<wens>
montjoie: one thing i run into is if you bring the interface down, then up again, it stalls
<wens>
montjoie: iperf running between cubietruck and cubietruck plus gives 300 Mbps / 500 Mbps and no errors :)
tipo has joined #linux-sunxi
<wens>
montjoie: ethtool -S shows rx_payload_error & rx_header_error going up
vagrantc has quit [Quit: leaving]
khuey|away is now known as khuey_
p1u3sch1_ has joined #linux-sunxi
p1u3sch1 has quit [Ping timeout: 276 seconds]
Gerwin_J has joined #linux-sunxi
Deskwizard has joined #linux-sunxi
reev has joined #linux-sunxi
khuey_ is now known as khuey|away
FergusL has quit [Quit: No Ping reply in 180 seconds.]
huehner has quit [Ping timeout: 276 seconds]
al1o has quit [Ping timeout: 276 seconds]
huehner_ has joined #linux-sunxi
specing has quit [Ping timeout: 276 seconds]
FergusL has joined #linux-sunxi
specing has joined #linux-sunxi
al1o has joined #linux-sunxi
<montjoie>
wens: I am interested in the other stat counter like rx_dma_stop and tx_stop_queue
<montjoie>
rx_dma_ua
<montjoie>
when tx_stop_queue raise, try to increase nbdesc_tx and for rx_dma_ua try increase nbdesc_rx and see if perf raise also
<montjoie>
basicly
<montjoie>
wens: rx_payload_error and rx_header_error are in fact counter about non-basic-IP frame that MAC cannot handle CRC (like ARP :)). I believe that I will perhaps can do better than just ignoring them latter
<montjoie>
a sort of "If they are set, just ask the kernel to do CRC itself"
plaes_ is now known as plaes
JohnDoe_71Rus has joined #linux-sunxi
asmir has joined #linux-sunxi
TheLinuxBug_ has joined #linux-sunxi
engideavr has joined #linux-sunxi
afaerber_ has joined #linux-sunxi
igraltis1 has joined #linux-sunxi
alexxy has joined #linux-sunxi
FloH has joined #linux-sunxi
[Awaxx]_ has joined #linux-sunxi
atsampso1 has joined #linux-sunxi
pstef_ has joined #linux-sunxi
specing has quit [*.net *.split]
afaerber has quit [*.net *.split]
adj__3 has quit [*.net *.split]
Nyuutwo has quit [*.net *.split]
igraltist has quit [*.net *.split]
edolnx has quit [*.net *.split]
atsampson has quit [*.net *.split]
alexxy[home] has quit [*.net *.split]
FlorianH has quit [*.net *.split]
TheLinuxBug has quit [*.net *.split]
[Awaxx] has quit [*.net *.split]
wigyori has quit [*.net *.split]
engidea-vr has quit [*.net *.split]
pstef has quit [*.net *.split]
jbrown has quit [*.net *.split]
arnd has quit [*.net *.split]
TheLinuxBug_ is now known as TheLinuxBug
adj__3 has joined #linux-sunxi
Nyuutwo has joined #linux-sunxi
specing has joined #linux-sunxi
reinforce has joined #linux-sunxi
jbrown has joined #linux-sunxi
wigyori has joined #linux-sunxi
edolnx has joined #linux-sunxi
reev has left #linux-sunxi ["Leaving"]
arnd has joined #linux-sunxi
[Awaxx]_ is now known as [Awaxx]
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
firnsy has quit [Ping timeout: 260 seconds]
firnsy has joined #linux-sunxi
argulp has joined #linux-sunxi
huehner_ is now known as huehner
foudubassan has joined #linux-sunxi
<topi`>
systemd 219 here on this ubuntu vivid image, so probably FW loading requires a kernel patch
<topi`>
I wonder how to backport it to 3.4 kernel :/
<KotCzarny>
if you can use armbian?
<topi`>
this *is* armbian, but with a loboris kernel
<topi`>
the only kernel where I could get the orangepi camera module to work
<topi`>
umm, no, sorry, the whole rootfs is actually from loboris' image
<KotCzarny>
hmm, you can try replacing udev-systemd with eudev
<KotCzarny>
(ie. systemd free uev)
<KotCzarny>
*udev
<topi`>
maybe I should just start patching the mainline so that the camera module would work with it
<KotCzarny>
yes, mainline is the right thing
<topi`>
it feels so stupid to do lots of work to get something as fundamental as FW loading work on an obsolete kernel
<topi`>
but without FW loading, that OPI is not very useful as a video capture device... since the only way to connect to the net is via ETH0
<topi`>
I have two different WiFi usb dongles and both require a FW image
<KotCzarny>
you can also rebuild kernel and include those firmwares in the kernel
<topi`>
oh, how is that possible?
<KotCzarny>
kernel has an option to do just that
<KotCzarny>
when you do make menuconfig it's in: device drivers -> generic driver options -> External firmware blobs
<KotCzarny>
there just put space separated list of file paths to include
<KotCzarny>
there are other options regarding firmware loading too
<KotCzarny>
for example on my laptop i have: ipw2200-bss.fw radeon/R600_rlc.bin radeon/R700_rlc.bin iwlwifi-3945-2.ucode
<KotCzarny>
and "Firmware blobs root directory" is valid only during kernel compile
<topi`>
right
IgorPec has joined #linux-sunxi
premoboss has joined #linux-sunxi
apritzel has joined #linux-sunxi
asmir has quit [Ping timeout: 244 seconds]
tyler-baker has joined #linux-sunxi
caog has joined #linux-sunxi
matthias_bgg has joined #linux-sunxi
ricardocrudo has joined #linux-sunxi
alexxy has quit [Ping timeout: 244 seconds]
alexxy has joined #linux-sunxi
JohnDoe71rus has joined #linux-sunxi
JohnDoe_71Rus has quit [Read error: Connection reset by peer]
<plaes>
and of course, you can adapt the fbtft patch
<willmore>
keesj, let me know how this works out. Putting a TFT on my H3 boards is on my list of things to do.
<keesj>
let me try that first
jbrown has quit [Ping timeout: 250 seconds]
jbrown has joined #linux-sunxi
<plaes>
willmore: you would need SPI driver for H3 first
bwarff__ has quit [*.net *.split]
HeavyMetal has quit [*.net *.split]
lemonzest has quit [Ping timeout: 250 seconds]
JohnDoe71rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
JohnDoe_71Rus has joined #linux-sunxi
egbert has quit [Ping timeout: 250 seconds]
egbert has joined #linux-sunxi
<willmore>
plaes, is that still on the ToDo list? Crud. Okay. I'll put that on the long list.
premoboss has quit [Ping timeout: 250 seconds]
asmir has quit [Read error: Connection reset by peer]
premoboss has joined #linux-sunxi
lemonzest has joined #linux-sunxi
andoma has quit [Ping timeout: 250 seconds]
<plaes>
willmore: H3 is still missing usb and eth ;)
andoma has joined #linux-sunxi
<plaes>
it might have A10-compatible SPI driver though.. but I don't know whether anyone has tried it
<willmore>
Isn't montjoie working on eth and I thought I saw someone else was working on the USB driver where the issue was the reset lines.
HeavyMetal has joined #linux-sunxi
HeavyMetal has quit [Changing host]
HeavyMetal has joined #linux-sunxi
<plaes>
yeah.. but both of these are not yet in mainline
asmir has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
ojn has quit [Ping timeout: 264 seconds]
<wens>
willmore: hans
<Deskwizard>
any recommanded board (if theres more than one) recommanded for H3 developpement? I could get in on that
<Deskwizard>
recommended even
ojn has joined #linux-sunxi
<plaes>
orange pi pc is quite cool
<montjoie>
willmore: the sun8i-emac driver is working
honx_ has quit [Ping timeout: 250 seconds]
honx has joined #linux-sunxi
HeavyMetal has quit [Ping timeout: 264 seconds]
<Deskwizard>
plaes, thanks, i'll have a look, anything much left to do with A33 ? I have some random tablet that uses it but if most is done I'll take apart for parts for my olinuxino-A20
<Deskwizard>
A33s*
<scream>
montjoie, so the one on your github works well now? :)
<willmore>
montjoie, *highfive* Well done!
<plaes>
ugh.. A33s is dead end
<willmore>
wens, hans is working on?
* willmore
only has Opi1, so the dual USB reset isn't as much of an issue here.
<plaes>
or was it A31
<plaes>
one of those had quite buggy silicon
<montjoie>
scream: yes my github sun8i-emac-wip should work
<KotCzarny>
well, a31 might be considered a dead end because of pvr gpu
<scream>
cool, will recompile now
jukivili has joined #linux-sunxi
<scream>
do you guys just use the armbian env to compile?
<montjoie>
gentoo crosscompiling
HeavyMetal has joined #linux-sunxi
HeavyMetal has quit [Changing host]
HeavyMetal has joined #linux-sunxi
<scream>
allright, i'll just use my jessie compile VM then :)
<scream>
so I just enable CONFIG_SUN8I_H3_EPHY ant that's it?
<montjoie>
basicly, what's your board ?
<scream>
opi one
<scream>
there's also SUN8I_EMAC
<scream>
:D
<plaes>
you need wens' tree
<Deskwizard>
plaes, yeah thats what I thought, especially with the 512mb ram, done, I'll get a H3 board, its precisely what I was looking for to avoid amlogic. ty.
<scream>
plaes, thanks, i'll try
<plaes>
A1x(s), A20, H3, A64 are those that might eventually work
<plaes>
A23 too
<plaes>
and of course the one that C.H.I.P. has
<Deskwizard>
thanks for the info, I already have an A20 board, waiting for Olimex on the A64, guess I'm close to having what I need hehe
<plaes>
R8 which is just repackaged A13
<willmore>
plaes, A83T, too?
<plaes>
I wouldn't bet on it ;)
<willmore>
Ouch.
<plaes>
the ones I named have already reached critical mass
<plaes>
A83T just needs a cheap board
<plaes>
which means lots of users :)
<Deskwizard>
uhm
<willmore>
Ahh, the current board is a bit expensive.
<Deskwizard>
and I'm looking for a project hehe
<Deskwizard>
i'll look into that ;)
<willmore>
deskwizard, SPI on H3? ;)
<Deskwizard>
willmore, LOL I just saw the orangepi prices, so yes, why not ;)
<Deskwizard>
SPI is only of the only thing I can MAYBE manage lol
<Deskwizard>
one of the only*
<Deskwizard>
considering I ordered the olimex board from digikey, one of every of them is still cheaper hehe
<willmore>
I'm helping. ;)
<Deskwizard>
willmore, deal, I'll choose which one to get and order today
<willmore>
$9 for an Opi1 is hard to beat.
<Deskwizard>
yeah, all of thoses prices are hard to beat tbh
<willmore>
agreed
<Deskwizard>
I almost ordered a bunch of odroid yesterday, thank god I didnt
<willmore>
Do that, too!
<KotCzarny>
whatever you choose, dont fall into dumb power regulator trap
<willmore>
C2 is a beast.
<Deskwizard>
willmore, yes, but I hate amlogic and they B.O. :P
<KotCzarny>
s/power/voltage/
<Deskwizard>
KotCzarny, could you please elaborate if you dont mind? I really dont want to fall into that one
<willmore>
Next batch of C2 should release on the 22nd.
<willmore>
KotCzarny, there's dumb and then there's just stupid. At least the Opi1 is just 'dumb'.
<KotCzarny>
deskwizard: arm chips consume power depending on supplied voltage (and to minor point load)
<willmore>
KotCzarny, * number of active cores
<KotCzarny>
on opipc one can set a table of freq--voltage of just any values
<Deskwizard>
yeah I think I see what you mean ;)
<willmore>
P=V^2*F (* cores) + constant stuff for being alive
<KotCzarny>
some recent boards use FIXED voltages
<Deskwizard>
OH
<Deskwizard>
wow
<Deskwizard>
thank you so much
<KotCzarny>
which in turn means ~90C in IDLE
<Deskwizard>
o.o
<Deskwizard>
O.O
<willmore>
Good: Programmable regulator
<Deskwizard>
yeah, I figured
<willmore>
Bad: dual voltage--at least you can idle
<willmore>
Ugly: One voltage to rule them all!
<Deskwizard>
so basically, I want to make sure the board I chose uses ones of the AXPs or similar, correct ?
<Deskwizard>
or do they run those at fixed values?
<Deskwizard>
that makes no sense... nvm
<KotCzarny>
deskwizard, just google before the buy for possible temperature problems
<Deskwizard>
KotCzarny, will do, thank you
<Deskwizard>
very much.
<Deskwizard>
willmore, which H3 board do you currently use if any ?
<Deskwizard>
I'd buy at least one of those if you have one
<montjoie>
scream: I dont think have done dts patch for opione, but adding it should be easy
<willmore>
deskwizard, I have two Opi1 boards and that is all. I am net to the sunxi world. I came here from the ODROIDS.
<willmore>
err, 'new'.
<Deskwizard>
Oh I see! Welcome aboard then :)
<Deskwizard>
I came from hating amlogic :P and stuck here ever since hehe
<willmore>
I haven't been active in kernel development for 20 years, so I'm a bit rusty.
<KotCzarny>
20 years?
<KotCzarny>
o.o
<willmore>
grep the kernel source for 'willmore'
<willmore>
I think I'm only in the Documentation folder anymore.
* KotCzarny
feels old
* willmore
is old
<willmore>
:)
<scream>
montjoie, how is it different from opi pc in the ethernet part?
<Deskwizard>
willmore, I never really was so no worries ;)
<willmore>
Gotta run. Need to teach elementary kids about math. Who knew reading a clock could be so hard?
p1u3sch1 has quit [Ping timeout: 276 seconds]
arnd has quit [Ping timeout: 276 seconds]
wigyori has quit [Ping timeout: 276 seconds]
bbrezill1 has quit [Ping timeout: 276 seconds]
mzki has quit [Ping timeout: 276 seconds]
p1u3sch1 has joined #linux-sunxi
engidea-vr has joined #linux-sunxi
<KotCzarny>
willmore: it's because of those arbitrary 0-60 scale
<Deskwizard>
willmore, aight, later! ty for the chat
engideavr has quit [Read error: Connection reset by peer]
tipo has quit [Ping timeout: 276 seconds]
book` has quit [Ping timeout: 276 seconds]
jelle has quit [Ping timeout: 276 seconds]
<KotCzarny>
and reusing the scale for different things
<montjoie>
scream: opione seems to have only 10/100 ethernet so the same as opipc
<willmore>
KotCzarny, for the kids I'm working with, they can't seem to get the 'look at the long minute had to see where in the hour you are and *then* look at the short hour hand to see which hour you are in."
<KotCzarny>
:)
<KotCzarny>
yeah, color the arms with different colors
<willmore>
Set the clock past :40 and they think you're in the next hour. *sigh*.
<KotCzarny>
kids notice colors easier than sizes
mzki has joined #linux-sunxi
<KotCzarny>
btw. you are found only in qd65xx.c
<willmore>
Could be. I'm colorblind, so I wouldn't think of that. ;)
<willmore>
KotCzarny, not in the in2000 anymore? Did they pull ISA drivers?
<KotCzarny>
even colorblind people see SOME colors
<willmore>
KotCzarny, sure, but I don't think as much about color as fully color abled people do.
<KotCzarny>
regrepping with * instead of *.c
<willmore>
Oh, yeah, that'll get more.
<KotCzarny>
even white/black make a difference
<willmore>
Look in the docs folder.
<willmore>
Should be in200 and in the md driver
<willmore>
err, in2000.
bbrezill1 has joined #linux-sunxi
<willmore>
I think I still have that board. Pretty decent ISA SCSI board--once they hammered out the firmware issues.
<KotCzarny>
yeah, in2000.txt
<willmore>
That was me, way back when.
<willmore>
Okay, gotta go.
<KotCzarny>
cya
arnd has joined #linux-sunxi
<scream>
montjoie, that's what I thought. the only difference is soldered port count and 512MB ram
book` has joined #linux-sunxi
<montjoie>
so you could use the same patch
<montjoie>
at least the same content
<montjoie>
(adding ephy and emac)
wigyori has joined #linux-sunxi
jelle has joined #linux-sunxi
jelle has joined #linux-sunxi
p1u3sch1 has quit [Ping timeout: 244 seconds]
p1u3sch1 has joined #linux-sunxi
IgorPec has joined #linux-sunxi
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
<scream>
montjoie, I still get net eth0: PHY already attached
<scream>
maybe i don't need to enable some phy drivers?
premoboss has quit [Read error: Connection reset by peer]
Gerwin_J has joined #linux-sunxi
<scream>
config
iamfrankenstein1 has joined #linux-sunxi
iamfrankenstein has quit [Ping timeout: 264 seconds]
TheLinuxBug has quit [Ping timeout: 264 seconds]
al1o has quit [Ping timeout: 264 seconds]
iamfrankenstein1 is now known as iamfrankenstein
TheLinuxBug has joined #linux-sunxi
<keesj>
plaes: I am no longer getting errors but also not much is happening on the spi bus
jelle has quit [Ping timeout: 264 seconds]
lemonzest has quit [Ping timeout: 264 seconds]
egbert has quit [Ping timeout: 264 seconds]
pulser has joined #linux-sunxi
pstef has quit [Ping timeout: 264 seconds]
pulser_ has quit [Ping timeout: 264 seconds]
matthias_bgg has quit [Ping timeout: 264 seconds]
igraltis1 has quit [Ping timeout: 264 seconds]
FergusL has quit [Ping timeout: 264 seconds]
Deskwizard has quit [Ping timeout: 264 seconds]
MackBoy has quit [Ping timeout: 264 seconds]
egbert has joined #linux-sunxi
igraltist has joined #linux-sunxi
FergusL has joined #linux-sunxi
al1o has joined #linux-sunxi
cnxsoft has quit [Quit: cnxsoft]
premoboss has joined #linux-sunxi
Nacho has quit [Ping timeout: 259 seconds]
pstef has joined #linux-sunxi
Nacho has joined #linux-sunxi
jelle has joined #linux-sunxi
jelle has joined #linux-sunxi
matthias_bgg has joined #linux-sunxi
<scream>
montjoie, sorry got it working with your .config
<apritzel>
Ramose_: probably in the kernel tree that you are using
argulp has joined #linux-sunxi
<scream>
it is in kernel tree. arch/arm/boot/dts/sun8i-h3-orangepi-pc.dtb
<Ramose_>
Its not there
doppo has quit [Ping timeout: 268 seconds]
<apritzel>
then your tree will probably not support it anyway
doppo has joined #linux-sunxi
<scream>
montjoie, thanks for the rgmii info. I will try to use the opiplus dts
<Ramose_>
oh sorry , its there , I was searching in wrong way.
<lvrp16>
is there a dts for plus 2?
oneinsect has joined #linux-sunxi
<oneinsect>
friends quick question
<oneinsect>
my boot.scr doesnt work however But if I run each load and bootm statement individually at the u-boot > prompt, everything works and Linux boots
<oneinsect>
Found U-Boot script /boot/boot.scr reading /boot/boot.scr
<oneinsect>
## Executing script at 43100000 reading boot/sun8i-h3-orangepi-pc.dtb **Unable to read file boot/sun8i-h3-orangepi-pc.dtb
<oneinsect>
however the same command works manually
<oneinsect>
load mmc 0:1 0x43000000 /boot/sun8i-h3-orangepi-pc.dtb works fine
<oneinsect>
any reason why it doesnt pick it up from boot.scr?
<KotCzarny>
43100000 or 0x43000000 ?
<oneinsect>
hmmm
<oneinsect>
kinda strange
<oneinsect>
why it says 0x43000000
<apritzel>
oneinsect: do you have the boot.scr wrapped with mkimage?
khuey_ is now known as khuey|away
<oneinsect>
yes ofcourse
<oneinsect>
mkimage C none A arm T script d boot.cmd boot.scr
<oneinsect>
plus my boot.cmd are nothing but the same commands i run manually
<oneinsect>
they work manually perfectly fine
<oneinsect>
but then dont load
<oneinsect>
when i create a script
<oneinsect>
wonder why
<apritzel>
oneinsect: try to load boot.scr to a different address
<apritzel>
something up higher, where it doesnt overlap with the DT for instance
<oneinsect>
hmm
<oneinsect>
a moment please
<oneinsect>
let me pastbin my
<oneinsect>
script
<longsleep>
how does u-boot actually load / import boot.scr ?
<apritzel>
yeah, as expected, it's no fun, just confusing due all this nesting
<oneinsect>
yes
<apritzel>
so for instance it does: mmc dev 0
<Ramose_>
what would be the machine ID for organe pi pc ?
hulu1522 has joined #linux-sunxi
<oneinsect>
1029
<oneinsect>
i set it in the boot.cmd script
<oneinsect>
without machine id in the boot.cmd it doesnt boot
<longsleep>
apritzel: i am just looking into adding boot.src support to the default Pine64 U-Boot environment, would you suggest to just source boot.scr if found or should i still load uEnv.txt first?
<oneinsect>
I have seen a similar issue like this myself. It turned out that the uEnv.txt had '\r\n' line endings (Windows style) instead of just '\n'. And so even though the file name appears correct in the printenv it was actually one character off. You can test it in U-Boot to see if the actual environment variable is set correctly without trailing characters by doing the load command with the variable instead of the expected value, e.
<oneinsect>
that is from above mailing list
<oneinsect>
but i dont see any errors?
<apritzel>
try to "echo something" in the boot.cmd to see if you have missing characters
Rondom_ has quit [Ping timeout: 264 seconds]
<Ramose_>
apritzel: didn't get your earlier point ?
apritzel has quit [Ping timeout: 244 seconds]
Rondom has joined #linux-sunxi
<Ramose_>
oneinsect: Can you point what is going wrong with linux boot up?
<oneinsect>
yes
<oneinsect>
you mean your case?
<oneinsect>
Ramose_: ?
<Ramose_>
Yes
<oneinsect>
okie
<oneinsect>
paste your
<oneinsect>
boot.cmd
Gerwin_J has quit [Quit: Gerwin_J]
<oneinsect>
and whats your board model?
<oneinsect>
which kernel are you booting?
<oneinsect>
mainline?
<oneinsect>
what about u-boot? is it also mainline?