ChanServ changed the topic of #linux-sunxi to: Allwinner/sunxi development - Did you try looking at our wiki? https://linux-sunxi.org - Don't ask to ask. Just ask and wait for an answer! - This channel is logged at https://oftc.irclog.whitequark.org/linux-sunxi
Daanct12 has joined #linux-sunxi
Danct12 has quit [Read error: Connection reset by peer]
ftg has quit [Read error: Connection reset by peer]
Daaanct12 has joined #linux-sunxi
grming has joined #linux-sunxi
Daanct12 has quit [Ping timeout: 480 seconds]
paulk-ter has quit [Ping timeout: 480 seconds]
chewitt has joined #linux-sunxi
vagrantc has quit [Quit: leaving]
szemzoa has quit []
grming has quit [Quit: Konversation terminated!]
vagrantc has joined #linux-sunxi
apritzel_ has quit [Ping timeout: 480 seconds]
cnxsoft has joined #linux-sunxi
vagrantc has quit [Quit: leaving]
junari_ has joined #linux-sunxi
rajkosto has quit [Read error: Connection reset by peer]
chewitt has quit [Quit: Zzz..]
apritzel_ has joined #linux-sunxi
apritzel_ has quit [Ping timeout: 480 seconds]
junari__ has joined #linux-sunxi
junari_ has quit [Ping timeout: 480 seconds]
JohnDoe_71Rus has joined #linux-sunxi
junari__ is now known as junari
<junari> Where should I place phy reset definition in DT? MAC, MDIO or PHY node?
rm_ has joined #linux-sunxi
<rm_> hello
<rm_> setting up an Orange Pi Prime (H5), got Armbian, replaced their 5.15 kernel with Debian stock 6.0, and I find that cpufreq stopped working
<rm_> https://linux-sunxi.org/Linux_mainlining_effort says should be supported since 5.9 in mainline for H5
<rm_> so what's missing?
<rm_> and tried to manually load the "cpufreq-dt" module, no effect
apritzel has joined #linux-sunxi
JohnDoe_71Rus has quit []
<apritzel> junari: the PHY reset belongs into the PHY node, so the one with the ethernet-phy-ieee802.3-c22 compatible
<apritzel> in contrast to clocks this already works
<apritzel> rm_: it looks like DVFS is not enabled in the DT for that board
<apritzel> rm_: look at sun50i-h5-nanopi-r1s-h5.dts to check what you need
<rm_> apritzel, does it look from my description or you already checked the actual DTS?
<apritzel> it's bascially including a cpu-supply property and including sun50i-h5-cpu-opp.dtsi, IIRC
<rm_> that's in kernel DTS, right, not u-boot?
<apritzel> rm_: I looked at the mainline kernel DT, not sure which one you are using, though
<rm_> thanks
<apritzel> U-Boot should be in sync with the kernel
<apritzel> (with some delay)
<rm_> u-boot seems to use some [at least] much simpler DTSes
<rm_> and names are all different
<rm_> wait, names are ok
<rm_> it's their config names that are a bit lacking a naming convention
paulk has joined #linux-sunxi
grming has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
junari_ has joined #linux-sunxi
junari has quit [Ping timeout: 480 seconds]
<apritzel> yes, the _defconfig names are a whole mess
<apritzel> the DTs are literally copied from the Linux tree, we just need to amend them a bit, to increase compatibility and work around some path issues (arch/arm/dts vs. arch/arm64/boot/dts/allwinner)
<apritzel> it's true that U-Boot *could* get away with much simpler DTs, but there is also no real problem with letting it see the whole thing, and just using a much smaller subset
<karlp> and it makes merging and syncing much simpler
<apritzel> indeed
<apritzel> speaking of which: can someone please review the DT update patches: https://lore.kernel.org/u-boot/20230121012546.9963-1-andre.przywara@arm.com/
<apritzel> and https://lore.kernel.org/u-boot/20230121012613.9989-1-andre.przywara@arm.com/ for those boards with only half the bits ;-)
<apritzel> just verifying that this matches the Linux version mentioned, minus the differences that we need to unfortunately live with
<apritzel> bonus points for coming up with a tool that automates those changes or at least verifies them
szemzoa has joined #linux-sunxi
Net147_ has joined #linux-sunxi
Net147 has quit [Ping timeout: 480 seconds]
cnxsoft has quit []
junari has joined #linux-sunxi
junari_ has quit [Ping timeout: 480 seconds]
macromorgan has quit [Quit: Leaving]
sauce has quit [Quit: sauce]
sauce has joined #linux-sunxi
junari has quit [Ping timeout: 480 seconds]
getfrom has joined #linux-sunxi
getfrom has quit []
getfrom has joined #linux-sunxi
getfrom has left #linux-sunxi [#linux-sunxi]
apritzel has quit [Ping timeout: 480 seconds]
apritzel_ has joined #linux-sunxi
apritzel_ has quit [Ping timeout: 480 seconds]
JohnDoe_71Rus has quit []
apritzel_ has joined #linux-sunxi
vagrantc has joined #linux-sunxi
grming has quit [Quit: Konversation terminated!]
grming has joined #linux-sunxi
ftg has joined #linux-sunxi
apritzel_ has left #linux-sunxi [#linux-sunxi]
apritzel has joined #linux-sunxi
<apritzel> rm_: mmh, but does that really work, without specifying a CPU power supply?
<rm_> current policy: frequency should be within 648 MHz and 1.15 GHz.
<rm_> cpufreq stats: 408 MHz:0.01%, 648 MHz:90.76%, 816 MHz:0.86%, 912 MHz:0.31%, 960 MHz:0.18%, 1.01 GHz:0.10%, 1.06 GHz:0.13%, 1.10 GHz:0.18%, 1.15 GHz:7.48% (14304)
<rm_> well and I ran two instances of "dd if=/dev/zero bs=1M count=40960 | md5sum" for a few minutes, and it didn't lock-up
<rm_> speaking of which, trying to install Debian locked up for me twice
<rm_> at random points, but while installing packages (i.e. load)
<rm_> that's with vanilla kernel without my patch
<apritzel> so if the schematic is right, there is a separate voltage regulator for the CPU cores
<apritzel> it's even an I2C controllable SY8106, and there is already a driver in the tree (used by other Orange Pi boards)
<apritzel> you would need to add this, I am actually somewhat surprised that DVFS worked, if the kernel doesn't know or can control the voltage
<apritzel> rm_: check the Orange Pi PC2 .dts, in the same directory, that should give you an idea what needs to be added
szemzoa_ has joined #linux-sunxi
szemzoa has quit [Remote host closed the connection]
grming has quit [Quit: Konversation terminated!]