Shirasaka-Hazumi has quit [Ping timeout: 276 seconds]
Shirasaka-Hazumi has joined #linux-sunxi
ninolein has quit [Ping timeout: 258 seconds]
ninolein has joined #linux-sunxi
cnxsoft has joined #linux-sunxi
tkaiser has joined #linux-sunxi
tkaiser has quit [Ping timeout: 250 seconds]
doppo has quit [Ping timeout: 250 seconds]
doppo has joined #linux-sunxi
Net147 has quit [Ping timeout: 276 seconds]
Net147 has joined #linux-sunxi
doppo has quit [Remote host closed the connection]
doppo has joined #linux-sunxi
doppo has quit [Ping timeout: 264 seconds]
doppo has joined #linux-sunxi
tkaiser has joined #linux-sunxi
Shirasaka-Hazumi has quit [Ping timeout: 244 seconds]
tkaiser has quit [Ping timeout: 244 seconds]
Shirasaka-Hazumi has joined #linux-sunxi
pekka10 has quit [Ping timeout: 260 seconds]
pekka10 has joined #linux-sunxi
vagrantc has joined #linux-sunxi
tkaiser has joined #linux-sunxi
tkaiser has quit [Ping timeout: 264 seconds]
klynastor has joined #linux-sunxi
iamfrankenstein1 has joined #linux-sunxi
iamfrankenstein1 has quit [Ping timeout: 240 seconds]
tkaiser has joined #linux-sunxi
IgorPec7 has joined #linux-sunxi
vagrantc has quit [Ping timeout: 250 seconds]
<klynastor>
Hi guys, new Pine64 user here. I'm trying out the 4.7 kernel using apritzel's "a64-v5i" branch. Is there a quick fix to get USB working like in "a64-wip"? I patched in the 2 USB patches from "a64-wip" but get this error on bootup: sun4i-usb-phy 1c1b810.phy: Err requesting id-det-irq: -22 / sun4i-usb-phy: probe of 1c1b810.phy failed with error -22
<klynastor>
(I meant a64-v5 branch)
tkaiser has quit [Ping timeout: 260 seconds]
|Jeroen| has joined #linux-sunxi
<|Jeroen|>
Hi!
oneinsect has joined #linux-sunxi
vagrantc has joined #linux-sunxi
<|Jeroen|>
any idea with deps this are ? warning: (ARCH_SUNXI) selects PINCTRL_SUN50I_A64 which has unmet direct dependencies (PINCTRL && ARCH_SUNXI)
zuikis has joined #linux-sunxi
IgorPec7 has quit [Ping timeout: 250 seconds]
tkaiser has joined #linux-sunxi
IgorPec has joined #linux-sunxi
tkaiser has quit [Ping timeout: 244 seconds]
Netlynx has joined #linux-sunxi
Dodger78 has quit [Read error: Connection reset by peer]
<apritzel>
|Jeroen|: and btw: you should use the default load addresses in U-Boot: makes your life easier and prevents subtle errors (though your addresses are OK here)
<klynastor>
apritzel: thanks for answering up above. I too am one of those people who disabled all the other ARCH's from my linux config. I had a problem with PINCTRL not being defined (even though it was a dependency for something else ARCH_SUNXI uses). I'm not well-versed in Kconfig, but I made it into a kernel option so I could turn on PINCTRL just to get it to compile.
<apritzel>
Ntemis: yeah, like this
<apritzel>
Ntemis: Ideally you would "git am" this as jelle said
<apritzel>
I think you have to add "From foo" as the first line, but still "git am" is quite picky
<apritzel>
patch -p1 < file
<apritzel>
is more robust
<apritzel>
but you loose the commit message with this
tkaiser has joined #linux-sunxi
<apritzel>
unfortunately montjoie seems to have removed his branch with those patches from his github
<apritzel>
NiteHawk: good point, you can even download a mbox from there
Igor2 has joined #linux-sunxi
<klynastor>
To anyone looking to strip out everything except ARCH_SUNXI from their defconfig, I put up my sample config (which works) at: http://www.winds.org/pub/pine64/
<Igor2>
hi; I have an orange pi with an USB OTG port; I am trying to configure it as a gadget/device (g_ether or g_serial); on the orange pi it seemingly goes well, but when I plug it in, there's no sign of detection of the other side on either systems
<Igor2>
I also echo 2 to /sys/bus/platform/devices/sunxi_usb_udc/otg_role
<Igor2>
(which causes good things: some kernel messages and lsusb output suggests the OTG port stops being a host port and is turning to a device port)
<Igor2>
any way to debug this?
<tkaiser>
Igor2: You're dealing with 3.4.x and a H3 device? Or the first OPi with A20?
<Igor2>
forgot to mention I use loboris (maybe this why my CPU runs hot)
<Igor2>
random related question: I have a standard usb male A <-> micro usb male cable, I normally use it with a nokia N900 where the nokia is a gadget and PC is the host - should this cable work with the orange PI (gadget) - PC (host) thing?
apritzel has quit [Ping timeout: 244 seconds]
<Igor2>
hah! the USBootPi was a great idea, it's detected by the PC now
Netlynx_ is now known as Netlynx
<Igor2>
so I am sure the cable and the orange PI board are good and I am doing something wrong with the kernel
<Igor2>
unfortunately I still have no idea how to debug this other than trying with the mass storage driver instead of g_serial or g_ether
<Igor2>
no luck
<GeneralStupid>
i told you that i had problems with USB. So i will try armbian based image. they use a newer kernel
vagrantc has joined #linux-sunxi
willmore has quit [Ping timeout: 246 seconds]
<GeneralStupid>
anyone tried a mainline kernel on an allwiner chip?
<GeneralStupid>
i would like to know if it works with hdmi, graphics, vpu and stuff
<jelle>
which chip
<GeneralStupid>
h3
* vagrantc
was running mainline v4.7-rc1 on an orangepi plus 2 and it crashed daily and now the it doesn't boot anymore
<jelle>
GeneralStupid: no video on the mainline I think
<GeneralStupid>
nice
<GeneralStupid>
so thats what "Although it depends on the SoC, the status of mainline kernel is now almost suitable for normal use. " means :D
<vagrantc>
granted, i was running it with builds regularly...
<GeneralStupid>
i will try the 3.4.112 from armbian.
<GeneralStupid>
but, its worth a try, you could just measure on you pinheader. I did that with these alligator clips ,. is that the correct translation?
<vagrantc>
i've got two others that have been running fine for months
<GeneralStupid>
ok then, try downclocking. H3 is not really made to run on 1.6Ghz
<GeneralStupid>
"It's rated to run at 1.296GHz, anything higher is overclocking (and usually overvolting). If you run it without heatsink, fan and proper dvfs settings, you risk overheating. "
<vagrantc>
depends on what continent you're on weather it's alligators or crocodiles
<GeneralStupid>
europe?
<vagrantc>
i guess i could just patch the .dts file to max out at 1.2GHz?
<jelle>
I wonder how reliable that orangepi pc openelec port works
<GeneralStupid>
vagrantc: i did that by i decided to use a userland tool for clocking -.-
Mr__Anderson has quit [Remote host closed the connection]
<vagrantc>
are there typically some /sys values you can check the running clock speed?
<GeneralStupid>
cpufreq-set -g userspace …
<vagrantc>
i couldn't find last i tried ... but maybe mainline needs more patches for that to work
<GeneralStupid>
cpufreq-info
<GeneralStupid>
thats what i use
<Amit_t_>
While compiling pine64 kernel , I am seeing this message make: *** No rule to make target `zImage'. Stop.
<Amit_t_>
I can build only Image file
<Amit_t_>
not the zImage or uImage
<vagrantc>
Amit_t_: i think you need Image in order to boot
<Amit_t_>
I need a uImage or zImage
Mr__Anderson has joined #linux-sunxi
<vagrantc>
Amit_t_: i haven't seen uImage or zImage used on arm64 platforms ... but admittedly, i've only seen 3
Archaon has joined #linux-sunxi
<Amit_t_>
vagrantc: with Image file what u-boot command I should , bootm is not working
<Amit_t_>
*I should use
<vagrantc>
GeneralStupid: with the stable opi2 boards i've got, cpufreq doesn't know anything about the running kernel
<vagrantc>
Amit_t_: booti
<vagrantc>
GeneralStupid: no or unknown cpufreq driver is active on this CPU
<Archaon>
Hi together, I'am trying to boot a mainline 4.6 kernel on the Cubieboard 4 I've got u-boot compiled with ftd support and try to boot the kernel with: bootm 0x
<Amit_t_>
vagrantc: I need to pass DTB also
<Archaon>
<kernel_addr> - <dtb_addr>
<vagrantc>
Amit_t_: help booti
<Amit_t_>
would booti 0x42000000 - 0x43000000 works for pine64?
<Amit_t_>
I mean are these address OK ?
<Archaon>
the last line on the serial uart console is: Uncompressing Linux.... done, booting the kernel.
<vagrantc>
if the u-boot support is done well, booti $kernel_addr_r - $fdt_addr_r
<GeneralStupid>
vagrantc: hmmm -.-
<GeneralStupid>
i think mainline iss no option for me :D sounds a bit buggy
<vagrantc>
GeneralStupid: mainline is a work in progress
<vagrantc>
GeneralStupid: those are running 4.4-rc6 kernels with some patches ssvb pointed me to
<vagrantc>
given how my experiments with 4.7-rc* have gone, i didn't want to upgrade those...
<vagrantc>
hopefully it's just a power supply issue
<Amit_t_>
vagrantc: Thanks I booted just stuck at filesystem
<Amit_t_>
vagrantc: Just stuck at this plymouth, [ 3.440292] init: ureadahead main process (959) terminated with status 5
<Igor2>
is there a way to download a recent 3.4.11x-sun8i source? I mean just download, not git clone something then apply 10 patches, just a simple tarball
lamer14656604926 has joined #linux-sunxi
<Archaon>
I now get some output from the kernel on cubieboard 4 with mainline 4.6 after adding earlyprintk to bootargs. Now botting fails with "Failed to allocate 0x2000 bytes below 0x0". Does someone know howto progress from here?
tkaiser has quit [Ping timeout: 252 seconds]
<oliv3r>
ssvb: i'm looking through our "documented" brom dump, and i'm wondering if the booted_from_media stuff was actually inteded to be at 0x28 or if it was a side-effect from something
<oliv3r>
ssvb: but i guess (will check when i have hardware and time, so monday i guess as I have some nand and mmc2 booted boards, that SUNXI_BOOTED_FROM_NAND 1 and SUNXI_BOOTED_FROM_MMC2 is 2
cptG_ has joined #linux-sunxi
<ssvb>
oliv3r: it's unlikely to be a side effect, because it's the only byte that gets modified in SRAM *after* the SPL is loaded, and this behavior seems to be the same on all Allwinner chips, though I have only tested it on A10, A13, H3 and A64 so far
cptG has quit [Ping timeout: 252 seconds]
<ssvb>
oliv3r: I haven't looked at the disassembly of the BROM code yet, but this is a indeed a good idea
<Igor2>
tkaiser, sorry, I meant a source tarball, not binaries - I'd like to configure & compile the kernel
<Igor2>
(normally I just download one from kernel.org, but reading armbian wiki pages I think I shouldn't use vanilla mainline on H3)
<oneinsect>
here the pastebin ...if can someone guide me...i use an armbian kernel and mainline u-boot...the thing it reads boot.scr but says file not found however when manually loaded it works
<oneinsect>
says Unable to read file /vmlinuz....however it works if manually load it ...load mmc 0:1 0x45000000 /initramfs-sunxi-new
<oneinsect>
anyone?
<tkaiser>
Igor2: Nope, use Armbian's build system, put your kernel config in 'userpatches' dir and let a kernel build. We currently have 113 patches on top of the latest BSP legacy kernel. I wouldn't miss them ;)
<apritzel>
Amit_t_: there is no zImage for arm64, as there is no kernel decompression
<ssvb>
so I wonder if this boot device identifier was something that was only introduced later in some BROM code update in more recent revisions of the A10 SoC?
<ssvb>
we do have some A10 revision checks in http://git.denx.de/?p=u-boot.git;a=blob;f=arch/arm/mach-sunxi/dram_sun4i.c;h=f7b4915037cd3eec58d7b45cff636133945d28f8;hb=HEAD#l72 after all
<ssvb>
does anyone have some really ancient Allwinner A10 device?
bmeneg has quit [Ping timeout: 276 seconds]
bmeneg has joined #linux-sunxi
<ssvb>
NiteHawk: thanks, this code snippet also looks somewhat interesting, however it shows that the boot0 code for some reason *writes* these identifiers to the header instead of checking them
<ssvb>
NiteHawk: is it some sort of a trampoline to chain load another boot0 from a different boot media?
allwinner has joined #linux-sunxi
<allwinner>
just wanted to know the progress of pine64 mainline kernel
<ssvb>
no, the mainline kernel will not include blobs (in fact the mali blobs are only in the userland)
<ssvb>
but you can always compile the mali kernel driver yourself and use it with the userland mali blobs
<NiteHawk>
ssvb: i haven't quite gotten the overall picture of AW's github code yet there... but isn't writing to the boot file header (that how i interpret that "bfh->") what you observe? oh. just realized i may be wrong here - this is boot0 code, and we see this write straight from the BROM already, right?
<ssvb>
NiteHawk: yes
<allwinner>
ok i m seeing that not many functionalities are available, so my only option is to use the current 3.10 kernel with full functionality - mali driver
<allwinner>
so just wanted to know if there is any guide to build mali myself on 3.10 kernel?
<ssvb>
allwinner: frankly speaking, mali adds very little practical value, unless you are into playing 3d games
<allwinner>
when i tried to watch a video it is not watchable
<ssvb>
you don't need mali for hardware accelerated video decoding
<oneinsect>
i think sometimes mmc could be slow? may be its not fast enough?
<oneinsect>
hence not found errors?
<oneinsect>
but that should not be the case ....
<KotCzarny>
i have 2 (2gb) cards that refuse to boot at all
<oneinsect>
strange
<KotCzarny>
but 1gb and 4gb and samsung evo+ ones work without error
<ssvb>
oliv3r: BTW, for the boot media code 0x2, the A64 BROM also stores some additional information in the high 4 bits (maybe that's MMC2 boot and some extra differentiation for handling eMMC boot partitions?)