domidumont has quit [Remote host closed the connection]
domidumont has joined #linux-sunxi
diego_r has joined #linux-sunxi
diego_r has quit [Read error: Connection reset by peer]
bwarff has quit [Ping timeout: 248 seconds]
<wens>
plaes tries to keep the WIP list updated
ricardocrudo has joined #linux-sunxi
cnxsoft has quit [Remote host closed the connection]
cnxsoft has joined #linux-sunxi
<lennyraposo>
last question of the night for me
<lennyraposo>
need to sleep
afaerber has joined #linux-sunxi
<lennyraposo>
how often is the wiki page updated?
mzki has joined #linux-sunxi
<plaes>
lennyraposo: depends..
<plaes>
I mostly update based on patch submissions to linux-sunxi list
<plaes>
but some people skip that list
<plaes>
and I also don't know the non-submission cases
apritzel has joined #linux-sunxi
<plaes>
apritzel: o/
<plaes>
do you know who's working on the A64 USB?
<apritzel>
plaes: sadly: no
<apritzel>
I was trying a bit lately
<apritzel>
it seems like it needs two reset lines triggered at once, which the current driver does not support
FDCX has joined #linux-sunxi
<plaes>
ah.. this is what Hans is working on...
<plaes>
lennyrapos here was interested in USB stuff
<wens>
great... the weird design :(
<wens>
the shared reset stuff is pending
<apritzel>
I had Hans patches applied, but still it didn't work
Nyuutwo has quit [Ping timeout: 240 seconds]
<apritzel>
the driver initialized properly, but plugging in anything didn't change a thing
<apritzel>
I think there are _two_ places where it needed two reset lines, Hans patches covering only one
cptG_ has joined #linux-sunxi
Nyuutwo has joined #linux-sunxi
<plaes>
aww
<plaes>
wasn't his patches for H3 too?
cptG has quit [Ping timeout: 276 seconds]
<apritzel>
H3 differs USB-wise from the A64
<apritzel>
so it's no real help
tomboy65 has joined #linux-sunxi
<plaes>
yeah.. "At least the EHCI/OHCI found on the Allwinnner H3 SoC needs multiple
<plaes>
...
<apritzel>
that is the one place I mentioned above
tomboy64 has quit [Ping timeout: 260 seconds]
<apritzel>
but the A64 has a second variety
* plaes
update Mainlining Effort page
poffs has quit [Remote host closed the connection]
poffs has joined #linux-sunxi
IgorPec has quit [Ping timeout: 268 seconds]
Net147 has quit [Ping timeout: 244 seconds]
Net147 has joined #linux-sunxi
cajg has quit [Quit: flingers xed]
cajg has joined #linux-sunxi
domidumont1 has joined #linux-sunxi
domidumont has quit [Read error: Connection reset by peer]
domidumont1 has quit [Quit: Leaving.]
domidumont has joined #linux-sunxi
vishnup has joined #linux-sunxi
yann|work has joined #linux-sunxi
IgorPec has joined #linux-sunxi
cajg has quit [Quit: no plugins]
cajg has joined #linux-sunxi
IgorPec has quit [Ping timeout: 240 seconds]
<matthias_bgg>
apritzel, do you know which uboot is best to use, if you want to avoid the crazy partition scheme from the bsp uboot?
<matthias_bgg>
is it yours or the one from longsleep?
<apritzel>
matthias_bgg: I think both are avoiding that
<apritzel>
if you are looking for most enabled features, I'd recommend longsleep's for the moment
<apritzel>
not so much for U-Boot features (I think upstream is better here)
<apritzel>
but since that enables the PMIC properly, you have more fun with your Linux afterwards
<apritzel>
I will push some new patches and a new image later this week
<apritzel>
that will give you bootz and booti support, for loading 32-bit and 64-bit kernels
IgorPec has joined #linux-sunxi
<wens>
montjoie: sun8i-emac driver on a83 seems to be working, though it is too noisy to use for now
bwarff has joined #linux-sunxi
<matthias_bgg>
apritzel, ok, thanks. from a legal perspective your u-boot is critical to distribute, as it uses libdram without providing the sources, right?
<wens>
no word on the sources yet?
<apritzel>
Allwinner told me this week ...
<apritzel>
which is only half way through, so there is still hope
<wens>
half way through?
p1u3sch1 has quit [Ping timeout: 264 seconds]
<apritzel>
wens: I got an email last week stating that they would release the information "next week"
<apritzel>
matthias_bgg: I build upstream U-Boot with ssvb's patches, but don't use the SPL
<wens>
oh
<apritzel>
matthias_bgg: instead I use their boot0, which we allowed to distribute
<matthias_bgg>
apritzel, but that is not your a64 branch in github, right?
<apritzel>
matthias_bgg: not in the moment, yes
<apritzel>
matthias_bgg: but this just involves to move the text offset by 1.5 KB to make room for the header that boot0 requires
<apritzel>
matthias_bgg: what is your time frame? Can you wait till the weekend?
IgorPec has joined #linux-sunxi
bwarff has quit [Ping timeout: 250 seconds]
libv_ has joined #linux-sunxi
libv has quit [Ping timeout: 246 seconds]
<matthias_bgg>
apritzel, I would prefer to get that up and working asap. I'm not really into uboot, but from what you explained, it seems to me as if I had to add 1.5 KB to CONFIG_SPL_BSS_START_ADDR
<matthias_bgg>
and revert the patch that adds libdram
<apritzel>
matthias_bgg: not SPL, actually you build without SPL at all
tomboy65 has quit [Ping timeout: 260 seconds]
<apritzel>
it's CONFIG_SYS_TEXT_BASE
Hahaburger has joined #linux-sunxi
<apritzel>
I usually build with SPL, but ignore the generated file and use u-boot-dtb.bin
<apritzel>
(I need the SPL version still for the FEL boot)
tomboy65 has joined #linux-sunxi
<matthias_bgg>
apritzel, yes I was looking on the wrong linker script. Ok, I will give it a shot. If a achieve anything, I will let you know :)
<matthias_bgg>
thanks for the help!
libv_ is now known as libv
Hahaburger has quit [Quit: Page closed]
<montjoie>
wens: thanks for the test:)
<montjoie>
wens: do you have modified something for make it works ?
<montjoie>
since you said I forgot to backport your RGMII modification
<wens>
montjoie: i'll update my branch a bit later
<montjoie>
no hurry, I need to first found why dmesg stuck the ssh connection (packet sent with bad length)
Keziolio has quit [Remote host closed the connection]
susan33 has joined #linux-sunxi
<susan33>
hi ; the manual build howto (http://linux-sunxi.org/Manual_build_howto) refers to A10, A13 and A20 and at the bottom, some others. The pine64 is based on a soc A64. The question is : is there a howto for this soc ?
<lennyraposo>
you need lonsleeps
<lennyraposo>
longsleeps current
<lennyraposo>
which relies on adeal of the 3.10.65 BSP 1.2 kernel from Allwinner
<susan33>
lennyraposo: is it an answer to me ?
<lennyraposo>
yes Susan33
<lennyraposo>
;)
<lennyraposo>
you need longsleeps u-boot and pine64 linux repos
<lennyraposo>
for the time being
<lennyraposo>
the pine64 is a WIP
ricardocrudo has joined #linux-sunxi
<lennyraposo>
from what I gather tus far (I'm new myself) is that mainline kernel needs development in the areas of USB/Ethernet
<lennyraposo>
and U-Boot requires the boot0 sources from allwinner
<lennyraposo>
but someone here is near completion on the ethernet driver (MAC) for the a64
<susan33>
do you have an url to these repo please ?
<apritzel>
I feel someone should put this info into the Wiki
<lennyraposo>
both pine and sunxi wikis?
<apritzel>
latter only so far
<lennyraposo>
I can do that
<apritzel>
I think the template has sections for both BSP based and upstream versions of U-Boot and kernel
<susan33>
nice. i find somewhere in my reads in howto that there were something done with buildroot. Is it the case for these repos too ?
<susan33>
(my distribution is based on buildroot)
<apritzel>
susan33: those repos are not so much about distributions, really
<apritzel>
it's the kernel and U-Boot (read: firmware)
<lennyraposo>
are you looking for an image?
<apritzel>
for the userland you can use virtually any arm64 distribution, as long as you use the proper kernel
<susan33>
apritzel: yes, i got this points. But my aim is to put them into packages of buildroot so that buildroot checkout this kernel and uboot and compile them
<apritzel>
AFAIK the compilation is a bit involved, since it's based on Allwinner sources and their build system
<apritzel>
don't know how this works together with buildroot
<apritzel>
we need information from Allwinner to get a proper U-Boot port
Keziolio has joined #linux-sunxi
<apritzel>
for the time being we are stuck with their boot0 firmware blob
<apritzel>
I think somewhere in longsleep's repo there should be information on how to create an image
<susan33>
ok. This is included into the longsleep repository or where should i take it ?
<susan33>
ok nice. i'm going to read all that.
<susan33>
my understand is that : with an sd card, and these 2 repo, and my rootfs, i should be able to boot something. i hope i'm right.
<apritzel>
somehow: yes ;-)
<apritzel>
but if your concern is mostly your rootfs, I suggest you just go with some image and add your rootfs to it
<apritzel>
building Allwinner's crap is no fun, really
<lennyraposo>
and from what someone told me the wrong way to go about it ;)
<susan33>
i've tried to take an image, but i didn't understand how to do after : i dd the img on the sd card : and after that, partition seemed messy
<apritzel>
susan33: welcome to Allwinner ;-)
<lennyraposo>
hehe
<apritzel>
this is some weird Android partitioning scheme, which is only half way legal
<apritzel>
some partition tools indeed refuse to work with it
<lennyraposo>
longsleep cleaned up for what it's worth
<lennyraposo>
even backported some u-boot aspects in allwinners u-boot rendtion
* apritzel
sees some strong indication that I should push an updated firmware image with upstream u-boot ;-)
<lennyraposo>
boot0 source would go a long way
<lennyraposo>
then ATF
<susan33>
apritzel: so, from an ubuntu computer, is there a way to copy my rootfs to the sd card on the partition 10 (as far as i've read, it's partition 10) ?
<apritzel>
oh, are you using my original image then?
<lennyraposo>
didn't umiddle create a bunch of script for creating debina ubuntu rootfs?
<apritzel>
susan33: so you have a rootfs ready?
<susan33>
susan33: no, not yet, in fact i don't know exactly what to change to make it work : i've a buildroot distribution. i guess i've to change the target to arm8, but i don't understand how i will be able to put things like the linux modules ;-(
<apritzel>
the target name is aarch64, I think
<susan33>
however, i've a buildroot (so a tool to build a rootfs) ready for rpi0,1, 2, and 3. my understanding is that i've just to change the target processor (but i'm still learning about that kind of things)
<apritzel>
don't know much about buildroot, but at least it seems to support arm64 (aka aarch64)
<apritzel>
susan33: what does buildroot actually create: does it dump all files into a directory?
<susan33>
yes, i don't think that making the rootfs (except the linux driver) is a problem. my problem is more the boot.
<susan33>
so, i take your image, and how to you actually copy the rootfs on the sd ? while the partition schema is not recognized.
<apritzel>
your Linux should see the partitions, at least
<apritzel>
lsblk or cat /proc/partition
<apritzel>
so you make a filesystem on sdx10: mke2fs -t ext4 /dev/sdx10
<apritzel>
(replace x with the proper letter your SD cards get)
<susan33>
i've not it tonight unfortunatly, but i remember that gparted didn't see the same things as /dev/mmcblk0p* ; i'll try this week end when i'll be back at home.
<susan33>
oh, in fact, sdx10 doesn't exist, i've to create it. ok.
<apritzel>
mmcblk0p is also possible, sdx is for USB based card readers
<susan33>
thanks
<apritzel>
wait
<apritzel>
you should check the output of cat /proc/partitions, somewhere there should be something with 10 partitions, presumably its mmcblk0p<n>
<apritzel>
yes, unfortunately it's a bit dated now
<apritzel>
so say you found the right partition (presumably it's /dev/mmcblk0p10 in your case):
<susan33>
ok, any problem. my aim is only to boot my rootfs for the moment. my target is to include the compilation process into buildroot, but as far as i understand ,for the last point, i should wait while it's moving regularly
<apritzel>
yes, it's quite in flux at the moment
<apritzel>
I will probably release a much saner base image soon
<susan33>
apritzel: a last point : who are you from the pine64 point of view ?
<apritzel>
do you mean whether I am connected to the Pine64 organisation?
<susan33>
yes
<apritzel>
no, I am just a concerned citizen ;-)
<apritzel>
but I push the Pine64 upstream kernel support
<susan33>
ok, nice. Thanks a lot for you work and your help.