mkimage?
<libv>
boot.cmd, mkimage
<nedko>
libv: it makes a binary script, right? why i need it?
<nedko>
i want to know the commmands in it...
<libv>
you can just strings it
<libv>
nedko: try with a boot.cmd and a boot.scr, and see if that works, then move on to uEnv.txt which i haven't used myself
<nedko>
libv: i already tried quite a few things...
<libv>
nedko: try creating the boot.cmd as described in our manual build howto
<quitte>
nedko: how do you know passing bootargs doesn't work?
<libv>
and altering to rootfs to read /dev/sda
<nedko>
quitte: because i dont see kernel printk()s
<quitte>
nedko: that doesn't mean assing bootargs didn't work
<nedko>
libv: the setup is quite different from the one in the wiki. for example i have only one partition, ext4 one
<nedko>
quitte: it means, because the bootars are: console=ttyS0,115200
<libv>
nedko: where is that partition?
<nedko>
quitte: and the same args on same kernel work on my other board
<libv>
nedko: on sd or on sata?
<nedko>
libv: on the microsd card
<nedko>
libv: i dont have sata so far on this board, and i know how to put rootfs on sata, it is just that i don't have sata disk for this board yet
<quitte>
nedko: unless ttyS0 is something different on the new board and the kernel does not know about it
<nedko>
quitte: both boards are A20...
<libv>
nedko: then why did you talk about sata?
<libv>
quitte: the script.bin should take care of that i guess
<nedko>
libv: because the uboot that works most on this board tries to boot from sata (i think so)
<libv>
quitte: "should" :)
<nedko>
libv: by default, i cleared most of the default env vars and it doesnt anymore
<quitte>
libv: do i understand that correctly: fex files are compiled into script.bin. and script.bin is similar to fdt, but u-boot "interprets" script.bin ?
<nedko>
libv: what git version should i use to have the "A20-OLinuXino_Lime" target? i don't see it in the wiki
<libv>
quitte: uboot just loads script.bin at a specific address
<libv>
nedko: master.
<nedko>
libv: thanks
* libv
refrains from typing out the next answer
<libv>
quitte: then the kernel is loaded at another, and the kernel reads in the script.bin
<nedko>
libv: what repo? the one i originally clonned u-boot-sunxi doesnt have a branch called master
<libv>
nedko: then it's sunxi, which is the default branch