<philectro>
which distro did you recommend currently for an A10?
adj_ has quit [Ping timeout: 260 seconds]
adj_ has joined #linux-sunxi
ricardocrudo has quit [Remote host closed the connection]
jstein_ has joined #linux-sunxi
jstein has quit [Ping timeout: 240 seconds]
jstein_ has quit [Remote host closed the connection]
tomboy65 has joined #linux-sunxi
interrobangd_ has quit [Read error: Connection reset by peer]
cnxsoft has joined #linux-sunxi
libv_ has joined #linux-sunxi
libv has quit [Ping timeout: 250 seconds]
apritzel has quit [Ping timeout: 248 seconds]
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
ninolein has quit [Ping timeout: 252 seconds]
ninolein has joined #linux-sunxi
<Deskwizard>
philectro: id suggest you have a look at armbian
<philectro>
Deskwizard: thanks
<Deskwizard>
np
p1u3sch1 has quit [Ping timeout: 260 seconds]
p1u3sch1 has joined #linux-sunxi
atenart has quit [Ping timeout: 272 seconds]
hipboi has joined #linux-sunxi
TheSeven has quit [Ping timeout: 252 seconds]
uwe_ has quit [Ping timeout: 250 seconds]
atenart has joined #linux-sunxi
TheSeven has joined #linux-sunxi
uwe_ has joined #linux-sunxi
uwe_ has quit [Ping timeout: 240 seconds]
arossdotme-planb has quit [Ping timeout: 272 seconds]
arossdotme-planb has joined #linux-sunxi
<wens>
GL830 USB-SATA gets around 26 MB/s (read) on the cubietruck plus
<Deskwizard>
wens: nice :)
uwe_ has joined #linux-sunxi
<Deskwizard>
wens: how does the write look like? (if you know)
<vagrantc>
is it about as awful as the orange pi plus2 ?
<vagrantc>
i almost wonder if i wouldn't get better speeds by plugging in an external usb-sata adapter
IgorPec has joined #linux-sunxi
IgorPec has quit [Ping timeout: 250 seconds]
JohnDoe_71Rus has joined #linux-sunxi
philectro has quit [Quit: Konversation terminated!]
vagrantc has quit [Quit: leaving]
IgorPec has joined #linux-sunxi
p1u3sch1 has quit [Ping timeout: 248 seconds]
p1u3sch1 has joined #linux-sunxi
engideavr has joined #linux-sunxi
KotCzarny has quit [Ping timeout: 272 seconds]
orly_owl has joined #linux-sunxi
KotCzarny has joined #linux-sunxi
<KotCzarny>
vagrantc: you will most likely will, especially with usb-uas
cnxsoft has quit [Read error: Connection reset by peer]
KotCzarny has quit [Ping timeout: 272 seconds]
tchiwam has quit [*.net *.split]
pietrushnic has quit [*.net *.split]
flok420 has quit [*.net *.split]
buZz has quit [*.net *.split]
aep has quit [*.net *.split]
tchiwam has joined #linux-sunxi
pietrushnic has joined #linux-sunxi
buZz has joined #linux-sunxi
tkoskine has joined #linux-sunxi
aep has joined #linux-sunxi
buZz is now known as Guest95008
tomboy65 has quit [Ping timeout: 240 seconds]
flok420 has joined #linux-sunxi
cnxsoft has joined #linux-sunxi
KotCzarny has joined #linux-sunxi
<KotCzarny>
doh.
<Deskwizard>
Welcome back ;)
<KotCzarny>
yeah, my connections seems flaky today, so, anyone could shine some light on my audio issue?
<Deskwizard>
KotCzarny: uhm I can't shine some light, but I can try to replicate if you want
<Deskwizard>
mainline ?
<KotCzarny>
4.5.0-rc3
<KotCzarny>
bpi-m1
<KotCzarny>
i can pastebin the .config too if you want
yann|work has quit [Ping timeout: 240 seconds]
<KotCzarny>
one suspisious thing is that sun4i_codec is used by 3 things, even if i dont have anything playing/using it (regular alsa, no evilaudio installed)
Wizzup has quit [Read error: error:1408F10B:SSL routines:SSL3_GET_RECORD:wrong version number]
Wizzup has joined #linux-sunxi
reinforce has joined #linux-sunxi
arokux has joined #linux-sunxi
<Deskwizard>
KotCzarny: TBH, I'm reading DTB for dummies atm... but indeed, suspicious
rellla has joined #linux-sunxi
_massi has joined #linux-sunxi
<Deskwizard>
that A31s tablet just has too many test pads for me not to play with it
<KotCzarny>
:)
KotCzarny has quit [Quit: Reconnecting]
KotCzarny has joined #linux-sunxi
Guest95008 is now known as buZz
<Deskwizard>
they also forgot some connectors and ICs ... ;)
<MoeIcenowy>
KotCzarny : it seems that the used is for the device nodes
<Deskwizard>
i'll have to fix that hehe
<KotCzarny>
moe: hum?
<MoeIcenowy>
snd_hda_intel 28672 6
<MoeIcenowy>
this is on my laptop
<KotCzarny>
i've always thought 'used' means something actually opened the device
<MoeIcenowy>
it has 6 references
<KotCzarny>
just creating device nodes shouldnt make it used
<MoeIcenowy>
KotCzarny: please feel free to rmmod -f
<MoeIcenowy>
(but now I should modprobe my snd_hda_intel back :-)
<KotCzarny>
rmmod -f, modprobe, audio still doesnt play
<KotCzarny>
and when i ctrl-c aplay i get: aplay: pcm_write:1939: write error: Interrupted system call
<MoeIcenowy>
I think interrupted system call is ok
<MoeIcenowy>
do you have any GUI?
<KotCzarny>
xdm, but testing via ssh (not logged in x)
<MoeIcenowy>
(An audacious will have a progress bar stick at 0:00:00 when the sound driver sucks
<MoeIcenowy>
(oh a remote audacious may be terrible
<KotCzarny>
killing xdm didnt make any difference
<KotCzarny>
moe: i'm using oscp, but testing with aplay
<MoeIcenowy>
have you turned the mixer?
<MoeIcenowy>
oscp?
<KotCzarny>
i have all toggles on in alsamixer (except 'mute')
<MoeIcenowy>
what kernel are you using?
<KotCzarny>
4.5.0-rc3
<MoeIcenowy>
oh mainline
<MoeIcenowy>
I have never a mainlineable A10/20 device
<MoeIcenowy>
only two A33s
<KotCzarny>
i've thought i will give it a try
<MoeIcenowy>
and now I'm developing on a bcm2836
<KotCzarny>
but seems its either broken, or my .config is broken
<MoeIcenowy>
is there anything terrible in `dmesg` ?
<tkaiser>
That's code. What about reality. To confirm dvfs is working you would need a powermeter, userspace governor and idle through different dvfs operating points.
<KotCzarny>
usually looking at temperature should give similar info
<tkaiser>
Different Vcore voltage is way more important regarding consumption than cpufreq
<KotCzarny>
remember your findings that voltage directly correlated with temp (and not mhz)
<KotCzarny>
su just doing temperature graphs should be enough
<tkaiser>
Sure, but you need more time and have to be aware that there are other temperature influences also (display engine for example: On A20 and H3 this makes 3.4¡C difference whether a display is connected or not)
<tkaiser>
s/3.4/3-4/
rubensm_ has quit [Ping timeout: 264 seconds]
<KotCzarny>
if one only wants to find out if dvfs is working or not, you can do it without x and in 5-20 minutes of time i guess
<tkaiser>
Also the display engine seems to be enabled after a reboot even if no display is connected the first 10 minutes. I fooled myself multiple times due to this
<KotCzarny>
:)
<KotCzarny>
it auto disables itself?
<tkaiser>
At least on H3. Haven't checked A20
Worf has joined #linux-sunxi
hipboi has quit [Quit: This computer has gone to sleep]
tomboy65 has joined #linux-sunxi
avph has quit [Ping timeout: 240 seconds]
avph has joined #linux-sunxi
matthias_bgg has quit [Ping timeout: 264 seconds]
matthias_bgg has joined #linux-sunxi
tomboy65 has quit [Quit: Uhh ... gotta go.]
rubensm has joined #linux-sunxi
KotCzarny has quit [Ping timeout: 256 seconds]
p1u3sch1 has quit [Ping timeout: 248 seconds]
p1u3sch1 has joined #linux-sunxi
KotCzarny has joined #linux-sunxi
apritzel1 has joined #linux-sunxi
simosx has joined #linux-sunxi
apritzel1 has quit [Ping timeout: 248 seconds]
hipboi has joined #linux-sunxi
KotCzarny has quit [Ping timeout: 256 seconds]
avph has quit [Ping timeout: 240 seconds]
avph has joined #linux-sunxi
KotCzarny has joined #linux-sunxi
KotCzarny has quit [Ping timeout: 256 seconds]
KotCzarny has joined #linux-sunxi
<KotCzarny>
meh
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
apritzel1 has joined #linux-sunxi
afaerber has quit [Quit: Ex-Chat]
adj_ has quit [Quit: Leaving]
pietrushnic has quit [Quit: Coyote finally caught me]
cnxsoft has quit [Quit: cnxsoft]
lamer14555459556 has joined #linux-sunxi
hipboi has quit [Quit: This computer has gone to sleep]
lamer14555459556 has quit [Client Quit]
dlan_ has joined #linux-sunxi
avph has quit [*.net *.split]
tkaiser has quit [*.net *.split]
dlan has quit [*.net *.split]
avph has joined #linux-sunxi
cosm has joined #linux-sunxi
afaerber has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
flok420 has quit [Remote host closed the connection]
flok420 has joined #linux-sunxi
apritzel1 has quit [Ping timeout: 248 seconds]
KotCzarny has quit [Ping timeout: 256 seconds]
Worf has quit [Quit: Konversation terminated!]
Nacho___ has quit [Ping timeout: 250 seconds]
KotCzarny has joined #linux-sunxi
KotC has quit [Ping timeout: 250 seconds]
KotC has joined #linux-sunxi
Andy-D__ has quit [Ping timeout: 248 seconds]
tkaiser has joined #linux-sunxi
<tkaiser>
wens: Did you tested the GL830 for a specific reason?
Andy-D__ has joined #linux-sunxi
iamfrankenstein has quit [Ping timeout: 248 seconds]
<wens>
tkaiser: uh, because it was on the cubietruck plus?
<tkaiser>
Did you also measured write speed?
<wens>
i don't have any empty disks
<KotCzarny>
usb stick?
<tkaiser>
I always use iozone
<KotCzarny>
or just write 100MB of trash?
<KotCzarny>
should be enough to measure crapspeed
<Deskwizard>
any latest pop music album should do.
<Deskwizard>
*giggles*
<wens>
the disk is ntfs... not a good test bed :p
<tkaiser>
Anyway: I got 15/30 MB/s on the Banana Pi M3 and others reported the same values for Orange Pi Plus (also using GL830)
<tkaiser>
Yep, ext4 would be better
<tkaiser>
I just tested USB disk speed with the Orange Pi One and the old Allwinner 3.4.110 kernel: Close to 38 MB/s write and 35 MB/s read using ext4
<tkaiser>
JMS567 USB-to-SATA bridge. In this setup was no UASP involved, with mainline kernel and UASP we get ~40 MB/s
<tkaiser>
I really don't understand why all sunxi devices with onboard USB-to-SATA bridge use the same crap
<longsleep>
apritzel: the zzz lines are from reset-sunxi and the xxx lines from sunxi-mmc
adj_ has joined #linux-sunxi
<longsleep>
apritzel: well, it crashed now 5 times in a row and then it booted fine
<maz>
longsleep: your system is weird. you're getting an undefined instruction on "addx1, x1, #0x1".
<maz>
longsleep: doesn't look too good.
<longsleep>
maz: yes, the errors are not consistent - the Kernel or DT is not full compatible and correctly set up yet
<longsleep>
maz: it crashes with all kind of different reasons and sometimes works just fine
<longsleep>
maz: the same board never crashes with the BSP Kernel, so i guess the hardware is fine
<maz>
longsleep: not sure DT has anything to do with it. seems more like memory timings or something equally bad.
<longsleep>
maz: well, the only thing i replace is the Kernel and DT. With the BSP the same boot0, ATF and u-boot work fine
<longsleep>
apritzel: for comparison, boot log where it booted fine (and still is running after a couple of minutes): http://paste.ubuntu.com/15083026/
<longsleep>
maz: if you have any suggestion what i could try to nail this down, let me know. else i leave it for apritzel to figure out when he is back :)
alexst has quit [Ping timeout: 256 seconds]
<Turl>
plaes: when choosing reason for deletion choose spam and clear the box explaining the contents, less traces for search engines :)
alexst has joined #linux-sunxi
<plaes>
Turl: yeah.. my bad :(
<KotCzarny>
undelete, then delete again?
khuey is now known as khuey|away
arokux_ has joined #linux-sunxi
yann|work has joined #linux-sunxi
ssvb has joined #linux-sunxi
<Turl>
don't worry much
<ssvb>
longsleep: do you have "--enable-fix-cortex-a53-843419" reported for your toolchain when you run "aarch64-linux-gnu-gcc -v"?
<Turl>
KotCzarny: even if you do the msg persists heh
<KotCzarny>
kill it with fire?
<KotCzarny>
change the user name in the db?
<Turl>
KotCzarny: nevermind, only link to it is RecentPages
<Wizzup>
KotCzarny: have you seen phones with A10 or A20 in there?
<KotCzarny>
wizzup: um, a[12]0 is not a phone chip? power usage an order too high?
<Turl>
KotCzarny: ugh
<lamer14555655395>
Can anyone with upload permission to dl.sunxi.org please put the schematics of Banana Pi and Pro there so I can relink from the Wiki? I wasn't aware that they can only be downloaded after registrating in LeMaker's forum. I've put them here temporarely: http://kaiser-edv.de/tmp/2GB89H/
<KotCzarny>
turl: add some fake page on wiki with the link in the footer with 'support phone' keyword to override?
<KotCzarny>
tkaiser: i think i got the bpi schematics without registering, my memory might play tricks though (or it's recent)
<ssvb>
longsleep: please paste the output of "aarch64-linux-gnu-gcc -v"
<longsleep>
ssvb: no, my toolchain does not seem to support that parameter. Do you think that could be required?
<ssvb>
longsleep: that's a rather critical cortex-a53 erratum which can cause data corruption if you are unlucky to have a certain sequence of instructions generated in the code
<KotCzarny>
tkaiser: k, you didnt add that piece of info
<NiteHawk>
btw: the registration requirement is new, when i originally d/l'ed the m1 schematic, that wasn't necessary
<longsleep>
apritzel: latest linaro toolchain seems to be 15.05 with gcc 4.9 - which one do you use?
<KotCzarny>
also, what nitehawk said
<tkaiser>
M1 sounds like SinoVoip. Should be theoretically the same as 'Banana Pi' from LeMaker but who knows. At least Pro (LeMaker) and M1+ (SinoVoip) differ slightly
<apritzel>
longsleep: I compile mine myself, so atm I am at 2.25.1 with GCC 5.3.0
<KotCzarny>
m1 is just a rename of original bpi
<KotCzarny>
to differentiate from other models
<apritzel>
I guess I will recompile the whole beast
interrobangd has quit [Quit: Leaving]
<NiteHawk>
no, i simply meant original "Banana Pi"
<NiteHawk>
(i.e. not "Pro")
<KotCzarny>
banana pi == banana pi m1
<apritzel>
but in general I doubt that this is due to an CPU errata
<apritzel>
easy reflex by software developers to blame the hardware ;-)
<apritzel>
but in 99.9 % it's just bugs ...
<longsleep>
well - i try another compiler just to be sure
<longsleep>
i am using gcc-linaro-aarch64-linux-gnu-4.8-2014.04_linux.tar.xz now until i figured out the naming and download folder scheme of theirs :/
IgorPec has quit [Ping timeout: 240 seconds]
<apritzel>
longsleep: but as said above: you compiled the BSP kernel with the same toolchain, didn't you?
<longsleep>
yes
<longsleep>
BSP never crashed a single time
<ssvb>
apritzel: unfortunately I had quite interesting experience with CPU bugs on early Cortex-A8
<ssvb>
it was me who has provided testcases for reproducing #628216 and #725233 to ARM :-)
<ssvb>
also thumb2 used to be quite broken, and required workarounds in the toolchain too
<longsleep>
ssvb, apritzel: The linaro toolchain which i use now has --enable-fix-cortex-a53-835769 --enable-fix-cortex-a53-843419
<longsleep>
ssvb: do i need to add that parameter to the Makefile now or is it sufficient when the toolchain is built with it?
<ssvb>
longsleep: you can add it to the makefile to be sure, but I think that it should be already used by default
<longsleep>
ssvb: ok thanks
<KotCzarny>
ssvb, does a20 need any erratas in 4.5 ?
<ssvb>
KotCzarny: AFAIK nothing needs to be workaround at least in the toolchain
<ssvb>
longsleep: you can try to run "aarch64-linux-gnu-gcc -v helloworld.c" and check what kind of command line options are passed to the linker
<ssvb>
longsleep: if there is "--fix-cortex-a53-843419" among them, then the workaround is enabled by default
<longsleep>
lets see if it helped, booting in a minute
<tkaiser>
ssvb: Since another new Orange Pi H3 board is announced to be available in a few weeks I thought about creating some sort of a 'meta page' with knowledge/stuff that applies to all H3 based Orange Pis.
<tkaiser>
What do you think? We already started to fill the wiki with duplicated stuff (well, me started with it when creating the page for Orange Pi One)
<longsleep>
ssvb, apritzel well it did not crash on first boot - thats something :)
<longsleep>
ssvb, apritzel no luck on second boot - similar crash as before
<longsleep>
so i guess its something else
* apritzel
thought so ;-)
<apritzel>
longsleep: btw: what timer frequency does your kernel report?
<apritzel>
25.16 MHz?
<longsleep>
yes
<longsleep>
Architected cp15 timer(s) running at 25.16MHz (virt).
<apritzel>
so you should find the "24<<20" in the ATF source and replace it with 24 * 1000 * 1000
<longsleep>
actually that is oyur kernel
<apritzel>
but your ATF :-P
<longsleep>
err right
<longsleep>
ok let me find that
<ssvb>
apritzel: I'm pretty sure that somebody relying on an unpatched toolchain is destined to encounter this errata sooner or later :-)
avph has quit [Ping timeout: 264 seconds]
<ssvb>
so it's always better to take proactive measures
<apritzel>
ssvb: yeah, but those errata are usually rare
<apritzel>
sure you want to fix them, but normally they don't strike when you think they do ;-)
<apritzel>
only when you have shipped the thing to the customer and he has just launched that rocket with it ...
<ssvb>
:-)
<ssvb>
the compiler bugs are also rare, but there have been at least a few nasty ones affecting the linux kernel too
avph has joined #linux-sunxi
<apritzel>
longsleep: can you change the three last clocks in the mmc node to all read <mmc0_clk 0>?
<apritzel>
(instead of 0, 1, 2, respectively)
<longsleep>
apritzel: sure
<longsleep>
apritzel: also for mmc1 ?
<longsleep>
and mmc2?
<longsleep>
or ignore for now?
<longsleep>
apritzel: well, it still boots and some crash happend but non fatal