mnemoc changed the topic of #linux-sunxi to: Allwinner/sunxi development discussion - did you try looking at our wiki? - Don't ask to ask. Just ask and wait! - - Logs at
hramrach has quit [Ping timeout: 272 seconds]
leviathanch2 has quit [Ping timeout: 252 seconds]
leviathanch2 has joined #linux-sunxi
Andy-D has quit [Remote host closed the connection]
t3st3r has quit [Quit: C ya later, all.]
deasy has quit [Remote host closed the connection]
leviathanch2 has quit [Ping timeout: 252 seconds]
kenny has joined #linux-sunxi
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
payne has joined #linux-sunxi
<kenny> it's late here, so I might have just messed up some config, but trying to boot hans/sunxi-devel with the latest code won't work for me on a CB2: . I don't think I changed anything since the last kernel build and there's some mmc errors in the log. Ideas?
FreezingAlt is now known as FreezingCold
[7] has quit [Disconnected by services]
TheSeven has joined #linux-sunxi
lioka has quit [Ping timeout: 252 seconds]
lioka has joined #linux-sunxi
tomboy64 has quit [Write error: Connection reset by peer]
tomboy64 has joined #linux-sunxi
jeremb_ has quit [Quit: Connection closed for inactivity]
JohnDoe_71Rus has joined #linux-sunxi
imcsk8 has quit [Quit: Reconnecting]
imcsk8 has joined #linux-sunxi
<wens> mripard: thanks for the review
<wens> some of my fixups were squashed onto the wrong commits
alexvf has quit [Ping timeout: 240 seconds]
uwe_ has quit [Ping timeout: 258 seconds]
HeHoPMaJIeH has joined #linux-sunxi
HeHoPMaJIeH has quit [Changing host]
HeHoPMaJIeH has joined #linux-sunxi
diego_r has joined #linux-sunxi
libcg has joined #linux-sunxi
sehraf has joined #linux-sunxi
amitk has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
bbrezillon has joined #linux-sunxi
payne has quit [Remote host closed the connection]
FreezingCold has quit [Ping timeout: 245 seconds]
<ccaione> moin
MadSpark has quit [Ping timeout: 240 seconds]
nicksydney has quit [Remote host closed the connection]
merbanan has quit [Ping timeout: 276 seconds]
nicksydney has joined #linux-sunxi
MadSpark has joined #linux-sunxi
merbanan has joined #linux-sunxi
blsd has joined #linux-sunxi
apo__ has joined #linux-sunxi
apo_ has quit [Ping timeout: 252 seconds]
notmart has joined #linux-sunxi
wickwire has joined #linux-sunxi
popolon has joined #linux-sunxi
shineworld has joined #linux-sunxi
lioka has quit [Changing host]
lioka has joined #linux-sunxi
jeremb_ has joined #linux-sunxi
_massi_ has joined #linux-sunxi
<wigyori> morning
smotocel69_ has joined #linux-sunxi
<shineworld> morning wigyori
smotocel69 has quit [Ping timeout: 252 seconds]
FR^2 has joined #linux-sunxi
avsm has joined #linux-sunxi
TuxboxGuru has quit [Read error: No route to host]
<shineworld> there is a doc with differences between A20 rev A and rev B ?
<shineworld> what change at kernel level ?
<shineworld> what could change a script.bin level (I guess in dma timings or like that)
<JohnDoe_71Rus> don't seen this docs
<JohnDoe_71Rus> hi
deasy has joined #linux-sunxi
deasy has quit [Remote host closed the connection]
philippe_fouquet has joined #linux-sunxi
deasy has joined #linux-sunxi
<mnemoc> experimental/sunxi-3.14 = v3.14.4 + all sunxi patches I could find a torvalds/master pushed
<mnemoc> (aiming at a mainlineish stable branch, and 3.14 was already announced as the next LTS)
blsd has quit [Remote host closed the connection]
avsm has quit [Quit: Leaving.]
<shineworld> cool, should experimental/sunxi-3.14-android to be used or is just for guru ?
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
<mnemoc> experimental/sunxi-3.14-android is the merge of experimental/sunxi-3.14 and google's experimental/android-3.14
<mnemoc> so expertimental^3 :p
<mnemoc> testing the android-less experimental/sunxi-3.14 first would be appreciated :)
<shineworld> this evening I will try to paly with it :)
<shineworld> play
<mnemoc> awesome
<mnemoc> thanks!
<wigyori> hey
<wigyori> seen this?
<shineworld> at now :)
<shineworld> I will be happy to see Android sources for HP7Plus....
<shineworld> but I don't know what allwinner devices has 4 x 1Ghz chip...
<wigyori> a31
Black_Horseman has quit [Quit: Zwi se logou mou!!!]
<shineworld> only 1Ghz ?
<shineworld> I know A31 works at 1.5Ghz
<mripard> shineworld: mine runs at ~1GHz
<shineworld> I'm used to works with Android Sources for CB2 from Cubieboard team (sdk 1.05) and I'm just curious to see other "customizations"
FDCX has quit [Ping timeout: 252 seconds]
<mnemoc> mripard: is the vger list already created? to get subscribed :)
<mripard> mnemoc: nope, didn't ask it yet
<mripard> it's still on my todolist :)
<mnemoc> ok
<shineworld> ah ... right ... also mine run at 1Ghz.... (I've got wrong doc from google) :
<shineworld> is a 13.3" tablet from Colorfly
blsd has joined #linux-sunxi
<shineworld> actually we use a lot of "gadgets" powered by allwinner
arete74_ has joined #linux-sunxi
arete74 has quit [Read error: Connection reset by peer]
<oliv3r> mripard: mnemocI asked for a vger list 3? months ago, it got silently ignored
<oliv3r> :(
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier.]
lgentili has joined #linux-sunxi
<lgentili> news: Hewlett Packard Unveils $100 HP 7 Plus Tablet Powered by AllWinner A31 Quad Core SoC Read more:
lgentili has quit [Client Quit]
FDCX has joined #linux-sunxi
<mnemoc> oliv3r: hi! what's the current state of the MX and mailman at ?
<mripard> oliv3r: what did you ask for?
montjoie_[home] has quit [Quit: Quitte]
<ssvb> oliv3r: hi there, did you have time to run some tests on your cubietruck?
philippe_fouquet has quit [Remote host closed the connection]
philippe_fouquet has joined #linux-sunxi
arete74 has joined #linux-sunxi
arete74_ has quit [Read error: Connection reset by peer]
leviathanch2 has joined #linux-sunxi
montjoie[home] has joined #linux-sunxi
Faisal has joined #linux-sunxi
deasy has joined #linux-sunxi
jemk has joined #linux-sunxi
jinzo has joined #linux-sunxi
leviathanch2 has quit [Ping timeout: 240 seconds]
<oliv3r> mnemoc: erm, exactly the same as before :) postfix is setup, ; postfixadmin works, smtp auth needs to be tested, but shouldn't require a lot
<oliv3r> mripard: for a mailing list for sunxi @vger
<oliv3r> ssvb: not yet, it was busy doing a yum upgrade till in the whee hours; i have it scheduled for tonight though ;)
<oliv3r> and i need a nother serial converter, my tx is broken atm :(
leviathanch2 has joined #linux-sunxi
<oliv3r> i do have lots of exciting things to share, in dec. :(
philippe_fouquet has quit [Remote host closed the connection]
philippe_fouquet has joined #linux-sunxi
philippe_fouquet has quit [Read error: Connection reset by peer]
philippe_fouquet has joined #linux-sunxi
cosm has joined #linux-sunxi
<ssvb> oliv3r: ok, thanks
smotocel69_ has quit [Ping timeout: 276 seconds]
rz2k has joined #linux-sunxi
<oliv3r> ssvb: just to recap; leave everything at 'default, except FAST_MBUS' then compile via .sh and run
<ssvb> oliv3r: yes, run it as root using something like "lima-memtester 100M" command
<oliv3r> rgr; will do tonight
<ssvb> oliv3r: this will use 100MB buffer size, setting the buffer size too large does not improve anything and only risks causing allocation failures
<mnemoc> oliv3r: I didn't remember the "before" state, that's why I asked ;-)
<oliv3r> ah
<oliv3r> ok well postfix should be fully functional and postfixadmin aswell
<oliv3r> you did have the login for that ;)
<oliv3r> don't ask me what your password was :)
<oliv3r> my browser luckly remembered mine :p
<mnemoc> ow
<oliv3r> so i might be able to change yours if you don't remember yurs :)
<mnemoc> yes, please
<oliv3r> is your username; you should be able to do a 'forgot password' thing actually
<mnemoc> cool, thanks
<oliv3r> next thing that needs to be done, is add an ssl cert and secure smtp n stuff
smotocel69 has joined #linux-sunxi
Nazcafan has joined #linux-sunxi
paulk-collins has joined #linux-sunxi
rz2k has quit [Read error: Connection reset by peer]
shineworld has quit [Quit: Sto andando via]
<mnemoc> oliv3r: we got a free "real" ( cert which is currently in use for https. hopefully we can recycle that into the smtp
avsm has joined #linux-sunxi
<wens> no reason we can't
<mnemoc> :)
<mripard> wens: have you seen that GMAC looks broken in linux-next on the A20 ?
<wens> for real? :(
<mripard> unfortunately, yes
<wens> this is built from linux-next?
<mripard> yes, see the title
<mripard> tag next-20140526
<wens> suspecting someone made the WOL interrupt mandatory
<wens> let me dig through some history
<mripard> I don't know, it doesn't seem to be the same failure for all the defconfig
HeHoPMaJIeH has quit [Remote host closed the connection]
<wens> I'll give it a try tomorrow
libcg has quit [Quit: dogecoin DDNPK9LJwEzXJK1f7hoMwjvGGGHF8gkBmw]
syeekick has joined #linux-sunxi
Faisal has quit [Read error: Connection reset by peer]
Faisal has joined #linux-sunxi
<mnemoc> Turl: experimental/sunxi-3.14 now compiles. but sunxi-common-regulators.dtsi fails to compile
<mnemoc> I mean. make uImage and make modules work fine. make dtbs files on the sunxi-common-regulators.dtsi files
<mnemoc> and I have no idea why :(
<mripard> because it's not meant to be compiled.
<mnemoc> fair point :) .... but why `make dtbs` is doing so?
<mnemoc> FATAL ERROR: Unable to parse input tree
<mnemoc> Error: /srv/build/amery/allwinner/sunxi-nightly/linux-sunxi-experimental-3.14/arch/arm/boot/dts/sunxi-common-regulators.dtsi:14.1-2 syntax error
<mnemoc> make[2]: *** [arch/arm/boot/dts/sun7i-a20-cubieboard2.dtb] Error 1
<mnemoc> sunxi-common-regulators.dtsi:14 is the opening / {
<ssvb> oliv3r: should be a good start, but we can improve it
<mripard> mnemoc: ah. is the code there somewhere?
<mnemoc> the last 4 commits are out of order because I cherry-picked them to fix building after the automatic backporting
<mnemoc> mripard: sunxi_defconfig
<mnemoc> same for multi_v7_defconfig
<mnemoc> I assume I forgot to cp something... but can't find what :(
<mnemoc> cherry-pick*
philippe_fouquet has quit []
tomboy64 has quit [Ping timeout: 272 seconds]
avsm has quit [Quit: Leaving.]
tomboy64 has joined #linux-sunxi
<mripard> mnemoc: hmmm, weird
<mripard> it builds just fine here
<mnemoc> uhm
<mnemoc> mripard: on that branch?
* mnemoc tries a fresh tree
<mripard> on sunxi-next
<mnemoc> mripard: can you try on experimental/sunxi-3.14 ?
<mnemoc> it's a backport from torvalds/master upon 3.14.4
* mnemoc didn't know `tig` supported --no-merges <3
leviathanch2 has quit [Ping timeout: 240 seconds]
bertrik has joined #linux-sunxi
leviathanch2 has joined #linux-sunxi
<mnemoc> mripard: nevermind, found it.
<mnemoc> a }; lost when "fixing" a conflict
<mnemoc> in arch/arm/boot/dts/sun7i-a20.dtsi
FreezingCold has joined #linux-sunxi
diego_r has quit [Ping timeout: 252 seconds]
fredy has quit [Excess Flood]
Guest57375 has joined #linux-sunxi
bonbons has joined #linux-sunxi
bgal has joined #linux-sunxi
jinzo has quit [Ping timeout: 245 seconds]
_massi_ has quit [Remote host closed the connection]
bbrezillon has quit [Ping timeout: 252 seconds]
jinzo has joined #linux-sunxi
Guest57375 is now known as fredy
gzamboni has quit [Ping timeout: 252 seconds]
avsm has joined #linux-sunxi
gzamboni has joined #linux-sunxi
bbrezillon has joined #linux-sunxi
jemk has quit [Quit: leaving]
FR^2 has quit [Quit: Connection reset by peer]
kuldeepdhaka has joined #linux-sunxi
bgal has quit [Ping timeout: 252 seconds]
Netlynx has joined #linux-sunxi
libcg has joined #linux-sunxi
nove has joined #linux-sunxi
FDCX has quit [Remote host closed the connection]
FDCX has joined #linux-sunxi
libcg has quit [Ping timeout: 252 seconds]
FDCX has quit [Remote host closed the connection]
FDCX has joined #linux-sunxi
dack has joined #linux-sunxi
wickwire has quit [Quit: Leaving]
<nove> another afternoon wasted with blobs, that api) is working with libcedarx, but api) that is a wrapper for the old api, fails in h264 decode
<dack> I'm currently using a cubieboard with linux running on the NAND. I was thinking of switching the filesystem to UBIFS. How are you supposed to modify configurations for u-boot-sunxi? Should I just manually edit include/config.h?
<dack> .. maybe modify boards.cfg with my own entry?
ninolein_ has quit [Ping timeout: 252 seconds]
<libv> dack: there is no need to use ubifs with sunxi-3.4
<libv> dack: libnand does the wear levelling for you
ninolein has joined #linux-sunxi
libcg has joined #linux-sunxi
<libv> dack: note that you get block devices nodes in /dev/ instead of mtd
<dack> libv: so, putting the root filesystem on the NAND with ext4 has no issues?
<libv> dack: it shouldn't have issues
<libv> dack: that's not to say that there never are going to be any
<libv> dack: libnand does the wear levelling for you, that's all you need to know
<dack> libv: okay, so what is libnand? that something in the kernel providing wear-leveling?
Andy-D has joined #linux-sunxi
<libv> dack: a nasty allwinner hack, but it does what linux doesn't do (for historic and supposed patent reasons, but my feeling is that it is more of the former than the latter)
<dack> libv: okay.. what kernel config controls that? CONFIG_SUNXI_NAND?
<dack> libv: ... just trying to understand where this "hack" is located and a bit of how it works
<libv> dack: you already have this working if you started with sunxi-3.4 and a defconfig
<libv> and yes, CONFIG_SUNXI_NAND
<dack> libv: I figured as much, just wanted to further my understanding.. ;)
<libv> dack: it lives under drivers/block/sunxi_nand/
<dack> libv: okay, thanks for all the help and info!
<libv> dack: more info is here btw:
<libv> although that doesn't really explain what libnand is or does.
montjoie[home] has quit [Quit: Quitte]
rndnick82OO0 has joined #linux-sunxi
leviathanch2 has quit [Ping timeout: 240 seconds]
mjaco has joined #linux-sunxi
montjoie[home] has joined #linux-sunxi
<mjaco> Hi all, I’m having some issues with my displays (connected through HDMI and VGA) waking up from sleep on my A20-OLinuXino. As in they don’t.
notmart has quit [Quit: notmart terminated!]
FR^2 has joined #linux-sunxi
jemk has joined #linux-sunxi
avsm has quit [Quit: Leaving.]
montjoie[home] has quit [Read error: Connection reset by peer]
CaptHindsight has quit [Ping timeout: 240 seconds]
Netlynx has quit [Quit: Leaving]
apxiii has joined #linux-sunxi
apxiii has quit [Quit: KVIrc 4.2.0 Equilibrium]
jinzo has quit [Ping timeout: 240 seconds]
CaptHindsight has joined #linux-sunxi
<syeekick> anyone got a kali image i can flash striaght to the nand of the cubie truck?
fredy has quit [Excess Flood]
<libv> syeekick: have you tried asking the kali people themselves?
<libv> and image, hah, dream on
<libv> you'll be lucky if you get a rootfs.
<libv> syeekick: i am amazed that someone with such little real linux/unix knowledge is interested in exactly this distribution
<libv> this does not make me too positive about your intentions with it.
fredy has joined #linux-sunxi
<syeekick> there is a rootfs for it
<syeekick> and an image
<syeekick> but it wont flash to the cubietruck
kuldeepdhaka_ has joined #linux-sunxi
kuldeepdhaka has quit [Ping timeout: 252 seconds]
<libv> syeekick: maybe "to flash" is not the verb you should be using here.
heffer_ is now known as heffer
<mjaco> My two displays are connected to an A20 OLinuXino using HDMI and VGA running Arch, they go into sleep OK, but they don’t wake up. Running ‚xset dpms force on’ doesn’t work. When the screens are working and I run ‚xset dpms force off’ the screens go standby, but if I then run ‚xset dpms force on’ nothing happens. Only a reboot turns them on again. Is there anything else I can try? I’m using the fbturbo drivers and the d
<mjaco> be two different fbs
cernui has joined #linux-sunxi
libv has quit [Ping timeout: 260 seconds]
<ssvb> mjaco: this looks like all the same popular bug, everyone is complaining about
<ssvb> mjaco: very likely in the sunxi-3.4 kernel disp driver
cernui has quit [Ping timeout: 252 seconds]
<mjaco> ow, okay. I didn’t find any more information about this, is there some bug tracker I can find the issue?
<mjaco> ssvb
<oliv3r> ssvb: lima tester compiling; will update you soon
<oliv3r> /bin/ld: cannot find -lm
<oliv3r> oh wow, that's pretty basic to fial :p
<oliv3r> /usr/lib/ does exist; lets see what makefile causes it; the build script does seem quite quiet
<sehraf> i have the same with -lrt the lib is there but it can't find it..
<ssvb> oliv3r: you can probably drop the -static option
<ssvb> oliv3r: I guess, /usr/lib/libm.a might be just missing
<oliv3r> ah
<oliv3r> possible
<oliv3r> i just did a math.h test program and -lm worked fine
<oliv3r> I probably need the -dev packages? Though I'd expect math.a be pretty standard
<oliv3r> then again, i'm a gentoo guy, i always have my -dev packages around :)
<ssvb> it suggests that "yum install glibc-static" might fix it
<oliv3r> probably
<oliv3r> i'll just build a dynamic
<oliv3r> Compilation succeeded, now you have the 'lima-memtester' binary.
<oliv3r> yay
<oliv3r> rebooting without the mem reserve line
mjaco has quit [Quit: mjaco]
<oliv3r> ssvb: the cube is a little choppy, is that okay?
<oliv3r> i see a little tearing, and i would guess it runs at about 25+ fps
mjaco has joined #linux-sunxi
<oliv3r> it just stalls every now and then (once every 30 seconds?) for 0.5 seconds maybe
<oliv3r> also X running behind it may not help it much
<ssvb> it just runs too fast, more like 300 fps
<oliv3r> i want a --fps! :p
<oliv3r> so i can see the fps
<oliv3r> is that using limare?
<ssvb> yes, no userland blobs
<oliv3r> oh wow nice
<oliv3r> what's the lima state atm anyway?
<ssvb> that was the fosdem 2013 milestone
<oliv3r> so simple, yet so fasenating
<oliv3r> (test still running btw)
<ssvb> or even earlier than that, fosdem 2013 had quake :)
<oliv3r> i haven't read a blog post from libv either; how's he doing anyway?
<mnemoc> fighting trolls in our wiki
<oliv3r> lol
<oliv3r> and code wise? :p
<oliv3r> There will be a 'ndh' paragraph in the book :)
<ssvb> oliv3r: if the test has not failed yet, then it is likely going to be fine
<oliv3r> ssvb: still going
<oliv3r> i'll let it finish (bit flip 120 it is at right now)
<ssvb> oliv3r: is the system totally unbootable with FAST_MBUS?
<oliv3r> what will be the next step?
<oliv3r> yes
<oliv3r> it crashes during boot
Black_Horseman has joined #linux-sunxi
Black_Horseman has quit [Changing host]
Black_Horseman has joined #linux-sunxi
<oliv3r> ssvb: sometimes it manages to get passed the boot; but crashes on anything intensive
<oliv3r> SoC is quite hot!
<oliv3r> dram is ok
<Black_Horseman> hola
rndnick82OO0 has quit [Remote host closed the connection]
<oliv3r> dram i can keep my finger on for a good 30 seconds and longer, and it doesn't hurt really
<oliv3r> SoC hurts after 10 seconds
<oliv3r> but that's me being tough, 5 seconds is allready 'long' :p my poor thumb :(
<oliv3r> ok its starting loop 2: now
<mnemoc> oliv3r: during your tests, can you give a try to experimental/sunxi-3.14 ?
<oliv3r> is that a mainline kernel?
<mnemoc> yes
<oliv3r> does it do mmc?
<mnemoc> it matches torvalds/master
<ssvb> oliv3r: it would be interesting to dump the values in the rslr0 and rdgr0 dram controller registers right after dram is initialized in u-boot
<ssvb> oliv3r: and compare the values with and without FAST_MBUS
<oliv3r> mnemoc: i wish i could; but i'm working on the book while watching a spinning cube and pestering ssvb
<mnemoc> ok
<oliv3r> ssvb: ok, dump rslr0 and rdgr0 from this fastless u-boot; then compile u-boot with fast_mbus and see them again
<oliv3r> ssvb: how do I dump those from u-boot again? My knowledge with regards to u-boot has abbandoned me
<mnemoc> if someone can confirm it works I can continue backporting sunxi-next stuff, and then sunxi-devel
<oliv3r> mnemoc: 3.14 will replace 3.10?
<mnemoc> i think so
<oliv3r> or is 3.14 3.4 + mainline combined?
<mnemoc> probably a sunxi-3.14-hybrid branch will come
<mnemoc> but mainline is mature enough to deserve it's own stable
<mnemoc> and 3.14 is LTS
<ssvb> oliv3r: if you are recompiling u-boot anyway, just adding printf at the end of dram initialization can also do it
<oliv3r> mnemoc: i thought 3.10 was lts
<oliv3r> aight
<oliv3r> lemme see if i can find it at all :)
<oliv3r> (its been awhile
<mnemoc> oliv3r: greg posted 3.14 was going to be LTS a week ago
<oliv3r> ah okay
<oliv3r> well then fuck 3.10 :p
<oliv3r> 3.14 will be MUCH easier to backport stuff to
<mnemoc> in which case it doesn't make much sense to work on an "old" LTS
<mnemoc> yup
<oliv3r> what I want for xmas now is, dma enabled NAND on an A10
<ssvb> oliv3r: which snapshot of u-boot are you using? I can probably provide the needed patches to save your time
<oliv3r> who wants cookies and milk for that chore?
<oliv3r> ssvb: erm head
avsm has joined #linux-sunxi
<mnemoc> ssvb: any tree/branch I should add to the u-boot nightlies?
<oliv3r> ssvb: please do :)
<ssvb> oliv3r: the current up to date 'sunxi' branch from u-boot-sunxi?
<oliv3r> i'm sad I missed the whole dram awesomesauce hacking
<oliv3r> ssvb: yep
<ssvb> oliv3r: ok, give me a few minutes
<oliv3r> 44910e96f
<nove> what about this for the wiki
rz2k has joined #linux-sunxi
<nove> is another complexity level, but could help editing for the ones that wikitext is a no
<oliv3r> ssvb: loop 3 btw :)
bgal has joined #linux-sunxi
<mjaco> Is there anywhere I can track the ‚display not waking up’ bug on git or something? I’ve looked through all issues in linux-sunxi and didn’t see one with the same issue as I have (as far as I can tell).
bonbons has quit [Quit: Leaving]
jemk has quit [Quit: leaving]
<ssvb> mjaco: somebody just has to debug and fix it
mjaco has quit [Quit: mjaco]
<syeekick> anyone know anything about the cubie truck >
* mnemoc looks at one on 30cm away of his mouse
dack has quit [Remote host closed the connection]
<mnemoc> s/on/one his desk/
libv has joined #linux-sunxi
<mnemoc> wb libv
<mnemoc> .be?
<libv> nope, flaky hw
<oliv3r> wb libv!
<oliv3r> i just orders a jesus a10 hdmi stick with 1gb ram
<oliv3r> 35 USD @ dx
<oliv3r> ssvb: loop4; looks like it sstable :p
<oliv3r> just waiting for your patch and i will integrate it
<syeekick> mnemoc, you ever installed kali on that?
<ssvb> oliv3r: on my Cubietruck it reports -
<oliv3r> i'm glad when my book will be done; and we can forward people to te book :)
<mnemoc> syeekick: no
<syeekick> you reckon its the possible?
<mnemoc> oliv3r: CC-By ?
<ssvb> oliv3r: if you get different results with and without FAST_MBUS, I think that would explain everything
<oliv3r> mnemoc: CC-by-give-me-money >: )
<oliv3r> mnemoc: i doubt i will make a profit
<ssvb> oliv3r: also it would be interesting to know if the results from your CT are not the same as mine :)
Black_Horseman has quit [Quit: Zwi se logou mou!!!]
Xaros has joined #linux-sunxi
Xaros has quit [Read error: Connection reset by peer]
Black_Horseman has joined #linux-sunxi
<ssvb> oliv3r: and also one more thing, the DQS gate training basically selects a very low accuracy initial approximation
<oliv3r> (i don't know that much about dram stuff :(
<oliv3r> it is slightly different
<syeekick> anyone got a guide i can follow that would be altered a bit to suit a kali install ?
<oliv3r> [ 9.548179] Unable to handle kernel NULL pointer dereference at virtual address 00000000
<oliv3r> BOOK
<ssvb> oliv3r: thanks a lot! looks like that was it
<oliv3r> BOOM*
<oliv3r> it's trying to do 'something' while booting though
<oliv3r> all leds are solid on :)
<ssvb> oliv3r: can you check one more thing? in the cubietruck dram_para struct, just set tpr3=0x072222
<oliv3r> WITH fast_mbus?
<ssvb> oliv3r: with and without it, the chances are that they are going to converge
<ssvb> oliv3r: basically, it's like the search for the missing boeing :) the hardware dqs training is something like identifying the large search area
<oliv3r> looks stable now
<oliv3r> oh no
<oliv3r> crash
<oliv3r> lots of crashes whilst trying to boot
<ssvb> oliv3r: and inside of this area, tpr3 provides a more precise location, here are some pictures -
<oliv3r> ssv with 'fixed' being the new tpr3
<ssvb> oliv3r: for the cubietruck, the best tpr3 settings are in the top left corner of the table (which is on the borderline)
<oliv3r> (page still loading :()
<oliv3r> nvm, refresh fixed it
<oliv3r> best being the savest?
<mnemoc> syeekick: i think no one here uses kali, or even know about it
<ssvb> oliv3r: yes, the 'green' areas are the most reliable
<oliv3r> ssvb: ok, so the normal mbus works normally again; running limatester
<oliv3r> so should i try the 'green' trp3?
jinzo has joined #linux-sunxi
<ssvb> oliv3r: 0x72222 is already 'green' from that table
<oliv3r> yeah was looking at cb1
<oliv3r> so next step?
<oliv3r> fast_mbus still breaks things
<ssvb> oliv3r: yeah, seems like we see that the hardware dqs training is providing unreliable results on your Cubietruck
<ssvb> oliv3r: and you are also getting "dqs_gating_delay=05060606", while I have "dqs_gating_delay=06060606"
<ssvb> oliv3r: one lane still seems to be off and this could make your board unreliable
<oliv3r> no way to override this?
<ssvb> oliv3r: we could try to force it to 06060606 and check if this fixes everything for you
<oliv3r> but then, is that a recommended course of action for all ct's
<ssvb> oliv3r: basically, you can try to just set the rslr0 and rdgr0 registers to the same values as reported in my log at
<oliv3r> but wouldn't that be some 'heavy tweaking'
<ssvb> oliv3r: yes, it's just as a quick test to confirm if it helps or not
<ssvb> oliv3r: as for what would be recommended, I don't know, we definitely need test results from more hardware to ensure that dram is configured correctly everywhere
<oliv3r> ok enabling fast_mbus and setting your rslr0 and rdgr0
<oliv3r> just set those values just after those printf's you added should be ok?
<ssvb> oliv3r: yes, very likely
uwe_ has joined #linux-sunxi
avsm has quit [Quit: Leaving.]
<ssvb> oliv3r: I hope we don't need to do a dll reset after this
<oliv3r> any easy way to check?
<oliv3r> looks stable
<oliv3r> but can't say if those changes have been taken
jinzo has quit [Quit: Leaving]
<ssvb> oliv3r: with tpr3=0x072222 and dcdc3 voltage set to 1.3V fex, you can probably safely use dram clock frequencies 504MHz and 528MHz and maybe even higher
<oliv3r> any easy way to check rslr0 from linux
<ssvb> oliv3r: we can modify a10-meminfo to report this information
<oliv3r> true that
<oliv3r> well it's pretty safe to assume those changes where activated?
<ssvb> if your board is stable with FAST_MBUS now (and even at much higher dram clock speeds), then I'm pretty sure they got activated
<ssvb> but so far it looks like the hardware dqs gate training is somehow misbehaving on your board and we might need to do something about it
<oliv3r> i wonder if we can :)
<oliv3r> i'll let lima-memtester loop tonight
<oliv3r> and will fix a10-meminfo to confir
<ssvb> thanks a lot!
<oliv3r> what memtester para is rslr0 and rdgr0 under?
FR^2 has quit [Quit: Leaving]
<ssvb> oliv3r: these registers are not in the dram_para struct (because they are autodetected by the hardware), I would just introduce a new 'dqs_gating_delay' variable
<ssvb> oliv3r: but I guess, I will just also update a10-meminfo myself because I'm already hacking this stuff
<ssvb> oliv3r: what is the dram clock frequency used by your Cubietruck now?
libcg has quit [Ping timeout: 255 seconds]
<ssvb> oliv3r: 432MHz is rather low :) jemk could clock dram in his CT almost to 600MHz, and mine tops at "only" 552MHz
nove has quit [Quit: nove]
<oliv3r> ssvb: fair nuff :) i'll pull your changes tomorrow :)
<oliv3r> ssvb: i'm using a stock dram_cubietruck with 0x73333 for tpr
<oliv3r> well and the other changes + fast_mbus
<oliv3r> sleep time :)
<ssvb> oliv3r: 432MHz is likely selected only because the DDR3 lane delays on the Cubetruck PCB are poorly compatible with the hardware DQS training in the dram controller
<oliv3r> loop2 is still going good
bgal has quit [Ping timeout: 240 seconds]
<oliv3r> i know nothing of these lanes and poor pcb mumbo jumbo!
<oliv3r> :(
<oliv3r> i'm not smart enough!
<oliv3r> anyway nn :)
<ssvb> oliv3r: ok, good night, hopefully we can run more tests tomorrow :)
<ssvb> oliv3r: I'll try to push more code to github
<ssvb> oliv3r: FYI, has some references to various PDFs about DDR3
paulk-collins has quit [Quit: Ex-Chat]
arete74 has quit [Ping timeout: 258 seconds]
Lumocolor has quit [Ping timeout: 252 seconds]
sehraf has quit [Quit: ... be part of it...]
cernui has joined #linux-sunxi
arete74 has joined #linux-sunxi
Andy-D has quit [Ping timeout: 255 seconds]
smotocel69 has quit [Ping timeout: 240 seconds]
kuldeepdhaka__ has joined #linux-sunxi
kuldeepdhaka_ has quit [Ping timeout: 252 seconds]
smotocel69 has joined #linux-sunxi
kuldeepdhaka__ has quit [Read error: Connection reset by peer]
Nazcafan has quit [Quit: Quitte]
Andy-D has joined #linux-sunxi
bertrik has quit [Remote host closed the connection]
bbrezillon has quit [Ping timeout: 276 seconds]
Lumocolor has joined #linux-sunxi
joga has quit [Ping timeout: 245 seconds]
joga has joined #linux-sunxi
syeekick has quit [Read error: Connection reset by peer]
Faisal_ has joined #linux-sunxi
Faisal has quit [Ping timeout: 258 seconds]
egbert has quit [Ping timeout: 240 seconds]
egbert has joined #linux-sunxi