<megi>
+ some kernel patch to disable DSI-0 output, because kmssink had some issue outputing to the correct plane or something like that
<megi>
and I didn't really felt like recompiling gst plugins to fix that
vagrantc has quit [Quit: leaving]
warpme_ has joined #linux-sunxi
tnovotny has joined #linux-sunxi
<gediz0x539>
megi: thank you so much
gediz0x539 has quit [Remote host closed the connection]
gediz0x539 has joined #linux-sunxi
gediz0x539 has quit [Ping timeout: 480 seconds]
prefixcactus has joined #linux-sunxi
linusw has joined #linux-sunxi
Mangy_Dog has joined #linux-sunxi
<apritzel>
ah, hi linusw, do you have any information on the cw1200 Wifi chip?
gediz0x539 has joined #linux-sunxi
warpme_ has quit [Quit: Connection closed for inactivity]
Mangy_Dog has quit [Ping timeout: 481 seconds]
Mangy_Dog has joined #linux-sunxi
qCactus has joined #linux-sunxi
prefixcactus has quit [Ping timeout: 480 seconds]
warpme_ has joined #linux-sunxi
ftg has joined #linux-sunxi
prefixcactus has joined #linux-sunxi
qCactus has quit [Ping timeout: 480 seconds]
<linusw>
a bit in my head but sadly no datasheet :(
<linusw>
I will check some old mails.
<linusw>
apritzel: what I know for sure is that the team making CW1200 was based in Greater Noida, India, and taken over by Mediatek in July 2013.
<linusw>
What I also know is that we had a hard time enabling it even on the reference designs because the CW1200 requires an external clock source, no internal clock provided.
<linusw>
In our case that clock came from the Bluetooth chipset which was lacking a driver, so we would have to fix that driver first to get this driver working.
<apritzel>
linusw: thanks! Do you know of other related chipsets? I think arnd found some other driver in mainline that looks similar too?
gediz0x539 has quit [Remote host closed the connection]
<arnd>
linusw: the driver that I found is drivers/staging/wfx/, which is for the silicon labs version of the hardware and actively maintained by one of their developers (Jérôme Pouiller <jerome.pouiller@silabs.com>)
<arnd>
linusw: any chance you confused mtk and silabs, or that mtk sold it on afterwards?
<arnd>
(trying to construct a family tree of who licensed or sold the IP to whom)
<arnd>
the silabs copyright lists "2017-2020", so maybe that is when they got it from mtk
<arnd>
silabs definitely have the firmware sources and are modifying them, but maybe every licensee has those
<linusw>
arnd: I am certain it was Mediatek back then but no idea what happened after that.
<arnd>
another plausible route would be that Energy Micro was already a licensee if ST-Ericssons wifi before this got acquired by silabs
<ndufresne>
try to set env GST_DEBUG=2, should give you a preceding error log with more precise info
cmeerw has joined #linux-sunxi
<ndufresne>
gediz0x539: note that kmssink support is missing, so you can't direct display (buffers needs to be detiled on CPU, high CPU usage)
<ndufresne>
if me (or someone else) can manage to make it work properly, patches are welcome of course
<ndufresne>
there is also an interesting in adding some texelFetch driven detiler in libgstgl, but not sure the ancient GPU supports it or not
jemk_ has joined #linux-sunxi
jemk has quit [Read error: Connection reset by peer]
fcas has joined #linux-sunxi
<fcas>
good afternoon everyone. I'm trying to add a Chip Select to SPI1 on device tree (sun8i-orangepi-pc-plus.dts) but I haven't been able to make it work until now, linux 5.4.69
<fcas>
tried to use &spi1_pins{pins = "PA15", "PA16", "PA14", "PA13","PC3"; };
<fcas>
PC3 should be CS1 for SPI1
<fcas>
tried to make it simple too, using just &spi{ .... cs-gpios = <0>, <&pio 2 3 GPIO_ACTIVE_LOW>; //CS0, PC3 }
choozy has joined #linux-sunxi
hlauer has joined #linux-sunxi
warpme_ has quit [Quit: Connection closed for inactivity]
gsz has joined #linux-sunxi
<apritzel>
fcas: this cs-gpios definitely does not work this way, you cannot mix internal CSs with GPIOs - the value must be all valid GPIO specifiers
Luke-Jr has quit [Ping timeout: 480 seconds]
<apritzel>
fcas: I don't think that the H3 actually has an internal CS1, for either SPI0 or SPI1? PC3 is definitely SPI0_CS
<fcas>
but my spi0 isn't enabled, I'm just using spi1
<fcas>
and I read that cs-gpios=<0> is working, I can read my first device
<fcas>
read and write to my first device on spi bus
<fcas>
when I try to use the second one, I can see the MOSI and CLK working
<fcas>
just the CS1 (PC3) doesnt work
<apritzel>
I think there was an issue with external GPIOs used as chipselects in the driver?
<apritzel>
ah, indeed this 0 in the cs-gpios list is a special trick for gpios (because it's never a valid phandle)
gsz has quit []
Luke-Jr has joined #linux-sunxi
<apritzel>
fcas: you must not enumerate PC3 in the SPI pins property, by the way
<apritzel>
because your line above won't work anyway, because it tries to switch to function "spi1", which is not defined for PC3
<apritzel>
so from skimming over both the generic and sun6i SPI code, it looks like it should work, using cs-gpios, because this is handled by generic SPI code
<apritzel>
just don't add PC3 to the spi pins, but just name it in cs-gpios
<apritzel>
fcas: and keep an eye on kernel messages (dmesg) and /sys/kernel/debug/gpio, to see if the GPIO is configured correctly