skaczor has quit [Remote host closed the connection]
rm has joined #linux-sunxi
leviathanch2 has quit [Ping timeout: 252 seconds]
kivutar has quit [Quit: Ex-Chat]
Fusing has joined #linux-sunxi
Black_Horseman has quit [Read error: Operation timed out]
Black_Horseman has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
Black_Horseman has quit [Changing host]
peterbjo1nx has quit [Ping timeout: 252 seconds]
arky has quit [Quit: Leaving]
vector80 has joined #linux-sunxi
bjoernb has quit [Read error: Operation timed out]
<vector80>
Hi everyone, I have a strange issue. I have many boards with A20 and exactly same hardware. And also I have many Sdcards with debian and android images inside. When I connect these DEBIAN SDcards to some of my boards, they fail to boot, and stop at "waiting for root device"
<vector80>
However, these boards can boot from Android SDcard perfectly.
<vector80>
When I connect same Debian SDcards to other PCBA, they also boot fine.
bjoernb has joined #linux-sunxi
<vector80>
What could be wrong with those PCBA's that fail to boot debian Sdcard ? Any idea ?
<libv>
vector80: _exact_ same hardware?
<libv>
that seems to clash with "other pcba"
<libv>
but it is weird that android manages, while we don't
<libv>
but... that just means that you should go spend some time on narrowing it down and on improving either our u-boot or our kernel
<libv>
ssvb: i am removing libump from the sunxi-mali repository, now that the r3p0 x11 binaries have fixed dependencies
<vector80>
3) I tried Android SDCARD on that board, it boots perfectly
<vector80>
What else that I can do ?
<oliv3r>
it took us a good month to enable support for that board :(
<oliv3r>
vector80: it must be some timing thing; which kernel are you using for android.
<vector80>
it is aw kernel
<vector80>
3.4.39
<oliv3r>
and what kernel are you using for linux?
<vector80>
for debian, I am using 3 different kernels, 3.4.61+, 3.475+, and 3.4.79+, all can boot perfectly
<vector80>
all from sunxi
<vector80>
In my dmesg messages, I see a lot of warnings about my eMMC, but the boards are running it perfectly
<vector80>
I really hate these "company" things, otherwise I would directly cargo these boards to you people
<oliv3r>
yeah mmc controller stuff
<vector80>
I feel you never used any A20 device with an eMMC ?
<oliv3r>
nobody here has for sure
<oliv3r>
but is the SDcard causing problems, or the eMMC
<vector80>
It is amazing
<vector80>
I could write my eMMC by booting from TFTP+NFS
<vector80>
should I try that ?
<oliv3r>
depends on where the problem lies
<vector80>
yes, I really can't understand why it waits for /dev/mmcblk1p2
<vector80>
it is there, and it has /sbin/init
<vector80>
and when I connect this Sdcard to my other board, it run perfectly
<vector80>
And I am sure tha hardware is also perfect, because it can boot android from Sdcard in 17seconds
<vector80>
:)
<vector80>
I just can't find out what is the difference between two PCBA.
<oliv3r>
i'm not sure why it's causing problems
<oliv3r>
so ONE board boots it fine, but the other board doesn't boot it
<vector80>
I also want to explain something
<oliv3r>
and the boards are identical?
<vector80>
I feel MMC part in Aw kernel and sunxi kernel are different
<vector80>
yes oliv3r, they are identical
<oliv3r>
they are
<vector80>
Should I try to merge them ?
<oliv3r>
android 3.4 is newer mmc driver and i think there are a few patches on the ML to port it over
<oliv3r>
they are very very similar
<vector80>
oliv3r: I have to explain to you, what I have done in past months,
<vector80>
My mass production is almost completed
<vector80>
Boards can run debian, ubuntu, fedora, ubuntu, lubuntu....etc
<vector80>
my fex file gets almost perfect
<vector80>
I finished my FOTA system
<vector80>
I completed making an SDcard that automatically updates eMMC
<oliv3r>
awesome
<oliv3r>
just need to do the new_device_howto :p
<vector80>
yes, but I must know, how to make sure that my sdcard will be able to run on all 7500pcs boards ?
<vector80>
As you see, it just failed on two PCBA
<vector80>
I can't let my production people to use this Sdcard, some PCBA will fail to upgrade
<vector80>
I must find out, what is the problem with sunxi MMC driver ?
<oliv3r>
knx is an open standard and there's opensource stuff for it; amazing
<libv>
vector80: only you can find that problem
<oliv3r>
vector80: i think so, and i think you are 'on your own' here. i'm not sure anybody has encounterd this high failure problem yet
<oliv3r>
vector80: mmc stuff is slowly going into the mainline kernel now
<oliv3r>
you can try the sunxi-devel or sunxi-next kernels
<libv>
perhaps it is a clocking or power issue
<oliv3r>
it won't do everything you need; so it's not 'that' usefull, but you can see if the mmc problems are gone
<oliv3r>
if not, what libv just said
<vector80>
Alright
<libv>
as boards never are 100.000% compatible
<vector80>
I will try to git clone them tonight
<oliv3r>
sd cards ARE known to be fidity; there are microcontrolelrs with firmware trying to work around manufacturing issues in their firmware
<vector80>
The bad point is, I just can't see any other message than "waiting for root device"
<oliv3r>
vector80: but notice, this is just for experimentation; you'll need to figure out a few new things (all on our wiki) such as the device tree etc you need
<oliv3r>
vector80: yeah i understand :S
<oliv3r>
it's a difficult problem
<vector80>
yes
<vector80>
In the product lifetime, they would never be running from sdcard
<vector80>
%99 percent they would be running of the eMMC
<vector80>
And I never see any board to fail on eMMC
<vector80>
I feel there could be some timing or power issue regarding my board's TFcard circuit
<mripard>
oliv3r: the MMC is not in sunxi-next
<vector80>
but if this is the case, why android Sdcard is allways booting perfectly on it..
<libv>
vector80: because of difference in timing, clocking or power
<oliv3r>
mripard: good catch :)
<oliv3r>
mripard: it was in sunxi-devel though right?
<vector80>
So, it could be either kernel driver, or the hardware itself...
<oliv3r>
albeit without dma support?
nedko has quit [Ping timeout: 240 seconds]
<mripard>
oliv3r: yes
<mripard>
but with DMA
<mripard>
it has its own dma controller
leviathanch2 has quit [Ping timeout: 240 seconds]
<oliv3r>
vector80: i guess you should start by diffing aw-3.4 and sunxi-3.4 mmc driver and see the differences. then if there's very few, see if those diffs are on the ML list
<oliv3r>
mripard: ah nice one!
<vector80>
oliv3r: cool idea
<vector80>
Let me find out the differences
<oliv3r>
vector80: and if you do the new_device_howto; your board atleast be documented; even if there is propriatery hardware, that's ok. it's still really good to have there. and it won't take much of your time; 5 - 10 mins. http://linux-sunxi.org/New_Device_howto
<vector80>
Ok, I am really really a stupid one, I never created a wiki page in my life, let me try to learn it
<vector80>
may be I can copy from sample one
<oliv3r>
vector80: you are supposed to copy the sample page; and then fill it in :)
<oliv3r>
the howto explains it all :)
<oliv3r>
let me know when you have created an account; so we can activate it :)
<oliv3r>
vector80: btw, I do recall some mmc bug fixes on the ML a few weeks ago; so it's quite possible you are bumping into this bug. so the diff should show you that at the very least
<vector80>
yeah,
<vector80>
I will check that too
<oliv3r>
I don' tthinkt he diff will be big at all
<vector80>
oliv3r: I copied the contents of the edit, now, how can I create a new page ?
<oliv3r>
if you click that, you should get a new blank page :)
<oliv3r>
but you need to have created an account on the wiki first :(
puck_ has joined #linux-sunxi
puck_ is now known as fabianz
t3st3r has quit [Remote host closed the connection]
<vector80>
Ok I started
<fabianz>
Hello everyone! I tried posting my issue to mailinglist, but my posts never appeared in googles list overview. As I don't want to spam the mailinglist (even if I can't see my posts) I thought I'd take my issue here
<vector80>
Let me add some
t3st3r has joined #linux-sunxi
<oliv3r>
vector80: ok i'll fix your account then
<oliv3r>
fabianz: ask
<oliv3r>
vector80: ok i authorized you now to edit pages; you may have to relogin, but i doubt it :)
<oliv3r>
vector80: and if you get really board; i'm sure you will have to document all your tasks at some point :) (for yourself and for whoever comes in next). generic instructions would be cool on the wiki :) (like the FOTA stuff, the SD card updater etc)
<fabianz>
I have a problem regarding kernel config for Cubietruck/Cubieboard 2 - my distribution (Arch Linux ARM) packages linux-sunxi kernel for usage with both boards. It seems from my tests that Cubietruck needs all of FB_SUNXI=y in order for any display output to work (both VGA & HDMI). However, if FB_SUNXI_LCD=y the Cubieboard 2 seems to refuse to boot because of the LCD driver causing a kernel oops.
<oliv3r>
vector80: documentation would be usefull for others, but also for yourself, inc ase you forget ;)
<fabianz>
The distribution devs told me it won't be possible to integrate changes as long as the LCD driver problem doesn't get fixed upstream, so that's why I'm taking my issue here
<oliv3r>
fabianz: what kernel version do you ahve? arch might be using an old one that has long been fixed; we have nightly kernels available, dl.linux-sunxi.org/nightlies from the top of my head
<oliv3r>
fabianz: on top of that, libv is very busy in working up a KMS driver; so your issue will be fixed in due time ;)
<oliv3r>
fabianz: but strange, as the LCD stuff and FB stuff are all in use by people I imagine
<fabianz>
Arch is using 3.4.79 from commit 8ea347bcb
<fabianz>
oliv3r: Yeah, that was also what I thought. Kind of strange issue with the LCD driver
<fabianz>
oliv3r: I should also mention I can't test the LCD driver oops because I don't own a Cubieboard 2 myself. It was just that this was reported to me by the distribution devs
<fabianz>
oliv3r: Another weird aspect is why the Cubietruck display output won't work if FB_SUNXI_LCD is not set to Y
<fabianz>
dl.linux-sunxi.org/nightlies gives me a 404
<oliv3r>
can't say; I don't use the LCD stuff a whole lot; only vga or serial
<fabianz>
You can also see my PR in my first comment, which was reverted due to the suspicious LCD driver problem
<libv>
fabianz: i clicked one link already, but still cannot see the oops
<fabianz>
There were no log information posted by the distribution devs - they just wrote "The LCD driver built in is causing a kernel oops and halting the board. " As I don't own a Cubieboard 2 I can unfortunately not provide further information about this
<libv>
another link, no oops
<libv>
fabianz: then how do you expect anyone to debug this?
diego_r has joined #linux-sunxi
<fabianz>
I'm sorry if I'm causing trouble. I'm mostly trying to figure this out because the Cubietruck display won't work at all (neither HDMI nor VGA) if the LCD driver doesn't get statically built into the kernel
ssvb has quit [Ping timeout: 253 seconds]
<fabianz>
and distribution devs told me they can't to it because of the LCD issue, so they told me to take it upstream
<fabianz>
Their answer was "Simply put: take it to them, they do have Cubieboard 2, and Cubietruck thus they can actually sort the problem."
diego71 has quit [Ping timeout: 246 seconds]
<Turl>
fabianz: disp is known to be... let's say, not good quality
<oliv3r>
Turl: have you ever installed a cross-compiler on debian (from debian)
<oliv3r>
so not the linaro stuff ;)
<Turl>
fabianz: having said that, it used to work as modules. It just required a certain loading order
<oliv3r>
jessie all has it; but for wheezy; there's no arm cross compiler
shineworld has joined #linux-sunxi
<Turl>
oliv3r: on debian, don't think so
<Turl>
there's emdebian but dunno
<Turl>
ubuntu has some packages on their repos
<oliv3r>
yeah i tried to add the repo; but there's all sorts of conflicts
<fabianz>
Turl: i'm just baffled vga & hdmi stop working if lcd isn't compiled into kernel.. i tried various module loading /blacklisting already
<Turl>
fabianz: you need to load lcd, then hdmi and then disp modules
<oliv3r>
i just add the jessie repo; install it from there; and remove the repo! win :)
<Turl>
fabianz: that code is a mess; hopefully we'll be able to forget all this pain once libv gets sunxi-kms in shape :)
<Turl>
oliv3r: don't do that. You're asking for trouble
<ccaione>
Turl: pong
<Turl>
ccaione: hey
<ccaione>
yo
<Turl>
ccaione: did you submit any build fixes for watchdog that are not on torvalds/master?
diego71 has joined #linux-sunxi
<ccaione>
bulid fixes? no. Why?
<oliv3r>
Turl: ok what sane way is there to get gcc-arm-none-eabi onto wheezy then :)
<Turl>
ccaione: because 0bxxx notation is not standard and stuff explodes :)
ssvb has joined #linux-sunxi
<oliv3r>
emdeb won't work, dependancy problems
<Turl>
oliv3r: or wget linaro, untar, add to PATH
<Turl>
:p
<ccaione>
Turl: uat? lol ... I tried that stuff
<fabianz>
Turl: okay - can you give example? is it necessary to load fbcon as module?
<ccaione>
new gcc maybe?
<fabianz>
libv: I requested log of kernel oops from distribution devs. Sorry for troubling you
ZetaNeta has quit [Ping timeout: 246 seconds]
<Turl>
fabianz: dunno about fbcon, but you can probably keep it builtin
<ccaione>
Turl: have you a bug report or something?
<fabianz>
Turl: okay, i will try with the order you suggested and report back :)
<Turl>
fabianz: just make sure that either A) you keep lcd, hdmi, disp built in or B) they're all modules and you load them in that order (via eg insmod or putting in /etc/modules)
<Turl>
ccaione: I was just setting CI for my stuff, testing with torvalds/master
<Turl>
and it was like, wat, why is this broken? :)
<fabianz>
Turl: Okay, so it's either all or nothing. ty!
<ccaione>
Turl: gcc -v?
<ccaione>
I compile master almost every-day with sunxi-defconfig ... it never exploded
<Turl>
ccaione: it's sparse's fault if anything; I don't think it's a gcc warning
<oliv3r>
ccaione: GNU C provides several language features not found in ISO standard C. (The -pedantic option directs GCC to print a warning message if any of these features is used.)
<oliv3r>
ccaione: how are kernel folks about using gcc extensions?
<ccaione>
ok, but the kernel doesn't already use a lot of gcc ext?
<oliv3r>
probably
<ccaione>
sooooo
<oliv3r>
but they'll probably say 'try to minimize gcc specifc extensions'?
<oliv3r>
i dunno
leviathanch2 has quit [Quit: No Ping reply in 180 seconds.]
<oliv3r>
i'm just guessin' here :)
leviathanch2 has joined #linux-sunxi
<ccaione>
Turl: what's your $CHECK of C=2?
<Turl>
oliv3r: it's hard to read with non capitals imo :p
<Turl>
oliv3r: how many f's did I write? quick! :p fffffff
<oliv3r>
Turl: personally; i use captials in hex when the 0x is not there :)
<ccaione>
hahahaha
<oliv3r>
Turl: how is that different from FFFFFFF
<oliv3r>
my dislexia won't let me read either! :p
<oliv3r>
but remember, developers are lazy, and ffff is less effort then FFFFF ;)
<Turl>
ccaione: apt-get install sparse; make C=2 ... :)
<Turl>
oliv3r: capitals are easier to read and count on non monospaced fonts :p
rz2k has joined #linux-sunxi
<ccaione>
aaaarch non monospaced fonts
vector80 has quit [Remote host closed the connection]
<ccaione>
do you write code in non-monospaced font? :)
<oliv3r>
Turl: real developers use non capital hexadecimals when prefixed with 0x :p
<oliv3r>
ccaione: he codes in word!
<oliv3r>
:p
<oliv3r>
jk jk
<Turl>
ccaione: yeah, comic sans :D
<Turl>
ccaione: mripard I'll send that patch in later today :P
<ccaione>
you will be ack-ed :)
<oliv3r>
if you use small letters for the hexes :p
<oliv3r>
concistency etc
<Turl>
oliv3r: :p
leviathanch2 has quit [Ping timeout: 252 seconds]
<Turl>
mripard: ccaione I also found out that multi_v7 built, because it didn't have the watchdog driver enabled. Is that queued on arm-soc or in need of another patch as well?
<fabianz>
Turl: your advice was the one i missed! loading fbcon, lcd, hdmi, disp, disp_ump, ump in exactly this order brought my display back to life again :)
<mripard>
Turl: I have no idea.
<ccaione>
Turl: another patch is needed. I haven't updated multi_v7
<Turl>
mripard: I don't see it on armsoc/for-next, I'll submit another one then
<fabianz>
Turl: I missed loading lcd in my module load order, probably because i haven't though about using lcd anyway
<Turl>
fabianz: :)
<fabianz>
Turl: Thank you very much :)
<oliv3r>
to quote turl, 'disp is ... kinda fugly'
<fabianz>
yeah I see now.. but I guess we can look forward to libv making progress on sunxi-kms
<Turl>
fabianz: you're welcome
<oliv3r>
i'm just happy turl has a big memory and remembers all these quirkcs
<oliv3r>
speaking of which; i was experimenting with kernels for the bookt he other day
<oliv3r>
and i was using fedora 19's (hansg's) kernel with the debian builds too
<oliv3r>
i then switched to one of the nightly kernels + modules, but gmac wasnt' autoloaded anymore
<oliv3r>
manually loading it made network work just fine; but why wouldn't the gmac be autoloaded anymore?
<oliv3r>
or was the old gmac driver forced via some other way too I now wonder ...
<oliv3r>
but the name surely wouldn't have changed?
<oliv3r>
or! it was compiled in!
razlept has joined #linux-sunxi
shineworld has left #linux-sunxi ["Leaving"]
<oliv3r>
Turl: does arch have something like gentoo's crossdev to install the gcc arm toolchain? or how would it be done under arch
<fabianz>
good to have someone like Turl with a big memory here on irc :)
razlept has quit [Client Quit]
<ccaione>
oliv3r: you have xdev in aur usually
<Turl>
oliv3r: dunno, maybe on AUR. I just wget linaro/cs and add to path :p
<fabianz>
oliv3r: i installed gcc arm toolchain by gcc-arm-linux-gnueabihf-bin package on AUR
<oliv3r>
Turl: the ugly way then; no arch in book then!
<oliv3r>
crossdev ftw :p
<Turl>
oliv3r: you mean gentoo is nice? *bursts into laughter*
<Turl>
oliv3r: j/k :p
<oliv3r>
i did the gcc-arm-linux-gnueabihf-bin AUR package; i just want to mention via a line or two how to do it in other distro's that are not deb/rpm based
<oliv3r>
so having aur and crossdev mentioned is 'nice' if anything
<Turl>
oliv3r: how tolerant are they with dates btw?
<Turl>
still haven't sent my comments back :p
<fabianz>
oliv3r: so thanks for the packages ;) came in very handy instead of wget'ing/adding to path/etc
<fabianz>
*package
<oliv3r>
Turl: for the review stuff; based on your 'pay' very leniant
<oliv3r>
i've sent in a few chapters a day or two to late :)
<oliv3r>
fabianz: erm, your welcome! For? :)
<Turl>
oliv3r: yeah I can imagine :p
<oliv3r>
Turl: chapter 7 is going really slow; only at page 2 :(
<Turl>
oliv3r: ouch
<oliv3r>
and i may not get much more beyond 6 or 7
motley has quit [Remote host closed the connection]
<oliv3r>
chapter 8 will not be about xbmc at all
<Turl>
oliv3r: what was it about?
<oliv3r>
chapter 7? compiling bootloader + kernel from source
<oliv3r>
using the bsp :)
<Turl>
ah
<Turl>
:)
<oliv3r>
so that should be interesting :)
<fabianz>
oliv3er: just because you mentioned that you did the aur package i suggested ;)
<oliv3r>
fabianz: ah okay :)
<oliv3r>
Turl: chapter 8 will be about toggeling leds via gpio's or something
<oliv3r>
Turl: xbmc was 'if feasable' and it's not at all. i hoped we'd be coding harder :p
<Turl>
oliv3r: UEXT? :p
<oliv3r>
i blame jemk and nove
<oliv3r>
will mention that too; good tip
<Turl>
oliv3r: it'd be odd to mention uext on a book titled cubieboard tho
<razlept>
i will merge the rtc-sun5i.c with my current kernel sources
<razlept>
and give it a try
pwhalen has joined #linux-sunxi
akaizen has joined #linux-sunxi
akaizen_ has joined #linux-sunxi
nedko has joined #linux-sunxi
bertrik has joined #linux-sunxi
bertrik has quit [Changing host]
bertrik has joined #linux-sunxi
akaizen has quit [Ping timeout: 240 seconds]
popolon has quit [Quit: Quitte]
_massi_ has quit [Remote host closed the connection]
notmart has quit [Quit: notmart terminated!]
enrico_ has joined #linux-sunxi
CaptHindsight has joined #linux-sunxi
CaptHindsight has left #linux-sunxi ["not here"]
leviathanch2 has joined #linux-sunxi
diego_r has quit [Quit: Konversation terminated!]
<fabianz>
So - seems like beer o'clock now ;) Have a nice evening everyone!
fabianz has left #linux-sunxi [#linux-sunxi]
leviathanch2 has quit [Ping timeout: 240 seconds]
leviathanch2 has joined #linux-sunxi
kivutar has quit [Quit: Ex-Chat]
leviathanch2 has quit [Ping timeout: 268 seconds]
pwhalen has quit [Read error: Operation timed out]
bertrik has quit [Ping timeout: 264 seconds]
bertrik has joined #linux-sunxi
bertrik has quit [Changing host]
bertrik has joined #linux-sunxi
pwhalen has joined #linux-sunxi
FDCX has joined #linux-sunxi
kuldeepdhaka has quit [Quit: Leaving]
souther has joined #linux-sunxi
<oliv3r>
wens: how do you mean not wired properly, the Vin is unstable/not right, so the axp knows that, always? but not wired right to dot he interupt etc thing?
pireaa has joined #linux-sunxi
Night-Shade has joined #linux-sunxi
FR^2 has quit [Quit: Connection reset by peer]
jemk has joined #linux-sunxi
FreezingCold has joined #linux-sunxi
shineworld has joined #linux-sunxi
<pirea>
libv libump works whitout libdri2 and .... etc?
pireaa has quit [Remote host closed the connection]
drfu_ has left #linux-sunxi [#linux-sunxi]
shineworld has quit [Ping timeout: 265 seconds]
<Turl>
oliv3r: punit hasn't been posting for a bit
<razlept>
Turl: i try the link you gave me for the rtc-sun5i.c but too many error @ compilation
<Turl>
razlept: why are you trying to add it yourself? it's already there
<libv>
pirea: yes, libump itself just needs libc
<oliv3r>
Turl: he actually helped people a little
<oliv3r>
everytime i see puneet post a message; i sigh and say 'oh puneet'
<oliv3r>
i think he joined IRC the other week too
<libv>
pirea: why do you ask?
<pirea>
libv because i had update 1 pkgbuild for arch arm
atsampson has quit [Quit: soundcard upgrade time]
<libv>
pirea: so arch has these things packaged?
<razlept>
Turl: i git pull the latest sources of sunxi-3.4 but its different than the file you point me above
<Turl>
oliv3r: dunno, but the less code we need the merrier I suppose :)
<oliv3r>
i'm kinda annoyed I was sick the night of the sunxi dinner
<oliv3r>
while hans and ian's effort are way more then admirable for trying to get u-boot mainlined
<oliv3r>
there is really so so so much to continue cleaning up on :))
<oliv3r>
and i should have done smaller incremental steps in committing my git stuffs :p
jemk has quit [Remote host closed the connection]
hramrach has quit [Remote host closed the connection]
hramrach has joined #linux-sunxi
rz2k has joined #linux-sunxi
<razlept>
Turl: uncommenting not wokring.. kernel crash while booting.. i will try the link you provide and report back
xavia has left #linux-sunxi ["PART #interlock :PART #arduino :PART #wnydev :PART #buffalolab :PART #docker :PART #symfony :PART #virtuvia :QUIT :Leaving."]
akaizen has quit [Remote host closed the connection]
akaizen has joined #linux-sunxi
bertrik has quit [Remote host closed the connection]
<oliv3r>
hno: i'm reading board.c and you state there, that we don't know how SPL was loaded. I noticed each compnent enables clocks and gates, i'm sure it doesn't 'close' them. Using FEL mode, we can see what the default gates are (probably all closed) and we could use that to see which gates are open, mmc, nand or SPI. figuring out if it was mmc0 or mmc2 will be harder I guess :)
<razlept>
hmm... cant make it
<oliv3r>
hno: this isn't an issue now, when we always boot from MMC anyway, but if we start adding SPI boot, and far in the future nand boot ...
Night-Shade has quit [Ping timeout: 268 seconds]
<wigyori>
mripard: i think that's from an old tree - already obsoleted
<wigyori>
mripard: and yes, that was to have full coverage - to let the rest of the patches apply
<oliv3r>
hno: one more thing; what's the difference between board.c/board_init() and sunxi/board.c/s_init(); which one is called first? and should the asm from the one move to the other to have it all in the same spot? (I imagine s_init() to go first
Xaros has joined #linux-sunxi
Xaros has quit [Remote host closed the connection]
Xaros has joined #linux-sunxi
Black_Horseman has quit [Ping timeout: 264 seconds]
Xaros has quit [Client Quit]
Black_Horseman has joined #linux-sunxi
<oliv3r>
ijc: i see in your v2 your removing a lot of the sun4i stuff; wouldn't it be wiser to get it properly tested and try to submit sun4i + sun7i (and sun5i should be doable aswell)? I know I have sun4i, sun7i hardware + a sun5i tablet (u-boot testing will be hard on that one :p)
Seppoz has quit [Remote host closed the connection]
<libv>
ssvb, (others): maybe it is time to rename Binary_Drivers as it is very Mali specific
<libv>
we perhaps want an Xorg page, explaining how to set up xf86-video-fbdev/fbturbo
<libv>
also, the prerequisites seem off
ganbold_ has quit [Remote host closed the connection]