Turl changed the topic of #linux-sunxi to: Allwinner/sunxi /development discussion - did you try looking at our wiki? https://linux-sunxi.org - Don't ask to ask. Just ask and wait! - https://github.com/linux-sunxi/ - Logs at http://irclog.whitequark.org/linux-sunxi
aalm has quit [Ping timeout: 252 seconds]
libv has quit [Ping timeout: 265 seconds]
libv_ has quit [Ping timeout: 244 seconds]
libv has joined #linux-sunxi
libv has quit [Ping timeout: 260 seconds]
libv has joined #linux-sunxi
libv has quit [Ping timeout: 244 seconds]
libv has joined #linux-sunxi
libv has quit [Ping timeout: 244 seconds]
jelly has quit [Remote host closed the connection]
jelly-home has joined #linux-sunxi
Guest15887 has joined #linux-sunxi
corvusmalus has quit [Ping timeout: 244 seconds]
climjark has left #linux-sunxi ["Leaving"]
libv has joined #linux-sunxi
ox0re has joined #linux-sunxi
libv_ has joined #linux-sunxi
vagrantc has quit [Ping timeout: 258 seconds]
libv has quit [Ping timeout: 244 seconds]
libv_ has quit [Ping timeout: 244 seconds]
popolon has quit [Quit: WeeChat 1.4]
ninolein_ has quit [Ping timeout: 258 seconds]
ninolein has joined #linux-sunxi
lamer14705894366 has quit [Ping timeout: 276 seconds]
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
perr has quit [Quit: Leaving]
libv has joined #linux-sunxi
vagrantc has joined #linux-sunxi
apritzel has quit [Ping timeout: 244 seconds]
kaspter has joined #linux-sunxi
Sakami_ has quit [Ping timeout: 264 seconds]
Da_Coynul has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
cnxsoft has joined #linux-sunxi
vagrantc has quit [Ping timeout: 240 seconds]
vagrantc has joined #linux-sunxi
Sakami_ has joined #linux-sunxi
ox0re has quit [Quit: WeeChat 1.5]
<TheLinuxBug> sweet Orange Pi PC running nicely now
<TheLinuxBug> now if only I can patch this image for 2e
libv_ has joined #linux-sunxi
libv has quit [Ping timeout: 244 seconds]
perr has joined #linux-sunxi
perr has quit [Changing host]
perr has joined #linux-sunxi
p1u3sch1 has quit [Ping timeout: 265 seconds]
p1u3sch1 has joined #linux-sunxi
kaspter has quit [Ping timeout: 244 seconds]
lerc has quit [Quit: No! Not that one!]
lerc has joined #linux-sunxi
fire219 has quit [Read error: Connection reset by peer]
vagrantc has quit [Quit: leaving]
libv has joined #linux-sunxi
libv_ has quit [Ping timeout: 250 seconds]
libv_ has joined #linux-sunxi
libv has quit [Ping timeout: 244 seconds]
iamfrankenstein has joined #linux-sunxi
libv has joined #linux-sunxi
libv_ has quit [Ping timeout: 244 seconds]
<KotCzarny> TheLinuxBug: please update wiki
<KotCzarny> with your method
merbanan has quit [Ping timeout: 258 seconds]
lamer14705894366 has joined #linux-sunxi
merbanan has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
Putti has joined #linux-sunxi
libv_ has joined #linux-sunxi
libv has quit [Ping timeout: 258 seconds]
kaspter has joined #linux-sunxi
libv has joined #linux-sunxi
libv_ has quit [Ping timeout: 250 seconds]
libv has quit [Ping timeout: 276 seconds]
IgorPec has joined #linux-sunxi
libv has joined #linux-sunxi
perr_ has joined #linux-sunxi
perr_ has quit [Remote host closed the connection]
reinforce has joined #linux-sunxi
perr has quit [Quit: Leaving]
perr has joined #linux-sunxi
dearfibonacci has joined #linux-sunxi
libv has quit [Ping timeout: 276 seconds]
libv_ has joined #linux-sunxi
Putti has quit [Ping timeout: 240 seconds]
dearfibonacci has quit [Quit: Leaving]
libv has joined #linux-sunxi
libv_ has quit [Ping timeout: 264 seconds]
libv_ has joined #linux-sunxi
lemonzest has joined #linux-sunxi
jelly-home is now known as jelly
libv has quit [Ping timeout: 244 seconds]
iamfrankenstein has quit [Quit: iamfrankenstein]
iamfrankenstein has joined #linux-sunxi
dearfibonacci has joined #linux-sunxi
lamer14705894366 has quit [Quit: jIRCii - http://www.oldschoolirc.com]
tkaiser has joined #linux-sunxi
IgorPec has quit [Ping timeout: 258 seconds]
<Net147> ssvb: I am trying to render 3D graphics to both LCD and HDMI at the same time using OpenGL ES and framebuffer drivers. I can set FRAMEBUFFER=/dev/fb0 or FRAMEBUFFER=/dev/fb1 environment variables to select which display output to use if I have a process for each display, but how would I use multiple displays at the same time in a single process? is there some EGL option to set the framebuffer to use?
<tkaiser> Pictures of Xunlong's H5 boards have been 'leaked': http://forum.armbian.com/index.php/topic/1762-opi-3-and-opi-pc2-with-h5/
kaspter has quit [Remote host closed the connection]
<KotCzarny> tkaiser: looks like any other orange pi
<KotCzarny> he he
<KotCzarny> just looked at the third pic... lol
<KotCzarny> seriously, what were they drinking? ;)
<MoeIcenowy> what's H5?
<tkaiser> MoeIcenowy: Maybe a H64 without PMIC support?
<tkaiser> And more USB host ports?
<wens> no PMIC again?
<tkaiser> wens: Can't spot any on the pictures
<wens> the layout for the opi pc 2 is funky
kaspter has joined #linux-sunxi
<MoeIcenowy> I think models without PMIC is weird
<MoeIcenowy> this is why I didn't buy a opi
<KotCzarny> is this seriously going to be named opi pc 2?
<tkaiser> MoeIcenowy: At least voltage regulation on the larger OPi with SY8106A (said to be discontinued) works good. But no battery support...
<MoeIcenowy> maybe it's possible to attach a axp223/axp803 to a H3/H5
<tkaiser> KotCzarny: Well, Steven's naming scheme has been always a miracle
<MoeIcenowy> if someone really wants it
<KotCzarny> miss/move a letter, bam, different board
<wens> you probably could, though does h3/h5 have a working RSB controller?
<wens> also you lose vendor BSP support if you add a pmic yourself
<MoeIcenowy> yes...
<MoeIcenowy> so none will do it
<ssvb> Net147: I don't know much about what the mali blob can or can't do under the hood, but you can always have multiple processes as an ugly workaround
<ssvb> tkaiser: hmm, these pictures look very interesting, have they really added SPI flash?
<tkaiser> ssvb: You mean U25?
<ssvb> tkaiser: yes
<ssvb> unfortunately the pictures are too blurry to read text on the chips
<tkaiser> ssvb: To be honest: No idea, IgorPec is in contact with Steven, maybe he can share some info too (if he knows anything and is allowed to)
<tkaiser> ssvb: Since I fear there's an USB hub on the board I looked at the traces to the USB ports. They seem to end nearby U25 but the pictures are really to blurry to spot anything real
<ssvb> tkaiser: but there must be at least some difference between A64 and H5, right?
<KotCzarny> hmm, maybe h5 are h64 cores that didnt make q&a
<KotCzarny> ;)
<KotCzarny> ie. hotter/slower
<KotCzarny> theypiled up and someonethought 'lets rebrand them and sellanyway!
<tkaiser> ssvb: PMIC at least and then I would suspect based on stuff told by SinoVoip that the H vs. A think is market segmentation
<tkaiser> s/think/thing/
<tkaiser> A83T 'allowed' to run Android 5.x while H8 (otherwise identical) has to remain at Android 4.4 for example
sunxi_fan has quit [Remote host closed the connection]
sunxi_fan has joined #linux-sunxi
indy has quit [Quit: ZNC - http://znc.sourceforge.net]
yann has quit [Ping timeout: 260 seconds]
montjoie has quit [Ping timeout: 250 seconds]
indy has joined #linux-sunxi
florianH has joined #linux-sunxi
TheSeven has quit [Ping timeout: 250 seconds]
TheSeven has joined #linux-sunxi
fdcx has quit [Read error: Connection reset by peer]
premoboss has joined #linux-sunxi
speakman has quit [Read error: No route to host]
TheSeven has quit [Ping timeout: 258 seconds]
TheSeven has joined #linux-sunxi
<tkaiser> Weird: There's really no publicly available information for H5 and all we have are memories in IRC log: https://irclog.whitequark.org/linux-sunxi/2016-08-01#17123486;
<tkaiser> Quad core confirmed by pictures of the SoC
Leepty has joined #linux-sunxi
kaspter has quit [Ping timeout: 250 seconds]
kaspter has joined #linux-sunxi
Mr__Anderson has joined #linux-sunxi
apritzel has joined #linux-sunxi
premoboss has quit [Ping timeout: 264 seconds]
fdcx has joined #linux-sunxi
yann has joined #linux-sunxi
apritzel has quit [Ping timeout: 244 seconds]
libv has joined #linux-sunxi
libv_ has quit [Ping timeout: 240 seconds]
paulk-aldrin has joined #linux-sunxi
foxx_ has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
apritzel has joined #linux-sunxi
<tkaiser> apritzel: Did you got a BPi M64 as dev sample?
<apritzel> tkaiser: yes
<apritzel> I am about to send patches including a sun50i-a64-bananapi-m64.dts
<apritzel> tkaiser: the board is pretty neat, though I didn't do any load tests yet
<tkaiser> apritzel: Well, it's pretty much the same as BPi M3 with missing crappy GL830 USB-to-SATA bridge and another SoC (still lacking USB ports). So from my perspective: Uninteresting :)
<apritzel> the eMMC on it is interesting, for instance, as is pretty much proves that we need patches for the MMC controller
<tkaiser> It's a Pine64+ with 2 more leds, an USB hub and eMMC. For twice the price I would suspect
<wens> it's not available yet
<apritzel> but I really like those details (that the Pine64 lacks): microB OTG socket, soldered reset/power buttons, clearly labelled UART pins, FEL button, ...
TheSeven has quit [Ping timeout: 258 seconds]
<apritzel> also it's more compact
<tkaiser> apritzel: Let's see how well PCB helps with heat dissipation compared to Pine64+ since the board design misses mounting holes for a huge heatsink
TheSeven has joined #linux-sunxi
arete74 has joined #linux-sunxi
<tkaiser> If H5 features same count of USB host ports than H3 then all other 64-bit Allwinner chips become instantly obsolete for my use cases ;)
<jelle> H5 is announced somewhere?
<jelle> oh nevermind
<tkaiser> jelle: Only pictures and rumours ;)
iamfrankenstein has quit [Quit: iamfrankenstein]
<tkaiser> apritzel: Did you had a closer look at the PCB regarding hand soldered stuff?
<apritzel> no, but I can take a look tonight
<tkaiser> (SinoVoip sent me a M2+ sample (PCB rev 1.0) a while ago where two small resistors were soldered to 24 pin camera connector which seems to prevent any camera use on my M2+)
OverCR has joined #linux-sunxi
pbouda has joined #linux-sunxi
tkaiser has quit [Ping timeout: 258 seconds]
tkaiser has joined #linux-sunxi
perr has quit [Quit: Leaving]
perr has joined #linux-sunxi
perr has quit [Client Quit]
ricardocrudo has joined #linux-sunxi
perr has joined #linux-sunxi
perr has quit [Client Quit]
libv_ has joined #linux-sunxi
iamfrankenstein has joined #linux-sunxi
libv has quit [Ping timeout: 244 seconds]
perr has joined #linux-sunxi
IgorPec has joined #linux-sunxi
libv_ has quit [Ping timeout: 258 seconds]
perr has quit [Client Quit]
<tkaiser> IgorPec: Do you have any information regarding the H5 boards and are allowed to share?
perr has joined #linux-sunxi
perr has quit [Changing host]
perr has joined #linux-sunxi
<jelle> dual octa-core!
perr has quit [Client Quit]
<IgorPec> tkaiser: i will try to get more info
OverCR has quit [Quit: Leaving.]
perr has joined #linux-sunxi
perr has quit [Changing host]
perr has joined #linux-sunxi
jernej has joined #linux-sunxi
<tkaiser> IgorPec: I'm a bit curious how 'software support' should look like for the H5 boards when zero information is available...
<IgorPec> i assume terible as always :)
<tkaiser> Maybe Allwinner is providing an Android image (given that H5 is pretty similar to A64 and also RTL8723BS seems to be used for WiFi/BT this shouldn't be that hard)
<IgorPec> yes, that's possible, but haven't found much info around and also Steven's response is usually slow
jernej has quit [Quit: Page closed]
<tkaiser> :) If they provide an Android image and if H5 and A64 share most IP blocks then maybe all that's needed to get longsleep's BSP stuff working is an exchanged boot0 BLOB
<IgorPec> yep, that's my guess too
<tkaiser> In case devs here are interested in developer samples maybe you could ask Steven if he's willing to provide some?
<IgorPec> sure, I will try to arrange
libv has joined #linux-sunxi
<apritzel> jelle: I'd prefer if SoC vendor would stop adding more cores, instead adding bandwidth for DRAM and I/O to actually feed those cores properly
<apritzel> jelle: also taking care of head and power consumption with those tightly packed cores
<apritzel> *heat*
<jelle> ahh yeah
mzki has joined #linux-sunxi
<apritzel> for instance in the moment even four cores suck at compiling because of the ~20MB/s SD card
<jelle> yup
<apritzel> also the DRAM speed is ridiculous (single channel 32-bit @ 667 MHz)
<tkaiser> apritzel: But the target audience of tablets and OTT boxes cares only about count of cpu cores since they buy numbers. And so SoC vendors replace an A31 (quad-core CPU and dual-core GPU) with an overheating A83T (with half the GPU cores which is dumb for Android usage) and people still buy the octa-core tablet since 8 is more than 4 ;)
<apritzel> tkaiser: I know
libv has quit [Ping timeout: 240 seconds]
libv has joined #linux-sunxi
<apritzel> I still have that pipe dream that someone listens to random people on IRC and builds a chip for them ;-)
<tkaiser> apritzel: Different target audience and things look different, eg. Marvell Armada 38x: Only dual core A9 and maxing out 3 SATA 3.0 lines and multiple GbE NICs
<foxx_> apritzel: compiling sucks even in case of zram/tmpfs usage?
<apritzel> tkaiser: yeah, I was tempted already by this 90$ Clearfog Base board ...
<apritzel> foxx_: it's not like you have plenty of RAM to sacrifice for storage either, right?
<foxx_> apritzel: more or less :)
<apritzel> and the more threads you give make, the more memory it will consume on it's own
<apritzel> I remember seeing OOMs with make -j4 and 1GB RAM in the past, for instance
<foxx_> i agree. but anyway zram/tmpfs helps a lot on small targets
kaspter has quit [Remote host closed the connection]
<Amit_T> apritzel: Does it ARMADA 38x SoC has pcie controller built-in ?
<apritzel> foxx_: for smaller projects: definitely
<apritzel> Amit_T: yes
<tkaiser> Amit_T: Multiple SERDES lines that can be configured in u-boot to be either PCIe or SATA
<MoeIcenowy> apritzel: agree
<MoeIcenowy> especially Chinese people cares about only "cores"
<MoeIcenowy> but not memories
<apritzel> foxx_: but the Linux kernel source alone is around 1GB, the build directory is even 2.6GB on my setup here
<MoeIcenowy> few people know about DDR memory and eMMC storage
<wens> MoeIcenowy: isn't that just marketing? :p
<MoeIcenowy> yes it's marketing
<Amit_T> Let me check if it's available in South ASIA ?
<MoeIcenowy> ordinary users cannot understand the complex words
<MoeIcenowy> in China even sometimes people cannot diff RAM size and solid storage size
<apritzel> does anybody know if there will be 64-bit CPU modules for those ClearFog boards?
<wens> MoeIcenowy: me thinks that applies to average users around the world
<apritzel> wens: just give me some Gigabytes ;-)
<tkaiser> apritzel: AFAIK no, so better look at Armada 8040 from the very beginning ;)
<tkaiser> apritzel: And in case you want to do serious development I found Solid-Run pretty supportive (read as: dev samples and superiour support available)
<apritzel> tkaiser: yeah, but given that they only have renderings atm
nemunaire has quit [Quit: quit]
<ssvb> apritzel: DRAM is single channel 32-bit 1333 MT/s in A64, which is not too bad
nemunaire has joined #linux-sunxi
<ssvb> apritzel: and Allwinner SoCs generally always had good DRAM speed / CPU speed ratio
<apritzel> ssvb: I am not really arguing that this particular package (CPU cores, I/O, DRAM) fits their target market quite well
<apritzel> I am just always shocked when it comes to hdparm -T numbers
<tkaiser> apritzel: Already shocked by sequential I/O? Then try random I/O also ;)
<ssvb> apritzel: there is a bit of confusion between MT/s and MHz, and some other board manufacturers are sometimes show the MT/s numbers as MHz as far as the DRAM is concerned
<apritzel> hdparm -T is not even I/O, this is DRAM access speed, really
<tkaiser> apritzel: Ah, ok, I confused -t with -T
<apritzel> tkaiser: I never know which is which as well, that's why I always give both -tT ;-)
<ssvb> apritzel: also Allwinner typically uses state of the art DRAM controller IP, which has a decent "real speed / theoretical peak" ratio
<tkaiser> And that's the reason I never use hdparm for stuff like that ;)
<ssvb> apritzel: as far as I know, it is better than what is used in Freescale and TI OMAP
<apritzel> tkaiser: well, it's just a quick run and the package is usually already installed
<apritzel> ssvb: I see that the numbers may not be as bad, still I am far from swapping my X230 for an Allwinner based board for git/compiling work loads ;-)
kaspter has joined #linux-sunxi
<ssvb> apritzel: well, blame it on missing SATA, USB3 or fast eMMC
<apritzel> ssvb: I do, but nobody listens :-p
<apritzel> at least nobody who produces cheap SoCs ...
<ssvb> apritzel: A64 is a tablet chip, let's hope that H5 is a better choice for devboards
<foxx_> apritzel: do you compile kernel on the board?
<apritzel> foxx_: my day has only 24 hours ;-)
<ssvb> tkaiser: at least I hope that H5 has a 64-bit CPU, lest we all have been trolled :-)
kaspter has quit [Remote host closed the connection]
<apritzel> foxx_: I even tend to turn on my old PhenomX6 for compilation jobs to save some minutes
<apritzel> here at work I get around 2-3 mins for a full arm64 defconfig (cross-)compile, everything longer tends to annoy me
montjoie has joined #linux-sunxi
<foxx_> apritzel: are you using ccache/distcc or this is standalone time?
<apritzel> foxx_: i7-4770 @ 3.4GHz plus 20GB RAM
<foxx_> just interesting to compare
<apritzel> the Android people here get bigger boxes ;-)
<foxx_> i7-3820 @ 3.60GHz / 8Gb RAM + ccache the same around ~2-3minutes to rebuild pine64 kernel from scratch
<KotCzarny> ~50 minutes to compile kernel on opipc
<KotCzarny> ;)
<foxx_> sounds crazy
<KotCzarny> its cool man, cool
<KotCzarny> not that i do kernel work, so its one-off usually
<KotCzarny> and some software requires kernel source etc
<foxx_> kernel source can be mounted by nfs or something
mzki has quit [Quit: leaving]
<wens> great... just found out that the a31s codec is not the same as the a31 :(
<plaes> :(
<montjoie> cross compile is the fatest way to compile kernel
<apritzel> montjoie: make -j50 on a Cavium box isn't too bad ;-)
<buZz> i just ordered a xeon E3 to compile faster
popolon has joined #linux-sunxi
<montjoie> I survived 8 years with only one core:) (3 damaged out of 4)
<KotCzarny> how one damages cores?
marble_visions has joined #linux-sunxi
<apritzel> buying very cheap from the back of a lorry?
<marble_visions> hi all, how can I disable the kernel buffer output to console (UARTx)? I've tried console=null in u-boot and dmesg --console-level 1
<apritzel> marble_visions: it should be sufficient to just not name it as a console on the command line
<apritzel> marble_visions: unless you kernel config has some default command line that would kick in
<KotCzarny> i bet his problem might be uboot ignoring his changes
<KotCzarny> s/uboot/kernel/
<marble_visions> apritzel: does that mean ' console="" ' ?
<apritzel> just no console at all ;-)
<marble_visions> oh, ok.
<marble_visions> let me try that.
<apritzel> I managed to keep the serial port silent with that
<apritzel> but you have to tell userland about it as well (no logins, no systemd splats, etc)
<marble_visions> btw, i don't go set this with a .config, i've grepped to see where the console string is specified: include/configs/sunxi-common.h
<marble_visions> that's for u-boot
<marble_visions> cuz i believe it gets passed to the kernel at some point during boot
<KotCzarny> does the change show in /proc/cmdline ?
<marble_visions> apritzel: that should be via /etc/rsyslog.conf and the like, correct?
<marble_visions> KotCzarny: i'll have a look.
Da_Coynul has joined #linux-sunxi
cptG_ has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
reinforce has joined #linux-sunxi
cptG has quit [Ping timeout: 250 seconds]
perr has quit [Quit: Leaving]
<marble_visions> removed the console arg compiletime, but it gets inserted anyway
<marble_visions> maybe from u-boot
<marble_visions> selected "don't use bootloader args" and will now test
<marble_visions> awesome, "cat /proc/cmdline" doesn't show a console arg
<marble_visions> now I probably need to shut dmesg up
mzki has joined #linux-sunxi
<KotCzarny> or upgrade uboot to be able to pass cmdline via boot.scr etc
<KotCzarny> (easier than recompiling kernel everytime)
<marble_visions> oooooooooooooooooooh
kaspter has joined #linux-sunxi
<marble_visions> of course.
<marble_visions> i was wondering where that gets inserted!
<marble_visions> from where*
<KotCzarny> old uboots (2011) have it compiled in and need recompiling, if you use anything newer you can configure uboot via file in boot partition
<marble_visions> KotCzarny: 2015 here, all working correctly
<marble_visions> I now need to figure out why I still get messages across, even though there is no "console" specified
aalm has joined #linux-sunxi
<KotCzarny> check what talks to it? kernel, systemd or syslog ?
<KotCzarny> you might have earlyconsole enabled too
<KotCzarny> youmight also have some getty on serial line
<marble_visions> i made a kernel module with a printk(KERN_ERR "blah\n") that i insmod manually
<marble_visions> i've commented out the getty from inittab
<marble_visions> unless tty1 gets mapped to ttyS0 somehow
<KotCzarny> gzip -dc /proc/config.gz|grep CONFIG_SERIAL_CORE_CONSOLE
<marble_visions> i've got early printk enabled. i can't uncheck it from menuconfig. but even as it is, how does it know about ttyS0? and the baudrate?
<KotCzarny> you have a config in kernel that tells it which serial line to use
<MoeIcenowy> wens: a31s and a31 codec are diferent??!!!
<MoeIcenowy> are they the same die?!
lemonzest has quit [Quit: Leaving]
<wens> MoeIcenowy: who knows :(
<wens> seems a31s is a mix of a31 and a23 for the codec
<MoeIcenowy> is there dedicated A31s user manual?
<MoeIcenowy> or we can only check out the source code?
diego71_ has joined #linux-sunxi
<wens> there's a manual, but it has incorrect info
diego71 has quit [Ping timeout: 260 seconds]
Guest15887 has quit [Ping timeout: 260 seconds]
<marble_visions> ha, there is CONFIG_PRINTK in the kernel options, if I can't get it to shut up
<Amit_T> apritzel: But cavium box only has 48 cores :)
corvusmalus has joined #linux-sunxi
<apritzel> I wouldn't say "only", but in general more make jobs than cores are a good idea to cover up for I/O
<Amit_T> May be you'r using dual socket having 96 cores :)
<apritzel> so if one process is waiting for a file to load, another process can make progress
<KotCzarny> apritzel: compile with sources on tmpfs?
<apritzel> KotCzarny: well, eventually all of the source code will be buffered anyways, so this really doesn't matter
<KotCzarny> not true
<KotCzarny> writes still go to disk sometimes
<KotCzarny> and sometimes ram runs out (means buffers dropped)
<apritzel> this box has 128GB of RAM
<KotCzarny> he he
AllwinnerH3help has joined #linux-sunxi
<KotCzarny> still, writes go to disk
<apritzel> I agree, but the difference isn't that much
<AllwinnerH3help> Hi guys! I need some help compiling, copying and loading Kernel Modules on the Allwinner H3 ARM chip. I have followed this guide - http://linux-sunxi.org/H3_Manual_build_howto - but when I try to load the modules using indmod I get errors like: "modprobe: ERROR: could not insert 'snd_seq': Invalid argument"
<apritzel> KotCzarny: also putting source into tmpfs is a quite theoretical approach
<KotCzarny> AllwinnerH3help: wrong kernelmodules in regard to kernel version
<KotCzarny> AllwinnerH3help: did you make make modules_install / loaded right kernel?
<KotCzarny> apritzel: nah, rsync keeps things under control
<apritzel> KotCzarny: you mean: edit, git commit, rsy<box dies> ;-)
<AllwinnerH3help> KotCzarny: I must admit that I am a complete noob when it comes to the Kernel. What I did is dd the nanopi-m1-core-qte-sd4g-20160701.img onto a fresh SD card and then compile the modules as described in the howto. Then I copied them onto the rootfs..
<KotCzarny> apritzel: how many times your box died? ;)
<AllwinnerH3help> KotCzarny: I did the modules_install and had "INSTALL_MOD_PATH=output" set
<apritzel> it suffices if it does once before I managed to save my precious edit
<KotCzarny> AllwinnerH3help: hmm, might besomething else, what is the command line for modprobe?
<KotCzarny> apritzel: still, if your box is prone todying,it might die as wellduring disk write :P
<apritzel> KotCzarny: In general I agree that you get better results when running from tmpfs, but you can always construct a micro benchmark that suits your expectations
<KotCzarny> apritzel: if i did some intense work on the sources, yeah. for now i'm fine with compiling my app on bpi-m1 with microsd
<apritzel> compiling kernels is what I do ~ 20 times a day, so this really matters to me
<apritzel> and then I don't have the sources on tmpfs
<AllwinnerH3help> KotCzarny: I get "modprobe: ERROR: could not insert 'snd_seq': Invalid argument"
<KotCzarny> AllwinnerH3help: but how do you load it? from bash?
<apritzel> Amit_T: in the old times people were recommending "make -j<2 * NR_CPUS>" even
<AllwinnerH3help> KotCzarny: yes, I load it from bash over a ssh connection. Now I get "modprobe: ERROR: ../libkmod/libkmod.c:578 kmod_search_moddep() could not open moddep file '/lib/modules/3.4.39-h3/modules.dep.bin'"
<KotCzarny> AllwinnerH3help: also, snd_seq is only important if you do any midi work, not needed for pcm audio
<KotCzarny> do: depmod -a then retry
<AllwinnerH3help> KotCzarny: Allright, now depmod doesn't give me any errors! What I did is copy the Kernel source code onto the /lib/modules/<kernel version>/source. Before there was a missing link from when the .img file was made.
<KotCzarny> depmod does all dependency tracking and saves it to file
<KotCzarny> usually its done automatically when you do make modules_install
<KotCzarny> but if you changed paths itmight go bad
<AllwinnerH3help> but where is it saved?
<KotCzarny> /lib/modules/kernelversion/modules.dep
<AllwinnerH3help> I am not compiling the modules on the devce but on my ubuntu 64bit laptop
<KotCzarny> or something
<AllwinnerH3help> Ah, alright I see what you mean.
<AllwinnerH3help> but know when I do modprobe snd-seq I get no errors but in the dmesg I don't see any information about the module being loaded..
<KotCzarny> do lsmod
<KotCzarny> ifits there, its loaded
<KotCzarny> but as i said, its only for midi, if you dont use it, you dont need it
<AllwinnerH3help> It looks like it is loaded, awesome!
dearfibonacci has quit [Quit: Leaving]
<montjoie> disappointement!, hardware spinlock are not used for "speeding" spinlock
<AllwinnerH3help> Do you know of any good guide that takes you step by step through compiling a kernel + module, setting up a file system, bootloader etc..
<AllwinnerH3help> nb I did follow http://linux-sunxi.org/H3_Manual_build_howto but found it very confusing
<KotCzarny> AllwinnerH3help: keyword: linux from scratch
<KotCzarny> AllwinnerH3help: but for sunxi related stuff see the wiki
<KotCzarny> keep in mind its about mainline (h3 isnt in mainline fully yet) but you might get some hints from it
pbouda has quit [Quit: Leaving]
<KotCzarny> in short, for board specific boot you need only uboot and kernel
<KotCzarny> rootfs will work as long itscompiled for arm (specifically armhf)
kaspter has quit [Remote host closed the connection]
<AllwinnerH3help> Allright, that makes sense. But given I use Using debootstrap, will I be able to generate the rootfs on my laptop and then copy it over to the root partition on the sd card I am using?
<KotCzarny> never used it, so: maybe
<AllwinnerH3help> What do you normally use?
<KotCzarny> armbian
<KotCzarny> (im lazy)
<KotCzarny> btw. for legacy kernel (3.x) you also should have proper script.bin at boot partition (uboot uses it)
<AllwinnerH3help> yup, I think I have that in the boot partition already
<AllwinnerH3help> I am able to boot up, no problem. It's just that I would like to be able to automate the process eventually with a script so I have to understand the underlying steps <)
<KotCzarny> if your board is on their list its easy to build whole image with one command
<AllwinnerH3help> sweet!
<KotCzarny> or peek through their build script and modify for your board
<AllwinnerH3help> Will I be able to pick extra modules that I want to install?
<KotCzarny> dont remember, but it alsobuilds kernel, so probably there is a way to supplement own .config
<tkaiser> AllwinnerH3help: Everything possible, see 'userpatches' dir and read through http://docs.armbian.com/Developer-Guide_Build-Preparation/
<tkaiser> AllwinnerH3help: And if you cross compile you can just let u-boot and kernel rebuild and then get in output/debs .debs to install with new kernel+modules on the board in question
<KotCzarny> tkaiser: does armbian have irc channel or only forum?
<tkaiser> forum
<AllwinnerH3help> Interesting, this is what I was looking for.
<AllwinnerH3help> But now I have a different problem related to "snd_hwdep" module. It gives me "snd_hwdep: Unknown symbol snd_unregister_oss_device (err 0)" and other similar errors when I try to load it..
<AllwinnerH3help> using modprobe that is
<tkaiser> AllwinnerH3help: If you have a Ubuntu 14.04 or 16.04 (64-bit) box then it's just one step away...
<KotCzarny> AllwinnerH3help: try armbian, might be easier/more compatible
<AllwinnerH3help> hmm, I think you are right. Will try it out..
<KotCzarny> also, unknown symbol means you forgot to copy some modules
<KotCzarny> if you do 'dmesg' you can try guessing which modules are missing by symbol names
<KotCzarny> but again, make modules_install should do everything for you, you can pass the variable for destination path too
<AllwinnerH3help> yeah exactly, but it do not find the modules that dmesg tells me that are "Unknown symbol", like snd_unregister_oss_device, snd_register_oss_device
<KotCzarny> i think the correct syntax is: make modules_install INSTALL_MOD_PATH=/some/path
<AllwinnerH3help> yes, I did that correctly
premoboss has joined #linux-sunxi
kaspter has joined #linux-sunxi
keh has quit [Quit: = ""]
libv_ has joined #linux-sunxi
mzki has quit [Quit: leaving]
libv has quit [Ping timeout: 258 seconds]
libv has joined #linux-sunxi
libv_ has quit [Ping timeout: 240 seconds]
cnxsoft has quit [Quit: cnxsoft]
Sakami_ has quit [Ping timeout: 244 seconds]
libv has quit [Ping timeout: 244 seconds]
Sakami_ has joined #linux-sunxi
libv has joined #linux-sunxi
vagrantc has joined #linux-sunxi
kaspter has quit [Ping timeout: 264 seconds]
reinforce has quit [Quit: Leaving.]
libv has quit [Ping timeout: 276 seconds]
libv has joined #linux-sunxi
libv has quit [Ping timeout: 258 seconds]
ganbold has quit [Quit: This computer has gone to sleep]
ganbold has joined #linux-sunxi
libv has joined #linux-sunxi
libv has quit [Ping timeout: 264 seconds]
libv has joined #linux-sunxi
libv has quit [Ping timeout: 244 seconds]
libv has joined #linux-sunxi
AllwinnerH3help has quit [Ping timeout: 250 seconds]
libv_ has joined #linux-sunxi
Putti has joined #linux-sunxi
jernej has joined #linux-sunxi
libv has quit [Ping timeout: 244 seconds]
Da_Coynul has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
cosm has joined #linux-sunxi
Da_Coynul has joined #linux-sunxi
Da_Coynul has quit [Ping timeout: 264 seconds]
montjoie has quit [Ping timeout: 252 seconds]
montjoie has joined #linux-sunxi
IgorPec has quit [Ping timeout: 240 seconds]
tkaiser has quit [Ping timeout: 252 seconds]
BenG83 has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
tkaiser has joined #linux-sunxi
scream has joined #linux-sunxi
<marble_visions> for some reason, te
<marble_visions> the console output does not honor dmes --console-level
<marble_visions> s/dmes/dmesg/
<marble_visions> cat /proc/sys/kernel/printk says 1 4 1 7, but I get kernel messages with priority 4
<marble_visions> and 3
IgorPec has joined #linux-sunxi
IgorPec has quit [Ping timeout: 276 seconds]
<vpeter> Seems gearbest used wrong price for my Orange Pi PC H3, no invoice and no sender informations. And now I have problem with customs. That's happen when you get something for almost free :)
tkaiser has quit [Quit: jIRCii - http://www.oldschoolirc.com]
<TheLinuxBug> hehehe
<TheLinuxBug> mine arrived
<TheLinuxBug> got it on Friday
<vpeter> Wondering if they will refund this cost... It will probbaly be more than a board itself :(
cosm_ has joined #linux-sunxi
<hp197> vpeter: if the customs price > (price of goods X 2), then you can also reject the package and order it again
vagrantc_ has joined #linux-sunxi
vagrantc_ has quit [Changing host]
vagrantc_ has joined #linux-sunxi
<KotCzarny> o.O
<KotCzarny> what would be the point?
<hp197> they wont like it if you decline the package, and will still try you to pay for it
<KotCzarny> also, it was promotion
<KotCzarny> which ended
<hp197> the point is that you might get it around customs the second time
<hp197> hit ratio of customs isnt anywhere near 100%
vagrantc has quit [Ping timeout: 244 seconds]
vagrantc_ is now known as vagrantc
<KotCzarny> depends on the value entered
<hp197> true, and the shipper
<hp197> I noticed for NL that if it is shipped with DHL then the customs hit ratio will be near 100%
<hp197> postnl is only 25%
<vpeter> hp197: About NL - if send to eu why customs?
<hp197> if its from outside EU and > €30,- you have to pay import taxes and VAT
<hp197> if its send from within the EU the it doesnt go through customs
<vpeter> Now I see what you wrote "for NL".
<KotCzarny> but opipc without any promo is sold for ~15usd
<hp197> ow, sorry... the Netherlands :)
<KotCzarny> you can send them screenshot from xunlong's shop on aliexpress
<vpeter> But I still have to pay some small fee because they have work with a package.
<KotCzarny> ffs
Nacho___ has joined #linux-sunxi
yann has quit [Ping timeout: 250 seconds]
Nacho has quit [Ping timeout: 260 seconds]
ricardocrudo has quit [Remote host closed the connection]
IgorPec has joined #linux-sunxi
Mr__Anderson has quit [Remote host closed the connection]
Netlynx has joined #linux-sunxi
Netlynx has quit [Client Quit]
Netlynx has joined #linux-sunxi
Netlynx has quit [Changing host]
Netlynx has joined #linux-sunxi
apritzel has quit [Ping timeout: 244 seconds]
paulk-aldrin has quit [Quit: Leaving]
Nacho___ has quit [Quit: No Ping reply in 180 seconds.]
Nacho has joined #linux-sunxi
gzamboni has joined #linux-sunxi
mpmc has quit [Ping timeout: 252 seconds]
reinforce has joined #linux-sunxi
premoboss has quit [Ping timeout: 250 seconds]
tithrion has joined #linux-sunxi
iamfrankenstein1 has joined #linux-sunxi
fredy_ has joined #linux-sunxi
lynxis_ has joined #linux-sunxi
menomc has joined #linux-sunxi
fest_ has joined #linux-sunxi
lynxis_ has quit [Client Quit]
tgaz_ has joined #linux-sunxi
dlan has quit [*.net *.split]
tgaz has quit [*.net *.split]
leio has quit [*.net *.split]
fredy has quit [*.net *.split]
lynxis has quit [*.net *.split]
mnemoc has quit [*.net *.split]
fest has quit [*.net *.split]
jelle has quit [*.net *.split]
dlan_ has joined #linux-sunxi
dlan_ has quit [Changing host]
dlan_ has joined #linux-sunxi
fest_ is now known as fest
tgaz_ is now known as tgaz
Inode has joined #linux-sunxi
lynxis has joined #linux-sunxi
leio has joined #linux-sunxi
jelle has joined #linux-sunxi
jelle has quit [Changing host]
jelle has joined #linux-sunxi
OverCR has joined #linux-sunxi
foxx_ has quit [Ping timeout: 276 seconds]
mpmc has joined #linux-sunxi
iamfrankenstein1 has quit [Ping timeout: 244 seconds]
fredy_ is now known as fredy
fredy is now known as Guest51642
arossdotme has joined #linux-sunxi
Netlynx has quit [Quit: Ex-Chat]
tithrion has quit []
iamfrankenstein1 has joined #linux-sunxi
foudubassan has joined #linux-sunxi
Inode has left #linux-sunxi [#linux-sunxi]
Gerwin_J has quit [Quit: Gerwin_J]
foudubassan has quit [Quit: Leaving]
paulk-collins has joined #linux-sunxi
foudubassan has joined #linux-sunxi
apritzel has joined #linux-sunxi
fire219 has joined #linux-sunxi
fire219 has quit [Changing host]
fire219 has joined #linux-sunxi
yann has joined #linux-sunxi
iamfrankenstein1 has quit [Ping timeout: 240 seconds]
DaQatz has quit [Ping timeout: 250 seconds]
cosm_ has quit [Ping timeout: 240 seconds]
reinforce has quit [Quit: Leaving.]
popolon has quit [Quit: WeeChat 1.4]
[Awaxx] has quit [Quit: "The smart is the one who realizes his mistakes, my friend."]
Putti has quit [Ping timeout: 264 seconds]
IgorPec has quit [Ping timeout: 240 seconds]
DaQatz has joined #linux-sunxi
[Awaxx] has joined #linux-sunxi
scream has quit [Remote host closed the connection]
paulk-collins has quit [Quit: Leaving]
mosterta has joined #linux-sunxi
Da_Coynul has joined #linux-sunxi
cosm has quit [Quit: Leaving]
foudubassan has quit [Ping timeout: 250 seconds]
corvusmalus has quit [Ping timeout: 244 seconds]
corvusmalus has joined #linux-sunxi
florianH has quit [Quit: Connection closed for inactivity]
Da_Coynul has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
vagrantc has quit [Quit: leaving]
ikmaak has joined #linux-sunxi
OverCR has quit [Read error: Connection reset by peer]
OverCR has joined #linux-sunxi
OverCR has quit [Client Quit]
jernej has quit [Ping timeout: 260 seconds]
Da_Coynul has joined #linux-sunxi
mosterta has quit [Ping timeout: 276 seconds]
Da_Coynul has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
apritzel1 has joined #linux-sunxi