<DocScrutinizer05>
freemangordon: if it wasn't for dmesg, you wouldn't even need root, the usb are all in group dialout
arcean has joined #neo900
kerio has quit [Read error: Connection reset by peer]
kerio has joined #neo900
sparetire has quit [Quit: sparetire]
paulk-aldrin has joined #neo900
<freemangordon>
DocScrutinizer05: great!
<freemangordon>
DocScrutinizer05: we should agree on onenand layout, or define a way for xloader to find the 2nd
<freemangordon>
opensuse? never touched it before
Pali has joined #neo900
<freemangordon>
Pali: hi!
<Pali>
freemangordon: hi!
<freemangordon>
Pali: what 0xFFFF needs to flash onenand?
<Pali>
0xFFFF does not support flashing "on device"
<freemangordon>
Pali: no, no, via usb
<Pali>
it uses nolo protocol over usb
<freemangordon>
Pali: so, I should load a piece of SW in SRAM that implements that nolo proto?
<DocScrutinizer05>
that's what coldflash is doing
<Pali>
yes, if you want to use current 0xFFFF
<freemangordon>
Pali: also, do you have an idea how xloader finds the 2nd stage loader?
<Pali>
via magic number I think
<freemangordon>
it parses the whole flash?
<Pali>
2nd nolo has magic header
<Pali>
you are flashing both xloader and nolo image at one time which is stored to mtd0
<freemangordon>
ok, but how 1st knows which offset on nand to read the 2nd?
<Pali>
wait
<DocScrutinizer05>
and end of $self
<Pali>
there is offset for it
<freemangordon>
:nod:
<Pali>
xloader has hardcoded offset from mtd0
<Pali>
err seconadary nolo has
<Pali>
it
<Pali>
so maybe also xloadader has it too
<freemangordon>
Pali: BTW, in case you need it, xloader is *NOT* loaded from offset 0x40208000. or rather the file is not loaded on that address
<DocScrutinizer05>
basically 0xFFFF should be able to load second to RAM and flash NAND from it, no matter what's the CPU
<Pali>
in nolo debug console there id mtd0 - xloader, mtd1 - nolo, mtd2 - cal, mtd3 - log, mtd4 - kernel, ....
<freemangordon>
Pali: there is 0x470 bytes header at start of xloader.bin
<Pali>
ok
<freemangordon>
Pali: so you need to load the file at address 0x4020800 - ox470
<freemangordon>
*0x470
<DocScrutinizer05>
hmm?
<freemangordon>
this is for sugned loaders, those for qemu have ony 8 bytes at start (loading address + size in bytes)
<freemangordon>
*signed
<DocScrutinizer05>
I'd really suggest trying to flash "maemo" aka NOLO to BB-xN
<freemangordon>
DocScrutinizer05: that is what I did yesterday, but in qemu
<DocScrutinizer05>
ROMBL should be compatible to N900
<freemangordon>
it is not
<DocScrutinizer05>
o.O
<freemangordon>
I mean
<freemangordon>
n900 1st stage doesnt like our nand id
<DocScrutinizer05>
hmm ok
<freemangordon>
n9 xloader boots, somehow, but can't find the 2nd stage
<DocScrutinizer05>
could we patch that?
<freemangordon>
I guess yes
<DocScrutinizer05>
we don't need to worry much about xloader signature, right?
<freemangordon>
exactly
<freemangordon>
but then n9 xloader checks the 2nd signature, afaik
<DocScrutinizer05>
patch NOLO xloader in fiasco, then coldflash the whole thing to BB-xM with 0xFFFF
<freemangordon>
won't fly, it won;t be able to verify the signatures
<DocScrutinizer05>
yes, N9 xloader must check 2nd stage sig
<DocScrutinizer05>
I meant N900
<freemangordon>
DocScrutinizer05: elaborate please, I am still having my first coffee
<freemangordon>
:)
<DocScrutinizer05>
I'm looking forward to my 1st coffee :-)
<freemangordon>
you mean to patch the n900 xloader and to flash it to bb-xm?
<DocScrutinizer05>
patch N900 xloader to comply with new NAND ID, then coldflash the whole thing to BB-xM_1GB
<freemangordon>
I can use nokia flasher as well for that
<DocScrutinizer05>
dunno, it's checking some device IDs
<DocScrutinizer05>
maybe not during coldflashing though
<freemangordon>
but then, flashing is - load a modified xloader, which implements the nolo flashing protocol, then pass the "real" xloader to it
<DocScrutinizer05>
yes
<DocScrutinizer05>
basically
<DocScrutinizer05>
actually I thnk it loads a 2nd aka NOLO to RAM too
<DocScrutinizer05>
and then NOLO does the flashing
<freemangordon>
I am afraid that fill fail, as the xloader in SRAM won't be able to verify the signature of xloader send via usb. but then, there were some checks if we run on GP or HS device, so it may work
<DocScrutinizer05>
prolly xloader is modified so it loads NOLO to RAM from USB instead NAND
<freemangordon>
Pali knows better :)
<freemangordon>
anyway, I'll try some stuff in qemu first
jonsger has joined #neo900
<DocScrutinizer05>
xloader during coldflashing gets loaded to SRAM by ROMBL. For coldflashing this will be a modified xloader that loads NOLO to SDRAM from USB instead NAND. This will be a modified NOLO that knows how to coldflash. No checksum calulation involved in all this aiui
<freemangordon>
hmm, ok
<DocScrutinizer05>
Pali: do you want an account on our lab remote workbench?
<freemangordon>
DocScrutinizer05: I don;t see 0xFFFF on that machine
<Pali>
DocScrutinizer05: what is on that lab?
<freemangordon>
Pali: BB-xm with 1G
<freemangordon>
:)
<Pali>
oo :-)
<freemangordon>
Pali: what we want to verify is if onenand works
<freemangordon>
it is identified ok, but we were not able to check if writing and reading work as well
<freemangordon>
and it is ~100% sure it is SW problem
<freemangordon>
but still, we should manage to do it, before doc orders quantities of those samsung PoP chips
<DocScrutinizer05>
nevertheless for all I know the rootfs filsesystem is successor(!) of hffs2
<DocScrutinizer05>
jffs2
<DocScrutinizer05>
and that's on top of raw NAND with OOB
<DocScrutinizer05>
aka MTD
<DocScrutinizer05>
aiui
<DocScrutinizer05>
see http://privatepaste.com/22f71fe0f4 - it gives a pretty good idea how stuff works, once you got that OOB is for managing bad blocks
<DocScrutinizer05>
for creating an image with nanddump you want
<DocScrutinizer05>
-o --omitoob Omit oob data
<DocScrutinizer05>
-b --omitbad Omit bad blocks from the dump
<DocScrutinizer05>
this way nandwrite can create new oob info based on what's actually found on the target NAND.
<DocScrutinizer05>
usually it shall consider blocks already marked as bad, and not write to them, but also tag and skip new bad blocks found during writing
<DocScrutinizer05>
generally you don't want to overwrite and thus erase oob data that tags already spotted bad blocks
<DocScrutinizer05>
unless ypou want restore an image of this very NAND after you did a nandtest
alifakhruddin has joined #neo900
<enyc>
interesting, I wonder why the OOB data is externally readably/writable at all, is'n't that just down to the internal NAND controller ??
<DocScrutinizer05>
OneNAND pretends to be NOR
<DocScrutinizer05>
the controller is intentionally not as smart as in usual flash storage
<kerio>
it's barely a controller
<kerio>
DocScrutinizer05: do you know if raw flash is available for normal PCs?
<DocScrutinizer05>
never heard of
<enyc>
always seems to come in the form of an SSD or whatever =)
<kerio>
i reckon the cpu would be faster than a dedicated controller
<enyc>
kerio: that said PC BIOSes are some sort of eeprom-like flash ;p
<kerio>
smarter, too
<enyc>
kerio: is that what the linux MTD drivers are to talk to ??
<kerio>
the MTD drivers deal with raw flash, yes
<kerio>
you need a specialized filesystem, and specialized tools
<enyc>
DocScrutinizer05: so whats all the inetrest in the Nokia flash tool? using the same flash part on neo900?
<DocScrutinizer05>
for now we want to flash the BB-xM
<DocScrutinizer05>
but i'm busy. Sorry
<DocScrutinizer05>
bbl tonight
<DocScrutinizer05>
Social Sunday hours
<DocScrutinizer05>
o/
<enyc>
=)
<alifakhruddin>
So grt to use this device.. My N900 still works after 5 years.. Just that i miss to plugin the headsets.. Some hardware issue.
<kerio>
my wifi doesn't work :(
<alifakhruddin>
Thats lifeless man
<alifakhruddin>
Now m using this as my work phone.. Lolz when i go for the meetings people take out BB Passport and stuff out of their pockets and then this beast arrives.. Hahha mini computer..
<alifakhruddin>
I wish to c Neo900 man.. Nokia did one thing right.. Created this N900..
<Humpelstilzchen>
nokia created n900, right but they never really sold it
<alifakhruddin>
Exactly man.. If they did there would have been a different line of devices like N900 for devs and geeks and Linux fans..
<alifakhruddin>
But luckily the community is the backbone of all this.
<alifakhruddin>
Just imagine N900 without TMO...
<Pali>
ubuntu decided to compile kernel without CONFIG_DEVMEM
<Pali>
which means no /dev/mem support
<alifakhruddin>
Ooooh
* Pali
is going to recompile kernel
<alifakhruddin>
I wonder why that Ubuntu edge project failed.. That device would have changed how the mobile phones r today
<alifakhruddin>
Such a shame
<Humpelstilzchen>
Pali: do they have uio?
<Pali>
uio?
<Pali>
~uio
<Humpelstilzchen>
Userspace I/O, works a bit like /dev/mem but customizeable
<Pali>
do not know it
alifakhruddin has left #neo900 [#neo900]
<Humpelstilzchen>
Basicly you can configure the memory area that is available with a device file. ALso it supports interrupts
che1 has joined #neo900
<Pali>
missing /dev/mem break lot of applications...
* Pali
is going to reboot machine...
Pali has quit [Remote host closed the connection]
Pali has joined #neo900
<Pali>
and /dev/mem is back
che1 has quit [Ping timeout: 264 seconds]
che1 has joined #neo900
alifakhruddin has joined #neo900
alifakhruddin has left #neo900 [#neo900]
<freemangordon>
Pali: just managed to mount rootfs from qemu image :)
<Pali>
anyway, here are commands how to *correctly* mount rootfs extracted from FIASCO image: http://pastebin.com/VNt5wR68
<Pali>
(for n900)
<Pali>
this is correct way
<Pali>
for mtd files (they are character devices with special ioctls) you should use nand*/flash* commands
<Pali>
and not dd or other programs designed for block devices!
<Pali>
mtdblock.ko is there for backward compatibility for programs which using block devices
<DocScrutinizer51>
Pali: freemangordon: is the lab server OK?
<Pali>
I just connected to it
<Pali>
I do not know how to use it for now
<Pali>
any commands what I can do?
<DocScrutinizer51>
ttyUSB0 is BB-xM
<DocScrutinizer51>
you could run 0xFFFF against that port
<Pali>
0xFFFF use USB, not TTY
<DocScrutinizer51>
sorry I wasn't able yet to config a sudo for dmesg
<Pali>
dmesg | tail is working
<DocScrutinizer51>
well, the BB is on USB
<DocScrutinizer51>
oooh
<DocScrutinizer51>
then run relaycard-reset* whatever and watch dmesg -w
<DocScrutinizer51>
you see BB BOOTROM enum
jonsger has quit [Ping timeout: 252 seconds]
<DocScrutinizer51>
see pastebin I psted a few hours ago
<Pali>
now I see probably usb cold flash mode in dmesg
<DocScrutinizer51>
and RS232 is also hooked up to PC serial
<Pali>
and then after few sec it disappear
<DocScrutinizer51>
yes
<DocScrutinizer51>
bootrom disapperas fast
<DocScrutinizer51>
to try boot xloader
<Pali>
where is xloader file?
<Pali>
and how to do that?
<DocScrutinizer51>
I don't have any
<DocScrutinizer51>
ask freemangordon
<DocScrutinizer51>
Nik did it
<DocScrutinizer51>
see GDC
<Pali>
on that server is missing git :-( can you install it?
<Pali>
I cannot clone 0xFFFF
che1 has quit [Remote host closed the connection]
<DocScrutinizer51>
Pali: I install everyting you want. just please send me /query to doc 05 so I dont forget it. Zypper search and info works without root
<DocScrutinizer51>
for zypper install you need root
<mvaenskae>
oh, i just see there is an "incentive" to increase the donation to 300eurs
<mvaenskae>
i didn't get those news D:
nox- has joined #neo900
<DocScrutinizer05>
freemangordon: the BB is USB-powered rght now. Please holler when you encounter weird brownout effects. (I doubt any will occur since 500mA should suffice wor a BB without USB gadgets attached)
<DocScrutinizer05>
mvaenskae: yeah, that's sort of linked to our webshop which got a tiny bit delayed since I been busy with the europa lab server
<DocScrutinizer05>
pans been to officially open up the shop today the latest. will prolly take 2 more days though
<DocScrutinizer05>
plans*
<DocScrutinizer05>
hmmmm
<mvaenskae>
DocScrutinizer05: good thing i did catch up a bit :)
<DocScrutinizer05>
Pali: freemangordon: wpwrak: I'll create a /share dir with 777 for stuff you want to see oublished on neo900.org/stuff
<DocScrutinizer05>
mvaenskae: we'll keep you updated as soon as it gets going
<mvaenskae>
DocScrutinizer05: that is very good :) i assume a final estimated price has also been found with 750 eurs, correct?
<DocScrutinizer05>
sounds reasonable
<DocScrutinizer05>
though we only will know for sure when the devices are ready :-)
<DocScrutinizer05>
but yeah, I still *hope* we can stay in range we originally announced
<DocScrutinizer05>
unlike Hamburg opera and Berlin airport and....
<mvaenskae>
true, but i like to give people estimates when i interest them in the project :)
<mvaenskae>
and tell them to hurry up :D
<DocScrutinizer05>
NB that's for the bare board aka NeoN
<DocScrutinizer05>
but I'm actually 'selling' used N900 in fine condition now
<mvaenskae>
true, i will source myself a nice looking n900 this month :) i cannot use my n900 sadly as the hardware is too broken :(
<mvaenskae>
what price? :)
<DocScrutinizer05>
when you absolutely want, you can get a N900 without NeoN board
<mvaenskae>
DocScrutinizer05: btw, go fetch some food :) i can wait
<DocScrutinizer05>
150.- for the N900
<DocScrutinizer05>
zilch accessories
<mvaenskae>
even battery?
<DocScrutinizer05>
that's atad over the price we pay, but then... this is a business
<DocScrutinizer05>
we haven't sourced batteries so far. I guess I could try to source 500 for 5.-/pce
<DocScrutinizer05>
but then, so can you
<DocScrutinizer05>
ebay is yoir friend
<mvaenskae>
DocScrutinizer05: yeah, i did move in with two study buddies of mine and one is from germany and has paypal :) so i can ask him to bet on some auctions and i will get it then in two weeks :)
<nox->
i guess the problem is finding good ones out of a 1000 fake/poor ones...
<DocScrutinizer05>
however NB we don't take back N900 to upgrade them with NeoN boards. We only offer the upgrade to devices we sourced and never left our facilities
<mvaenskae>
DocScrutinizer05: that will otherwise be a hassle i guess :)
<DocScrutinizer05>
yes, way too much hassle
<DocScrutinizer05>
and double customs
<DocScrutinizer05>
or triple
<DocScrutinizer05>
plus the original one for our sourcing
<DocScrutinizer05>
you better get a ebay N900 then
<DocScrutinizer05>
err, a second N900
<DocScrutinizer05>
it's cheaper than have the one N900 sent arounf 4 times through customs
<DocScrutinizer05>
UPS/DHL alone will be 60 bucks
<DocScrutinizer05>
no way
<DocScrutinizer05>
I can source a 500 N900 easily, I've been told
<mvaenskae>
500 n900 devices?!
<mvaenskae>
holy schnikes
<DocScrutinizer05>
so ~60 more than we would need worst case, when *every* customer would want one
<mvaenskae>
DocScrutinizer05: cool though, that will make the neo900 exist in a full setup
<mvaenskae>
that is very exciting news
<DocScrutinizer05>
that is crucial for the project
<mvaenskae>
indeed, i can finally tell people they can expect a full device
<DocScrutinizer05>
I been very worried about availability of the ultraspecial Nokia parts
<DocScrutinizer05>
now I sleep waaay better than 2 months ago
<DocScrutinizer05>
a megaton of thanks to xiangfu and his company
<DocScrutinizer05>
what a great guy
<nox->
what happened?
<DocScrutinizer05>
nothing new. Just the securing of 500 N900 I mentioned already 4 weeks ago
<nox->
ah
<mvaenskae>
what keyboard layout will they have?
<DocScrutinizer05>
well, dunno
<DocScrutinizer05>
prolly US
<DocScrutinizer05>
unless they come with something different
<useretail>
FYI: Multi-Mode, Multi-Band Spectrum Sensor for Cognitive Radios Embedded to N900
<DocScrutinizer05>
we will try to keep national variants of keymat according to user requests
<freemangordon>
DocScrutinizer05: I was able to download xloader via rs232, by using pserial
<freemangordon>
:)
<DocScrutinizer05>
DUH WHAT?
<DocScrutinizer05>
freemangordon: \o7
<DocScrutinizer05>
:-))
norly has quit [Quit: Leaving.]
norly has joined #neo900
<DocScrutinizer05>
WTF?!?! even a spacer frame
<DocScrutinizer05>
useretail: awesome
<DocScrutinizer05>
ooh, also a megaton to useretail for all the help
<DocScrutinizer05>
+ of thanks
<DocScrutinizer05>
>>Multi-Band Spectrum Sensor for Cognitive Radios << sophisticated stuff, even to me it *almost* sounds like snakeoil
<DocScrutinizer05>
only almost
<DocScrutinizer05>
when it's a hoax then a brilliant one
<DocScrutinizer05>
when not then it's even more brilliant
<DocScrutinizer05>
"Nokia Research Center" -- *sigh*
<DocScrutinizer05>
anybody ever heard of a Microsoft Research Center? I guess there must be one, but you're frequently more happy when you don't hear from it
<DocScrutinizer05>
actually I can't think of anything but kinect that ever came from this direction and been worth a second look
<DocScrutinizer05>
and prolly kinect also been developed by a small innovative company and then simply scooped up by big M$ [SIC]
<DocScrutinizer05>
I guess the 8 bit are incl parity
<freemangordon>
it is, as it was able to download xloader through serial :)
<mvaenskae>
they are using 9bit to represent a byte :D
<DocScrutinizer05>
nomenclature differs there
<nox->
its always 8 bits, just the 8th bit is either data or parity
<nox->
so either 7 + parity or 8 w/o
<DocScrutinizer05>
:nod:
<nox->
iirc
<DocScrutinizer05>
man stty
<DocScrutinizer05>
some UARTS can do s+8+P-stop
<nox->
oh
<DocScrutinizer05>
depends
<DocScrutinizer05>
it's however VERY unusual
<DocScrutinizer05>
at least I never seen it being used
<nox->
*nod*
<freemangordon>
ok, back to qemu :)
<DocScrutinizer05>
standard ASCII is 7bit anyway
<DocScrutinizer05>
you even see in every email
<DocScrutinizer05>
man ascii
sparetire has joined #neo900
<DocScrutinizer05>
>> ASCII is the American Standard Code for Information Interchange. It is a 7-bit code. Many 8-bit codes (such as ISO 8859-1, the Linux default character set) contain ASCII as their lower half. The international counterpart of ASCII is known as ISO 646.<<
<DocScrutinizer05>
pigeons: nice one - and exactly the kind of stuff I'd expect to come from Microsoft Research. Pkus awsome mice in pink now, of course
<DocScrutinizer05>
plus*
<DocScrutinizer05>
with a name that took 9 month to evaluate worldwide - for the mouse
<pigeons>
:)
alifakhruddin has joined #neo900
alifakhruddin has quit [Client Quit]
<kerio>
DocScrutinizer05: can i get a megaton of neo900s
<DocScrutinizer05>
we need to further shrink the detonator first ;-)
<kerio>
no like
<kerio>
can i buy 10^9 kilograms' worth of neo900s
<DocScrutinizer05>
well, I don't think there exist that many
<DocScrutinizer05>
but depending on payment modalities I'm absolutely sure there are ways I could offer and deliver them to you nevertheless
<DocScrutinizer05>
you could even chose display size and case
<DocScrutinizer05>
and we could easily get to a price level of < USD 4000 / kg
<kerio>
how about an IOU
<DocScrutinizer05>
I however wonder how you'll manage to deliver every human being their individual Neo900
<DocScrutinizer05>
since that fits pretty exactly
<kerio>
step 1: build a raft out of neo900s
<DocScrutinizer05>
6 * 10^9
<DocScrutinizer05>
maybe 7
<DocScrutinizer05>
seems US national finances could pay for such project in one year already - given nothing else spoils the plan, like expense for armament/military
arcean_ has joined #neo900
arcean has quit [Disconnected by services]
arcean_ is now known as arcean
jonsger has joined #neo900
<wpwrak>
(partity) seems you're talking about the 3730 ? found 8 + parity in the manual, too. exotic indeed
<freemangordon>
wpwrak: yes
<freemangordon>
wpwrak: 115200 8E1?
<wpwrak>
yes, that's what the manual says
<wpwrak>
(had to look it up - all i remembered was "oh, that's unusual" :)
<freemangordon>
wpwrak: any clue if I can load xloader through usb?
<DocScrutinizer05>
shall I hook up the rigol to see what's actually up there?
<freemangordon>
yeah, good idea
<freemangordon>
but it seems to be what manual says
<freemangordon>
115200 8E1, as pserial deffinitely sends a file to the device
<DocScrutinizer05>
then that's it
<DocScrutinizer05>
will ponder if it's easy to attach scope somehow
<DocScrutinizer05>
but actually the rigol isn't exactly great in analyzing the serial protocol parameters, it's rather a try&error game until the data it shows looks reasonable
<freemangordon>
well, there is *some* activity shown in minicom, unfortunately it is garbage
<DocScrutinizer05>
freemangordon: I noticed stty often throws an error "not all settings could get done" - on first invocation with new parameters. A seconf call with 100% identical parameters dosn't throw such error
<DocScrutinizer05>
serial/uart is a bitch
<wpwrak>
freemangordon: hmm no, don't have any hand-on experience with the chip yet. according to the manual, it should be quite possible, though. now, the question is what protocol they use ..
<DocScrutinizer05>
freemangordon: loading xloader through USB is exactly what the whole coldflashing/serial-boot is made for
<DocScrutinizer05>
that's prolly for BeagleBoard where USER button could get configured by resistor jumpers so when pressing it it switches to alternative SYSBOOT_5
<freemangordon>
ok
<freemangordon>
but I still don't understand why there is nothing on the serial port
<freemangordon>
not romboot, but xloader when started
<DocScrutinizer05>
yep
<DocScrutinizer05>
that's however a question how xloader is configured
vakkov has quit [Ping timeout: 246 seconds]
SylvieLorxu has joined #neo900
<DocScrutinizer05>
freemangordon: hey, I have an unpimped BB-xM. Would that help?
<DocScrutinizer05>
it however has no mod on SYSBOOT_x
<freemangordon>
yeah, it may help. at least I can check if serial functions as it should
vakkov has joined #neo900
<DocScrutinizer05>
shall I swap? - NB remote reset not supported
<freemangordon>
well, do not swap then, I'll play with that one
<DocScrutinizer05>
k
<freemangordon>
the current one that is
<DocScrutinizer05>
I could hook up the other one in parallel, for USB and even reset when I find a way to plug the two wires to P9:27,28(?) or solder them to board as I did on this one
<DocScrutinizer05>
just RS232 I'm out of spare ports for now, until I find that damn USB<->serial adapter
<DocScrutinizer05>
seems I lost it 1h after I bought it, before I even brought it home
<freemangordon>
:)
<DocScrutinizer05>
sometimes things are really cursed, first they gave me a USB<->PS2*2 adapter, the I swap it for a RS232 adapter and that vanishes
<DocScrutinizer05>
anf yes I think UART3 is the one you usually want for this
<freemangordon>
hmm, finally some signs of life :)
<DocScrutinizer05>
wow
<freemangordon>
yeah, got echo in minicom, but not exactly what I type :D
<DocScrutinizer05>
sporadic random errors sound like poor cable
<DocScrutinizer05>
systematic errors are clearly a config issue
<freemangordon>
hmm, looks like there is really some problem with the cable
<freemangordon>
or I can;t find the right settings for rs
<DocScrutinizer05>
how many errors (% of chars) you get?
<freemangordon>
see QW�RTYU��P{}!S4����;�:qw%rtyu9/p1234567890�11111111112222222222
<DocScrutinizer05>
hmmm
<freemangordon>
this is 9600 8E1
<DocScrutinizer05>
is this with echo?
<kerio>
use screen
<freemangordon>
I managed to load something in the device, this is what get echoed
<DocScrutinizer05>
hmm
<freemangordon>
I hope this is not your loopback :D
<DocScrutinizer05>
nope
<DocScrutinizer05>
shall I connect loopback?
<freemangordon>
yeah, now it is supposed to accept kermit file download, but it fails
<DocScrutinizer05>
?
<DocScrutinizer05>
shall I connect loopback?
<freemangordon>
no
<freemangordon>
I am pretty sure your pc rs port works
<DocScrutinizer05>
the loopback includes cable
<freemangordon>
ok, but what we will test?
<DocScrutinizer05>
see if you get what you type
<freemangordon>
ok
<DocScrutinizer05>
mompls
<freemangordon>
yep, echo is fine
<freemangordon>
115200 8N1
<DocScrutinizer05>
looopback attached
<freemangordon>
yeah, saw it
<freemangordon>
everything is echoed back as it should
jonsger has quit [Quit: jonsger]
<DocScrutinizer05>
let me connect a external power supply, might help
<freemangordon>
DocScrutinizer05: what will happen if rx and tx are short on that bb?
<DocScrutinizer05>
errr
jonsger has joined #neo900
<freemangordon>
you said you soldered some wires
<DocScrutinizer05>
no idea
<DocScrutinizer05>
nah, that soldering will not affect RS232
<freemangordon>
ok
<DocScrutinizer05>
one is soldered directly to reset swich pin, the pther to a huge GND pad
<freemangordon>
hmm, what about grounding?
<freemangordon>
I mean - if pc and bb are connected to different outlets
<DocScrutinizer05>
wouldn't know where exactly connect GND to
<freemangordon>
no, see ^^^
<DocScrutinizer05>
BB is not connected to any outlet
<freemangordon>
how it is supplied?
<DocScrutinizer05>
USB
<freemangordon>
and it is connected to the pc?
<DocScrutinizer05>
yes
<freemangordon>
ok, not that one then
<DocScrutinizer05>
but USB might be weak, particularly for RS232
<freemangordon>
could you attach bb instead of loopback?
<DocScrutinizer05>
that's why I said I ,ight connect dedicated PSU
<DocScrutinizer05>
yes, sure
<DocScrutinizer05>
mompls
<freemangordon>
connect where, sorry, don;t know your wiring :)
<DocScrutinizer05>
done
<DocScrutinizer05>
supply BB via 5V barrel jack PSU
<DocScrutinizer05>
is what I thought about
<freemangordon>
ah, got it now :)
<DocScrutinizer05>
actually I just see in schematics that inserting barrel actually switches off USB VBUS supply rail
<Wizzup>
dedicated psu for the bb seems sensible
<DocScrutinizer05>
The TPS65950 is used on the BeagleBoard to provide power with the exception of a 3.3V regulator which is used to provide power to the DVI-D encoder and RS232 driver and an additional 3.3V regulator to power the USB Hub. In addition to the power the TPS65950
<nox->
usb hub gets 3.3v?
norly has quit [Quit: Leaving.]
<DocScrutinizer05>
nox-: sorry?
<DocScrutinizer05>
looks like, yeah
<nox->
".. and an additional 3.3V regulator to power the USB Hub. .."
<nox->
oh hm
<DocScrutinizer05>
dunno where from that thing opened in okular: Beagle_SRM_XM_C2_0_0.pdf
<DocScrutinizer05>
prolly next to the schematics
<DocScrutinizer05>
anyway freemangordon I powered the board from wallwart now
<freemangordon>
ok
jonsger has quit [Quit: jonsger]
<freemangordon>
DocScrutinizer05: I am starting to think that more than mmc iface is broken on that BB
<freemangordon>
maybe after all the idea to connect vanilla BB is not that bad
<freemangordon>
as I have some scripts and bins to load
<freemangordon>
which are supposed to bring u-boot console on serial port
<DocScrutinizer05>
ok
<freemangordon>
will you do it now, or tomorrow?
<DocScrutinizer05>
done
<freemangordon>
thanks
<freemangordon>
lets see
<freemangordon>
I can't reset it via relay, can I?
Pali has quit [Read error: Connection reset by peer]
<DocScrutinizer05>
no
<freemangordon>
could you do ti for me?
<DocScrutinizer05>
sure
<freemangordon>
usb cable is connected too, ain't?
<DocScrutinizer51>
sure
<freemangordon>
"device not accepting address 80, error -71"
<freemangordon>
in dmesg
<freemangordon>
bad cable?
<DocScrutinizer51>
dunno
<freemangordon>
broken device?
<freemangordon>
once again please, download failed this time
<DocScrutinizer05>
changed USB cable
<freemangordon>
yeah
<freemangordon>
:)
<DocScrutinizer05>
?
<freemangordon>
reset it
<freemangordon>
ok, download done
<freemangordon>
lets see
<freemangordon>
is it connected to the same serial port?
<DocScrutinizer05>
yes
<DocScrutinizer05>
of course, there is only one serial port on Europa
<freemangordon>
hmm, ok. could you reset it for the (I hope) last time?
<DocScrutinizer05>
np
<freemangordon>
again not enumerated
<freemangordon>
?
<DocScrutinizer05>
?
<freemangordon>
needs one more reset, it was not enumerated
<DocScrutinizer05>
k
<freemangordon>
seems that one has problematic usb
<DocScrutinizer05>
well, it is: a B device, not reworked on SYSBOOT, and might have older RMBOOT and maybe even some code on NAND
<freemangordon>
nothing happens
<DocScrutinizer05>
I can't recall the other had enumerated
<freemangordon>
sure, but it worked once
<freemangordon>
sure, in dmesg :)
<freemangordon>
it was recognized as USB device
<freemangordon>
see dmesg
<DocScrutinizer05>
I think reset doesn't do the trick
<DocScrutinizer05>
unplugging USB did
<freemangordon>
shit, my ssh session got broken in the meanwhile :(
<DocScrutinizer05>
they are cheap. get a new one
<freemangordon>
ok, lets try it for the last time for today (disconnect the cable). it will either work or not :)
<freemangordon>
Waiting for USB device vendorID=0x451 and productID=0xD00E:
<freemangordon>
Dev #89: Texas Instruments - OMAP3630
<freemangordon>
DDump:Expected to write 4096, actual write -32 Broken pipe
<freemangordon>
send file failed
<freemangordon>
:(
<freemangordon>
lets continue tomorrow
<freemangordon>
seems today is not the day
<DocScrutinizer05>
ok
<DocScrutinizer05>
mow it works fine with reset again
<DocScrutinizer05>
weird thing
<DocScrutinizer05>
hmmm
<DocScrutinizer05>
[111320.742118] usb 1-2.3: new high-speed USB device number 93 using xhci_hcd
<DocScrutinizer05>
vs
<DocScrutinizer05>
[111423.775383] usb 1-1: new high-speed USB device number 94 using xhci_hcd
<DocScrutinizer05>
maybe that USB port on PC is better
<DocScrutinizer05>
must be a USB3 port
<DocScrutinizer05>
if you want, I can press reset button a few more times
arossdotme has quit [Ping timeout: 276 seconds]
arossdotme has joined #neo900
arcean has quit [Read error: Connection reset by peer]
<DocScrutinizer05>
fine that the setup is made for upgrade :-) noth boards are hooked up now. Find two new commands for reset when you search for the 'old' one. RS232 so far linked to 'plain' board. Pinped board will get its own USB<->rs232 adapter in less than 12h
<DocScrutinizer05>
s/noth/both/
<DocScrutinizer05>
pimped: [113673.345432] usb 1-2.3: new high-speed USB device number 106 using xhci_hcd
<DocScrutinizer05>
plain: [113874.191986] usb 1-1: new high-speed USB device number 107 using xhci_hcd
* DocScrutinizer05
looks forward to device number 128 and 256