<mripard>
and you'll need earlyprintk in your bootargs when booting your kernel
<hramrach_>
hmm, meed to grep for those messages, moment
<mripard>
what messages?
<mripard>
this is in menuconfig
<hramrach_>
config messages
<hramrach_>
yes
Renard has joined #linux-sunxi
philipp__ has joined #linux-sunxi
philippe_fouquet has quit [Ping timeout: 258 seconds]
<hramrach_>
ok. I guess I found a problem with the config
<hramrach_>
will try a new kernel. thanks
<mripard>
like what problem?
<hramrach_>
like not having DEBUG_KERNEL
<arokux1>
mrnuke: otg works in 3.4, in case you do not know
<hramrach_>
sometimes
<hramrach_>
mripard: I think there is a config problem there. without DEBUG_KERNEL you get DEBUG_LL option but not the sunxi entry
Guest72468 has quit [Ping timeout: 272 seconds]
bengal has joined #linux-sunxi
deasy has joined #linux-sunxi
<mripard>
hramrach_: I don't know what kernel version you're using, but if I don't select CONFIG_EXPERT here, that in turn selects DEBUG_KERNEL, I don't see DEBUG_LL at all.
<hramrach_>
something like 3.18
naobsd has quit [Ping timeout: 246 seconds]
Guest72468 has joined #linux-sunxi
<mripard>
I'm testing on 3.18 as well.
<mripard>
and DEBUG_LL depends on DEBUG_KERNEL there
<mripard>
so unless there's a bug in Kconfig itself, you cannot really end up in the case you mentionned
<hramrach_>
hmm
<hramrach_>
I reused old config and merged same nand tree bu this should not affect this basic config part
bengal has quit [Ping timeout: 256 seconds]
kurain has quit [Quit: kurain]
kurain has joined #linux-sunxi
<hramrach_>
I have no CONFIG_EXPERT and when I remove CONFIG_DEBUG_KERNEL I no longer have the DEBUG_LL menu. So some old config reuse glitch I guess
kurain has quit [Client Quit]
<mripard>
how do you edit your config?
naobsd has joined #linux-sunxi
<hramrach_>
vim
<hramrach_>
make oldconfig
<mripard>
you realise how broken it is, right?
<hramrach_>
equally broken to the situation when you upgrade to kernel with different features and keep config
<hramrach_>
actually the option you get without DEBUG_KERNEL is different one. SERIAL_EARLYCON_ARM_SEMIHOST
<hramrach_>
so there is no bug. just two different but seemingly similar options
<mripard>
and someone that uses a tool that completely ignores any dependency.
tomboy64 has quit [Ping timeout: 250 seconds]
<hramrach_>
there is oldconfig to fix the dependency
<hramrach_>
given the obscure places AW put the kernel options in 3.4 and how broken the interaction between options was there there was no way other than vim, oldcconfig, vim to check and fixupo, oldconfig until the option stick in the right choice. It should be better with mainline but menuconfig has no search afaik
Renard has quit [Remote host closed the connection]
<hramrach_>
hmm, I have a real problem here now
<hramrach_>
tried to transplant the nand entry into cubieboard1 dts and it does not work
<hramrach_>
I guess I have sunxi-nand-v7 on top of 3.18 or something like that
<libv>
or, loosely translated, answer my question.
<furmuwon>
thank you ^^
<furmuwon>
excuse my poor english...
<libv>
...
<furmuwon>
libv : are you staff of aw?
<libv>
furmuwon: none of us are.
<libv>
furmuwon: read the front page of our wiki.
<libv>
furmuwon: why don't you state what you are _really_ trying to achieve
<libv>
it's called "asking proper questions"
<furmuwon>
sorry i cant understand
<furmuwon>
What's wrong with my question?
premoboss has quit [Quit: Sto andando via]
<libv>
you are trying to achieve something.
<libv>
oh, nm
<libv>
furmuwon: good luck,.
<furmuwon>
Because of curiosity.
<libv>
bs.
<libv>
furmuwon: please contact your allwinner representative.
<furmuwon>
what is the means bs?
<libv>
bullshit.
<furmuwon>
why contact to aw representative?
Renard has joined #linux-sunxi
<furmuwon>
What did I do wrong?
<libv>
a) you do not know how to ask questions to get solutions b) you clearly do not want to tell us what you are trying to achieve c) because you asked whether i worked for allwinner, which only served to cement my suspicion that you are doing something with a commercial intention for which you are not willing to divulge any details
<libv>
furmuwon: so stop wasting the sunxi communities time and go talk to your allwinner representative
Black_Horseman has quit [Remote host closed the connection]
<furmuwon>
Simply just curious..
<libv>
bs again.
<furmuwon>
libv : If you feel bad, i am soory
<furmuwon>
sorry
<jmcneill>
wow
<hramrach_>
furmuwon: boot0 is stored at start of nand. if you disassembled and understood the brom you would know that
<libv>
furmuwon: i do not feel bad at all, i just read your backlog and knew what you were about, and when you were given a chance to get to the issue you are trying to solve, you refused to do so.
<furmuwon>
hramrach : but no seem the boot0 code in nand
<furmuwon>
I try to "nand read 0x50000000 0x0 0x200"
leviathanch2 has quit [Ping timeout: 244 seconds]
<hramrach_>
it's not visible unless you read raw nand blocks. which no version of u-boot I am aware of can do
jmcneill has quit [Quit: Leaving]
<hramrach_>
in other words if you use Allwinner driver it hides boot0 from you
<Seppoz>
is there a way to access the system console via ssh
<furmuwon>
I think that read the raw nand block offset:0
leviathanch2 has joined #linux-sunxi
<Seppoz>
all tht output given on ttyS0?
<furmuwon>
not see the boot0 code and magic key..
<hramrach_>
if it does so it should contain a fragment of boot0 but it probably does not
<hramrach_>
because raw data contains stuff like checksum and padding but that command looks like reading continuous data blocks without that stuff
<hramrach_>
Seppoz: no. ssh starts way later than ttyS0. you can use a hardware serial to ethernet convertor (about the price of a Cubieboard - or another Cubieboard for that matter ) to capture the serial data and send it over network
<hramrach_>
or if you expect error only in latter boot stages you can try netconsole
<hramrach_>
but very early messages (before ethernet starts) cannot be captured by netconsole
<furmuwon>
hramrach_ : In short, to read the boot0 code from nand, Need to the checksum?
<hramrach_>
if you are interested only in working system messages then netconsole and syslog settings should give you that anywhere you like
<hramrach_>
furmuwon: the checksum is normally handled by the driver. but in absence of driver you need to handle it yourself
<hramrach_>
probably if you booted experimental mainline kernel with nand patches you could read boot0 but it is not known what it does to the rest of the data in nand
tomboy64 has joined #linux-sunxi
<furmuwon>
hramrach_ : thank. I just know that the boot0 is where stored in nand.
<furmuwon>
libv : are you alive?
<furmuwon>
In fact, I have been working for the company.
<libv>
?
<furmuwon>
but this asking is only my curiosity
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
<libv>
?
<furmuwon>
I'm not feeling well because of you
<hramrach_>
that happens to a lot of people
<libv>
well, keep up the lie then.
<libv>
nobody wakes up in the middle of the night and out of the blue wonders where boot0 on nand lives.
<furmuwon>
libv : go and apologize to me
<libv>
furmuwon: haha
_massi has joined #linux-sunxi
<plaes>
mripard: I saw that you added cpufreq to the .20 plan
<plaes>
mripard: could you also add link to patches on the mainlining effort page
<mripard>
No patches have been posted yet
<mripard>
but once they are, yeah, sure
<plaes>
ok, cool
<plaes>
I've been trying to keep this page up-to-date so all the WiP stuff has links to patches
<mripard>
yes, I saw that :)
<mripard>
thanks btw
<furmuwon>
libv : are you manage in the sunxi community?
<libv>
furmuwon: nope, i do not think that anyone can really claim that atm in this community.
<libv>
furmuwon: why?
<libv>
or are you once again going to lie and claim curiosity
<furmuwon>
atm?
<furmuwon>
what is the atm?
<hramrach_>
furmuwon: if you are uncomfortable about what people say there is /ignore command
<furmuwon>
hramrach_ : thank you. I will do.. next time
leviathanch2 has quit [Ping timeout: 255 seconds]
<hramrach_>
however, it's better if you can get along without that
<hramrach_>
there are some people here who work for a company and disclose what they are working for and for which company
whitesn has joined #linux-sunxi
<furmuwon>
Because of my poor english skills, I cant under stand the "nobody wakes up in the middle of the night and out of the blue wonders where boot0 on nand lives."
<hramrach_>
then it makes impression they really want to be part of the community and not only leech knowledge for their commercial career without giving anything back
<furmuwon>
what is the means?
<libv>
hramrach_: i would've been happy if furmuwon just stated what he wanted to achieve, along the lines of "booting from nand" or "replacing uboot on nand" or build livesuit images.
<hramrach_>
it means that it is unlikely that you would want to know where boot0 is stored without first trying to solve more basic problems about the device you work with
<hramrach_>
indeed, putting your problem in context makes it also possible to suggest more fitting solution than the one exact piece of information which may in the long run prove useless if you are taking the wrong approach
<libv>
but we've been through that before.
<hramrach_>
but that depends on telling at least what you are working on
<libv>
i was amused by the "i'm going to tell on you" and "where is your mommy anyway" bit though :)
<furmuwon>
hramrach_ : Sorry, I am not possible to join the conversation because of low level english
<furmuwon>
Also did not understand one thing.
<furmuwon>
hramrach_ : thank for your kindness
<hramrach_>
furmuwon: libv suggests that you probably want to know where boot0 is stored as a part of a project
<hramrach_>
if that is so telling more details about the project might allow us to give you more useful information about other parts of the project
<furmuwon>
I only use the script from AW to make Android image
<hramrach_>
well, if the development is already finished and you only build images for finished product there is not much to do anymore
<furmuwon>
hm..
<hramrach_>
Still it might be interesting if you told us what product it is and what features it has. If the features are interesting enough for somebody to get the device you get beta testing for free ;-)
<hramrach_>
there are numerous devices using allwinner chips and we cannot know all
<furmuwon>
I want to make the product (CDR , Tablet) using A20
<furmuwon>
Currently, only being studied.
<hramrach_>
what is CDR?
<buZz>
CDRecordable
<buZz>
constant datarate?
<furmuwon>
Using the hummingbird board kit
<furmuwon>
CDR is recoding machine
<hramrach_>
A20 is full feature chip so if you do not plan to have HDMI output on your tablet you might cut costs by using a13 or a23
<furmuwon>
if i cant not control the A20, I will not use the A20
<furmuwon>
I will fine the another chips
<furmuwon>
find
<hramrach_>
it definitely is controllable
<libv>
furmuwon: would it have killed you to have said "i have built android images, and flashed them, and i am now wondering where uboot lives on nand"?
<libv>
furmuwon: this is the context we were asking for all along
<hramrach_>
but be sure to test performance with the screen size you plan for your tablet
<libv>
as opposed to "can you help me please?" "i am just curious"
<furmuwon>
libv : What are you trying to say?
<libv>
*sigh*
<furmuwon>
libv : for my company, I studied the A20
<hramrach_>
you can presumably feed it into google translate to get more difficult word translated
<hramrach_>
Seppoz: is that on the board on which USB boot did not work?
afaerber_ has joined #linux-sunxi
<hramrach_>
Seppoz: you know, there is an USB testsuite your board should pass to be considered USB enabled :>
paulk-collins has joined #linux-sunxi
<hramrach_>
it was you with the custom board, right?
afaerber has quit [Ping timeout: 264 seconds]
afaerber_ is now known as afaerber
<Seppoz>
hramrach_: yea but usl is so easy
<Seppoz>
you cant really do much wrong
<Seppoz>
also what is confusing me is that it works right
<hramrach_>
I don't have a FTDI chip but with pl2303 (or something like that) I have no error
<Seppoz>
let me attach it to another port
<hramrach_>
maybe paste the part where the errors start
Andy-D has quit [Ping timeout: 265 seconds]
<Seppoz>
connected to another port and the messages are gone
<hramrach_>
also even if it's easy your board seems to have more problems than usual so there possibly is something wrong
<Seppoz>
i wonder if it has somehting to do with OTG
<Seppoz>
on none-otg it seem to be just fine
<hramrach_>
quite likely. these messages look like otg
<Seppoz>
also i experienced problems with wlan devices on otg ports
Andy-D has joined #linux-sunxi
<Seppoz>
but we dont use otg, its hardwired to host and configured as host only
<hramrach_>
then usb boot cannot work
<Seppoz>
what?
<Seppoz>
we dont boot form usb
<hramrach_>
if the host detection is correct it cannot work as device
<Seppoz>
we set it to host in fex
<hramrach_>
yes but you complained that usb boot does not work some time back, right?
<Seppoz>
no
<Seppoz>
never used usb boot
<hramrach_>
then I confused with somebody else, sorry
<Seppoz>
yes it was connected to 1-1 which is OTG
<Seppoz>
not its on 3-1 and no more of those messages
<hramrach_>
I wonder where I put my mni-USB otg cable
<Seppoz>
also we do not use the most recent kernel
<Seppoz>
the one we use is like 6 month old
<Seppoz>
were there any fixes
<hramrach_>
not sure but upgrading to current is always good idea
<hramrach_>
what is your kernel version?
<Seppoz>
3.4.61+
Andy-D has quit [Quit: Alive/Dead]
paulk-collins has quit [Remote host closed the connection]
<hramrach_>
hmm, I have probably OTG configured as device only because I had problems with OTG detection
<hramrach_>
no response with OTG cable
arete74 has quit [Ping timeout: 264 seconds]
<hramrach_>
actually, it appears broken with my kernel version
<hramrach_>
I have 3.4.75
<hramrach_>
the port is configured as OTG and works as device
<hramrach_>
3.4.79
arete74 has joined #linux-sunxi
popolon has joined #linux-sunxi
cnxsoft has quit [Remote host closed the connection]
FR^2 has quit [Quit: Connection reset by peer]
wickwire has quit [Ping timeout: 256 seconds]
tomboy64 has quit [Remote host closed the connection]
tomboy64 has joined #linux-sunxi
vishnu_ has joined #linux-sunxi
sehraf has quit [Read error: Connection reset by peer]
sehraf has joined #linux-sunxi
HeHoPMaJIeH has quit [Quit: Konversation terminated!]
wickwire has joined #linux-sunxi
<hramrach_>
furmuwon: what kind of tablet are you looking to make? presumably some budget 7-8" one since you are looking on a20?
leviathanch2 has joined #linux-sunxi
vishnu_ has quit [Remote host closed the connection]
philipp__ has quit [Remote host closed the connection]
leviathanch2 has quit [Ping timeout: 272 seconds]
HeavyMetal has quit [Ping timeout: 258 seconds]
HeavyMetal has joined #linux-sunxi
Skaag has quit [Quit: Leaving.]
HeavyMetal has quit [Ping timeout: 258 seconds]
konradoo77 has joined #linux-sunxi
HeavyMetal has joined #linux-sunxi
konradoo77 has quit [Ping timeout: 244 seconds]
domidumont has quit [Read error: Connection reset by peer]
wickwire has quit [Read error: Connection reset by peer]
bonbons has joined #linux-sunxi
wickwire has joined #linux-sunxi
konradoo77 has joined #linux-sunxi
konradoo77 has quit [Ping timeout: 264 seconds]
konradoo77 has joined #linux-sunxi
paulk-collins has joined #linux-sunxi
Skaag has joined #linux-sunxi
rds has joined #linux-sunxi
<rds>
merry Xmas community!
<rds>
stopped by to say, that I am really surprised how well the A80 is selling!
<rds>
At current prices, it was a non brainer it would sell like hot cakes!
<rds>
Also, the KMS driver only next Xmas, I guess!
<rds>
and finally, amazing how smart AllWinner is, after 4 years on the market, there still no stable release available!
<vpeter>
surprised in positive way or negative?
<Wizzup>
it this point it could all be ironic
<Wizzup>
at*
<vpeter>
I guess so.
<hramrach_>
dunno bout Asia
<hramrach_>
but I don't see sudden rush of A80 devices in shops in Europe
<hramrach_>
I have this feeling that both AW and RK stumbled. RK with 3288 did not go far enough so it was easy but in the long run they might have problem offering feature-competitive devices. AW with A80 tried to go too far it seems. Also the AW SoC line is very unbalanced which is one of its problems. RK has pretty much all SoCs with every feature on and off. The feature matrix for AW is very random
<hramrach_>
but to the point with current 3.4 and cubieboard 1 OTG detection does not work
gianMOD_ has joined #linux-sunxi
bonbons has quit [Quit: Leaving]
<hramrach_>
can my OTG cable be defective? I only found one
<hramrach_>
hmm, ok
<hramrach_>
found another cable and that one works. that was easy
gianMOD has quit [Ping timeout: 244 seconds]
bonbons has joined #linux-sunxi
domidumont has joined #linux-sunxi
<rds>
I was joking.... A80 sells must be really low!
<rds>
Do you really think that Asia has the capability to buy these expensives boards ? Hell, no!
<hramrach_>
yes, although I cannot point my finger on what they did wrong the chip seems to be a dropout
<hramrach_>
well, Asia has the capability to design and sell more devices so A80 might have a share there for all I know (or not)
<rds>
AW chips has only one compelling reason, pricing. If the dev boards were priced at US$9.99, people might had a compelling reason to buy
Zboonet has joined #linux-sunxi
<rds>
that is where I am expecting the next spin of R-Pi to be priced at!
<hramrach_>
the price of the devboard is not the point. Prospective device maker will buy a devboard for a few $k
<hramrach_>
but looks like they failed to target existing market or create new one for the A80 chips
<rds>
understood, but building a nice community around the platform is very important, to have stable releases of SW
<rds>
so people can make useful things with the SOC's
<hramrach_>
they rely on their own 'releases'
<rds>
well, their releases works "as good as many chinese thngs"
<hramrach_>
the devboards with hackable sw are negligible share of their or anyone elses market
konradoo87 has joined #linux-sunxi
Seppoz has quit [Remote host closed the connection]
Seppoz has joined #linux-sunxi
<hramrach_>
and consumers just want a tablet or another device that works as good as any other Chinese thing
<hramrach_>
if it's supposed to have QA and testing it's device manufacturer's job
konradoo77 has quit [Ping timeout: 244 seconds]
<rds>
no, people buy things that works. Example APPLE and Sansung
<hramrach_>
for some meaning of 'works' and for triple or quadruple the price, yes
<hramrach_>
I was seriously considering a Samsung tablet until I realized their one-connector design prevents you from connecting a screen and an USB device at the same time
<hramrach_>
unless you want to use an USB screen
<rds>
yes, I agree with you that they are very priced, but there should be a balance between price x quality
<rds>
with the A80, All Winner has none
domidumont has quit [Remote host closed the connection]
<hramrach_>
well, in Chinese tablet market quality is something that is added as an extra feature by device manufacturer. But they do not seem to have much interest so far
<hramrach_>
I would be somewhat mildly interested in the actual performance of the A80 chip but since the graphics subsystem is cripple by their 3D engine choice I could seriously use it as a router at most
<hramrach_>
not that I have 3d working on any of the cheap tablets ATM but it's very well possible to get it there and run some vintage 3D games
<rds>
hramrach, nice to chat with you!
<rds>
hopefully, things change and this community is able to get a reliable release of AW Soc's software!
rds has quit [Quit: Page closed]
<hramrach_>
I sometimes get very chatty when I do not feel like doing anything useful ^.^
<hramrach_>
sure, if AW got their act together and did useful software releases of all the stuff that is common for every AW device the manufacturers wishing to make quality software would have easier job. It's their market and their chips and their business plans. It's not like you can tell them what to do. They will not listen even if you were absolutely right
<hramrach_>
if you are really into it license some IP, go to China and start making decent chips ^.^
<nove>
rellla: this is not to say that you should start writting testcases for a driver that doesn't exist yet, but wait and can start learning, because when if the driver come to realization, it will be very helpful to have one more pair of eyes
whitesn has quit [Ping timeout: 252 seconds]
whitesn has joined #linux-sunxi
Renard has quit [Remote host closed the connection]
Zboonet has quit [Remote host closed the connection]
gianMOD_ has quit [Remote host closed the connection]
<buZz>
wiki also needs more pictures :D
<draconian>
well i was wondering in a crash course to get familiar with stuff like .flex u-boot n stuff
Guest72468 has joined #linux-sunxi
<draconian>
could u plz guide me to somewhere that explains it from the start
wickwire_ has quit [Read error: Connection reset by peer]
<buZz>
linux-sunxi.org has a -lot- of information draconian
<buZz>
look around, some pages are pointed out on the main page aswell
vishnu has quit [Quit: Leaving]
<draconian>
thank you guys i hope u dont get bothered by my dumb questions but i am really willing to dig deeper
<draconian>
i am a bit confused as i tried many sd card images on a Q8 type tablet and it didnt boot
<draconian>
so i thought it might be specifically build for a different configuration and simply my tablet isnt compatible
bengal has joined #linux-sunxi
<nove>
"cedrus" is a good name (can't think of other better), but because we don't now the really name of the hardware (allwinner is calling this video engine only as "video engine")
<nove>
this is a odd situation, as in the mainline kernel other hardware is called by it name, also it would be good to give a name in that allwinner doesn't fell aliened by it("as those annoying foss people gave a wierdo name to my precious video engine")
<hramrach_>
or is that only the userspace library?
cinatic has quit [Ping timeout: 252 seconds]
<nove>
hramrach_: is "phoenix" in the a20 datasheet, and then just called "video engine"
<hramrach_>
many references are only VE, especially in kernel so you caould make it something like sunxi-ve too ;-)
<nove>
hramrach_: cedar also come from "/dev/cedar_dev" in which there is comments with "cedar hardware"
<hramrach_>
they also have phoenixcard and phoenixsuite so that might be typo and if not that's reusing the name way too much
<hramrach_>
it would be understandable if they branded everything theirs as phoenixsomething but that's clearly not the overall theme
<hramrach_>
so they named it cedar at some point at least
<nove>
hramrach_: i want that this video engine conquer the world so not sunxi in the name will be better (this engine is so simples compared to others of the same kind)
<hramrach_>
it should have sunxi in the name as the vendor of the first hardware it was spotted on
Andy-D has quit [Ping timeout: 245 seconds]
<hramrach_>
and you cannot make it conquer the world since the VEs seem to be very SoC specific. Never heard of two vendors using same VE so far
<hramrach_>
not that I could tell, really. They are always mysterious no-name IP block
<hramrach_>
compared to which VE is it simple, though?
<hramrach_>
which other have you looked at closely
konradoo77 has joined #linux-sunxi
gianMOD has joined #linux-sunxi
<nove>
all of then, that i saw (a superficial "saw" not in detail)
cinatic has joined #linux-sunxi
<mrnuke>
arokux1: I know a lot of stuff should work in 3.4, but I tend to stray away from heavily branched kernels