qschulz has quit [Remote host closed the connection]
qschulz has joined #u-boot
mmu_man has quit [Ping timeout: 256 seconds]
apritzel has quit [Ping timeout: 248 seconds]
thopiekar has joined #u-boot
thopiekar_ has quit [Ping timeout: 264 seconds]
<sjg1>
brujah: That seems to be a bug, and in fact that test is not running in CI. For now you can do 'ln -s build-sandbox_vpl/image.bin' and try again
<macromorgan>
question... how close to mainline Linux does U-Boot configurations need to be? I would like to support a "pseudo-device" for a family of devices in U-Boot, whereas Linux will support each device individually.
<macromorgan>
The thought is that U-Boot will boot a generic devicetree and the board.c file will have a function that passes an environment parameter that detects the correct device. The Linux boot.scr or whatever can then use this to load the correct devicetree.
<macromorgan>
The "pseudo-device" would be the Anbernic RGxx3, and it would support the Linux devices of RG503, RG353P, RG353M, RG353Vv1, RG353VSv1, RG353Vv2, and RG353VSv2
<macromorgan>
I will import the rk356x.dtsi, rk3566.dtsi, and rk3566-anbernic-rgxx3.dtsi devicetrees, but will deviate from Linux beyond that
<macromorgan>
I can read an ADC channel to know if it's an RG503, RG353P(or RG353M, same device in a different shell), or RG353V. I can read an I2C address to know if it's an RG353V or RG353VS. I'll have to port the DSI and DSI-DPHY drivers, but then I can know if it's a v1 or v2 of the 353V/VS
<Tartarus>
What's the binding, exactly? You can have the u-boot "psuedo-device" bind on the same compats as the linux driver would?
<broonie>
If they're compatible enough for the binding to have a fallback compatible they could still share the DT.
redbrain has quit [Read error: Connection reset by peer]
<macromorgan>
bindings in Linux are "anbernic,rg503", "anbernic,rg353p", "anbernic,rg353v", "anbernic,rg353vs", "anbernic,rg353v-v2", and "anbernic,rg353vs-v2". I haven't added "anbernic,rg353m" to Linux yet, but it's identical in everything but form factor to the rg353p.
<macromorgan>
the devices differ in form factor, a handful of buttons, LCD panel, touchscreen presence, and eMMC presence
<macromorgan>
I recall there being a way to embed multiple dtbs in an SPL fit image, I'll have to read into that more
jaganteki has joined #u-boot
redbrain has joined #u-boot
mmu_man has joined #u-boot
apritzel has quit [Ping timeout: 252 seconds]
jaganteki has quit [Quit: Client closed]
vagrantc has quit [Quit: leaving]
dossalab has joined #u-boot
dossalab has quit [Client Quit]
niska has quit [Quit: Leaving]
prabhakarlad has quit [Quit: Client closed]
jaganteki has joined #u-boot
prabhakarlad has joined #u-boot
prabhakarlad has quit [Client Quit]
bern has joined #u-boot
niska has joined #u-boot
pgreco_ has quit [Ping timeout: 248 seconds]
jaganteki has quit [Quit: Client closed]
ccf has joined #u-boot
<ccf>
Ahoj, for rk3399, using u-boot and TF-A mainline leads to DMA probe issue in linux-kernel: "OF: amba_device_add() failed (-19) for /dma-controller@ff6d0000". Any ideas?
ldevulder has quit [Read error: Connection reset by peer]
ldevulder has joined #u-boot
jaganteki has joined #u-boot
dossalab has quit [Quit: dossalab]
ldevulder has quit [Quit: Leaving]
mmu_man has joined #u-boot
mmu_man has quit [Ping timeout: 268 seconds]
mmu_man has joined #u-boot
jaganteki has quit [Ping timeout: 260 seconds]
goliath has quit [Quit: SIGSEGV]
mmu_man has quit [Ping timeout: 268 seconds]
naoki has joined #u-boot
mmu_man has joined #u-boot
ikarso has joined #u-boot
pgreco has joined #u-boot
ebach has quit [Ping timeout: 264 seconds]
naoki has quit [Remote host closed the connection]
<marex>
ccf: git grep 19 include | grep errno
<hanetzer>
that's one thing I don't like about ghidra... you can, for example, 19 to ERRNO_WHATEVER in their disassembly/decompiler view, but you can't set -19 to -ERRNO_WHATEVER :/
<Forty-Bot>
marex: I prefer `errno 19` :)
<ccf>
marex: It's ENODEV, but when using a different u-boot and TF-A then the DMA device gets probed fine by the kernel. any ideas why?
<marex>
Forty-Bot: well now that ... is awesome
<marex>
TIL
<marex>
ccf: could it be the device is not enabled ?
<marex>
ccf: I had a quick look into Linux drivers/amba/bus.c amba_device_add and there is a large comment about it
<marex>
also, there are only two sites which return -ENODEV