thefloweringash has quit [Ping timeout: 276 seconds]
<gnarface>
suniel: did you try it not in a VM?
junak[m] has joined #linux-sunxi
kayterina has joined #linux-sunxi
thefloweringash has joined #linux-sunxi
psydruid[m] has joined #linux-sunxi
suniel has quit [Ping timeout: 240 seconds]
suniel has joined #linux-sunxi
<suniel>
gnarface: not in a VM? meaning
ullbeking has quit [Ping timeout: 260 seconds]
z3ntu has joined #linux-sunxi
Benjojo has quit [Ping timeout: 240 seconds]
<gnarface>
suniel: GL_VENDOR: VMware. Inc.
<gnarface>
suniel: this part seems to indicate it detected VMWare's virtual devices, instead of the hardware you're actually developing for
<suniel>
glmark2 command is trigerred on the target(based on allwinner r16 cpu). target has a image generated from yocto build
ullbeking has joined #linux-sunxi
charco has quit [Ping timeout: 245 seconds]
<gnarface>
could it be the build environment that set that? i don't know, i just know VMware is not ARM hardware
jstefanop has joined #linux-sunxi
charco has joined #linux-sunxi
ullbeking has quit [Max SendQ exceeded]
mossroy has quit [Quit: Leaving]
<apritzel>
isn't that a red herring, and this Mesa code was just once provided by VMware?
charco has quit [Ping timeout: 260 seconds]
Benjojo has joined #linux-sunxi
ullbeking has joined #linux-sunxi
charco has joined #linux-sunxi
jonkerj has quit [Remote host closed the connection]
jonkerj has joined #linux-sunxi
maz has quit [Remote host closed the connection]
maz has joined #linux-sunxi
freemangordon1 has joined #linux-sunxi
freemangordon has quit [Read error: Connection reset by peer]
dlan has quit [Ping timeout: 265 seconds]
dlan has joined #linux-sunxi
kaspter has quit [Remote host closed the connection]
tnovotny_ has joined #linux-sunxi
kaspter has joined #linux-sunxi
gediz539 has joined #linux-sunxi
tnovotny has quit [Read error: Connection reset by peer]
hlauer has quit [Remote host closed the connection]
gediz0x539 has quit [Remote host closed the connection]
nashpa has quit [Quit: Going away]
x5eb has joined #linux-sunxi
_0x5eb_ has quit [Quit: Goodbye!]
akaWolf has quit [Ping timeout: 265 seconds]
x5eb is now known as _0x5eb_
libv_ has quit [Ping timeout: 240 seconds]
suniel has quit [Ping timeout: 240 seconds]
akaWolf has joined #linux-sunxi
hlauer has joined #linux-sunxi
libv has joined #linux-sunxi
sunshavi has quit [Ping timeout: 265 seconds]
nashpa has joined #linux-sunxi
* gnarface
doesn't know
buzzmarshall has joined #linux-sunxi
gaston1980 has quit [Ping timeout: 252 seconds]
reinforce has quit [Quit: Leaving.]
gediz539 has quit [Remote host closed the connection]
ldevulder has quit [Remote host closed the connection]
sunshavi has joined #linux-sunxi
ldevulder has joined #linux-sunxi
choozy has joined #linux-sunxi
tnovotny_ has quit [Quit: Leaving]
JohnDoe_71Rus has joined #linux-sunxi
matthias_bgg has quit [Ping timeout: 268 seconds]
<jernej>
VMware is just author of softpipe (it would seem)
<jernej>
and softpipe is obviously SW implementation
<jernej>
so issue is that mesa for some reason doesn't find lima driver
<jernej>
suniel: make sure card0 is sun4i-drm and card1 is lima
<jernej>
other than that, I don't know
<tuxd3v>
jernej he is with a outdated mesa, also it is reporting GL ES 3.1, but the hardware only supports GL ES 2.0
<jernej>
tuxd3v: reported version is due to softpipe implementation
gaston1980 has joined #linux-sunxi
<tuxd3v>
jernej, agree
<tuxd3v>
the maximum I achieved was around ~330fps for Mali 400, with no vblank mode..
<tuxd3v>
in glxgears
cmeerw has joined #linux-sunxi
matthias_bgg has joined #linux-sunxi
<tuxd3v>
I was even able to play supertuxlart@~20-25fps, but not all tracks only few, as I received a warning that they recomend GL ES 3.1(hardware only had GL ES 2.0.)
<tuxd3v>
supertuxkart*
vpeter has joined #linux-sunxi
Avamander[m]1 has quit [Quit: Idle for 30+ days]
e1z0_ has quit [Read error: Connection reset by peer]
e1z0 has joined #linux-sunxi
<MoeIcenowy>
jernej: VMware is really the author of softpipe. The original author is Tungsten Graphics, who developed the whole Gallium3D; then Tungsten got bought by VMware
<MoeIcenowy>
tuxd3v: BTW I think using GL non-ES may behave better on STK
<MoeIcenowy>
although the optimization feature is not implemented
<MoeIcenowy>
in GL1/GLES1, there's a feature called "alpha test"
<MoeIcenowy>
which gets dropped on newer GPUs, and emulated by using "discard;" in fragment shader
<MoeIcenowy>
Mali-400 supports alpha test
<MoeIcenowy>
(although our driver seems to be lack of this, because no one cares this GL1 feature)
<MoeIcenowy>
when STK is running on GLES2, it uses the alpha test emulation with discard, to deal with some textures (e.g. the tree)
<MoeIcenowy>
(and discard will prevent some Mali-400 optimization feature
<MoeIcenowy>
(this is what I found when I debug STK on Lima some months ago
<MoeIcenowy>
BTW complex STK tracks behave much worse than simple ones
<MoeIcenowy>
I cannot get some very complex ones run at 30fps even on RK3399 (Mali-T860 MP4, Panfrost)
tuxillo has quit [Remote host closed the connection]
<anarsoul|2>
MoeIcenowy: IIRC alpha testing is implemented
<anarsoul|2>
but it's likely not used by gles1 emulation code
<anarsoul|2>
well, pipe cap is actually set to 0
random_yanek has quit [Ping timeout: 250 seconds]
Mangy_Dog has joined #linux-sunxi
random_yanek has joined #linux-sunxi
parabyte has joined #linux-sunxi
<parabyte>
i got a unknown h2 or h3 tv board, iv tried loading armbian and some debian ports on it, what can people suggest i do to get a distro running on this?
<parabyte>
any documentation recommended?
<apritzel>
parabyte: a serial port would be probably very helpful
gaston1980 has quit [Ping timeout: 260 seconds]
specing has quit [Remote host closed the connection]
specing has joined #linux-sunxi
indy has quit [Ping timeout: 240 seconds]
camus has joined #linux-sunxi
kaspter has quit [Ping timeout: 240 seconds]
camus is now known as kaspter
sunshavi has quit [Read error: Connection reset by peer]
<parabyte>
apritzel, my end goal is to get rid of android completely and adapt libreelec to run on the board
<parabyte>
also i enjoy the educational exercise always good to learn something new
<apritzel>
parabyte: yes, that is a noble goal, and achievable, I'd say
libv has quit [Quit: leaving.]
<apritzel>
from experience it is best to dump as much from the installed system as possible (logs, any device trees, etc)
<parabyte>
what do you suggest i do today with this board apritzel ? oh also i have fel access to it and oddly booting via that means hangs the board too?!
<parabyte>
apritzel, how do you suggest i go about dumping the contents of the installed system?
libv has joined #linux-sunxi
<apritzel>
parabyte: well, for a start it would be sufficient to dump the bootlog of the Android that is on there
<apritzel>
I assume it spits out something on the serial while booting?
<parabyte>
yeah i piped that to a text file on my laptop
<apritzel>
parabyte: good, if you could put that into some pastebin it would be helpful
<parabyte>
give me just a few moments wife requires me!
tuxillo has joined #linux-sunxi
<apritzel>
parabyte: so U-Boot hanging after the SPL is somewhat odd (at least on an H3), have you checked that you built it correctly?
<apritzel>
and the box does really have 1GB of DRAM? It might be a DRAM problem, but the driver didn't complain
<apritzel>
you might try to lower the DRAM frequency, that sometimes helps
<parabyte>
oh yes i didnt set up the ram when i built uboot correctly i sort of went with defaults and yeah you got a point
<parabyte>
oh hang on does the uboot on these h3 cpus do the same as the old a20's you like set up a sdcard with a image on it and the boards uboot booted from it?
<parabyte>
is that a thing people still do?
kaspter has quit [Read error: Connection reset by peer]
kaspter has joined #linux-sunxi
<parabyte>
apritzel, i figured it out, i will get back to you later after i play with ram frequency
<apritzel>
parabyte: the boot process hasn't changed much since the A10 days: the BROM reads some sectors starting at sector 16 from the SD card into SRAM and executes that if the magic and checksum match
<parabyte>
apritzel, so what methods can you think of i can use to dump its flash via the command line?
<parabyte>
via uart
<parabyte>
i would grab firmware off the web but its abit of a chance to find the correct firmware for these things online
<apritzel>
I would try to get mainline U-Boot running, then you should be able to access everything from there
<apritzel>
assuming this box has eMMC and not raw NAND flash
<apritzel>
parabyte: make beelink_x2_defconfig looks like a good start, it's using the same DRAM frequency
<apritzel>
and since it's a TV box as well, chances are you can reuse much of the DT
<parabyte>
ah fantastic thats the only board image iv not tried, if that works from a precompiled image that be a good result!
<apritzel>
jernej: mmh, let me have a try. I was wondering about the clocks for the PHY device anyway
parabyte has quit [Remote host closed the connection]
<apritzel>
jernej: it seems like the PMU devices are gated by the EHCI clock anyway, which we don't model properly at the moment. I think it just works because the EHCI init triggers the PHY init
parabyte has joined #linux-sunxi
<parabyte>
apritzel, i tried the beelink x2 armbian image and i get this error
<parabyte>
apritzel, Failed to set core voltage! Can't set CPU frequency
<parabyte>
just out of interest why is that error happening? what does it mean exactly in the context of uboot?
<parabyte>
obviously i can read lol but i mean why cant it set the frequency?
<jernej>
parabyte: if you really want to move forward and stop guessing, you should extract FEX from Android
<jernej>
that would tell you almost everything you need to know to write proper U-Boot config file and device tree file
<jernej>
regarding that message - it means that CPU voltage regulator is not found
mmarc__ has quit [Remote host closed the connection]
<jernej>
that's not too surprising, many H3 designs don't have it
mmarc__ has joined #linux-sunxi
<jernej>
just disable it in U-Boot config
mmarc__ has quit [Remote host closed the connection]
mmarc__ has joined #linux-sunxi
<apritzel>
jernej: just odd, the beelink defconfig doesn't define a regulator, so this message shouldn't appear
mmarc__ has quit [Remote host closed the connection]
<apritzel>
jernej: not sure if Armbian deviated here?
mmarc__ has joined #linux-sunxi
mmarc__ has quit [Remote host closed the connection]
mmarc__ has joined #linux-sunxi
<jernej>
Armbian updates many U-Boot configs, mainly to lower DRAM frequency
mmarc__ has quit [Remote host closed the connection]
mmarc__ has quit [Remote host closed the connection]
<parabyte>
apritzel, yeah i can ill get onto it shortly
camus has joined #linux-sunxi
cnxsoft has quit [Read error: Connection reset by peer]
kaspter has quit [Ping timeout: 240 seconds]
camus is now known as kaspter
cnxsoft has joined #linux-sunxi
mmarc__ has joined #linux-sunxi
netlynx has quit [Quit: Ex-Chat]
gsz has quit [Ping timeout: 240 seconds]
mmarc__ has quit [Ping timeout: 260 seconds]
indy has joined #linux-sunxi
tuxillo has quit [Ping timeout: 252 seconds]
matthias_bgg has quit [Ping timeout: 240 seconds]
reinforce has joined #linux-sunxi
reinforce has quit [Client Quit]
stipa has joined #linux-sunxi
mmarc__ has joined #linux-sunxi
<stipa>
have aynone had luck with RTS5411 usb3 hub chip plugged into H6 USB3 port on opi lite 2 ? I have some trouble with Realtek r8152 usb ethernet nic crashing...
<stipa>
i have compiled the freshest driver from the realtek for the r8152 but it's the same
<stipa>
r8152 usb ethernet nic is plugged into the RTS5411 based usb3 hub
<stipa>
some of the kernel errors are: r8152 Tx status -71 , r8152 get_registers -71 , r8152 2-1.3:1.0 eth0: set_registers -71
<stipa>
sometimes it works but for a short time
sunshavi has quit [Ping timeout: 252 seconds]
<tuxd3v>
stipa, Realtek r8152 usb ethernet nic crashing, is known to crash a lot specially if you push high on bandwith, it will reset and atribute a new random maccaddress..
<stipa>
it made no difference from the previous 2.14 one
<stipa>
it works better on a genesys logic non powered usb3 hub, don't know the chip
<tuxd3v>
check if macaddress changes on reset
<stipa>
tuxd3v: yeah, i'll do that
<stipa>
thanks
<apritzel>
stipa: is this some dodgy off-tree kernel module driver for the r8152?
<apritzel>
if yes, what's wrong with the mainline driver?
<stipa>
apritzel: idk if it's dodgy, it's the one which comes with the armbian for the opi lite 2, but it seems the same as this one from realtek now that's running on the opi lite 2 now, nothing wrong
<stipa>
i hoped it'll fix the issue
<apritzel>
in my experience most off-tree drivers are just plain horrible
<apritzel>
the only reason to use them is if there is no alternative
<apritzel>
but there is a driver in mainline Linux
<stipa>
there is but
<stipa>
it doesn't work
<stipa>
there's an issue somewhere
sunshavi has joined #linux-sunxi
<stipa>
driver for the usb3 hub is in the kernel from 2.6
<stipa>
so it should be stable
<stipa>
dunno
<stipa>
i suspect it's the H6 usb3
<stipa>
not liking RTS5411 chip
<apritzel>
well, IIRC USB3 hubs are not trivial (if done correctly), and there are quite some broken ones out there
<apritzel>
is the r8152 an USB 2.0 chip?
<stipa>
3
<apritzel>
did you try the hub + Ethernet on something else, like a PC or laptop?
<stipa>
i'll do it tonight
<stipa>
the stuff is running internet now
<apritzel>
and is the hub powered (externally)?
<stipa>
i'm temporary using another non powerd usb hub where stuff work
<apritzel>
I doubt that those typical dev board ports can drive every device reliably, power-wise
<stipa>
apritzel: , yeah, not much juice from those ports
<apritzel>
the schematic looks reasonable, though, but one never knows
<stipa>
yeah
<stipa>
i should get one of those usb am meteres...
<stipa>
amp*
<stipa>
or more, even better
<apritzel>
stipa: are you powering the hub? is the cable/adaptor used for that reasonable?
<apritzel>
I don't trust micro USB when it comes to power delivery beyond 500mA
<stipa>
looks, good, outputs 5V
<stipa>
but there should be enough power from the board
<stipa>
for the ethernet nic wifi nic
<stipa>
usb*ž
<stipa>
i want powered one because of the hard disk
<stipa>
for the NAS
<stipa>
but something is fucky
<apritzel>
I have seen micro USB cables that crash boards when they draw more than ~0.5A, so you might want to try to rule out power issues as much as possible
<stipa>
yeah
<stipa>
if there is no power connected to the hub it draws power from the board, no obvious problems there
hlauer has quit [Ping timeout: 240 seconds]
<stipa>
it behaves the same with power and without
<stipa>
sometimes it spits errors right on boot, sometimes it crashes after a shor time of internet usage
<stipa>
the ethernet nic crashes*
<stipa>
i have no idea what that hub is doing, maybe it's switching power by software or soething...
<stipa>
maybe it briefly cuts power to the usb devices, dunno
<stipa>
ok, i'm going to check, bbl
stipa has quit [Quit: WeeChat 3.0.1]
jstefanop has quit [Remote host closed the connection]