Schimsalabim has quit [Read error: Connection reset by peer]
Schimsalabim has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
hexdump01 has joined #linux-sunxi
hexdump0815 has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
JohnDoe_71Rus has joined #linux-sunxi
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
aggi has quit [Remote host closed the connection]
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
aggi has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
aggi has quit [Remote host closed the connection]
aggi has joined #linux-sunxi
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
Schimsalabim has quit [Read error: Connection reset by peer]
Hypfer is now known as Guest2419
Guest2419 has quit [Read error: Connection reset by peer]
Hypfer has joined #linux-sunxi
warpme has joined #linux-sunxi
warpme is now known as Guest2420
warpme_ has joined #linux-sunxi
warpme_ is now known as warpme
warpme has quit [Ping timeout: 480 seconds]
Guest2420 has quit [Read error: Connection reset by peer]
warpme has joined #linux-sunxi
warpme is now known as Guest2423
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
Guest2423 has quit []
warpme has joined #linux-sunxi
warpme has quit []
JohnDoe_71Rus has quit [Ping timeout: 480 seconds]
kepstin has quit [Remote host closed the connection]
kepstin has joined #linux-sunxi
Schimsalabim has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
Schimsalabim has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
Schimsalabim has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
warpme has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
Schimsalabim has quit [Read error: Connection reset by peer]
Schimsalabim has joined #linux-sunxi
digetx has quit [Ping timeout: 480 seconds]
dsimic is now known as Guest2433
digetx has joined #linux-sunxi
dsimic has joined #linux-sunxi
Guest2433 has quit [Ping timeout: 480 seconds]
gsz has joined #linux-sunxi
digetx is now known as Guest2434
digetx has joined #linux-sunxi
Guest2434 has quit [Ping timeout: 480 seconds]
apritzel has joined #linux-sunxi
warpme has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
<electricworry>
apritzel: Good morning! I learned how to apply patches from a mbox, so I'm more ready to start reviewing. That patch series you sent over for me (v5 of Allwinner DE33). Can I just apply that against linux master? What are "the 6.12 drm-next patches" mentioned in patch 0/X?
<apritzel>
electricworry: yes, give it a try. 6.12 drm-next should be part of the tree already
<electricworry>
Thanks. I'm not entirely comfortable with DTs yet. These new patches, are these in a state that I could add them to the Orange Pi Zero 3 device tree and have them enabled for a live test - regardless of how broken they might be at the moment? And do you know if I'm likely to see any graphics output at this point?
<apritzel>
no, those patches are just the core display engine support. For the OPi Zero3 you would need HDMI patches on top
<apritzel>
tokyovigilante has them in his repo, I believe
<electricworry>
Gotcha!
<apritzel>
it helped me to go through this series patch by patch, trying to figure out what's going on
<apritzel>
because if you are interested in learning about *drivers*, this is the way to learn
<apritzel>
I mean learning about the device registers and such is only going to teach you how the Allwinner display engine is built - not sure that's too useful
<electricworry>
Thanks apritzel. I appreciate your patience. I'll read through the patches.
Hypfer is now known as Guest2438
Guest2438 has quit [Read error: Connection reset by peer]
Hypfer has joined #linux-sunxi
apritzel has quit [Remote host closed the connection]
smaeul_ has joined #linux-sunxi
<electricworry>
TCON is a timing controller?
smaeul has quit [Ping timeout: 480 seconds]
<MoeIcenowy>
electricworry: yes, it sends display signal to HDMI controller, DSI controller or directly to RGB/LVDS
warpme has joined #linux-sunxi
<electricworry>
Thanks. What does "IP block" mean in the context of "Allwinner Display Engine IP block"?
<MoeIcenowy>
IP core - a reusable/licensable part of a chip
<electricworry>
Ok, thanks. I'm still quite confused about the boundaries that separate what is ARM vs Allwinner. e.g. A board like OrangePi Zero2 has an Allwinner H616, which is based on (includes? or is designed based on a reference spec of?) ARM Cortex-A53 and Mali-G31.
<electricworry>
Does the Display Engine 3.3 utilise the Mali GPU? i.e. sit on top of it, providing an abstraction unique to the Allwinner SOC? Or is it just the Mali GPU rebadged? Or is it discrete from and nothing to do with the Mali "IP block"?
<electricworry>
And am I asking stupid questions?
warpme has quit []
BroderTuck has joined #linux-sunxi
<BroderTuck>
DisplayEngine is for 2d, Mali is for 3d, different things
<electricworry>
Ok, thanks. I'll forget about Mali/3D as irrelevant. :)
digetx has quit [Ping timeout: 480 seconds]
digetx has joined #linux-sunxi
apritzel has joined #linux-sunxi
<apritzel>
electricworry: Allwinner acts as an integrator: they take IP blocks (complex circuits, if you like) from different companies, or use their own, and combine them into a SoC
<apritzel>
the Cortex-A53 and Mali G31 cores are Arm IP, AW takes them unaltered, just are allowed to configure them, for instance choose the cache size, or whether they contain the crypto extension
<apritzel>
they combine this with their own (we think) DisplayEngine, and combine them with some apparently altered or hacked HDMI controller IP core from some other company
<electricworry>
Thanks. Getting clearer. I've given the v5 DE33 patch series a good read and it seems sane. I can see there were a couple of suggestions but it looks like it's well engineered to avoid regressions while introducing additional features.
<electricworry>
How can I help it along? Boards I have: Orange Pi Zero Plus2 (H3/sun8i/Cortex-A7), Orange Pi Zero 2 (H616/sun50i/Cortex-A53), Orange Pi Zero 3 (H618/sun50i/Cortex-A53). Is it useful for me to test the DE33 patch series on all three of these boards to confirm no obvious regressions?
<apritzel>
Mali is a pure memory-to-memory device: it reads data (textures, vectors, etc) from one part of memory, and renders that into another part of memory, in a configurable format matching the framebuffer encoding
<electricworry>
A-ha! That's clever.
<jakllsch>
I mean, technically any GPU is that... it's just graphics cards have seperate memory
<apritzel>
jakllsch: yes, but the PC habit of offering everything from one hand as a "graphics card" often confuses people, and they think Mali is the "graphics card", and would be needed for displaying anything
<electricworry>
apritzel: Exactly how I imagined it.
<electricworry>
This is very helpful.
<apritzel>
and the funny thing is that it involves IP from at least three different vendors: DE from Allwinner, Mali from Arm, and HDMI from Designware (we think)
<apritzel>
without those patches you wouldn't see anything on an HDMI monitor
<apritzel>
electricworry: also the hardware video encoders and decodes are again separate IP, partly again from other companies
<apritzel>
and they work similarly like Mali: consuming the compressed video stream, outputting frames directly into the framebuffer (for the decoders)
<apritzel>
regarding the patches: I agree they look good on a first glance, but maybe you can dig deeper, and use that to understand parts of the graphics stack
<apritzel>
and hopefully it clicks at some point, and you really understand what they do: and then can provide a review
<apritzel>
what's yet unresolved are some hacks in the HDMI patches, I think, so if you want to dig in there ...
Schimsalabim has quit [Read error: No route to host]