buzzmarshall has quit [Quit: Konversation terminated!]
GNUtoo has quit [Ping timeout: 240 seconds]
GNUtoo has joined #linux-amlogic
Danct12 has quit [Remote host closed the connection]
GNUtoo has quit [Remote host closed the connection]
Danct12 has joined #linux-amlogic
GNUtoo has joined #linux-amlogic
Danct12 has quit [Read error: Connection reset by peer]
Danct12 has joined #linux-amlogic
Danct12 has quit [Quit: Leaving]
GNUtoo has quit [Remote host closed the connection]
GNUtoo has joined #linux-amlogic
ldevulder has quit [Quit: Leaving]
GNUtoo has quit [Remote host closed the connection]
ldevulder has joined #linux-amlogic
Danct12 has joined #linux-amlogic
GNUtoo has joined #linux-amlogic
<xdarklight>
filt3r: just seen the reply - thanks! Liang is typically quick to respond, so let's see what he thinks :-)
Danct12 has quit [Quit: Leaving]
GNUtoo has quit [Remote host closed the connection]
GNUtoo has joined #linux-amlogic
GNUtoo has quit [Remote host closed the connection]
GNUtoo has joined #linux-amlogic
<narmstrong>
filt3r: the problem is Amlogic doesn't store the bad eraseblocks and the bootloader the same way it's done with mainline linux
<narmstrong>
filt3r: so you'll need to either implement the same bad erase block in Linux or in the vendor u-boot, or port the NAND driver to mainline U-boot
<narmstrong>
filt3r: for how they store the bootloader in NAND is not clear, it seems they store it in raw mode without using ECC
<narmstrong>
filt3r: for this you'll need to find out how to setup the bootloader partition in Linux
ndufresne0 has joined #linux-amlogic
tomeu6 has joined #linux-amlogic
<filt3r>
narmstrong: thanks for the info, now that you mentioned it, i faintly remember seeing something like this a few years ago when working with the vendor kernel/u-boot
<filt3r>
but yeah we will have to adjust the Linux handling as we have devices in the field so a flash layout change is not possible :(
GNUtoo has quit [Remote host closed the connection]
GNUtoo has joined #linux-amlogic
tomeu6 has quit [Ping timeout: 246 seconds]
ndufresne0 has quit [Ping timeout: 255 seconds]
vagrantc has joined #linux-amlogic
jernej has joined #linux-amlogic
buzzmarshall has joined #linux-amlogic
elastic_1 has joined #linux-amlogic
elastic_dog has quit [Killed (mercury.libera.chat (Nickname regained by services))]
ldevulder has quit [Quit: Leaving]
vagrantc has quit [Quit: leaving]
<tortoise>
Question about the radxa zero which is g12a (s905y2) the device tree doesn't include ge2d and I'm wondering if this is because this SoC doesn't have that or if it is just simply not enabled?
<xdarklight>
tortoise: the GE2D driver which we have currently only supports the A113D (AXG) SoC. I think the GE2D on that SoC is slightly different than the GE2D IP on other SoCs, so it would require some work (Neil probably knows more)
<tortoise>
ok, thats interesting, I may be able to just use the panfrost driver's vkCmdBlitImage instead
jernej has quit [Ping timeout: 256 seconds]
<cphealy>
tortoise: out of curiosity, what is the use case you have for the GE2D?
<tortoise>
so I'm still learning and I might actually not need it at all. my uses were pixel format conversion 1) taking NM12 (yuv) from decoder and converting that RGBA and 2) scaling. Both of these might be fine to do on GPU since I'm using external texture but I'm still reading up on this part
<cphealy>
ack
<cphealy>
My recollection with YUV is that there are display HW planes in the SoC that directly support YUV. The GPU certainly supports YUV too. Both GPU and HW planes should support scaling too.
<tortoise>
yeah, I'll going to see how it goes with GPU for now. I'm trying to build out some userspace applications, that I can understand, which use this hardware . My hopes are once I understand that side I can be helpful contributor to the kernel mainlining effort.
<cphealy>
The GE2D is exposed via the V4L2 API, right? If so, it would be the gstreamer v4l2 mem2mem convert plugin I think.
<xdarklight>
cphealy: correct, GE2D is exposed via V4L2 to userspace
<xdarklight>
cphealy: according to https://lwn.net/Articles/832862/ it announces itself as "Video Memory-to-Memory" device (so I think it's reasonable to assume that the mem2mem plugin would be used with it)