<naobsd> github/linux-rockchip description is "an open source developers' community for Rockchip SoC based devices" and I'm always waiting *BSD (and other non-linux) developers ;)
<naobsd> there is "dedicated to providing linux support" on (linux-)sunxi.org :)


<naobsd> sorry, I cannot find good info/page now :(
<naobsd> yup
<naobsd> well, /proc/cpuinfo output on somewhere...
<naobsd> just 4 Cortex-A5...
<naobsd> oh, no, no hybrid
<naobsd> hramrach_: it's FAQ ;) google it
<naobsd> next problem: how to avoid "hey mom! my daddy has new toys! I want new toys too!"
<naobsd> I think "no thank you" immediately
<naobsd> anyway, probably she already gave up to understand my hobby
<naobsd> no idea ;)
<naobsd> "hey! you already have same SoC, buy another one!"
<naobsd> I hope she don't learn about detail of arm socs
<libv> naobsd: with me it just vanishes in the pile
<libv> naobsd: she can tell the difference?
<naobsd> then my wife will not claim "oh! you bought new garbage!?"
<naobsd> same case is important
<naobsd> collect exact same case but different SoC
<naobsd> I want to collect MK802-style devices
<naobsd> I only have A10/A20
<naobsd> well
<naobsd> I don't have any A3x
<naobsd> (please don't ask who want wired network)
<naobsd> then tablet can have ethernet connector!
<naobsd> ethernet should use another form connector such as Lightning
<naobsd> info is taken from allwinner web, not datasheet
<naobsd> http://linux-sunxi.org/Allwinner_SoC_Family table is updated a little


<ssvb> naobsd: I mean, the problem is entirely on my end :) no need to move it anywhere, thanks
<ssvb> naobsd: yes, my download speed is not that great
<naobsd> ssvb: local speed issue? or I can try another server at us/nl
<ssvb> naobsd: thanks, downloading, it would take about 2 hours with my internet speed
<ssvb> naobsd: but the 2D engine still seems to be in A31s, and a thin tablet does not strictly need GMAC :)
<naobsd> well
<naobsd> oh
<naobsd> then... +gmac please ;)
<libv> naobsd: stick these sort of things in the wiki btw
<naobsd> ah, thanks
<naobsd> right?
<naobsd> -rtp (what is it?)
<naobsd> -2nd lvds
<naobsd> -2nd rgb lcd
<naobsd> -mipi dsi
<naobsd> -mipi csi
<naobsd> -2nd nand
<naobsd> -2nd ddr ch
<naobsd> -ddr3u
<naobsd> -gmac
<naobsd> diff A31 A31s
<naobsd> ah, please wait another 10min or so for uploading
<ssvb> naobsd: yes, sure, thanks
<naobsd> ssvb: download finished. do you need it?
<naobsd> ssvb: no
<ssvb> naobsd: do you have any devices with this SoC?
<naobsd> ssvb: if you need, I'll put somewhere
<naobsd> ah, I should make dummy user for it ;)
<naobsd> I often run "rm -rf ~/.mozilla" :)
<naobsd> ssvb: same for me. so I started firefox(non-used for daily purpose) and downloading now :)
<ssvb> naobsd: yes, but it wants me to install a browser plugin, which raises a big red flag :)
<naobsd> ssvb: link to "mega" seems alive


<naobsd> oh, lime2...


<naobsd> mripard_: really thanks!
<mripard_> naobsd: sorry, I forgot to push the branch
<naobsd> mripard_: where can I see A20 Lime dts you merged?


<naobsd> I see
<naobsd> merge A10/A20 Lime page, is not good?
<naobsd> btw A20 Lime uses same board as A10 Lime
<naobsd> after I do some linux-rockchip things...
<naobsd> I'll maintain A20 Lime page
<naobsd> GMAC is used in upstream
<naobsd> and I got "merged" reply from mripard for Linux
<naobsd> btw, A20 Lime support is already in u-boot-arm/sunxi custodian
<naobsd> olinuxino_m <-> oli_micro
<naobsd> name for micro is not consistent too
<naobsd> I think so
<naobsd> yes
<naobsd> is Dimitar Gamishev olimex person?
<naobsd> I think lime2 is not available yet
<naobsd> I posted s/EMAC/GMAC/ patch for A20 Lime (not Lime2)


<naobsd> can anyone merge s/EMAC/GMAC/ patch for linux-sunxi u-boot for A20 lime?


<naobsd> olimex's lime2 is not sold yet


<naobsd> ijc: ping
<naobsd> how should I write a patch to add A20 LIME config?
<naobsd> there are 2 groups, SUNXI BOARD and CUBIEBOARD2 BOARD
<naobsd> in board/sunxi/MAINTAINERS in u-boot custodian tree,
<naobsd> hmm
<naobsd> :)
<libv> naobsd: that's clear from the patch itself :)
<naobsd> that patch is for u-boot
<naobsd> ah, I forgot to mention


<naobsd> but I don't have any debug message for now
<quitte> naobsd: the only other one i can think of that could possibly matter is operating voltage.
<naobsd> but capacity is not only parameter which is used by driver to configure controller
<naobsd> that uSD is 2GB
<quitte_> naobsd: once the sd-card is initialized there are no differences in single block read and writes between any. is that card remarkable in any way compared to the others? for example: is it huge?
<naobsd> it may be bad sdcard, but I have no idea for now
<naobsd> I can read that uSD on PC, sha1 is same...
<naobsd> software issue...?
<naobsd> it happens both cb2 and a20 lime
<naobsd> s/fals/fatls/
<naobsd> hmm, one of uSD card behave strangely... fatls works, next fatload failed(error reading cluster), next fals failed too :(
<naobsd> ah, wiki is already updated :)
<naobsd> wens: anyway, thank you for your guide! I understand procedure
<naobsd> I subscribed, then I sent a patch...
<naobsd> hmm, "Your message to U-Boot awaits moderator approval"
<naobsd> hipboi: thanks, I see, but I'm feeling something strange :)
<hipboi> naobsd, it's overwritten by livesuite when flashing
<naobsd> are these params ignored or overwritten by bootloader?
<naobsd> I didn't try any official image from olimex yet
<naobsd> reading NDH...
<naobsd> anyway, I sent patches for a20 lime
<naobsd> is u-boot maillist moderated?
<naobsd> now I learned how to "git send-email", subscribed linux-arm-kernel, let's send patch...
<wens> naobsd: you could say based on a10 lime and keep the original copyright, and add your name
<naobsd> about a20 lime patch, should I add copyright line into .dts? it's basically copy&paste thing from dts for a10 lime (I think "no" for this kind of work(?))


<naobsd> good night
<naobsd> zzz
<naobsd> only some functions are tested for now
<naobsd> here is patches:
<naobsd> sleeply, I'll try tomorrow
<wens> naobsd: at least to mripard_ and linux arm kernel ML
<naobsd> I'm not sure about linux patch procedure well, I need to learn about it ;)
<naobsd> where should I send changes for a20 lime for linux mainline? linux-sunxi group?
<naobsd> will be merged in 3.17? or 3.18?
<libv> naobsd: yes, patch was on the ml 2 weeks ago
<naobsd> I thought kernel config was fixed ;)
<naobsd> codepage cp437 not found
<naobsd> everything is new. (too new for me ;)
<naobsd> 3.17-rc1 is booted. axp20x is loaded w/ modified dts.
<naobsd> let's try w/ mainline kernel...
<naobsd> interrupt-parent = <&nmi_intc>; interrupts = <0 8>;
<wens> naobsd: oh crap, forgot to change the interrupt description for the axp, see the other sun7i boards
<naobsd> wens: I see, thanks
<wens> naobsd: since mripard_ hasn't pushed the new sunxi-next, you should probably use master at v3.17-rc1
<naobsd> [ 0.892448] axp20x 0-0034: Failed to request IRQ 0 for axp20x_irq_chip: -22
<naobsd> ok, a20 lime is running with mainline u-boot and sunxi-next kernel
<naobsd> I guess dram_a20_olinuxino_l.c can be used for mainline w/o any change...
<naobsd> let's try a20 lime
<naobsd> ok, gmac works fine
<naobsd> maybe: it just complain about ethaddr
<naobsd> Error: dwmac.1c50000 address not set.
<naobsd> anyway, U-Boot 2014.10-rc1-00476-gbf4ca38 is booted on cb2
<naobsd> I thought dtb is referred while building mainline u-boot, but it isn't.
<naobsd> why CONFIG_FTDFILE="sun7i-a20-cubieboard2.dtb" is defined in u-boot config
<naobsd> I just wondered
<quitte> naobsd: in uboot you load the dtb to a memory address just like a kernel image
<naobsd> dtb string seems to be just for default setting...
<naobsd> let's try cb2 at first...
<naobsd> well, put .dtb compiled from linux tree to mainline u-boot tree?
<naobsd> it's time to learn about mainline u-boot ;)
<naobsd> ...slowly...
<naobsd> fetching now...
<naobsd> oh
<naobsd> may I use pull req on github? or format-patch to ML?
<naobsd> I see. I can confirm only A20 Lime
<naobsd> gmac is preferred on u-boot too?
<naobsd> for A20 Lime, I changed it to _GMAC (w/o RGMII of course) and it works
<naobsd> there are some A20 boards with SUNXI_EMAC in boards.cfg in u-boot-sunxi
<naobsd> it seems working on A20 lime
<naobsd> can all A20 board with ethernet use gmac?
<naobsd> hmm.
<naobsd> +label = "a10-olinuxino-lime:green:usr";
<naobsd> -label = "a20-olinuxino-lime:green:usr";
<naobsd> wens: thanks, I also made almost same one :)
<naobsd> and currently I don't have fex, it's no NAND model, I also need to download image from olimex
<naobsd> yes, I just build u-boot-sunxi for a20 lime
<libv> naobsd: nm then
<naobsd> if my board is not broken (not yet powered!)
<naobsd> probably I can update some bits on wiki
<naobsd> then
<naobsd> sorry
<naobsd> it's not lime2, it's A20 Lime
<naobsd> ah
<libv> naobsd: ndh page still needs work
<naobsd> now trying to make it from dts for Lime(1)
<naobsd> :)
<naobsd> if (own_lime2 && have_dts_for_it) then reply()
<naobsd> (now I'm looking at linux-sunxi/sunxi-next)
<naobsd> does anyone have .dts for Lime2?


<libv> naobsd: i have 3 haipad m701-rs lying about, in varying states
<naobsd> libv: I still have HSG X5A too ;)
<naobsd> oh, is it time to talk about happy crappy chinese tablets? ;)


<naobsd> bring from radxa -> exact same copy
<naobsd> github/linux-rockchip is not radxa's repo, but there are some branches bring from radxa's repo


<naobsd> libv: I see, thank you
<libv> naobsd: such things are really what wikis are for, so next time, look through the wiki pages you followed, completely. If that fails, then come and ask, as then something is wrong with the wiki page
<naobsd> ccaione: yes
<naobsd> I'm always use Japanese ;)
<naobsd> oh :)
<Turl> naobsd: you got a follower :)
<naobsd> (I'm addicted)
<naobsd> libv: sorry for silly question!!! ;)
<naobsd> libv: yes, few minutes ago, after fall into pitfall!
<libv> naobsd: did you check the common pitfalls section of the binary drivers page?
<naobsd> all effort were already done by olegk0 and this community :)
<naobsd> I spent only few hours :)
<naobsd> I just use oglek0's code(he did it on rk3066) and sunxi wiki/code :)
<Turl> naobsd: there were some people asking about it on the radxa channel, you may want to document it
<Turl> naobsd: heh :)
<naobsd> finally I can use mali on linux on radxa rock(rockchip rk3188) by using sunxi information/code, thanks :)
<Turl> naobsd: that should do the trick, yep
<naobsd> I have to move /usr/lib/arm-linux-gnueabihf/mesa-egl
<naobsd> I learned.
<naobsd> I have to learn more about mali and gles thing, sorry
<naobsd> glmark2-es2 says "libEGL warning: DRI2: failed to open lima (search paths /usr/lib/arm-linux-gnueabihf/dri:${ORIGIN}/dri:/usr/lib/dri)", then it renders 3D with mesa software rasterizer
<naobsd> "AIGLX error: dlopen of /usr/lib/arm-linux-gnueabihf/dri/lima_dri.so failed" this should be expected, no problem...
<naobsd> oh well...
<naobsd> ump/mali/drm/mali_drm is compiled, loaded. fbturbo is running
<naobsd> libv: where is lima_dri.so?
<naobsd> I just started it few hours ago, I misunderstand something...
<naobsd> it seems lima_dri is needed
<naobsd> libv: I'm trying X/mali on linux on rockchip with ssvb's fbturbo
<libv> naobsd: yes?
<naobsd> libv: ping


<naobsd> a23! very interesting...


<naobsd> not sure, I'll check tonight
<naobsd> probably it was around 90Mbps or so
<naobsd> some days ago I tried iperf with allwinner's emac driver on cb2
<ganbold_> naobsd: if you have time can you test emac on cb1 or cb2 using netperf?
<ganbold_> naobsd: ping


<Tsvetan> naobsd yes, lot of things to check
<naobsd> under cpu load
<naobsd> Tsvetan: please check power consumption too
<naobsd> I really have interest to A20-LIME


<naobsd> ganbold: probably many drivers have similar setup code. I don't know well about recent FreeBSD ethernet driver structure, but probably unicast/broadcast part should be set in init/reset, and multicast/promisc part should be set in ioctl path
<ganbold__> naobsd: thanks, now I need to port set_rx_mode to freebsd
<naobsd> it seems 3.4 Android kernel
<naobsd> I just tried A20 SDK2.0 3.4 kernel on CB2... booted, ethernet works
<naobsd> ganbold: you can refer 3.3 wemac driver in github/cubieboard2
<naobsd> ganbold: and compared with emac driver in allwinner's A20 SDK2.0 3.4 kernel, they are same
<naobsd> ganbold: I compared latest cb2/ct 3.3 driver and github/cubieboard2 driver, they are basically same
<naobsd> rx setup func in 3.3 wemac driver seems good for me, it looks like very standard
<naobsd> it takes time
<naobsd> ganbold: shoud I check RX mcast filter more?
<naobsd> but I'm not sure packets are properly filtered
<naobsd> I can see packets while tcpdump is running
<naobsd> IPv6 is working well with 3.3 kernel for cb2
<naobsd> my brain is XXX...
<naobsd> I built fel version...
<naobsd> ah,
<naobsd> where is u-boot-sunxi-with-spl.bin...
<naobsd> hmmmmmmm
<naobsd> IPv6 needs multicast, tcpdump uses promisc, it should be easy ;)
<naobsd> if you need I'll setup cb2...
<naobsd> probably no
<ganbold__> naobsd: did you try to put interface to promisc mode?
<naobsd> I used 3.3 kernel on cb2, unicast/broadcast setting should work, but I didn't care about multicast...
<ganbold__> naobsd: does it work?
<naobsd> (btw, repo in github/cubieboard2 shouldn't have latest code)


<naobsd> slapin: did you erase your SD card before flashing new image?
<naobsd> well...
<naobsd> slapin: how did you erase configuration for using falcon mode?
<slapin> naobsd: it should not, but it is.
<naobsd> (I'm talking about configuration, not about source tree/compile time thing)
<naobsd> slapin: then why you use falcon mode? it should not enabled by default, as you said.
<oliv3r> naobsd: ah, yeah the commit subject seems wrong
<slapin> naobsd: NO
<oliv3r> naobsd: ^
<naobsd> slapin: yes. you want(ed), right?
<slapin> naobsd: if he wants it, but not by default
<naobsd> I think falcon can be used if user want(configure) it
<naobsd> which commit disables falcon?


<naobsd> yeah, it's big thing for us. we should add information about it on our wiki :)
<mnemoc> naobsd: I only mean that for us it's a big thing.... for them, it's just another toy
<naobsd> I think it should be good to put more information on linux-sunxi wiki, but I don't think it's wrong to put info about xen arm thing on xenproject wiki
<naobsd> prebuilt image? is there any missing piece on wiki on xenproject?
<kenny> naobsd: I'm trying to get a prebuilt image posted from the author of http://wiki.xenproject.org/wiki/Xen_ARMv7_with_Virtualization_Extensions/Allwinner
<naobsd> oh, xen on sun7i, very interesting


<naobsd> on some RK30/31. some(RK3168) has SGX
<naobsd> Mali400-MP4
<naobsd> unruly horse, it seems I need to do more work to stable it ;)
<ssvb> naobsd: which mali version is used for RK?
<naobsd> oh, lima, I want to try it :)
<libv> naobsd: so you might want to use those too, as soon as things start trickling out
<libv> naobsd: i also am in the process of creating hacked mesa packages which are a precursor for the lima driver
<naobsd> (NEON part for now)
<naobsd> very nice :) I want to use it on my RK environment too :)
<libv> naobsd: yes
<naobsd> preparing fbturbo dpkg?


<naobsd> gpio section should be for gpio driver
<arokux2> naobsd: do you know what those gpio section could mean?
<naobsd> oh, dmesg said ehci1... I guess ehci2 was 2nd instance
<arokux2> naobsd: wifi is connected to ehci2.
<naobsd> arokux: well, ehci1? do you have any physical USB port connected to ehci1?
<naobsd> arokux: if it works with 3.4, board specific power thing should not be needed because it seems there is no such information in fex