<arokux2>
ssvb: I cannot help myself resist from hacking... so I thought I could try to take a look at usb otg, since everything else is kind of done or started to work on...
<arokux2>
ssvb: what do you think?
konradoo77 has quit [Ping timeout: 272 seconds]
<netlynx>
Is anybody working on the tv decoder?
konradoo77 has joined #linux-sunxi
<arokux2>
netlynx: search the mailing list, there was some guy talking about tv things
pwhalen has joined #linux-sunxi
esperegu has joined #linux-sunxi
<ssvb>
arokux2: well, you are more than welcome to do some hacking to improve usb otg support :-) and the mainline kernel needs it badly
<arokux2>
ssvb: do you have some tips for me? some kind of a roadmap?
<ssvb>
arokux2: I'm not a usb person myself and still need to read a bunch of usb specification documents to get a better idea about how this hardware works
<arokux2>
ssvb: I know nothing :) so it means it will take ages
<ssvb>
arokux2: ping wens and maybe some other people who tried to work on the usb otg code earlier
<ssvb>
maybe jukivili has some usb otg mainline plans too
esperegu has quit [Ping timeout: 260 seconds]
<netlynx>
arokux2, thx.
netlynx has quit [Remote host closed the connection]
<arokux2>
ssvb: is has been long time since jukivili was here, so I guess no
<ssvb>
and ccaione was also doing something
konradoo77 has quit [Ping timeout: 244 seconds]
<arokux2>
ssvb: and what are you working on?
konradoo77 has joined #linux-sunxi
ninolein has quit [Ping timeout: 245 seconds]
ninolein has joined #linux-sunxi
<ssvb>
arokux2: universal and (mostly) device independent wrapper around u-boot to handle user friendly and (mostly) foolproof installation and reconfiguration :-)
<ssvb>
libv: not sure what we can or should do for sunxi-3.4 kernel, because the current USB OTG support out of the box is not that great
<libv>
ssvb: i know
<libv>
ssvb: and yeah, but i know where to find you :p
<libv>
and since you are not taking these devices apart, ndh can never be finished :)
bonbons has quit [Quit: Leaving]
w00tc0d3 has quit [Read error: Connection reset by peer]
w00tc0d3 has joined #linux-sunxi
konradoo87 has quit [Ping timeout: 255 seconds]
konradoo77 has joined #linux-sunxi
<hno>
Turl, very interesting. Claims SATA support. And comes with heatsink so it might even be able to use the CPU...
Renard has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
bengal has quit [Quit: Leaving]
Gerwin_J has joined #linux-sunxi
blsd1 has quit [Remote host closed the connection]
Andy-D has quit [Remote host closed the connection]
konradoo77 has quit [Ping timeout: 240 seconds]
konradoo77 has joined #linux-sunxi
<hno>
Hm... this smells very much like obfuscation rather than trying to keep IP private. Are they ignoring licensing requirements in the third-party blocks they use? If so then there may be not only GPL to worry about.
petr has quit [Ping timeout: 272 seconds]
petr has joined #linux-sunxi
Gerwin_J has quit [Ping timeout: 250 seconds]
libcg has quit [Quit: libcg]
konradoo77 has quit [Ping timeout: 246 seconds]
wickwire has quit [Remote host closed the connection]
bertrik has quit [Remote host closed the connection]
<hno>
If this is true then sun8i have a GMAC and it looks the same as A20 one but with scrambled register order.
<hno>
sun9i looks to be the same as on the A20.
FDCX__ has joined #linux-sunxi
FDCX_ has quit [Ping timeout: 246 seconds]
rafaelMOD has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
F1skr has quit [Quit: WeeChat 1.0.1]
<arokux2>
hno: why do we all bother about this mess? :( I mean allwinner in general..
* arokux2
in bad mood after reading Allwinner code
arokux2 has quit [Remote host closed the connection]
<bsdfox>
can anyone help me out with the proper strategy for this-- I'm using nixio to access a spidev device in lua (openwrt). The slave device I'm working with requires 16-bit reads and writes (8-bit address, 8-bit data) so I'm able to write to the device no problem spidev:write(string.char(addr, data)) however this function only returns the number of bytes written.. The spidev:read() function only takes the number of bytes to read then writes 0x
<bsdfox>
00 dummy bytes.. if I write the address then try to read one byte the /CS line is pulled high in the middle and the read returns nothing
<bsdfox>
I think there might be a way to use pipe() or fork() to access the data but my attempts have been feeble and I need confirmation that it's an approach that might have potential :)
<hno>
bsdfox, what do you mean by "write the address"?
hipboi has joined #linux-sunxi
<hno>
SPI do not have any addresses.
<bsdfox>
hno, I have to write the address of the register I want to read
<hno>
So the SPI device you are talking to want you to write the register address and then it returns data in the same transaction?
<hno>
if so then you need full duplex SPI.
<bsdfox>
hno, yes. so I send 0x06 then 0xnn(dummy byte) and the value in register 0x06 is returned when I send the dummy byte
<hno>
Or.. actually half duplex work too if chained with no CS change flag.
<bsdfox>
hno, I am full duplex
hipboi_ has quit [Ping timeout: 255 seconds]
<hno>
in either case you need to use the ioctl() interface. The simple read/write spidev interface cannot do this in one transaction.
<bsdfox>
hno, so there isn't a way to grab the data from a write using fork() or pipe()? bummer. I'm quite convinced that it can't be accomplished using read()
<hno>
Doesn't help. Still boils down to a read() which gets translated into it's own SPI transaction.
<hno>
read/write on the device is half duplex by design, and each read/write is a separate transaction.
<hno>
for anything else (which is pretty much everything) you need to use the ioctl() spidev interface.