issueat has quit [Remote host closed the connection]
akaizen has quit [Remote host closed the connection]
akaizen has joined #linux-sunxi
petr has quit [Ping timeout: 252 seconds]
akaizen has quit [Ping timeout: 240 seconds]
petr has joined #linux-sunxi
petr has quit [Ping timeout: 255 seconds]
popolon has quit [Quit: Quitte]
akaizen has joined #linux-sunxi
petr has joined #linux-sunxi
egbert has quit [Ping timeout: 264 seconds]
egbert has joined #linux-sunxi
maksimlin has quit [Remote host closed the connection]
ganbold_ has joined #linux-sunxi
mmarker has quit [Read error: Connection reset by peer]
mmarker has joined #linux-sunxi
mmarker has quit [Ping timeout: 272 seconds]
xavia has joined #linux-sunxi
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
alexst has quit [Ping timeout: 244 seconds]
alexst has joined #linux-sunxi
alexst has quit [Ping timeout: 244 seconds]
eagles0513875_ has quit [Ping timeout: 260 seconds]
eagles0513875_ has joined #linux-sunxi
[7] has quit [Ping timeout: 248 seconds]
TheSeven has joined #linux-sunxi
xavia has quit [Quit: Leaving.]
netchip has quit [Ping timeout: 240 seconds]
w00tc0d3 has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
alexst has joined #linux-sunxi
alexst has quit [Ping timeout: 248 seconds]
joedj_ has quit [Quit: uNF]
joedj_ has joined #linux-sunxi
joedj_ has quit [Remote host closed the connection]
tomcheng76 has quit [Remote host closed the connection]
amitk has joined #linux-sunxi
w00tc0d3 has quit [Ping timeout: 264 seconds]
w00tc0d3 has joined #linux-sunxi
gzamboni has joined #linux-sunxi
kuldeepdhaka has joined #linux-sunxi
amitk has quit [Ping timeout: 248 seconds]
amitk has joined #linux-sunxi
ecelis has quit [Ping timeout: 272 seconds]
tomboy64 has quit [Remote host closed the connection]
Gerwin_J has quit [Quit: Gerwin_J]
tomboy64 has joined #linux-sunxi
afaerber has quit [Ping timeout: 244 seconds]
kuldeepdhaka has quit [Read error: Connection timed out]
afaerber has joined #linux-sunxi
kuldeepdhaka has joined #linux-sunxi
tomcheng76 has joined #linux-sunxi
sehraf has joined #linux-sunxi
HeHoPMaJIeH has joined #linux-sunxi
hypno has quit [Ping timeout: 272 seconds]
hypno has joined #linux-sunxi
amitk has quit [Ping timeout: 260 seconds]
amitk has joined #linux-sunxi
ganbold_ has quit [Ping timeout: 264 seconds]
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
rgenoud has joined #linux-sunxi
mtaubert has joined #linux-sunxi
<mtaubert>
hi
mtaubert is now known as cybin
FR^2 has joined #linux-sunxi
ecelis has joined #linux-sunxi
bbrezillon has joined #linux-sunxi
<cybin>
i've found the module hdmi_cec in drivers/video/sunxi/hdmi. is that what i need to get cec support for a20 based boards? how do i use that module after loading?
libcg has joined #linux-sunxi
HeHoPMaJIeH has quit [Quit: Leaving]
libcg has quit [Remote host closed the connection]
libcg has joined #linux-sunxi
popolon has joined #linux-sunxi
popolon has joined #linux-sunxi
diego_r has joined #linux-sunxi
libcg has quit [Read error: Connection reset by peer]
libcg_ has joined #linux-sunxi
libcg_ has quit [Remote host closed the connection]
kuldeepdhaka has quit [Read error: Connection reset by peer]
Quarx has quit [Client Quit]
Andy-D has quit [Remote host closed the connection]
_massi has joined #linux-sunxi
notmart has joined #linux-sunxi
HeHoPMaJIeH has joined #linux-sunxi
HeHoPMaJIeH has joined #linux-sunxi
kuldeepdhaka has joined #linux-sunxi
xshock has joined #linux-sunxi
kuldeepdhaka_ has joined #linux-sunxi
kuldeepdhaka has quit [Ping timeout: 264 seconds]
<xshock>
in the logs of this channel i have found references to Allwinner F1C100 chip, and someone named amx got some code running on it. Does anything have any information about F1C100 and where I should start digging about it?
eagles0513875_ has quit [Ping timeout: 255 seconds]
hypno has quit [Ping timeout: 260 seconds]
xshock has quit [Quit: Page closed]
eagles0513875_ has joined #linux-sunxi
akaizen has quit [Remote host closed the connection]
<ddc>
F-series processors From 2007 to 2011, Allwinner introduced its F-series processors, F10, F13, F18, F20, F1E200, F1C100, F20. This series runs Allwinner's self-developed operating system Melis2.0, which is now mainly used in vehicle multimedia system, E-ink reader, Visual intercom system etc.
ddc has quit [Ping timeout: 272 seconds]
hypno has joined #linux-sunxi
notmart has quit [Ping timeout: 240 seconds]
notmart has joined #linux-sunxi
Quarx has joined #linux-sunxi
<rellla>
uh. A33 was new to me. is this kind of a Mali'd A31?
<ssvb>
oliv3r: it is also very interesting to check if different a10 boards can use the same calibration settings for the temperature sensor
Net147 has quit [Quit: HydraIRC -> http://www.hydrairc.com <- The professional IRC Client :D]
alexst has quit [Ping timeout: 244 seconds]
<oliv3r>
ssvb: well the FLIR cam should be highly accurate
<ssvb>
oliv3r: that's cool, one problem is that there is some difference between the case temperature (you are measuring this) and the junction temperature (the temperature sensor is likely measuring this)
<ssvb>
but hopefully the Psi-JT value is small enough and can be ignored
<oliv3r>
seems to correlate quite linearly though
<oliv3r>
right, first the jpeg test
<ssvb>
if we look at the tables from that article, Psi-JT is listed to be between 0.5 and 1.0 C/W (for 5 watts power consumption it would be 5 degrees difference at most)
<oliv3r>
even if this happens only 1/1024 times, it's 1 to many
<oliv3r>
so 960 Mhz is the new default for our boards :p
<oliv3r>
kinda explains why they might use that as a default clock for the A20 :)
<ssvb>
these are the safe voltages for reliable operation of 'sun4i_poorlime' on the plot
<oliv3r>
this is default vcc btw
<oliv3r>
ssvb: you mean you are working on that? nice
<oliv3r>
so running at v1.4 volt
<ssvb>
the plot shows that the 'cruise' speed for a10-lime seems to be around 864MHz, after that point increasing the clock speed needs large increases of the core voltage
<hypno>
what are your requirements for reliable operataions?
<ssvb>
oliv3r: however if you read that discussion thread fro mthe linux-sunxi mailing list, we suspect that the A10-Lime PCB may just have an unusually high voltage drop on the power line between the PMIC and the SoC
<ssvb>
oliv3r: because my Cubieboard1 and Mele A1000 with the same A10 SoC both passed this test at 1.4V and 1008MHz
<ssvb>
oliv3r: if you can confirm that more than one A10-Lime consistently fail the test, then this is a pretty much good explanation
<ssvb>
hypno: the blue (sun4i), orange (sun5i) and brown (sun7i) lines show the default cpufreq operating points, which are hardcoded in the linux-sunxi kernel sources
jebba has quit [Quit: Leaving.]
FreezingCold has joined #linux-sunxi
alexst has joined #linux-sunxi
<ssvb>
hypno: the cyan line (the practical test results from my A10-Lime) going higher than the blue line (default sun4i cpufreq settings) means reliability troubles
dack has joined #linux-sunxi
<ssvb>
hypno: what kind of hardware do you have? are you interested to running some tests on it?
<oliv3r>
ssvb: ok i got 8 lime's here, let me test them in turn
<ssvb>
oliv3r: if you confirm the problem on more than one A10-Lime, then we might probably want to talk with Tsvetan and ask about the power line resistance on the A10-Lime PCB
<oliv3r>
rgr
<oliv3r>
testing it now
alexst has quit [Ping timeout: 264 seconds]
<ssvb>
the fix would be to just increase the core voltage to compensate the excessive voltage drop
<ssvb>
which would also require us to tailor the cpufreq tables specifically for each device
<oliv3r>
which would suck, better fix the pcb :p
<oliv3r>
but where exactly is the voltage drop happening?
<ssvb>
well, it's the elementary physics :) you have the wire between the AXP chip and the A10 chip used to supply power, and it has some resistance
nabblet has quit [Quit: leaving]
<oliv3r>
hmm, this LIME does work at 1008
<ssvb>
the voltage drop on this wire is the current multiplied by this resistance
<oliv3r>
aye
<oliv3r>
so the wire causes the voltage drop?
<wens>
correct
<ssvb>
the higher is the current (on heavy CPU load), the higher is the voltage drop on the wire, so the SoC is in fact getting less than 1.4V
<oliv3r>
ah a PCB wire
<oliv3r>
they should fix that in rev D
<oliv3r>
but this lime works
<oliv3r>
not sure if its temperature / PSU related
<oliv3r>
i power via OTG now
<ssvb>
well, just try it on more devices
<ssvb>
in any case, if the test fails, this means that the device is unreliable in these conditions
<ssvb>
or maybe somebody could run the test on more A10 based devices, it could be not A10-Lime PCB specific after all
<ssvb>
I got 1 fail out of 3 Allwinner A10 devices
<ssvb>
but first install git, ruby and libjpeg-turbo ...
<ssvb>
and enable the userspace governor in the kernel
<ssvb>
I see that this already sounds complicated enough to discourage people :)
<oliv3r>
it builds libjpegturbo by itself :)
<ssvb>
right
<ssvb>
has it done this for you?
* ssvb
thought that fedora was one of the first distros to switch to libjpeg-turbo
<lioka>
ruby. ohnoez
<ssvb>
why not? it's a nice universal programming language, very useful for fast prototyping
<ssvb>
I'm actually seriously considering to switch to mruby, because it allows to build static programs with no dependencies
<oliv3r>
ssvb: it did, v 1.3 something
<oliv3r>
python!! :p
<ssvb>
meh
<ssvb>
python typically needs more code to do the same job compared to ruby, and python also has weird indentation rules
kuldeepdhaka has quit [Ping timeout: 248 seconds]
<ssvb>
at least this was my impression when I made the choice between these two languages :)
<oliv3r>
:p
<oliv3r>
ssvb: the failed lime is a revA
<oliv3r>
the working (so far) lime's are lime-4gb revC
<ssvb>
oliv3r: ok, mine is also revA, maybe Tsvetan has already fixed this issue :)
<ssvb>
do you have any other revA boards?
<hypno>
ssvb: informative, thanks
<hypno>
ssvb: i have an unused cubietruck (A20) i'm planning to fiddle with
<hypno>
i'm more of an hardware guy so i don't know too much of which stability test sw's for linux is any good for actual usage
<hypno>
i just remember from when i used to overclock computers a thousand years ago that most stress test softwares either didn't load the hardware enough or handled errors too well
enrico_ has joined #linux-sunxi
<hypno>
so i am a but wary of using stress tests for evaluating clock and vcore settings
<hypno>
bit|
<hypno>
"
rz2k has joined #linux-sunxi
<oliv3r>
ssvb: nope, the revA board was an old freebe
<oliv3r>
i never used it until now :p
xavia has joined #linux-sunxi
<lioka>
ssvb: it would be nice to put back original governor, btw
<ssvb>
lioka: you are right, thanks
<ssvb>
got any results?
nabblet has joined #linux-sunxi
<lioka>
ssvb: nope, not at home currentry
paulk-collins has joined #linux-sunxi
kuldeepdhaka has joined #linux-sunxi
<wens>
mripard: for some unknown reason, I'm getting 2 copies of _some_ of your at91 patches
notmart has quit [Quit: notmart terminated!]
<mripard>
wens: you don't want to know the reason ... :)
<oliv3r>
ssvb: so far 4 LIME-4G revC have passed the test
<oliv3r>
mripard: you made a pboboo!
<ssvb>
oliv3r: ok, so far looks like only revA boards are defective
<oliv3r>
yep up to 6 now
<ssvb>
do you have any revB?
<oliv3r>
nope
<oliv3r>
maybe olimex can run this test and see what's up with several rev A, B and well not C boards
<ssvb>
I have also checked the schematics for revA, revB and revC for the ZQ resistors, and they are different everywhere!
<oliv3r>
i noticed that with an SD card, it's very easy to test
<oliv3r>
just log in, run test, log out, swap sd, repeat
<oliv3r>
well the jpeg decode test fails due to l2 cache problems no?
jebba has joined #linux-sunxi
<ssvb>
well, the L2 cache is the component that apparently fails, and it likely fails because of the insufficient core voltage, and the voltage is likely insufficient because of the voltage drop on the power line between A10 and AXP
<oliv3r>
wens: how is USB OTG power going IN to the board?
<oliv3r>
does that go through the AXP?
<wens>
oliv3r: through the AXP, and some bypass channel (through a diode) for 5v power to stuff like sata and usb
<oliv3r>
if i download it via wget, rename it to jpg, eog shows the proper res too
kuldeepdhaka has quit [Ping timeout: 272 seconds]
<oliv3r>
ssvb perfect 5V in, 0.3 amps idle
<oliv3r>
running jpeg test @ 1008 mhz is 0.48 amps
<oliv3r>
i'll do the rest in a table withtemps tomorrow
<wens>
well according to the board designs, you really didn't have to do that
<wens>
but it's possible exten wasn't enabled, thus you're not getting boosted 5v from ipsout
libcg has quit [Ping timeout: 240 seconds]
afaerber has joined #linux-sunxi
<wens>
but then you wouldn't have 3.3v either
FDCX has joined #linux-sunxi
<oliv3r>
i don't follow, but i gotta catch a train
<oliv3r>
talk to oyu later
<oliv3r>
maybe scrollback to see my earlier screenies to see what i did (with the heat camera shots)
<wens>
the pages are still small, maybe the service doesn't like me
<wens>
anyway, bedtime, talk tomorrow
nabblet has quit [Ping timeout: 260 seconds]
FDCX has quit [Remote host closed the connection]
FDCX has joined #linux-sunxi
tomboy64 has quit [Ping timeout: 264 seconds]
tomboy64 has joined #linux-sunxi
rgenoud has quit [Quit: Leaving.]
bonbons has joined #linux-sunxi
xinj has quit [Ping timeout: 240 seconds]
nabblet has joined #linux-sunxi
bonbons has quit [Quit: Leaving]
notmart has joined #linux-sunxi
bonbons has joined #linux-sunxi
xinj has joined #linux-sunxi
FDCX has quit [Remote host closed the connection]
xinj has quit [Ping timeout: 240 seconds]
FDCX has joined #linux-sunxi
FDCX has quit [Remote host closed the connection]
seppel has joined #linux-sunxi
FDCX has joined #linux-sunxi
dack has quit [Ping timeout: 264 seconds]
bbrezillon has quit [Ping timeout: 255 seconds]
enrico_ has quit [Quit: Bye]
bertrik has joined #linux-sunxi
akaizen has joined #linux-sunxi
bertrik_ has joined #linux-sunxi
bertrik has quit [Ping timeout: 260 seconds]
leviathanch has joined #linux-sunxi
_massi has quit [Quit: Leaving]
dack has joined #linux-sunxi
amitk has quit [Quit: leaving]
netlynx has joined #linux-sunxi
netlynx has joined #linux-sunxi
akaizen has quit [Remote host closed the connection]
akaizen has joined #linux-sunxi
FR^2 has quit [Quit: Connection reset by peer]
alexst has joined #linux-sunxi
akaizen has quit [Ping timeout: 240 seconds]
<lioka>
ssvb: 1056 and up skipped, 1008 and down all ok
<ssvb>
lioka: looks good
<ssvb>
technically, you can unlock higher clock frequencies via something like "echo 1008000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq" (use a higher clock frequency instead of 1008)
<ssvb>
just to see where it fails :) or don't do this if you don't like the idea of overclocking for testing purposes even for a short period of time