Turl changed the topic of #linux-sunxi to: Allwinner/sunxi development discussion - Don't ask to ask. Just ask and wait! - See http://linux-sunxi.org | https://github.com/linux-sunxi/ | Logs at http://irclog.whitequark.org/linux-sunxi
nabblet has quit [Quit: leaving]
FDCX has quit [Ping timeout: 240 seconds]
FDCX has joined #linux-sunxi
kuldeepdhaka has joined #linux-sunxi
xinj has joined #linux-sunxi
mmarker has joined #linux-sunxi
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
seere has joined #linux-sunxi
seere_ has quit [Ping timeout: 252 seconds]
xinj has quit [Read error: Connection reset by peer]
pseudomind has joined #linux-sunxi
xinj has joined #linux-sunxi
xinj has quit [Quit: xinj]
xinj has joined #linux-sunxi
wingrime1 has joined #linux-sunxi
hipboi_ has quit [Read error: Connection reset by peer]
hipboi_ has joined #linux-sunxi
ganbold_ has quit [Ping timeout: 276 seconds]
TheSeven has quit [Ping timeout: 276 seconds]
TheSeven has joined #linux-sunxi
xavia has left #linux-sunxi ["PART #interlock :PART #arduino :PART #wnydev :PART #buffalolab :PART #docker :PART #symfony :PART #virtuvia :QUIT :Leaving."]
kuldeepdhaka has quit [Ping timeout: 252 seconds]
kuldeepdhaka has joined #linux-sunxi
pseudomind has quit [Quit: pseudomind]
pseudomind has joined #linux-sunxi
pseudomind has quit [Quit: pseudomind]
kuldeepdhaka has quit [Ping timeout: 245 seconds]
kuldeepdhaka has joined #linux-sunxi
ganbold_ has joined #linux-sunxi
ganbold_ has quit [Ping timeout: 276 seconds]
deffrag has joined #linux-sunxi
FreezingCold has quit [Ping timeout: 250 seconds]
discopig has quit [Quit: Bye]
discopig has joined #linux-sunxi
pseudomind has joined #linux-sunxi
pseudomind has quit [Quit: pseudomind]
pseudomind has joined #linux-sunxi
deffrag_ has joined #linux-sunxi
deffrag has quit [Ping timeout: 245 seconds]
pseudomind has quit [Client Quit]
pseudomind has joined #linux-sunxi
pseudomind has quit [Client Quit]
paulk-aldrin has joined #linux-sunxi
Quarx has joined #linux-sunxi
Superpelican has joined #linux-sunxi
pseudomind has joined #linux-sunxi
pseudomind has quit [Client Quit]
<Superpelican> buZz: Do you happen to know if Ubuntu kernels have special requirements?
<Superpelican> I mean are there any options that I should enable or disable in while configuring with make menuconfig for Ubuntu?
<Superpelican> buZz: I was already installing Owncloud etc. when I wanted to enable the firewall, turns out my custom kernel doesn't have iptables support :(
<hramrach> well, iptables are not special. they are standard and obviously required for firewalling
<hramrach> unlike Fedora Ubuntu does not require overly special kernels
<Superpelican> yes, but I meant that I now have to compile and configure a new kernel
<Superpelican> but before I can do that I have to know if Ubuntu has any special requirements regarding kernels ;)
<Superpelican> My current custom kernel was originally configured and compiled for Mer
<hramrach> then it will likely work
<Superpelican> And I remember having to manually set some options in the .config to be compatible with Mer
<Superpelican> Does Ubuntu need any similar things?
<Superpelican> yes
<hramrach> I never had problems with booting Ubuntu with custom kernel
<Superpelican> but the problem is, my custom Mer kernel doesn't have iptables support ;)
<hramrach> if you leave out iptables you cannot do firewalling but Ubuntu still boots
<Superpelican> :nod:
<Superpelican> but I'm going to run a web server
<Superpelican> so I'd prefer to have a firewall
<hramrach> then add the iptables options and you will
<hramrach> still you can run a web server fine without one
<Superpelican> Yes, but how safe is that?
<hramrach> you want the web server accessible so what use is firewall, anyway
<Superpelican> well...
<Superpelican> I don't want every single port to be accessible
<hramrach> if you don't open them they aren't
<Superpelican> hramrach:And did you base your custom Ubuntu kernels on the default Debian/Ubuntu config, or a vanilla Linux config?
<Superpelican> I could use the base Ubuntu config and modify it
<hramrach> linux-sunxi config, vanilla config, Debian config w/e
<Superpelican> But then I can't use the linux-sunxi config :(
<Superpelican> If I could merge them somehow
<hramrach> it's not worth the effoer
<hramrach> effort
<Superpelican> hramrach: So you think the linux-sunxi config will suffice for Ubuntu?
<hramrach> most of the Ubuntu config is drivers and those are for different arch
<Superpelican> ok
<hramrach> yes. It should suffice. Some features might not work - eg. it does not have firewall either
<Superpelican> then I'll just base my config on the linux-sunxi config
<hramrach> but you should be able to boot using that
<Superpelican> And while I am at it, I might as well switch to 3.4 :)
<Superpelican> My custom Mer kernel was built last summer
<hramrach> or 3.10
<Superpelican> when it was still advised to use 3.0
<Superpelican> How stable is 3.10?
<hramrach> when you don't need screen it's probably better choice
<Superpelican> So the framebuffer/console won't work either when using 3.10?
<hramrach> yes, that's a no-go. But you get serial, network, mmc, sata, usb host and newer Linux architecture overall
<hramrach> unfortunately the nand does not work yet
<hramrach> and the OTG usb is very experimental if it's working at all
<Superpelican> oh
<Superpelican> if OTG USB doesn't work
<Superpelican> then I'll stick to 3.4
<Superpelican> it's a tablet, so I don't have SATA
<hramrach> it that so important?
<Superpelican> :)
<Superpelican> so I'm dependent on USB OTG for storage
<hramrach> well, if it has only the OTG port then it sucks
<Superpelican> Why?
<Superpelican> It's only for personal use, so I don't need extreme bandwidth
<Superpelican> ;)
<hramrach> the host ports would work but they are not wired on the device
<Superpelican> you mean the SATA ports?
<hramrach> no, the other two USB ports
<Superpelican> BTW can I also edit configuration options that haven't changed with the new version when using oldconfig?
<Superpelican> like say, iptables support ;)
<hramrach> not with oldconfig
<Superpelican> Can you also edit .configs with defconfig?
<hramrach> but with config or menuconfig or editing the config with a text editor
<hramrach> no defconfig replaces it
<Superpelican> So menuconfig can also edit existing .configs?
<hramrach> yes, and so can any text editor. Anything you delete oldconfig will ask again
<Superpelican> ok
<Superpelican> then I'll just run oldconfig first and then edit with menuconfig
<hramrach> deleting stuff with a text editor and then doing oldconfig is probably one of the fastest ways to configure ;-)
bertrik has joined #linux-sunxi
bertrik has quit [Changing host]
bertrik has joined #linux-sunxi
wingrime has quit [Ping timeout: 258 seconds]
wingrime1 has quit [Ping timeout: 240 seconds]
<Superpelican> hramrach: Is btrfs safe on 3.4?
<Superpelican> Or is the BTRFS version independent of the kernel version?
<hramrach> I have no idea
<hramrach> the usual fs to use is ext4
Superpelican has quit [Ping timeout: 240 seconds]
Superpelican has joined #linux-sunxi
<Superpelican> hramrach: Do you happen know what system call auditing support exactly is?
<Superpelican> I don't think it is necessary for me
<Superpelican> Should I enable or disable "enable sunxi_is_foo() even for other mach-sunNi (SUNXI_MULTIPLATFORM)"?
wingrime1 has joined #linux-sunxi
<Superpelican> hmm it seems to be an option that the kernel will also be enable to run on other sunXi platforms
<Superpelican> and not only on sun4i
<Superpelican> As I will only be running my kernel on 1 device I'll just disable it
<Superpelican> torindel: Should I enable Architected timer support (ARM_ARCH_TIMER) ?
ganbold_ has joined #linux-sunxi
<Superpelican> nvm if it's just a support option, I'll just enable it ;)
rz2k has joined #linux-sunxi
deasy has joined #linux-sunxi
<Superpelican> hramrach: What should I set as default kernel command string?
<Superpelican> hmm I think I'll go with 0x40008000 as entry point, because the Tiny Core uImage (which I was able to successfully boot on my tablet) used that as entry point/load address
Superpelican has quit [Quit: Page closed]
wingrime has joined #linux-sunxi
wingrime has quit [Ping timeout: 240 seconds]
wingrime1 has quit [Ping timeout: 240 seconds]
kuldeepdhaka has quit [Ping timeout: 240 seconds]
wingrime1 has joined #linux-sunxi
Superpelican has joined #linux-sunxi
<Superpelican> hramrach: I followed this http://linux-sunxi.org/Linux_Kernel, but there are this no uImage in ach/arm/boot and there are also no modules in output/lib/modules/3.4.79/
<Superpelican> hramrach: also I'm getting lots of cp: cannot stat '': no such file or directory errors
<Superpelican> hramrach: http://pastebin.com/mYyxMzM2
wingrime has joined #linux-sunxi
deffrag__ has joined #linux-sunxi
deffrag_ has quit [Ping timeout: 240 seconds]
<hramrach> then something went wrong like you are out of disk space or something
<libv> Superpelican: that addition that you made to the inet_1 page... why did you add this like this, and why did you not edit the manual build section?
<Superpelican> libv: Well I can change it if you want, don't exactly understand what you mean though with manual build section
<Superpelican> libv: BTW how would I attach files, like say the script.bin and script.fex?
<Superpelican> hramrach: I definitely have enough disk space
<libv> Superpelican: that is mentioned in the new device howto
<Superpelican> hramrach: I checked again and it seems to be OK that there are no modules, as all are built-in
<Superpelican> hramrach: But I don't understand why there is no uImage
<Superpelican> libv: ok will check
<Superpelican> hramrach: there are only the dirs bootp, compressed and dts in arch/arm/boot and the files an install script (install.sh) and the Makefile
bertrik_ has joined #linux-sunxi
bertrik_ has quit [Changing host]
bertrik_ has joined #linux-sunxi
bertrik has quit [Ping timeout: 252 seconds]
<Superpelican> buZz: Do you have any experience with the above?
<buZz> lets see
<buZz> not sure which part you are refering to ..
<buZz> so many questions ;)
<Superpelican> buZz: That make'ing hasn't resulted in a uImage ;)
<hramrach> Superpelican: uImage may be missong if you don't have u-boot tools
<buZz> did you do 'make uImage' ?
<buZz> and that
<Superpelican> I followed http://linux-sunxi.org/Linux_Kernel
<Superpelican> so yes
<Superpelican> first make -j4 ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- uImage modules and then make ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- INSTALL_MOD_PATH=output modules_install
<buZz> i havent crosscompiled for sunxi
<buZz> i just do it all native
<hramrach> it works for me
<Superpelican> On device or using qemu?
<buZz> on device
<Superpelican> My previous custom kernel was compiled using qemu (Mer SDK)
<hramrach> but some cross-toolchains are broken and you will get file not found errors for gcc binaries
<hramrach> pastebin blocks me so I cannot see the error you get
<Superpelican> ok, I'll use another paste service then
<Superpelican> BTW I do have the u-boot-tools package installed
<Superpelican> (and can use the mkimage command)
<Superpelican> but I don't have the "u-boot" package installed
<Superpelican> (Ubuntu 13.10)
<buZz> its just so supereasy to fuck up crosstoolchains :/
<buZz> too much effort imho, when native device can do it just fine (just not so fast)
<Superpelican> buZz: I just installed with the package manager :)
<Superpelican> hramrach: http://pastie.org/9095117
<Superpelican> so the crosscompile toolchain should be fine
<buZz> cp: cannot stat ‘drivers/ata/sw_ahci_platform.ko’: No such file or directory
<buZz> compile failed?
<Superpelican> but later on it does say INSTALL for all those .ko's
<buZz> no
<buZz> its reversed
<buZz> it says INSTALL -first-
<buZz> and -then- fails
<Superpelican> buZz: Can I just make' again with the same source tree?
<buZz> yep
<Superpelican> ok
<Superpelican> then I'll just do that and hope for the best
<Superpelican> buZz, hramrach: Interesting, make now fails due to a unused variable
<Superpelican> "make: *** [drivers] Error 2"
<buZz> some functions might not have the proper dependancy checking
<buZz> espec. if you handedit the .config
<Superpelican> buZz: I haven't manually edited the .config
<Superpelican> Should I now manually edit the Makefile like suggested in http://forums.gentoo.org/viewtopic-t-934822-start-0.html
<Superpelican> ?
<buZz> ehw, i wouldnt edit makefiles just yet
<Superpelican> What should I do then?
<buZz> i would look at config , maybe you have something enabled that just doesnt compile
wingrime has quit [Ping timeout: 240 seconds]
<Superpelican> buZz: Well the rtl8188eu driver seems to cause lots of multiple definition, first defined here messages
<Superpelican> ...for about all its functions, at least that's what it looks like
<buZz> do you need it?
<Superpelican> well it's an ethernet driver, right?
<Superpelican> My tablet doesn't have ethernet
<buZz> its some networking chip
<Superpelican> But I don't know if WiFi depends on it?
<Superpelican> I do need the RTL8192CU to work
<Superpelican> (WiFi)
<buZz> not sure if they depend on each other, menuconfig should tell you
<torindel> Superpelican: thats wifi too
<Superpelican> yes
<Superpelican> torindel: Was just searching on Google for it, it indeed seems to be a WiFi chip too
<Superpelican> Used in some TP-Link USB solution
<Superpelican> don't think I need it then
<Superpelican> buZz: So I can just run menuconfig again on the same source tree?
<buZz> yes
<torindel> yep
<torindel> if you dont change things too much you dont even need to make clean even
<torindel> -one even
<torindel> so it will continue from where it failed more or less
<Superpelican> torindel: I'll just play it safe and run make clean first ;)
<torindel> copy your .config
<torindel> to be sure
<Superpelican> torindel: What do you mean?
<Superpelican> menuconfig says it has written the configuration to the .config
<torindel> Superpelican: clean afair as i remember didnt remove it, but distclean did ;]
<Superpelican> and then I ran make clean
<Superpelican> torindel: But is my .config still there then after make clean?
<torindel> if it is youre ok
<torindel> just dont overdo with cleaning cause make distclean and make mrproper removes it
<Superpelican> BTW I'm running make uImage modules again :)
<torindel> but makes source like you would extract it
<Superpelican> so far so good
<Superpelican> well make clean reported that it just cleaned some dirs
<Superpelican> like kernel and firmware
<Superpelican> so I don't think it removed the .config
<Superpelican> torindel: Looking at the make output, it seems the .config indeed wasn't removed :)
<torindel> Superpelican: i usually do mrproper when cleaning ;]
<torindel> (and the name is old joke)
<Superpelican> I know
<Superpelican> It's a brand name IIRC
<Superpelican> BTW make just finished :)
<Superpelican> now going to run the install modules output command
<Superpelican> torindel, buZz, hramrach: Still no uImage in arch/arm/boot though :(
<buZz> did you look for errors during the compile?
<Superpelican> no errors during compile
<Superpelican> just some unused variable warnings again
<Superpelican> I do remember having problems with make uImage though with my previous kernel last summer
<Superpelican> IIRC I ended up manually creating a uImage
<Superpelican> first creating a normal image and then converting that to a uImage with u-boot-tools IIRC
<Superpelican> buZz: Should I just run make, then make modules and then make output modules_install?
<buZz> make uImage
<buZz> not make
<buZz> what does 'make output' do?
<Superpelican> buZz: Seems like running make again without cleaning first always results in "make: *** [drivers] Error 2"
<Superpelican> buZz: "make: Nothing to be done for `output'."
<buZz> doesnt exist :P
* buZz afk, bbl
tat has joined #linux-sunxi
<tat> i have a cubietruck with an A20 running sunxi kernel, if don't use my keyboard or mouse for a longer period the hdmi port goes black, if i hit the keyboard id does not turn on again, for a shorter period of time that is not a problem, cause the screen turns back on when i use the keyboard or mouse,
<tat> where could i start looking for that issue, or has anyone come accross that thing bevore
<tat> btw the system is still runnning smoth i can log in with ssh
<buZz> you can disable it turning black from being idle
<buZz> with xset
<buZz> (if you are in X
<buZz> )
<buZz> xset s off
<buZz> i guess
<buZz> or it might need xset -dpms
wingrime_ct has joined #linux-sunxi
<wingrime_ct> ls
<wingrime_ct> ls
lauri__ has joined #linux-sunxi
<wingrime_ct> libv: are you known someone who works on power vr video decoder IP
<wingrime_ct> libv: there is some chance, that we have such IP
lauri__ has quit [Client Quit]
lauri has quit [Quit: leaving]
<torindel> huh: uptime: 21:41:03 up 365 days, 22:35, 1 user, load average: 2.39, 2.69, 2.74 on some of my A10's ^^
lauri has joined #linux-sunxi
<Superpelican> wingrime_ct: IIRC GNU are working on RE'ing the PowerVR driver
<Wizzup> There is any progress on that at all?
<Superpelican> they don't seem to have made a whole lot of progress though...
<Superpelican> wingrime_ct: You should check out http://en.wikipedia.org/wiki/Free_and_open-source_graphics_device_driver
<wingrime_ct> Superpelican: nicks and channels
<Superpelican> #powervr
<Superpelican> I don't know anything about the project though
<wingrime_ct> I should try
<Superpelican> torindel, hramrach: http://pastie.org/9095261
<Superpelican> make failed with that
<buZz> i hope the gpu on A80 gets a foss driver someday
<Superpelican> buZz: http://pastie.org/9095261
<torindel> buZz: yea esp powervr along with mali are most popular gpu's for arm now
<Superpelican> torindel: What about Adreno?
<torindel> Superpelican: theres more powervr/malis out there, would have to google for some comparasion chart ;]
<buZz> Superpelican: which kernel are you compiling anyway?
<Superpelican> buZz: linux-sunxi 3.4(.79)
<buZz> well guess that should work .. maybe that driver is just broken in that kernel
wingrime_ct has quit [Read error: No route to host]
<Superpelican> buZz: but at first I didn't get that error
<Superpelican> buZz: ...and it's rtlwifi, I need that
<Superpelican> buZz: LOL, make clean and make'ing again 'solved' the problem
<torindel> Superpelican: mrproper ftw ^^
geecko has joined #linux-sunxi
<Superpelican> torindel: I just ran "make clean" no make mrproper
<torindel> Superpelican: trying to continue after changing things is fine as long as it wouldnt change autogenerated headers or build options ;]
<Superpelican> torindel, hramrach, buZz: still no image in arch/arm/boot though
<torindel> since not always things are rebuilt after changing headers like your problem here
wingrime has joined #linux-sunxi
<Superpelican> torindel, hramrach, buZz: This time I just ran "make", then "make modules" and finally "make INSTALL_MOD_PATH=output modules_install"
<Superpelican> this time there are .ko's in output/lib/modules though
<Superpelican> but still no image
<torindel> Superpelican: do you have any of those files: ./vmlinux arch/arm/boot/zImage arch/arm/boot/Image arch/arm/boot/compressed/vmlinux ?
<Superpelican> torindel: no
<Superpelican> no vmlinux, no zImage and no Image
<Superpelican> nothing
<torindel> run make again and pastebin output somewhere (you dont need clean since you havent changed anything)
<torindel> and Image should be rebuilt anyway
<torindel> or you can just do make menuconfig and exit
<torindel> it will change kernel built number
<torindel> build*
<Superpelican> torindel: make'ing without running make clean first results in the driver 2 error
<torindel> o_O
<buZz> are you even compiling for ARM ;)
<buZz> did you pastebin the .config yet?
<Superpelican> "/home/jurre/CherryServer/linux-sunxi/arch/arm/boot/"
<Superpelican> oops
<Superpelican> "make -j4 ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf-" <- this is what I meant
<torindel> Superpelican: i compile natively on my build cluster ;]
<Superpelican> You actually have a build cluster :O
<Superpelican> only have Core i3 M
<torindel> i prefer not to crosscompile as its error prone
<Superpelican> how fast is crosscompiling on an A10?
<torindel> ends up in either: software build systems assuming something from build host, generating bad headers, using headers from host system instead of arm ones etc
<Superpelican> I could try with the Mer SDK again
<torindel> Superpelican: i dont crosscompile, i compile natively ^^
<Superpelican> :nod:
<Superpelican> but Mer SDK can use scratchbox2 (qemu)
<torindel> Superpelican: scratchbox is better than crosscompiling yourself but you still wont have things exactly as you want ;] (and yes i toyed with it its old maemo thing)
<Superpelican> torindel: maemo uses scratchbox(1), Mer uses scratchbox2
<torindel> Superpelican: yes buts its basicly same basis does it not? qemu+compilator+modifying build system of packages enough to not assume too much from build host
<Superpelican> torindel: Ok, so should I try the Mer SDK?
jemk has joined #linux-sunxi
<torindel> doubt it will help you with kernel
<torindel> if you're crosscompiling it correctly
wingrime-ct has joined #linux-sunxi
<wingrime-ct> jemk:
<wingrime-ct> [vdpau] EOSD: putbits failed: An invalid handle value was provided.
<wingrime-ct> [vdpau] EOSD: error when creating surface: An invalid/unsupported VdpRGBAFormat value was supplied.
<wingrime-ct> jemk: seems mplayer's ssa mode are bad
<wingrime-ct> jemk: err, I mean -saa
xavia has joined #linux-sunxi
nicksydney has quit [Remote host closed the connection]
nicksydney has joined #linux-sunxi
<hramrach> Superpelican: what arch are you building for/
<hramrach> and what crosstoolchain?
<Superpelican> hramrach: Allwinner A10 (ARMv7hl/armhf)
<Superpelican> hramrach: just the normal gcc gnueabi toolchain
<Superpelican> installed with apt-get (Ubuntu)
<Superpelican> gcc 4.8
<hramrach> that should work
<Superpelican> hramrach: I'm cloning the linux-sunxi 3.4 branch again
<Superpelican> and then going to copy over my .config and just try again
<Superpelican> with a clean source tree
<hramrach> why not gnueabihf? not that it matters much for the kernel but anyway hf is the current toolchain
<Superpelican> and if crosscompiling with the clean source tree doesn't work, I'll try the Mer SDK
<Superpelican> which is installing in the meantime
nabblet has joined #linux-sunxi
<hramrach> this is the config I used last for a sun4i kernel build http://sprunge.us/KIND
<hramrach> it should at least produce a kernel
<hramrach> if not something on your system is likely broken
<Superpelican> hramrach: that link almost crashes my browser :(
<hramrach> your browser sucks :/
<Superpelican> latest Chromium
<hramrach> use wget then
<Superpelican> (33)
<Superpelican> nah it already loaded it
<Superpelican> the browser just became a bit unresponsive while loading the web page ;)
<Superpelican> hramrach: So basically what kind of config is it?
<hramrach> something random that I Used for last test build
<Superpelican> So it's not exacty tuned?
<Superpelican> Does it have all the GUI and video bloat?
<Superpelican> Or just framebuffer/console?
<hramrach> there is no gui bloat in kernel
<libv> gui bloat?
<Superpelican> I meant CedarX etc.
<Superpelican> which isn't very useful for a web server ;)
<hramrach> well, you can consider it tuned for booting on sun4i hardware
<Superpelican> BTW it's done cloning the branch
<Superpelican> so I can try compiling with the clean source tree :)
<libv> what is wrong with defconfig?
<hramrach> dunno
<hramrach> does not have iptables?
<Superpelican> no it doesn't have iptables support
<Superpelican> and I also like to tweak the config a bit
<libv> then first get it building and working with defconfig, and verify that it actually works
<Superpelican> now running "make -j4 ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf-"
<libv> then add iptables, and verify again
wingrime-ct has quit [Ping timeout: 276 seconds]
<Superpelican> with the clean source tree
<Superpelican> libv: Do you think that my .config is related to the no-uImage problem?
<hramrach> probably
<libv> it sounds like a likely candidate
<hramrach> some config settings don't work
<torindel> hramrach: he does have hardfloat toolchain ^^
<Superpelican> ok
<hramrach> yes, I see it
* torindel prefers armv7a-hardfloat-linux-gnueabi CHOST
<libv> torindel: armhf works fine here
<libv> so don't unneccessarily confuse this situation even more
<torindel> libv: yea i know it probably does, mine just have minimal arm family binaries will run in it
<torindel> Superpelican: dont forget to tee or > output this time
<torindel> so you can pastebin it
<Superpelican> torindel: oops :D "Your paste cannot be larger than 64 kb. Sorry."
<torindel> Superpelican: get other paste service
<jemk> wingrime: mplayer wants to create alpha-only surface, we don't support this yet
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
<wingrime1> jemk: good to know
<Superpelican> torindel, hramrach, buZz: http://paste.kde.org/pq7lnv8j0
<libv> hrm, the inet_1 page is really pointless
<Superpelican> libv: I know
<Superpelican> still have to fix it
<libv> there's nothing there that i didn't add apart from some pictures, and some probably very outdated and useless ft5x info
<Superpelican> BTW I have already spotted some errors in the make output
<Superpelican> make fails 2 times in the driver section
<Superpelican> but continues!
<torindel> Superpelican: :(.data+0x9fc): multiple definition of `Rtl8192CUAGCTAB_1T_HPArray'
<Superpelican> for example near the end: "make: *** [drivers] Error 2"
<jemk> wingrime1: supporting the surface wouldn't be a problem, but to mix it with other surfaces g2d driver would need to be extended (or softwarerendering)
<wingrime1> jemk: I add feature matrix to wiki
<Superpelican> and again there are no images in arch/arm/boot
<Superpelican> I don't understand
<wingrime1> jemk: also, I just get strange hung with g2d enabled + fbturbo
<Superpelican> I can't be doing something too special
<torindel> Superpelican: your build failed again
<hramrach> Superpelican: you need to make rtl8192cu and rtlwifi moduke
<hramrach> module
<torindel> Superpelican: youll get image only if build finishes
<hramrach> these two drivers define the same symbol
<wingrime1> jemk: I need investegate. who is bad, DRAM timings, or MMC is broken
<Superpelican> hramrach: what do you mean with "make rtl8192cu and rtlwifi"?
<hramrach> module
<hramrach> in config
<Superpelican> you mean I should disable the rtl8192cu and rtlwifi drivers?
<Superpelican> but I need those
<wingrime1> jemk: http://linux-sunxi.org/Cedrus feel free add profiles and any cedar features, as roadmap
<hramrach> are they built as modules or built-in?
<Superpelican> built-in this time
<hramrach> that's why your build fails
<hramrach> you build as built-in stuff that only works as module
<Superpelican> ?
<Superpelican> Build I could do 'y'
<Superpelican> s/build/but
<libv> Superpelican: you do not need both
<hramrach> you can but it does not work
<Superpelican> so only rtl8192cu and no rtlwifi?
<libv> you either need one or the other, or does your tablet have 2 wifi modules?
<hramrach> that too but it's hard to tell in advance which will work
<libv> build a defconfig kernel and see which one sticks
<Superpelican> libv: http://pastebin.com/Yab6tyYb
<Superpelican> from stock android
<libv> Superpelican: and?
<Superpelican> So do you think my tablet has 2 WiFi modules?
<Superpelican> I don't think it has 2
<torindel> libv: you managed to confuse him anyway ;]
<Superpelican> my previous kernel for Mer had internet connection with a 8192cu driver
<Superpelican> don't know which one though
<Superpelican> it was a module
<libv> Superpelican: so why are you building rtlwifi?
<Superpelican> don't know
<hramrach> also why are you building it into the kernel?
<Superpelican> I thought rtlwifi was generic?
<Superpelican> hramrach: I thought why not, let's ditch the modules ;)
<wingrime1> Superpelican: #powervr are looks dead
<hramrach> do you know how to build the wifi firmware into the kernel so the built-in driver actually works?
<buZz> not all drivers -can- be built-in to the kernel
<Superpelican> wingrimel: could be true, I never joined it ;)
<libv> wingrime1: as is the pvr reing project
<Superpelican> Why do they even allow the option to mark them as built-in in menuconfig then/
<libv> Superpelican: this is broken allwinner added code
<wingrime1> libv: just need find some one who saw any code / blob for PoverVR video decoder IP
<libv> Superpelican: which we have inherited and which we are slowly cleaning up
<wingrime1> libv: there is not so much vatiants what cedar could be
<hramrach> it should technically work but rtl8192cu is cut&paste of older rtlwifi module with some extensions for sunxi platform or something like that
<hramrach> so they conflict and cannot be in the kernel at the same time
<Superpelican> ok
<hramrach> also the firmware is a problem which can be taken care of if you know what you are doing but it's easier to just let the drivers live as modules
<Superpelican> ok
<Superpelican> hramrach: TBH I can't find the rtlwifi option in menuconfig
<libv> defconfig really isn't the worst
<hramrach> it's probably named differently
<Superpelican> there are 2 RTL8192C(U) though
<libv> just add iptables and the necessary iptables modules
wingrime has quit [Ping timeout: 245 seconds]
wingrime has joined #linux-sunxi
<Superpelican> " Realtek RTL8192CU/RTL8188CU USB Wireless Network Adapter" and "Realtek 8192C USB WiFi for SW"
Gerwin_J has joined #linux-sunxi
<torindel> second one is probably old allwinner stuff
<libv> Superpelican: normally, the help shows you which module name is actually built
<Superpelican> ok the 1st option is RTLWifi
<Superpelican> "Selects: FW_LOADER [=y] && RTLWIFI [=y] && RTL8192C_COMMON [=y]
<Superpelican> So what should I *disable*?
uwe_ has quit [Ping timeout: 252 seconds]
uwe_ has joined #linux-sunxi
<Superpelican> torindel: Should I disable the rtlwifi driver?
<torindel> Superpelican: both are same driver, first one is newer, second one is older (with old changes from allwinner)
<Superpelican> so I should disable rtlwifi?
<torindel> "Realtek 8192C USB WiFi for SW"
<Superpelican> So I should disable the above?
<torindel> yes
<Superpelican> because the above "Selects: WIRELESS_EXT [=y] && WEXT_PRIV [=y]"
<Superpelican> the other driver selects rtlwifi
<torindel> hramrach, libv: i think it would be better to add option "Allwinner lefovers (dont enable unless you know what are you doing)" and make all deprecated stuff in kernel depend on it
<Superpelican> I am confused now
<torindel> Superpelican: disable "Realtek 8192C USB WiFi for SW" and compile kernel again
<Superpelican> hmm I can't remember what I disabled :(
<Superpelican> anyway at least I modularized the remaining driver
<libv> torindel: i think it would be better if people, for sunxi-3.4 don't stray too much from defconfig
<torindel> libv: that too ;]
<torindel> libv: but there are group of ppl who always wants to make things "their way"
<torindel> and not all of them know what they are doing
<hramrach> for that the defconfig would have to include stuff people use
<hramrach> also it's not always obvious which driver is old and which new and which of the old and new happens to work on a particular piece of hardware
<hramrach> libv was the one having trouble with USB WiFi on some tablet iirc
<Superpelican> I'll copy the .config and open it with menuconfig and check which one I disabled
<libv> which i documented on our wifi page.
<libv> of course.
<Superpelican> libv: which WiFi page, I read the WiFi page but didn't see anything about that
<libv> anything about what?
<Superpelican> Did read something about PM issues with the rtl8192cu though if that is what you mean
<Superpelican> libv: about what you documented
<libv> yes, that was the issue i was having.
<Superpelican> ok
<libv> it's a wiki
<libv> it comes with a history of who added what
<libv> that is the bit of the wiki that tells you what i documented.
<Superpelican> not very experienced with wiki's ;)
<Superpelican> libv: And with what driver?
<Superpelican> rtlwifi (the old allwinner stuff) or with the other driver?
<Superpelican> ok make just finished
<Superpelican> let's have a look
<libv> Superpelican: why do you ask this?
<libv> Superpelican: the answer will not help with your kernel config/build issues
<Superpelican> just wanted to know what your experiences are with the 2 different drivers
<Superpelican> BTW there are now images in arch/arm/boot :)
<libv> udev chooses the driver for you, if you enable the usbc
<libv> then there is the catch-22 bit, the usbc will only be enabled if a wifi driver is loaded, and the fex instructs it to enable a specific usbc
<libv> so you just load a random driver, and if it's the right one, no other driver will be loaded, if it is the wrong one, udev will load the right one for you
<Superpelican> ok I just verified that I indeed disabled the rtlwifi driver
<Superpelican> so that's OK
<Superpelican> now going to run make modules
<Superpelican> buZz, torindel, libv, hramrach: OK now got the modules and images
bertrik_ is now known as bertrik
<Superpelican> Should I now just convert the zImage or Image to a uImage with mkimage?
<Superpelican> Or should I run make again with uImage?
nicksydney has quit [Remote host closed the connection]
nicksydney has joined #linux-sunxi
<Superpelican> torindel: Can I continue using my old rootfs/setup or should I start all over again to be safe?
<buZz> i would run make uImage in the beginning
<Superpelican> buZz: So you say that I should avoid manually converting the zImage to a uImage?
<buZz> i just dont see any advantage to it
apo has joined #linux-sunxi
<Superpelican> buZz: Well I have already created a uImage with mkimage, so...
<Superpelican> ...is it worth it doing everything again?
<buZz> no :)
<Superpelican> buZz: And should I start with a fresh rootfs?
<buZz> vs what?
<Superpelican> or can I just replace the uImage and modules and continue?
<Superpelican> buZz: ^
<buZz> i dont get what you are asking
<Superpelican> buZz: Well yesterday I create a SD card with my previous custom kernel and a linaro server rootfs
<Superpelican> buZz: then I started installing owncloud etc.
<Superpelican> and this morning I discovered that my custom kernel didn't have iptables support...
<Superpelican> buZz: so I started building a new kernel
<Superpelican> buZz: So should I just continue using my SD card (setup) and just replace the uImage and the modules with my new ones?
<buZz> FYI, you could have started with the kernel config in /proc/config.gz <- to start with your -current- config
<buZz> but yeah i would just use your new kernel on that rootfs
geecko has quit [Remote host closed the connection]
<Superpelican> hmmm
<Superpelican> I'm considering starting all over again regarding the rootfs
<Superpelican> because I had problems with Owncloud anyway
<Superpelican> and maybe they are related to having no firewall
<Superpelican> buZz: I think I'll just back up my current set up with dd and start over
<Superpelican> buZz: And just follow http://linux-sunxi.org/Bootable_SD_card :)
<buZz> why not ;)
<buZz> are you running owncloud on a tablet?
<Superpelican> buZz: yes
<Superpelican> the device is unusable with Android
<Superpelican> so why not?
<Superpelican> saves some money spent on a oLinuxino or RPi/Beaglebone
<Superpelican> :)
<Superpelican> LOL: ""I don't care about any of this, make my device work.""
geecko has joined #linux-sunxi
<Superpelican> libv: How do I attach the script.bin and script.fex to the wiki page?
<Superpelican> libv: And BTW is that even legal?
wingrime1 has quit [Ping timeout: 276 seconds]
<Superpelican> it is a stock android script.bin
<Superpelican> so isn't it the manufacturer's IP?
<buZz> i dno, i thought you should run owncloud on a server
<Superpelican> buZz: Well I'm not an expert, but I don't think Owncloud with nginx and sqlite should put extreme loads on the tablet ;)
<Superpelican> there are even people running it on 256 mb RPi's ;)
<buZz> no thats not what i mean
<buZz> i mean, dont you want your tablet to be a tablet
<buZz> and not a server
<buZz> guess i have some oldfashion ideas ;)
<Superpelican> buZz: well it's unusable with Android
<buZz> so?
<Superpelican> buZz: And I don't want to bother compiling CM for it ;)
<Superpelican> buZz: I was porting Nemo Mobile to it
<buZz> i dont think it will perform better with linux running server systems ...
<Superpelican> well at least it will be useful again
<buZz> but you make a point
<Superpelican> instead of sitting in some drawer
<Superpelican> :)
<buZz> i should ask that friend with a A10 tablet to gift it to me
<Superpelican> ok apparently wine thought it was a good idea to start IE7 when it needs to update some config :/
<Superpelican> libv: That's the .fex of the original/stock android script.bin
<libv> Superpelican: i told you before to read through the new device howto
<Superpelican> libv: there is already u-boot-sunxi support and I have read the new device howto
<Superpelican> libv: http://linux-sunxi.org/Inet_1 it's on the wiki page
<libv> Superpelican: so why are you still asking about attaching the fex to our wiki?
<Superpelican> libv: I thought it might be useful for someone who wants to create his own image
<libv> Superpelican: *bzzz*
<Superpelican> libv: as there is no hwpack for the Inet 1
<libv> read through that again
<libv> or for the first time, which seems much more likelyu
<Superpelican> libv: which .fex should I add?
<libv> ?
<Superpelican> the original stock android .fex
<Superpelican> or a fixed one
<libv> Superpelican: the one you just grabbed off the device
<Superpelican> ok
<libv> what do you mean, fixed one?
<Superpelican> well I also have a version of the original that .fex that I edited
<Superpelican> I added some missing information
<libv> what information was missing?
<Superpelican> doubt it was necessary
<Superpelican> can't quite remember
<Superpelican> it's about 3/4 year ago
<libv> ...
<libv> Superpelican: take the one from android, and use that for your manual build.
<Superpelican> ok
<libv> Superpelican: also, can you find out how to identify your device from android?
<libv> it is probably different than the protabs identification strings
<Superpelican> libv: You mean in the about page?
<Superpelican> in the Android Settings app?
<libv> yes
<Superpelican> libv: I have already collected quite some information and files from the stock android install
<Superpelican> including kernel modules
<Superpelican> IIRC also build.prop
Gerwin_J has quit [Quit: Gerwin_J]
<libv> http://linux-sunxi.org/Inet_86vz has an example of 2 android id sections for the same device
<Superpelican> libv: I also have some additional images for the wiki, you can see them here: http://superpelicanblog.wordpress.com/porting-mer-corenemo-mobile-to-the-cherry-m1007/
<Superpelican> libv: from the "About tablet" screen of the Android Settings app: http://pastie.org/9095606
<Superpelican> libv: also I have the lsmod (on Android) dump on pastebin that I already mentioned earlier
<Superpelican> "init.sun4i.rc"
<Superpelican> "build.prop"
<Superpelican> the original "u-boot.bin"
<Superpelican> "linux.ini"
<Superpelican> "boot.ini"
<Superpelican> and all kernel modules
<Superpelican> libv: "git: 'send-email' is not a git command. See 'git --help'."
geecko has quit [Ping timeout: 245 seconds]
<Superpelican> libv: looks like I need the sendemail package ;)
<Superpelican> libv: it still doesn't work
Gerwin_J has joined #linux-sunxi
xinj has quit [Ping timeout: 240 seconds]
<Superpelican> torindel: mount ${card}${p}2 /mnt/ : "mount: you must specify the filesystem type"
pseudomind has joined #linux-sunxi
Quarx has quit []
<libv> Superpelican: did you partition your sd card as prescribed?
<libv> and did you then, as described, create a filesystem on that partition?
<libv> these really are very trivial little issues
<Superpelican> libv: yes I all did that
<Superpelican> libv: But I just did everything over again, and now it suddenly does work :/
<libv> so you must have missed a step.
<Superpelican> libv: this time I used the first partition command though instead of the 2nd
<Superpelican> maybe that's it
<libv> ...
<libv> you do understand the difference between the two, right?
<Superpelican> libv: well not exactly, it's close to black magic for me ;)
<Superpelican> libv: I thought they were just 2 different ways to express the same
xavia has quit [Quit: Leaving.]
<Superpelican> libv: BTW I can't send the patch
<Superpelican> libv: Keep getting "git: 'send-email' is not a git command. See 'git --help'."
<megal0maniac_afk> Also, silly maybe, but you did have /dev/ before mount ${card}${p}2, right?
<Superpelican> megal0maniac: I substituted ${card}${p}2 for /dev/sdb2 in all the commands ;)
<Superpelican> anyway my sd card is done now
<Superpelican> let's hope it works...
<Superpelican> it doesn't work... :(
<Superpelican> the tablet boots Android
<Superpelican> torindel, hramrach, buZz, libv: ^, any ideas where I should start debugging?
<megal0maniac_afk> Superpelican: Gotta check :)
<libv> Superpelican: you probably messed up copying the spl or u-boot blobs to the sdcard
<Superpelican> libv: Could the problem also be related to my boot.scr?
<Superpelican> libv: Why do they use 0x48000000 as entry point and not 0x40008000?
<libv> wingrime: nice one on the marketing cleanup
<Superpelican> libv: But how can I mess up the blobs?
<Superpelican> libv: I'm sure I did it right
<libv> Superpelican: you had problems with your sd-card before
<libv> Superpelican: you can switch grab those areas from the sdcard and then md5sum the result
<libv> s/switch //
<megal0maniac_afk> Haha :) "Remove marketing bullshit"
<Superpelican> libv: I created a compined bin and did sudo dd if=u-boot-sunxi-with-spl.bin of=/dev/sdb bs=1024 seek=8
<libv> well, for some reason your device sees your boot0 as invalid and reverts to android
<Superpelican> ok
<Superpelican> libv: But what do you mean with "switch grab"
<libv> remove the word switch
<Superpelican> libv: but still what do you mean with "grabbing those areas from the sd card"?
<libv> Superpelican: read up on dd
<Superpelican> libv: But why do they use 0x48000000 "fatload mmc 0 0x48000000 uImage || ext2load mmc 0 0x48000000 uImage boot/uImage \n bootm 0x48000000" as load address instead of 0x40008000, I thought you should use 0x40008000 as load address on sun4i
<libv> you can use whatever you like, as long as the areas do not overlap
<libv> and it is not sun4i dependant
<libv> you just told u-boot from where the kernel should be loaded
<libv> and...
<libv> if you mess that bit up, the boot is stopped.
<libv> you do not get to go to android
<libv> u-boot will sit there and wait for user input
<libv> so you are definitely barking up the wrong tree there
<Superpelican> libv: And what if u-boot can't find the script.bin?
<libv> same thing
<Superpelican> libv: And for some reason "LOST.DIR" keeps being created on /dev/sdb1...
<Superpelican> libv: Can I just rewrite the combined bin, without doing everything over again?
<Superpelican> libv: ok I have just rewritten the u-boot-with-sunxi-spl.bin
<Superpelican> libv: now I get a black screen
<Superpelican> libv: Can you call that progress? Or have I know just messed up the whole sd card?
Gerwin_J has quit [Quit: Gerwin_J]
<yang> Hi, I am wondering what is the status of free (reverse engineered) port for RPI ?
<yang> With a totally free distribution
<yang> and if there is a way to test it and contribute with bug reports at this stage, yet ?
Superpelican has quit [Ping timeout: 240 seconds]
<buZz> yang: wrong channel?
<libv> a free, reversed engineered, port of sunxi to rpi?
SuperpelicanJoll has joined #linux-sunxi
<buZz> i noticed allwinner does their development on FPGAs
<buZz> did someone leak those binaries yet? :P
SuperpelicanJoll has quit [Ping timeout: 240 seconds]
ykchavan has joined #linux-sunxi
FunkyPenguin has quit [Ping timeout: 240 seconds]
FunkyPenguin has joined #linux-sunxi
deffrag has joined #linux-sunxi
deffrag__ has quit [Ping timeout: 240 seconds]
joedj has quit [Ping timeout: 276 seconds]
joedj has joined #linux-sunxi
SuperpelicanJoll has joined #linux-sunxi
fredy has quit [Excess Flood]
SuperpelicanJoll has quit [Client Quit]
fredy has joined #linux-sunxi
FunkyPenguin has quit [Ping timeout: 240 seconds]
FunkyPenguin has joined #linux-sunxi
FunkyPenguin has quit [Excess Flood]
FunkyPenguin has joined #linux-sunxi
megal0maniac_afk is now known as megal0maniac
heretic_ has joined #linux-sunxi
heretic_ has quit [Remote host closed the connection]
FreezingCold has joined #linux-sunxi
Superpelican has joined #linux-sunxi
<Superpelican> libv: ok you definitely can't just rewrite the u-boot-with-sunxi-spl.bin: ls /mnt/ (with /dev/sdb1 mounted): http://pastie.org/9095955 ;)
<libv> ?
<libv> what did you just do?
Gerwin_J has joined #linux-sunxi
<Superpelican> libv: well I just ran the first dd u-boot-with-sunxi-spl.bin command again
<Superpelican> libv: the screen of the black stayed black while attempting to boot
<Superpelican> libv: So I mounted it again on my pc
<Superpelican> libv: (/dev/sdb1) and ran ls /mnt ;)
<Superpelican> apparently you can't just run the dd u-boot-with-sunxi-spl.bin command again
<torindel> Superpelican: do you have correct partition table and do you dd it in right place? ;p
<Superpelican> it'll completely mess up the sd card
<Superpelican> torindel: I'm now creating the sd card all over again
<torindel> pastebin your partition table on it when you're done
<torindel> and paste your dd command before you run it
<Superpelican> torindel: http://pastie.org/9095983
<Superpelican> already done :)
<torindel> also ls your u-boot-with-sunxi-spl.bin
<torindel> ls -la
<torindel> Superpelican: ok you can partition your card again :P
<torindel> and make some free space before first partition this time maybe ;p
<torindel> to be sure
<Superpelican> torindel: and how do I make some free space before the first partition?
<torindel> Superpelican: dont start at 0mb
<torindel> use 1mb
<Superpelican> torindel: but the card was partitioned correctly?
<Superpelican> torindel: And how would I do that?
<torindel> Superpelican: run normal fdisk, make new dos partition table (o), create new partition (n), partition (p), number one (1), start sector 2048, +20MB size
<torindel> change type (t), partition 1 (1), fat32 (b)
<Superpelican> torindel: why +20MB?
<torindel> pick any size you want
<Superpelican> torindel: And how do I start at 1024 kb?
<torindel> just make some space before first partition (like sector 2048)
<torindel> it asks you: start, end
<torindel> add second partition afterwards too
paulk-aldrin has quit [Read error: Connection reset by peer]
<torindel> with remaining of the card
<torindel> and default type
paulk-aldrin has joined #linux-sunxi
<Superpelican> IIRC the last sd card setup that I made (that did work) was also created with fdisk
<Superpelican> but I can't seem to remember what guide I followed last summer :/
nabblet has quit [Ping timeout: 240 seconds]
<torindel> and when you dd your u-boot-with-sunxi-spl.bin do it in correct place :P
<Superpelican> torindel: And that is?
<Superpelican> torindel: And how do I correctly add the 2nd partition?
<torindel> 8kb from begining of medium if i remember correctly
<torindel> Superpelican: n, p, 2, enter, enter
<Superpelican> ok
geecko has joined #linux-sunxi
<Superpelican> torindel: And I can dd the u-boot-sunxi-with-spl.bin even after creating the partition table?
<torindel> yep
<torindel> how big is your u-boot-sunxi-with-spl.bin?
<torindel> you write it like this: dd if=u-boot-sunxi-with-spl.bin of=/dev/sdb bs=1024 seek=8
<Superpelican> torindel: 260,3 kb
bertrik has quit [Remote host closed the connection]
<torindel> Superpelican: write it to sd like i written
<Superpelican> torindel: Isn't that way to big?
<Superpelican> 260 kb?
<torindel> its ok
<torindel> 2048 sectors you have before your first partition = 1MB
<Superpelican> ok
Superpelican has quit [Quit: Page closed]
FunkyPenguin has quit [Ping timeout: 276 seconds]
FunkyPenguin has joined #linux-sunxi
FunkyPenguin has quit [Ping timeout: 240 seconds]
FunkyPenguin has joined #linux-sunxi
FunkyPenguin has quit [Ping timeout: 245 seconds]
FunkyPenguin has joined #linux-sunxi
FunkyPenguin has quit [Excess Flood]
FunkyPenguin has joined #linux-sunxi
rz2k has quit []
megal0maniac is now known as megal0maniac_afk
FunkyPenguin has quit [Ping timeout: 245 seconds]
FunkyPenguin has joined #linux-sunxi
jemk has quit [Quit: leaving]
FunkyPenguin has quit [Ping timeout: 276 seconds]
FunkyPenguin has joined #linux-sunxi
wingrime has quit [Ping timeout: 240 seconds]
<Turl_> hm, wtf
<Turl_> We're writing to let you know that the group you tried to contact (cubieboard) may not exist, or you may not have permission to post messages to the group. A few more details on why you weren't able to post
ykchavan has quit [Read error: No route to host]
FunkyPenguin has quit [Ping timeout: 245 seconds]
FunkyPenguin has joined #linux-sunxi
FunkyPenguin has quit [Ping timeout: 240 seconds]
FunkyPenguin has joined #linux-sunxi
bertrik has joined #linux-sunxi
bertrik has quit [Changing host]
bertrik has joined #linux-sunxi
FunkyPenguin has quit [Ping timeout: 250 seconds]
FunkyPenguin has joined #linux-sunxi
ykchavan has joined #linux-sunxi
FunkyPenguin has quit [Excess Flood]
FunkyPenguin has joined #linux-sunxi
vicomte88 has joined #linux-sunxi
<vicomte88> bonsoir a tous
<vicomte88> Good evening all was, I have a question about the drivers CAN BUS peter chen? I do not understand why ADR of BASE is not like documentation
<vicomte88> #define CAN_BASE0 0xF1C2BC00 for drivers
<vicomte88> the doc . 0x011C2BC00
paulk-aldrin has quit [Quit: Ex-Chat]
<vicomte88> an idea for my ?
patapovich has joined #linux-sunxi
FreezingCold has quit [Ping timeout: 245 seconds]
<patapovich> vicomte88: it looks like the one used in the driver is right
<patapovich> probaly error in documentation
<vicomte88> I'm not on the doc cortex said the same thing ..
<vicomte88> 0x01C2BC00 sorry
<vicomte88> is the 0xF I do not understand
<vicomte88> is linux kernel is tea that imposes register
<vicomte88> ?
<vicomte88> like MMU
<vicomte88> not MMU for corteA7 A20
<vicomte88> ok cool the link
<patapovich> afaik the one with 0xf is virtual adress
<vicomte88> virtual adress ok thank you ... where it occurs it is virtual adr
<vicomte88> ?
<vicomte88> in the doc ?? no find my
<patapovich> check the link i gave you
<vicomte88> yes thank you I saw ca alright help me
<Turl_> mnemoc!
<vicomte88> but the virtual address is found how? in general
<Turl_> vicomte88: add 0xf0000000 to physical address
FreezingCold has joined #linux-sunxi
<Turl_> (at least on sunxi, where there is a static mapping for the register section)
<vicomte88> ok a big thank you
<Turl_> you're welcome
wingrime has joined #linux-sunxi
patapovich has left #linux-sunxi [#linux-sunxi]
vicomte88 has quit [Ping timeout: 276 seconds]
Xaros has joined #linux-sunxi
wingrime has quit [Read error: Connection reset by peer]
Black_Horseman has quit [Ping timeout: 250 seconds]
Web-aptosid876_ has quit [Quit: Leaving]
FreezingCold has quit [Ping timeout: 245 seconds]
ganbold_ has quit [Remote host closed the connection]
FreezingCold has joined #linux-sunxi