mnemoc changed the topic of #arm-netbook to: EOMA: Embedded Open Modular Architecture - Don't ask to ask. Just ask! - http://elinux.org/Embedded_Open_Modular_Architecture/EOMA-68 - ML arm-netbook@lists.phcomp.co.uk - Logs http://ibot.rikers.org/%23arm-netbook or http://irclog.whitequark.org/arm-netbook/ - http://rhombus-tech.net/
kaspter has quit [Quit: Instantbird 1.1]
kaspter has joined #arm-netbook
P4R4N01D has quit [Quit: Konversation terminated!]
<furan> I love that you can't brick these things
DEAT_ is now known as DEAT
destinal-away is now known as destinal
bobbio has joined #arm-netbook
orly_owl has quit [Ping timeout: 246 seconds]
orly_owl has joined #arm-netbook
ka6sox is now known as ka6sox-away
destinal has quit [Quit: Bye]
destinal has joined #arm-netbook
ka6sox-away is now known as ka6sox
destinal has quit [Changing host]
destinal has joined #arm-netbook
bobbio` has joined #arm-netbook
bobbio has quit [Ping timeout: 246 seconds]
bobbio` has left #arm-netbook [#arm-netbook]
ibrah has quit [Remote host closed the connection]
dfletcher_ has quit [Changing host]
dfletcher_ has joined #arm-netbook
dfletcher_ is now known as drgreenthumb
HeHoPMaJIeH has joined #arm-netbook
tzafrir_laptop has quit [Ping timeout: 252 seconds]
Turl has quit [Ping timeout: 272 seconds]
Turl has joined #arm-netbook
menomc has joined #arm-netbook
lundman has quit [*.net *.split]
mnemoc has quit [*.net *.split]
specing has quit [*.net *.split]
termleech has quit [*.net *.split]
arete74 has quit [*.net *.split]
robws has quit [*.net *.split]
specing has joined #arm-netbook
termleech has joined #arm-netbook
arete74 has joined #arm-netbook
robws has joined #arm-netbook
tzafrir_laptop has joined #arm-netbook
killerkink has joined #arm-netbook
<killerkink> for those of you who are interested in cwm touch support on a10
<ZaEarl> killerkink, nice. what touchscreens have you tried it on? does dragging work right? our goodix driver only supports the old dragging api
<killerkink> this is not my work, i just forward port to cm9 and a10. :)
<killerkink> its not drag based. just icon based(up/down/back/select)
<ZaEarl> no long menus requiring dragging?
<killerkink> no. its just virtual buttons on the screen
Quarx has joined #arm-netbook
<ZaEarl> killerkink, i see. buttons on the side of the touchscreen. works great.
<killerkink> yes, work very well. simple concept
<killerkink> but it does not work on resistive single touch
<killerkink> i got a dirty hack for that but its on another tablet
<killerkink> zaearl: useful for those a10 without hard buttons
<ZaEarl> even with hard buttons, a lot of them are poorly configured.
lundman has joined #arm-netbook
Triffid_Hunter has quit [Excess Flood]
Triffid_Hunter has joined #arm-netbook
<killerkink> true. with this it can be use across all a10 tablet
<killerkink> without the need to compile the recovery keys for each
<killerkink> exception is to load the correct touch driver
just4dos has joined #arm-netbook
killerkink has quit [Quit: Page closed]
<lundman> wooo
P4R4N01D has joined #arm-netbook
Vayu has joined #arm-netbook
<CIA-121> rhombus-tech: Nino master * r045df9c84ff3 /allwinner_a10/orders/Ninux.mdwn:
bobbio has joined #arm-netbook
<bobbio> ls
hipboi has joined #arm-netbook
<RaYmAn> hi
<hipboi> hi
<lundman> lo
<menomc> lo
Quarx has quit []
Quarx has joined #arm-netbook
<hipboi> Hardkernel shipped my odroid today
bobbio has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
<RaYmAn> Cool!
<menomc> hipboi: what about the sexy cubieboards? :)
hipboi has quit [Quit: Leaving]
cubieboard has joined #arm-netbook
<cubieboard> what?
<cubieboard> who is calling me?
<cubieboard> it will be available by the end of this month
<cubieboard> second revision of the hardware
<menomc> \o/
<RaYmAn> cubieboard: are you familiar with JTAG on a10/a13 using that microsd breakout board? :)
<cubieboard> yes
<cubieboard> take a look at the script.fex
<RaYmAn> cubieboard: how do I go about using it? :S hehe. There is a lot of unnamed pins on the 14 pin connector
<cubieboard> RaYmAn:
<cubieboard> [jtag_para]
<cubieboard> jtag_enable =1
<cubieboard> jtag_ms = port:PB14<3>
<cubieboard> jtag_ck = port:PB15<3>
<cubieboard> jtag_do = port:PB16<3>
<cubieboard> jtag_di = port:PB17<3>
<cubieboard> just set the port to the value and jtag is configed
<RaYmAn> okay, so disable sdcard, change jtag to be on PF
<RaYmAn> cubieboard: do you have any idea what pins are what on the microsd breakout 14-pin connecter? I'm assuming the included 2x2pin 0.05" pitch to 4pin 0.1" pitch cables are sufficient? :)
<cubieboard> the 14pin connector is standard arm jtag header
<cubieboard> many arm jtag debbger uses it
<RaYmAn> fair enough - I'm kind of lacking any standard connectors, so planning to hook it up with jumper cables :)
<RaYmAn> s/connectors/cables/
<ibot> RaYmAn meant: fair enough - I'm kind of lacking any standard cables, so planning to hook it up with jumper cables :)
Quarx has quit []
<Mazon> cubieboard with sata ? - or is that breakout sata ?
<cubieboard> what is breakout sata?
<Mazon> it says: 96 extend pin including i2c, spi, lcd, sata, sensors, ..
<cubieboard> there are picture there
<Mazon> looks very neat
<Mazon> just not seeing any sata connector ?
<menomc> cubieboard: who does setup jtag with that? I don't see any reference in the kernel
<cubieboard> in the bootloader i think
<menomc> cubieboard: does the stock bootloader read the script.bin file?
<menomc> uhm
menomc is now known as mnemoc
<mnemoc> Mazon: http://linux-sunxi.org/images/2/24/Cubieboard.jpeg left side above the extesnion header
<specing> mnemoc: you mean right?
<hno> mnemoc, there is jtag config data in boot0, and I think boot1 also parses it from script.bin.
<mnemoc> specing: yes, sorry. right :)
<hno> hm.. does not seem so.
<Mazon> mnemoc, aha, thanks - for some reason I thought that was another usb - great then
<mnemoc> Mazon: afaik in rev.B that will be esata, to solve the power problem
<hno> cubieboard, do you have any uSD breakout boards left?
<cubieboard> cubieboard will be sold with uSD breakout board
<cubieboard> there is no jtag port on cubieboard
<cubieboard> but there are two sd slot, one can be used as jtag
<mnemoc> cubieboard: I think he wants one for hacking his current devices :)
<mnemoc> cubieboard: the break out board I mean
<cubieboard> yes, hno
<hno> A cubieboard with uSD would solve that as well :)
<hno> s/uSD/uSD breakout/
<ibot> hno meant: A cubieboard with uSD breakout would solve that as well :)
<cubieboard> hno: no problem
cubieboard has left #arm-netbook ["Leaving"]
cubieboard has joined #arm-netbook
<mnemoc> hno: do you know if your fel-boot works in A13 devices?
<hno> mnemoc, it should.
<mnemoc> i'll find another card then
<mnemoc> still no luck :<
<mnemoc> i starting to believe this uSD slot is bad
<RaYmAn> mnemoc: got a link to a fel-boot .img that I can dd to an sd card and try on my a13 devices?
<mnemoc> 1m
<mnemoc> sha1 4a7d346550df1111eac06f7958cca7765036d23b
<RaYmAn> bah, assuming I can find a working microsd, lol
<mnemoc> :)
<hno> www.hno.se/code/A10/ always have a copy
<RaYmAn> 'k, ty
<mnemoc> same sha1 as mine
* RaYmAn waits for backup of microsd to finish
<Mazon> esata > sata, for my usage
<Mazon> looks great
<hno> esata is sata + slightly other connector + hotplug functionality required on the host for compliance.
<Mazon> oh, thought it was ONLY a connector thing
<phh> well hot host handles hotplug
<phh> s/hot/all/
<ibot> phh meant: well all host handles hotplug
<phh> and also esata has wider voltage support for the signals
<mnemoc> Mazon: esata -> sata + power .... there are some very cheap adapters for that usage
<phh> mnemoc: not according to the spec.
<mnemoc> phh: it was added later
<phh> mnemoc: i meant it's not just that
<phh> it's way more than that.
<mnemoc> sure, it's hotplug too
<mnemoc> i suppose it will enable multipliers too...
<mnemoc> for my use case the only but is how to power the drive without another PSU... and esata solves that, so that is what I value most of that interface :)
* mnemoc very sad about having a dead uSD slot on the a13-olinuxino :<
<RaYmAn> did you check script.bin setup, just in case?
<RaYmAn> I dunno how hte low-level boot stuff works but..
<mnemoc> this is far earlier than uboot or script.bin
<WarheadsSE> afaik, then low level doesnt read the bin @ boot
<RaYmAn> I know, but still worth verifying
<mnemoc> RaYmAn: this is the whole bootloader code: https://github.com/amery/sunxi-tools/blob/master/fel-boot.c
<mnemoc> ((void (*)(void))0xffff0020)();
<RaYmAn> mnemoc: yeah, I know
<mnemoc> I'll try to get uart on android tonight
<RaYmAn> Sometimes it's good to assume everything you know is wrong ;)
<mnemoc> =)
<WarheadsSE> aye
<hno> mnemoc, could be wiring error. uSD have not been verified yet.
<hno> you could play with pio to check the pinout.
<mnemoc> yes, good idea, I'll make a dump on my next break
<hno> dump?
<mnemoc> pio register
<hno> WarheadsSE, BROM do not read script.bin. It only probes the available boot devices in a predefined order (SDC0, NAND, SDC1, SPI)
<WarheadsSE> right, spl doesnt check the dram timings either though
<hno> mnemoc, you need a little more than just dumping the registers. Alsn need to set them to input.
<hno> WarheadsSE, check against what?
<WarheadsSE> against, the .bin.. idk
* WarheadsSE is still pissed @ gooseberry
<WarheadsSE> We're pretty sure is a dram timing issue causing the reset
<mnemoc> hno: yes, I know, but first need the dump to see it's initial state ;-)
<mnemoc> hno: and to pass to `pio`
<hno> WarheadsSE, what DRAM does it have? And do you have an matching android image for it?
<WarheadsSE> it has an image in nand
<WarheadsSE> I haven't spent a ton of time on it yet
<mnemoc> isnt' that the board of the ly-f1?
<WarheadsSE> and I dont have the board in front of me at the moment
<WarheadsSE> A721
<WarheadsSE> yes, though, very much almost identical
<hno> WarheadsSE, do you have UART?
<WarheadsSE> yeah
<WarheadsSE> brb
<WarheadsSE> back, yes we do
<WarheadsSE> But the device is not in front of me atm
<WarheadsSE> its a home, i am at the office
just4dos has quit [Quit: Leaving]
just4dos has joined #arm-netbook
just4dos has quit [Remote host closed the connection]
just4dos has joined #arm-netbook
<hno> mnemoc, default A13 PIO is in github.
<mnemoc> hno: uh, received your board??
cubieboard has quit [Quit: 暂离]
devxdev has joined #arm-netbook
devxdev has quit [Quit: ...]
<mnemoc> hno: does your uSD work?
<WarheadsSE> hno: why ask about the wart? get console & dump nand?
<RaYmAn> mnemoc: hmm, should one usually completely remove the mux definitions when disabling a module or is it sufficient to just set it to unused/disabled?
<mnemoc> RaYmAn: there is no such thing as "disabled" in the PIOs
<RaYmAn> ok, so I need to set anything using the same pins to another port
Vayu has quit [Ping timeout: 268 seconds]
<mnemoc> they'll remain as inputs if you don't touch them
<RaYmAn> well, what I mean is. I have the PF00,01,03 and 05 set in the JTAG part, but they are also used elsewhere, so that would cause issues
<RaYmAn> I can't seem to find the wiki page describing how to get e.g. UART over microsd :/
<mnemoc> you'll need to relocate the other feature or disable it, setting foo_used to 0
<RaYmAn> I was so sure I saw one at some point but..
gimli has joined #arm-netbook
<mnemoc> RaYmAn: search for hacking A10 devices on elinux.org
<RaYmAn> right, that's what I was forgetting - elinux.org :P
<RaYmAn> thanks
<mnemoc> RaYmAn: and the original wiki page in RT's wiki about hacking the mele
<mnemoc> yw
<RaYmAn> mnemoc: any idea what the last three params in the port block is? The default default defeault, and sometimes <1> or <0>
QingPei has joined #arm-netbook
<mnemoc> function, pull up/down, drive level and value
<mnemoc> pull up/down is boolean, enable-disable
<mnemoc> value is boolean too, but only meaningful when function is output
<mnemoc> drive level I think goes up to 3
<RaYmAn> perfect, thanks
<mnemoc> yw
<mnemoc> RaYmAn: btw, it would be awesome if you could add some of this to the linux-sunxi wiki :)
<RaYmAn> some of what? setting up uart? or the jtag stuff
<mnemoc> both :p
<RaYmAn> bah :P
<RaYmAn> that sounds like work
<mnemoc> ^^
Quarx has joined #arm-netbook
tzafrir_laptop has quit [Ping timeout: 276 seconds]
P4R4N01D has quit [Ping timeout: 246 seconds]
P4R4N01D1 has joined #arm-netbook
just4dos has quit [Quit: Leaving]
just4dos has joined #arm-netbook
Quarx has quit []
QingPei has left #arm-netbook [#arm-netbook]
tzafrir_laptop has joined #arm-netbook
arokux_h has joined #arm-netbook
<CIA-121> rhombus-tech: John master * redf25206b347 /allwinner_a10/orders/jguy.mdwn:
P4R4N01D1 has quit [Quit: Konversation terminated!]
pwhalen has quit [Ping timeout: 264 seconds]
pwhalen has joined #arm-netbook
gimli has quit [Quit: Verlassend]
<RaYmAn> success! Of some sort anyways
<RaYmAn> jtag> detect
<RaYmAn> IR length: 4
<RaYmAn> Chain length: 1
<RaYmAn> Device Id: 01001011101000000000010001110111 (0x4BA00477)
<rz2k> omg
<RaYmAn> now, If I just had any clue how to actually use jtag ;)
<rz2k> document that on wiki please
<rz2k> connect gdb server
<rz2k> if you have one
<RaYmAn> doesn't that require it having some sort of code?
<RaYmAn> lol
<RaYmAn> this is using urjtag
<rz2k> also you can read/write memory directly
<rz2k> I remember I had good article aboutt kernel debugging with j-link, I will try to find it tomorrow, now sleep.
<rz2k> good night
<RaYmAn> jtag> peek 0x00004000
<RaYmAn> error: illegal state: Bus missing
<RaYmAn> so somehting isn't quite right
<RaYmAn> night
eFfeM has joined #arm-netbook
<mnemoc> RaYmAn: kudos!
<RaYmAn> apparently it's important to connect VDD as well :P
<RaYmAn> when dealing with JTAG
<mnemoc> :D
<mnemoc> wiki wiki :)
<RaYmAn> yeah
<RaYmAn> I'm not sure it'd be very easy without the microsd breakout though
<RaYmAn> though, I guess I could test that, heh
<mnemoc> first document the easy path ;-)
<RaYmAn> I think I managed to flash a broken script.bin, so this actually seems to be default behavior
<RaYmAn> I will document it (i have a few notes =P), but I would like to uhm, confirm it actually works properly first
<RaYmAn> which is hard when you are mostly guessing when it comes to jtag :P
<mnemoc> :D
nibb__ has joined #arm-netbook
nibb_ has quit [Read error: Connection reset by peer]
<RaYmAn> mnemoc: isn't there a way to make those code sections nice? It looks kind of silly that each line gets "highlighted" by itself: http://linux-sunxi.org/JTAG
* Turl asks wikipedia-knowledgeable buddy
<mnemoc> RaYmAn: we have syntaxhighlighter_3.0.83
<mnemoc> RaYmAn: but no 'ini' highlighter available
<RaYmAn> yeah, I can live with that, just, it looks bad that there is seperate "boxes" around each line
<RaYmAn> highlight was a poor choice of word
<mnemoc> there is documentation about how to implement grammars
<Turl> you can just pad the stuff
<Turl> like
<Turl> blah
<Turl> some code
<Turl> blah blah
<Turl> some more
<Turl> it'd make a quote like box
<mnemoc> RaYmAn: use <pre> for multiline
<Turl> or use <source> which doesn't seem to work
<mnemoc> quotes work too
<RaYmAn> yay, pre does it
<RaYmAn> much better
<RaYmAn> :)
<mnemoc> the highlighter extends <pre>
nibb__ has quit [Read error: Connection reset by peer]
<mnemoc> we really need a ini grammar
nibb__ has joined #arm-netbook
<Turl> mnemoc: I can write one
<Turl> where's the docs mnemoc?
<Turl> on ini grammars
<mnemoc> [section]
<mnemoc> ; comment
<mnemoc> key = value
<mnemoc> very simple :)
<Turl> mnemoc: someone already did apparently :P https://github.com/borisguery/shBrushIni/blob/master/shBrushIni.js
<Turl> code reuse ftw ;)
<RaYmAn> interestingly enough, it actually leaves room for using serial at the same time (over microsd port)
<mnemoc> Turl: let me try it
dyoung is now known as dyoung-away
dyoung-away is now known as dyoung
<mnemoc> f*, destroyed my script to generate the .php :<
eFfeM has quit [Quit: Leaving.]
<mnemoc> Turl: works :)
<mnemoc> RaYmAn: use `<pre class="brush: ini">`
<RaYmAn> uuuh, fancy, but not quite correct :P
<RaYmAn> it seems to be stuffing in a ; after <
<mnemoc> it seem we need a Fex :<
<mnemoc> seems*
<mnemoc> Turl: want to do it? :)
<mnemoc> or fix this ini
<RaYmAn> is < not support in ini? :P
<RaYmAn> seems odd
<Turl> RaYmAn: it is
<Turl> but ; == comment
<RaYmAn> there's no ; in there
<RaYmAn> the highligter is inserting it
<Turl> yeah there is
<Turl> <;4
<mnemoc> not in the source
<RaYmAn> exactly
<mnemoc> no idea what's injecting it
<Turl> mght be a derpage on &lt; and &gt;
<mnemoc> good catch
<RaYmAn> heh
<RaYmAn> surprising it still manages to display the < correct
<Turl> mnemoc: try this
<Turl>         regex : /;.*/g,
<Turl> to
<Turl>         regex : /^;.*/g,
<Turl> that's the comment regex
<mnemoc> done
<mnemoc> but that's cheating :)
<Turl> ;)
<Turl> you could use         regex : /(^|\s);.*/g,
<mnemoc> sounds like a post from God^WRob Pike about the origin of dot files
<mnemoc> instead of using strcmp to exclude "." and ".." they were lazy and made name[0] != '.'
<Turl> heh so that's the origin of dot files? :)
<Turl> what about file~ ?
<mnemoc> file~ is vi's
<mnemoc> err
<mnemoc> .oO
<Turl> 'filename~' is the standard way of hiding files on GUI viewers
<Turl> or 'standard' I guess
<Turl> :P
* mnemoc didn't know that
<mnemoc> but I use $f~ for temp files in scripts
<RaYmAn> hm, I think we are going to need some jtag info from tom to get anywhere with this
<ZaEarl> ~ is for a backup of the edited file
<RaYmAn> guessing doesn't seem like a sane way to do it
<Turl> mnemoc: there's mktmp & family :)
<mnemoc> ZaEarl: yes, but the doubt is who made it first
<mnemoc> Turl: nah, $f~ is easier when composing a $f which you only want to replace when it's finished :p
<Turl> mnemoc: (18:49:55) Turl: you could use         regex : /(^|\s);.*/g,
<Turl> less cheating :P
<mnemoc> Turl: that's already active ;-)
<RaYmAn> still issues but:
<RaYmAn> Info : JTAG tap: sun4i.dap tap/device found: 0x4ba00477 (mfg: 0x23b, part: 0xba00, ver: 0x4)
<RaYmAn> Info : sun4i.cpu: hardware has 6 breakpoints, 2 watchpoints
<RaYmAn> then loads of errors :/
<Turl> did you write the sun4i yourself
<RaYmAn> yeah
<Turl> ah
<Turl> ver 0x4 makes sense though
<Turl> :P
<RaYmAn> that device id seems to be used by a lot of devices
<RaYmAn> so
<Turl> RaYmAn: maybe your uSD connector has logic on your own?
<Turl> on its own*
<RaYmAn> doubtful
<RaYmAn> it's pretty simple
<RaYmAn> at most it has a few resistors
<Turl> yeah that kind of stuff
<Turl> maybe your bus thingy doesn't like it
<RaYmAn> oh, nah, this is the "real" microsd breakout
<RaYmAn> with a 14-pin jtag connector
<RaYmAn> so it should be fine
<RaYmAn> I think it's just that you can't really autodetect most of jtag stuff :/
<Turl> 14 pin?
<Turl> out of a uSD? o.O
<RaYmAn> lol
<Turl> doesnt uSD have like 5 pins?
<RaYmAn> only TDI, TMS, TCK and TDO are used here
<RaYmAn> VDD and GND comes from "regular" microsd stuff
<Turl> ah, like half of it are grounds :)
<RaYmAn> > sun4i.cpu arp_halt
<RaYmAn> target state: halted
<RaYmAn> cpsr: 0x60000153 pc: 0xffff01f4
<RaYmAn> MMU: enabled, D-Cache: disabled, I-Cache: disabled
<RaYmAn> target halted in ARM state due to debug-request, current mode: Supervisor
<RaYmAn> woah :D
<WarheadsSE> woo
<mnemoc> RaYmAn: http://linux-sunxi.org/JTAG/sun4iconfig <--- with line numbering :) see our wiki isn't THAT dully? ;-)
<RaYmAn> hehe
<mnemoc> (and doesn't break copy&paste
<RaYmAn> not bad :)
<RaYmAn> so uhm
<RaYmAn> any good registers to dump to verify it works?
<WarheadsSE> Can the uboot-allwinner from hno use uEnv.txt, or does it require boor.scr
<RaYmAn> like, to verify I actually get to dump memory :P
<mnemoc> PIO>
<mnemoc> ?
<WarheadsSE> (aka does it required the wrapped)
<RaYmAn> mnemoc: mm, well, it's 90% zeroes, isn't it?
<RaYmAn> > sun4i.cpu mdw 0x1c20800
<RaYmAn> 0x01c20800
<RaYmAn> 00000000
<RaYmAn> so I need to calculate the offset and stuff ,P
<hno> WarheadsSE, it reads environment from raw blocks, and boot.scr from the filesystem.
<mnemoc> script.bin will have lots of text
<RaYmAn> I'm not sure script.bin is loaded at this point
<hno> actually it's the default environment that have commands for reading boot.scr in the default boot action.
<WarheadsSE> hno: yes, I am just checking for the capability of non-wrapped uboot script
<mnemoc> RaYmAn: hasn't "booted" yet?
<WarheadsSE> without digging into the code..
* WarheadsSE is being lazy
<RaYmAn> mnemoc: well, I happened to break script.bin at the same time I think
* WarheadsSE also currenly laid up with back problems
<mnemoc> RaYmAn: :)
<hno> RaYmAn, if you have memory access then try dumping FFFF4000
<mnemoc> BROM?
<hno> Yes.
<RaYmAn> nope, that's not working
<mnemoc> WarheadsSE: no, only boot.scr or raw env
<RaYmAn> hno:
<RaYmAn> > sun4i.cpu mdw 0xFFFF4000
<RaYmAn> Address translation failure
<RaYmAn> so probably missing some setup
<hno> RaYmAn, Linux running?
<RaYmAn> well, register dumping works at least:
<RaYmAn> > sun4i.cpu mdw 0x1C208B4
<RaYmAn> 0x01c208b4
<RaYmAn> 00404044
<WarheadsSE> mnemoc: by choice, or by code? hno?
<RaYmAn> that'st he same as the expected value for PF PIO setup
<hno> WarheadsSE, what?
<RaYmAn> hno: no, as I said, I kind of broke that :P
<RaYmAn> before I got it working, lol
<RaYmAn> it's actually in fel mode right now
<hno> RaYmAn, odd.. should be no address translation then.
<mnemoc> WarheadsSE: afaik uEnv.txt is deprecated
<WarheadsSE> idk mnemoc im not deep in uboot. I know it works for some devices, including the RPi & beaglebone
<WarheadsSE> thats just an example form a patch for beaglebone
<mnemoc> WarheadsSE: seduce hno
<RaYmAn> I have no idea how to bring the cpu out of halt again :/
<mnemoc> in gdb it should be `c` (continue)
<RaYmAn> in telnet
<RaYmAn> lol
<mnemoc> .oO(telnet?)o
<RaYmAn> openocd has a telnet interface
arokux_h has quit [Remote host closed the connection]
<WarheadsSE> ah
<mnemoc> i see
<hno> WarheadsSE, should be easy to add uEnv.txt support.
<mnemoc> RaYmAn: openocd documentation suggests `resume`
<WarheadsSE> hno: I agree, I just haven't dug in and wanted to check
* WarheadsSE is literally laid out atm..
<WarheadsSE> stupid back
<mnemoc> typing to the ceiling?
<hno> I think the version we use do support "env import"
<traeak> WarheadsSE: no special laptop rigging so you can work on your back?
<WarheadsSE> just a chromebook propped on one knee
<traeak> chromebook ?
<traeak> hope that was free :-p
<WarheadsSE> it works for me since most of my work is done remotely on ARM devices via ssh
tuliom has joined #arm-netbook
<RaYmAn> hno: with linux at least partially running (not sure what breaks - but script.bin issue) - > sun4i.cpu mdw 0xFFFF4000
<RaYmAn> 0xffff4000
<RaYmAn> ea000006
<mnemoc> strange you couldn't access it earlier
<RaYmAn> well, there were some weird issues tbh
<RaYmAn> mmu stuff
<RaYmAn> at a later point, I suddenly couldn't access 0x00004000 at all, but 0x40000000 worked fine
<mnemoc> RaYmAn: I guess if you enable ramconsole you can see the kernel log within openocd
<RaYmAn> so it must be mmu
<hno> WarheadsSE, ok, think I got uEnv.txt support in now. Would you be able to test?
<WarheadsSE> lol.. I just was putting it in myself :P
* WarheadsSE hits pull & pull
<WarheadsSE> lmao
<hno> pushed
<WarheadsSE> And yes, I should be able to spend a few minutes at my desk with a serial console
<hno> I kept boot.scr support as well.
<WarheadsSE> k
<WarheadsSE> slightly derpy.. make sun4i right
nikk has joined #arm-netbook
<WarheadsSE> hmm
<WarheadsSE> you have #define CONFIG_BOOTCMD 2x
<WarheadsSE> I'll see if it stays happy
<hno> fixed.
<WarheadsSE> again, slightly derpy in the memory area (meds) make command: make sun4i ?
<hno> make CROSS_COMPILE=toolchainprefix sun4i
<WarheadsSE> local :)
<WarheadsSE> :O
<WarheadsSE> odd, barfed on me about no rule to make crc32.c
<WarheadsSE> o0 how out of it am I
<hno> There is lib/crc32.c
<WarheadsSE> No rule to make target `/root/uboot-allwinner/uboot-allwinner/lib/crc32.c', needed by `crc32.o'. Stop.
<WarheadsSE> oh, ass.
<WarheadsSE> it didnt see the directory move.
* WarheadsSE facepalm
<hno> move?
<WarheadsSE> < local fs.
<hno> "make distclean"?
<WarheadsSE> ty
<hno> RaYmAn, that's right for BROM.
<hno> RaYmAn, if 0x40000000 works then kernel have been started and MMU set up.
<RaYmAn> hno: in fel mode, I'm getting a lot of Error: Address translation failure
<RaYmAn> so I guess something is weird
<hno> Wonder if sunxi have an debug DCC channel.
<hno> RaYmAn, that's wierd indeed. There is no MMU enabled in fel mode.
<RaYmAn> target halted in ARM state due to debug-request, current mode: Supervisor
<RaYmAn> cpsr: 0x60000153 pc: 0xffff01ec
<RaYmAn> MMU: enabled, D-Cache: disabled, I-Cache: disabled
<RaYmAn> it claims there is
<hno> Hm.. odd.
<RaYmAn> maybe it's just setup differently
nikk has quit [Quit: there is intelligent life out there...]
<hno> RaYmAn, maybe it does enable MMU in fel. There is quite a bit of cp15 magics in the fel setup routine.
<mnemoc> helpful :)
<RaYmAn> I *seem* to be able to single-step through the pio-to-sram function
<RaYmAn> so that part works, heh
<RaYmAn> (and loading the function to sram)
<hno> you loaded it via jtag?
<RaYmAn> yes
<hno> very good
nibb__ has quit [Read error: Connection reset by peer]
<RaYmAn> I'm assuming gdb isn't just pretending to step through it
<hno> what device are you running on?
<RaYmAn> mk802
<RaYmAn> mainly because I had it around, and I don't care too much if I broke it
<hno> do you still have u-boot in nand? Would be very interesting to see how it reacts on a normal boot
<RaYmAn> I do, but I need to uhm, reflash or something
<RaYmAn> the script.bin has broken or something
<RaYmAn> at least, it stopped showing up in lsusb/adb
<RaYmAn> I don't have it connected to any screen or anything
<RaYmAn> just usb cable
<Turl> livesuit? :P
<hno> Right. You may even need to build an jtag enabled image so you have an boot0 that enables jtag.
<WarheadsSE> hno: perfectly worked.. i had to run commands by hand cause i forgot to wipe my env on the sdcard
<RaYmAn> hm
<hno> and stop it at u-boot prompt.
<RaYmAn> Turl: yeah - I need to reboot into windows then :/
<RaYmAn> hno: jtag is enabled at boot, so I really just need something that doesn't disable it :D
<hno> No idea what boot0 does if jtag is not enabled.
<hno> in boot0
<RaYmAn> well, it seems to work at least partially if I boot "normally"..e.g. I can halt still
<Turl> RaYmAn: you don't need to 'reboot'
<hno> you should be able to reset the ARM core also.
<Turl> livesuite works good on a vm :)
<ZaEarl> Turl, have you tried building a LiveSuit img?
<ZaEarl> It would be really nice to have a full CM9 LiveSuit img.
<Turl> nope
<RaYmAn> Turl: I don't have a windows vm
<WarheadsSE> hno: any thoughts on why this gooseberry is being stupid?
<Turl> ZaEarl: a dd-able flashing SDCard might be better
<ZaEarl> that would be great too
<Turl> esp. if we need repartitioning
<Turl> mnemoc: btw how's repartitioning going? you were making a tool I think :)
<hno> WarheadsSE, because it's a tablet?
<WarheadsSE> hehe
<WarheadsSE> I mean, we cant get this uboot to work on it at all
<WarheadsSE> well, it *sort of* works, then just craps/reboots whenever it wants to
<Turl> mnemoc: once your tool works I'll make a bootable repartition+flashing sdcard
<mnemoc> Turl: not mine, but the tool is in sunxi-tools
<Turl> and it works? :)
<mnemoc> afaik
<Turl> it is to be used in device?
<mnemoc> author claims to have used it :)
<mnemoc> as long as you don't touch nanda everything is claimed to wrok
<mnemoc> work
nibb__ has joined #arm-netbook
<Turl> mnemoc: anything needed on kernel land?
<Turl> I suppose you need to export the full nand as a device?
<mnemoc> yes, but that's also merged
<Turl> great
<RaYmAn> hno: can you break the builtin u-boot if you have a serial console?
<hno> RaYmAn, on mele you can.
<WarheadsSE> yeah, on mele
<hno> mnemoc, should not be a problem to mess with nanda.
<Turl> mnemoc: were the build sd card docs on RT or linux-sunxi?
<Turl> I need the uboot steps :P
<RaYmAn> elinux I think
<mnemoc> Turl: uboot steps are in uboot's wiki in github
<hno> and copied with various degrees of errors everywhere else..
<mnemoc> :)
<Turl> which uboot wiki? hno's, mnemoc's? :P
<mnemoc> Turl: did you see the description of amery/uboot ?
<Turl> no
<Turl> I just recall there's those two
<mnemoc> "please use https://github.com/hno/uboot-allwinner instead!"
<mnemoc> mine is merely a mirror
<mnemoc> of hipboi's