<rm>
same CPU benchmark across a wide range of architectures
<andoma>
morning
cheng has quit [Ping timeout: 255 seconds]
popolon has joined #arm-netbook
<oliv3r>
so an arm cortex-a8 @1GHz (could be A10) performs abou tthe same as a 1.6GHz single core atom
<oliv3r>
or a single core 1.5GHz C7
<oliv3r>
interesting
popolon has quit [Ping timeout: 252 seconds]
QingPei has left #arm-netbook [#arm-netbook]
Sv has quit [Quit: Sv]
cheng has joined #arm-netbook
Sv has joined #arm-netbook
Gumboot_ has quit [Ping timeout: 260 seconds]
phh has joined #arm-netbook
popolon has joined #arm-netbook
Gumboot has joined #arm-netbook
arete74 has joined #arm-netbook
cheng_ has joined #arm-netbook
pawel5870 has joined #arm-netbook
Sv is now known as discopig
arete74 has quit [Ping timeout: 255 seconds]
<Maqs>
:-) system freeze after ioctl(4, 0x4a, 0xbe910638) on /dev/disp :-)
arete74 has joined #arm-netbook
<Maqs>
linux 3.0.42, a10
tzafrir_laptop has quit [Ping timeout: 265 seconds]
tzafrir_laptop has joined #arm-netbook
cheng_ has quit [Quit: Leaving]
cheng has quit [Remote host closed the connection]
cat_n9 has quit [Ping timeout: 252 seconds]
<rm>
I wonder if I could run dual displays from an MK802
<rm>
primary would be 1920x1080 from HDMI
<rm>
and a secondary with 1280x1024 from USB DisplayLink
<rm>
(well, I certainly could)
cat_n9 has joined #arm-netbook
rellla has quit [Ping timeout: 255 seconds]
rellla has joined #arm-netbook
<Maqs>
rm: do you have an mk802, too?
<rm>
yes, lots :D
<Maqs>
were you able to build u-boot for them?
<rm>
I didn't bother
<rm>
just got a pre-built u-boot
<Maqs>
ok.. i have one image with u-boot running and able to load my kernels.. but it crashes when using hardware decoding.. and one image with u-boot running but not loading my kernels.. this one works with hardware decoding :-)
<rm>
would like to say "and have no problems" but actually nope, one of the three MK802 that I have does not boot from SD
<rm>
images of what?
<Maqs>
whole sd card images
<rm>
with which OS?:)
<Maqs>
linux.. linaro, ubunto
<Maqs>
ubuntu
<rm>
hm I see
<rm>
so you have h/w decoding working there, that's impressive :)
<Maqs>
both armhf, i built xbmc in an armel chroot
<Maqs>
all 1080p and 720p videos i tried so far worked just fine.. but the kernel was missing support for dvb and usb audio hardware -.- :D
<mnemoc>
Maqs: might you try the stage branch?
<mnemoc>
s/might/may/
<ibot>
mnemoc meant: Maqs: may you try the stage branch?
<rellla>
anyone tested disp_mode=4 in script.bin? should it already work for hdmi and vga simultanously for 1080p60 and 1280*1024 with same content at different screens?
<Maqs>
mnemoc: i'll try it
<mnemoc>
rellla: i doubt the current driver supports to use both at the same time
<Maqs>
the strange thing is that the system just freezes after that ioctl call, without any klog output etc.
<andoma>
Maqs: which ioctl() is that?
<mnemoc>
Maqs: you might be affected by the pll4 bug if you don't have an A10 revC
<Maqs>
sunxi-3.4 or 3.0?
<mnemoc>
same as you are using now
<Maqs>
andoma: ioctl(4, 0x4a, 0xbe910638) on /dev/disp
<rellla>
mnemoc: but it would be worth a try nevertheless ;-)
<Maqs>
so stage/sunxi-3.0
<mnemoc>
rellla: :)
<mnemoc>
Maqs: yes
<Maqs>
andoma: there were three other calls before that one after opening /dev/disp
<Maqs>
i need to get a better internet connection.. even shallow clones take ages
<mnemoc>
Maqs: i thought you already had a working tree
<Maqs>
i have, but it's a shallow clone only
<mnemoc>
:)
<mnemoc>
cheater
<rellla>
Maqs: you've got h/w decoding working on armhf with vpu? cedarx? armhf?
<rm>
<Maqs> the strange thing is that the system just freezes after that ioctl call, without any klog output etc.
<andoma>
0x4a is SET_PARA .. so i would just start sprinking with printk()s inside there to see what goes on
<Maqs>
rellla: armel chroot on armhf. xbmc
<rm>
pft, entirely natural :D
<rm>
for the quality code that we have in there
<rm>
and in aspects of h/w design perhaps too
<andoma>
everytime i look at that code i want to claw out my eyes
<Maqs>
as i worked as a computer science teacher for two years, i guess i've seen worse code.. ;-)
arete74 has quit [Ping timeout: 268 seconds]
<oliv3r>
as a computer science student, I appologize :(
<rm>
I wonder is this really a DAC and works with any HDMI device
<oliv3r>
in theory, it should; based on the specs. i wouldn't be supprised if it simply used the analog wires :p
<oliv3r>
though it's pricey enough to be 'real'
arete74 has joined #arm-netbook
<Maqs>
lol oliv3r :-)
<oliv3r>
what I personally need, is a vga -> scart (rgb) cable, i have a vga cable allready cut up so just need to solder it; but a working one would be easier :)
<oliv3r>
Or, find out why svideo is so 'colorful' :S
<Maqs>
rm: just to let you know, this seems to be the adapter that killed some raspberry pis..
<oliv3r>
libv: i found soem schematics that solders some transistors to combine sync signals, I don't know current driver/X status
<rz2k>
good to hear some communication from chinese guys.
<libv>
ah, ok, right, or some electronics
<oliv3r>
libv: so ... can it actually work these days (on some intel chip, have to check); or do I have to find out why svideo output is giving me some wrong colors on r300g?
<libv>
oliv3r: for our a10, you should be able to one day set an fb/kms option to achieve this
<libv>
r300g?
<libv>
why g?
<oliv3r>
libv: regular PC
<libv>
isn't that a gallium driver?
<oliv3r>
yes :)
<libv>
what has a gallium driver got to do with display stuff?
<oliv3r>
libv: i use an old PC as an xbmc box, or plan to for now
<libv>
oliv3r: this is a modesetting issue
<libv>
oliv3r: or cabling
arete74 has quit [Ping timeout: 256 seconds]
<oliv3r>
libv: cabling should be fine, i'll try again to find out what's wrong with svideo if i can't get scart -> vga working
<libv>
oliv3r: from my excrutiating personal experience: radeon tv encoders aren't worth bothering
<oliv3r>
the TV specifically accepts RGB, the OSD even says '[}-> RGB'
<libv>
oliv3r: we got told from the start of the freeing ati project that noone cares anymore, and the knowledge or documentation is not available anyway
<oliv3r>
but if i can't get that to work properly, i'll have to go for radeon+tvout (svideo)
<oliv3r>
libv: yeah, i know, analog is deing :)
<oliv3r>
but that's the TV kind I have up in the bedroom :p
<libv>
oliv3r: somewhere along the way, when the some wankers started competing with radeonhd, ati was actually feeding them as much info as possible
<oliv3r>
if i have enough $$ for a new tv in the living room, i'll move that one upstairs, which has hdmi etc in
<oliv3r>
libv: why would ati do that?
<libv>
when we then put it on the agenda again, we ended up wasting our time
<libv>
oliv3r: amd + suse versus ati + redhat and other loud elements
<libv>
amd bought ati and needed to integrate ati, and ati resisted
<oliv3r>
ah, so it was really ATi, not AMD then?
<libv>
yes
<oliv3r>
as far as I know, or hope to know, is that AMD is really trying to be a good FOSS citiczen
<libv>
ati got forced to work with the suse guys and to help create an open source driver which we at suse wanted to do preloads with
<libv>
amd is
<libv>
former ati employees usually arent
<libv>
we still have fglrx only
<oliv3r>
:(
<libv>
the ati open source driver is a figleaf
<oliv3r>
what do you mean, only?
<oliv3r>
oh yeah
<oliv3r>
the figurativily olive branch
<oliv3r>
I guess it takes time, when i have to belive bridgeman anyway
<libv>
bridgman is the root of most of the evil there
<libv>
he always played double games and never ever did what he was first tasked to do: provide us with information and documentation
<libv>
oliv3r: have you seen any documentation out of amd on radeon recently?
<oliv3r>
libv: i think that's a little beyond my grasp :(
<libv>
oliv3r: the register documentation went away around the time i got booted from suse
phh has quit [Ping timeout: 255 seconds]
<oliv3r>
hmm, on the phoronix forums he was always very Pro foss
<libv>
(when novell culled 20% of developers here in nue, and the socially stronger got booted first according to german law)
<libv>
this is also the time when amd was in deep financial troubles and sold of its foundries
<libv>
there were no register documents produced after that
<oliv3r>
this was when? 2009?
<libv>
we at suse only received documentation on the explicit condition that we got the last draft to go public, right before the last checks were made on it by legal
<libv>
but we demanded free documentation
<oliv3r>
on 15th of august the SI ISA documentation was delivered
<libv>
i still have a few docs that never made it out
<libv>
oliv3r: aha, well spotted
<libv>
oliv3r: what part of amd produces those docs?
<libv>
oliv3r: people handpicked from ati, from amd and hired specifically to do the gpgpu stuff
<libv>
so a brandnew team, with no bonds or political ties with ati
<oliv3r>
ah
<libv>
and those docs never ever passed through bridgmans hands
<oliv3r>
so you do say they are slowly improving still yes?
<libv>
ati still is a huge organization
<oliv3r>
well he is no longer on the driver team
<libv>
amd still has no full control
<oliv3r>
on the radeon team*
<libv>
yes, but the damage was done
<oliv3r>
well, sad, but true; amd relies heavly on ATi for income
<libv>
there is only a figleaf driver
<libv>
and suse ships on preloaded laptops with the fglrx driver
<libv>
and noone can really fix the perl code that is xf86-video-ati or the radeon kms
cat_n9 has joined #arm-netbook
<libv>
it's just hopeless, and impossible to return from there
<libv>
with r500 we had a brand new display engine, AMD willing to crack things open at ATI and very keen on hearing suse engineers ideas and seeing them crack open ati for them
<libv>
clean start all round
<libv>
and then the politics started kicking in
<libv>
and some people who still like to affiliate themselves big time with free software were keen on participating in the shit throwing contest
<libv>
even though they very directly ended up hurting free software
<libv>
they went with ati, produced badly tested and unmaintainable crap code, and made big statements
<oliv3r>
:(
<oliv3r>
so what about r600?
<oliv3r>
wasn't that the first round of 'the better new'?
<libv>
airlied, alex deucher, matthew garrett, daniel stone, adam jackson, benjamin herrenschmidt, etc... quite a large portion of the X.org community really
<libv>
they all played a very dirty game, just so they would be seen to run the show
<libv>
there is no other reason that remains
<oliv3r>
sad really
<libv>
no, same display engine
phh has joined #arm-netbook
<libv>
but due to atombios, and hdmi audio happening, r600 had a few registers introduced in the middle of register blocks, and the upper half of the register blocks then shifted upwards
<oliv3r>
so all hope is with intel then? :)
<libv>
atombiso could hide all that shit anyway... fail.
<libv>
yes/no
<oliv3r>
i still prefer AMD cpu's :p
<libv>
but there are only three vendors
<libv>
intel, who has been spending a lot of money on open source
<libv>
amd, who will keep fglrx and has the figleaf driver to not have too much bad marketing
<oliv3r>
OR! arm with mali gpu and your driver :D
<libv>
and nvidia, which provides a rock solid closed source driver and actually delivers what linux gamers need
<libv>
if you are a gamer, go for intel cpu and nvidia graphics
<libv>
otherwise, just buy intel
<libv>
forget amd.
<libv>
yes, in the arm world everything is still open
<oliv3r>
i'm not a hardcore gamer; but AMD + fglrx/r300g /r600g works pretty good for me
<libv>
and the assholes that killed radeonhd just for their own "fame", are not in that area anyway
<oliv3r>
i _was_ in a good mood until ^ :p
<oliv3r>
i'd still buy an AMD CPU over an intel cpu though, gotta go with the underdog :)
<libv>
buy arm.
<oliv3r>
i've read 'somewhere on the internet' that ARM isn't that oss friendly either
<libv>
radeonhd still is lightyears ahead in structure and in exposing actual hw features, lightyears ahead of kms today
<oliv3r>
what about their GPU's; if your code would reach performance or even surpass it, any chance of that stance of them changing?
<libv>
they will need to give out gpgpu info anyway
<libv>
with the t604
<libv>
at least some info
<libv>
and then optimisation
<libv>
and the command stream stuff should be easier for t604 than with earlier gens
<libv>
so there is much more hope of freeing arm, yes
<oliv3r>
well lets hope and pray (and work on it :D)
<libv>
and then, we do not get to deal with two equally sized companies, and one trying to claim ownership over the other
<libv>
but wait and see
<oliv3r>
well what I do like about 'arm', is that anybody can make a CPU/SoC with an arm core in it, and while some glue/shim code would be obviously required, it should still work the same
<libv>
but arm already did not end up supporting lima in any way
<oliv3r>
and well defined
<oliv3r>
see, arm isn't that OSS friendly
<oliv3r>
only what is required to get it working
<rz2k>
oliv3r: they dont do even that, there is still no documentation on buf1/2 implementation in new mali-400 release
<rz2k>
r3p1-rel1
<lundman>
hmm
<mnemoc>
please don't forget to test the stage branch so I can merge the buf2 fix (and the rest of it)
<rz2k>
atleast they dumped their new mali_drm (with bunch of files that should not go to public release like build script that points to various kernel dirs)
<rz2k>
mnemoc: works for me (tm)
<rz2k>
:p
<mnemoc>
good :)
<rz2k>
still need a word from android users
<oliv3r>
rz2k so of course we use that to gain more knowledge about the innter workings?
<oliv3r>
rz2k: i'm not that far yet, in the world of the A10 :p
cat_x301 has quit [Ping timeout: 256 seconds]
<rz2k>
dont know about that, libv said that he didnt dig mali kernel drivers.
<rz2k>
also it isnt A10 world, only A10 thing there is platform definition that we got with r2p4 package from Tom :p
rellla has quit [Quit: rellla]
<libv>
well, ideally egl is knowledgeable about memory and windowing systems and such, and this is the most non-portable userspace bit
<libv>
other bits are libc and others abi differences
<libv>
but apart from that, the userspace should be reasonably universal, save mali kernel api version
rellla has joined #arm-netbook
<discopig>
is there any way to mount the nand on which android OS is from inside linux? /dev/nanda and others only seem to be boot files and the 2gb user partition used on android
<discopig>
on the hackberry a10
rellla has quit [Client Quit]
<oliv3r>
rz2k: a10-world I ment, building my own kernel for it, compiling any sources for it (android or native), i just installed a CM10 rom and 'use' it atm :)
<mnemoc>
discopig: in android it's /dev/block/nanda
<oliv3r>
gotta work on setting up uSD boot to start playing within the 'A10-world' :D
<discopig>
i'm on debian right now
<discopig>
but i want to mount the NAND
<discopig>
i dont know if it's possible
<mnemoc>
discopig: sure it is
<oliv3r>
discopig: you want to mount the nand in its entirity? you know there's about 9 partitions on the nand?
<discopig>
yeah
<discopig>
that's fine
<oliv3r>
so I assume you want to mount one of the partitions of the nand?
<discopig>
i see them all in /dev/block now
<mnemoc>
in linux they are /dev/nand*
<oliv3r>
usually, /dev/block/nandd (if my head serves me right) is /system, which is the core of android
<discopig>
yeah i just wasnt sure which ones existed and which didnt
<mnemoc>
all exist
<discopig>
because most of them wouldnt mount other than nandk nanda and nandi
<discopig>
but it works now
<oliv3r>
some are formatted vfat, some are ext4 and some are something weird I don't understand yet, mmc (with vfat ontop)
<mnemoc>
discopig: they use "raw" partitions for u-boot's env, the kernel, and as closed-bootloader arguments
<mnemoc>
iirc misc has arguments for their bootloader, to pick u-boot or recovery
<oliv3r>
oh and that yea too! :D
rellla has joined #arm-netbook
rellla has quit [Client Quit]
drachensun has joined #arm-netbook
<oliv3r>
what's the max frequency of the mali400 in the A10? anybody know?
<oliv3r>
381 MHz sound reasonable?
<mnemoc>
i think i've seen 400
rellla has joined #arm-netbook
<oliv3r>
but the datasheet says 381 MHz is max!! :p
<oliv3r>
ok, so it's not an absolute, just wanted to know that :)
<mnemoc>
rellla: hi, can you test if your problem with pll4 is solved in the stage branch?
cat1 has joined #arm-netbook
<mnemoc>
Hexxeh: hi, does chromeos use X11? raw fb? something else?
cat1 has quit [Ping timeout: 255 seconds]
<grevaillot>
mnemoc: according to a friend who just tested chromeos in a vm, there's an X running.
<mnemoc>
grevaillot: great!, so one should be able to "repurpose" the closed libs in the exynos5 chromebook and run a normal linux system without much trouble
zeug has joined #arm-netbook
sspiff has quit [Remote host closed the connection]
tzafrir_laptop has quit [Ping timeout: 246 seconds]
<rellla>
mnemoc: i can do yes, but later...
<mnemoc>
rellla: thanks
<jelly-home>
oliv3r: would it not be called mali381 then
<oliv3r>
does the 400 relate to its core speed? or is mali400 simply the model number?
<mnemoc>
i don't think it related to the core speed
<mnemoc>
its
<oliv3r>
i agree :p
<mnemoc>
in the .fex files I have here clkdiv is usually 3, but there are 4s and 2s too
QingPei has joined #arm-netbook
<mnemoc>
docu says that is against pll4
<oliv3r>
I haven't reached the mali400 clk register yet, but I think you can div it anywhere from 1 - 16 from PLL 3-7
<Maqs>
techn_: do i have to enable kernel debugging as well?
<RaYmAn>
yeah, that blog hasn't been updated for ages
<techn_>
Maqs: dunno.. but propably not :p
<Turl>
RaYmAn: why? looks good to me :P
<RaYmAn>
yeah
<mnemoc>
RaYmAn: so your hope is based on a diagram describing (again) their intentions about the display?
<RaYmAn>
well, no, they said that is the specs for their display
<RaYmAn>
They did realize the crappy ass lcd of adam was a major fail ;)
<mnemoc>
the only reason everyone had to buy their device was the pixelqi display...
<RaYmAn>
I really tried pushing them to go atmel mxt for touchscreen (with pen input!), but it's apparently impossible to get from atmel unless you're really really big :(
<RaYmAn>
mnemoc: indeed, but it'll still be interesting :)
<RaYmAn>
though, I must admit, i probably wouldn't buy one myself.
<RaYmAn>
mnemoc: also, the issue with pixelqi is that they couldn't keep up :(
<mnemoc>
didn't 3M get involved?
<RaYmAn>
no idea
<RaYmAn>
I don't have a pixelqi variant so never really cared
<RaYmAn>
I do have like 3 non-pqi :/
<mnemoc>
:<
<RaYmAn>
meh, this exy5 kernel is hacky :/ They have dt but then lots of hardcoded device-specific hacks in the code :( (runtime checking from dt though)
<RaYmAn>
it comes with a 2 year *prepaid* dataplan (with 100mb transfers/month)
<mnemoc>
100mb?
<Turl>
that ATT one? :P
<mnemoc>
V
<RaYmAn>
verizon
<Turl>
ATT made one for.. N7 was it?
<Turl>
kindle fire HD
<RaYmAn>
mnemoc: ah, those specs say eMMC storage
<Turl>
it has no moving parts from what I read
<Turl>
fanless
<RaYmAn>
If I were to get one of these, I'd kind of definitely want 3G I think
<mnemoc>
the uk reseller is claiming SSD
<RaYmAn>
though, maybe not
<RaYmAn>
lol
<RaYmAn>
solid state drive is..broadly defined Isn't it?
<mnemoc>
right
<mnemoc>
it's not strictly a lie
<RaYmAn>
I meant, technically it *is* SSD, I'd think
<mnemoc>
16GB considering it has SD and usb3 is not that bad
<RaYmAn>
200GBP
<RaYmAn>
so only around uhm, 70$ more expensive *sigh*
<RaYmAn>
Though, tax/vat on import would probably cost at least that
* mnemoc
hopes his forwarded "small personal package" to be ignored by the spanish customs, as usual
<RaYmAn>
:P
<RaYmAn>
it's such a pity this wasn't out a month or two ago - my mom was just on holiday there :/ could have brought it back, which, as long as she opened it & charged it would have definitely gotten past :P
<RaYmAn>
meh, the 3g version is 279GBP
<RaYmAn>
that's just stupid
cat_n9 has quit [Ping timeout: 246 seconds]
<mnemoc>
hopefully one can just connect a normal 3g mini pci-e/usb module once received :p
<RaYmAn>
yes
<RaYmAn>
It will be interesting to see how open it is
<rz2k>
techn_: I've fixed r3p1-rel1 drm for 3.0.x and tested - our libs are outdated, they seriously reworked pixmap stuff in the lib, without proper ioctls going from lib to drm secureid thing fails even more horribly than when we didnt have it at all
<rz2k>
also we can just forgot that rel1 exists, drivers are still same crap.
<rz2k>
also now I'm confused, why drm has "memory working"-ioctls when they have their shiny UMP?
cat1 has quit [Ping timeout: 245 seconds]
<techn_>
rz2k: I was thinking it was kms/drm driver skeleton.. but since there were really old codes.. it's propably not
<techn_>
Maqs: I have no idea why it fails :(
<techn_>
Linaro 12.09 updated to Linaro 12.10 :/
<techn_>
yes it is.. not to investigate what's improved/regression :)
<Maqs>
i'll try lichee-3.0.8-sun4i now and see if that works
ibrah has joined #arm-netbook
<techn_>
atleast there is some dependency fails :(
<techn_>
libgbm dependencies broken :(
cat_n9 has joined #arm-netbook
<techn_>
libgbm1 (= 9.0-0ubuntu1) but 9.0.0-7 is to be installed
<techn_>
oh.. even x wont start with mali .. they have updated X.org
<techn_>
so driver must be recompiled
<techn_>
ok.. it wont compile
cat has joined #arm-netbook
cat is now known as Guest8545
Guest8545 is now known as cat_x301
<shivansps>
my serial adapter just arrive, time to open the Novo 7 Aurora and figureout wth is wrong with never kernels, as well i gona try to figureout linux boot on JZ4770
<rz2k>
techn_: they had iso-codes dependency broken somehow
<rz2k>
tons of packages died
<rz2k>
just wait till it will be fixed.
<mnemoc>
shivansps: JZ4770 is mips
<shivansps>
i know, but ingenic has the u-boot source, kernel source (old kernel source), and im sure i can use some debian mips rootfs
<rz2k>
ainol leaked their jz4770 kernel, just fyi.
<shivansps>
and a media player source that uses his vpu too
<RaYmAn>
I always wanted one of those mips devices, heh
<RaYmAn>
have they gone down in price yet
<RaYmAn>
?
<shivansps>
the jz4770 2.6.36 kenerl is avalible on ingenic web
<shivansps>
yeah but almost everyone adds hdmi on jz4770
cat_x301 has quit [Quit: Leaving.]
cat_x301 has joined #arm-netbook
bsdfox has quit [Read error: Connection reset by peer]
bsdfox has joined #arm-netbook
bsdfox has joined #arm-netbook
bsdfox has quit [Changing host]
bsdfox has quit [Read error: Connection reset by peer]
bsdfox has joined #arm-netbook
bsdfox has quit [Changing host]
bsdfox has joined #arm-netbook
simosx has joined #arm-netbook
simosx has quit [Changing host]
simosx has joined #arm-netbook
bsdfox has quit [Write error: Connection reset by peer]
bsdfox has joined #arm-netbook
bsdfox has quit [Changing host]
bsdfox has joined #arm-netbook
ZaEarl has joined #arm-netbook
bsdfox has quit [Write error: Connection reset by peer]
bsdfox has joined #arm-netbook
bsdfox has quit [Changing host]
bsdfox has joined #arm-netbook
bsdfox has quit [Read error: Connection reset by peer]
bsdfox has joined #arm-netbook
bsdfox has quit [Changing host]
bsdfox has joined #arm-netbook
Almamuetya has quit [Ping timeout: 246 seconds]
bsdfox_ has joined #arm-netbook
bsdfox_ has joined #arm-netbook
bsdfox_ has quit [Changing host]
bsdfox has quit [Ping timeout: 252 seconds]
cat_x301 has quit [Ping timeout: 265 seconds]
cat_n9 has quit [Ping timeout: 264 seconds]
cat_n9 has joined #arm-netbook
cat_x301 has joined #arm-netbook
Guest53506 is now known as toxicpsion
toxicpsion has quit [Quit: Ex-Chat]
toxicpsion has joined #arm-netbook
toxicpsion has quit [Client Quit]
<oliv3r>
mnemoc: i'm reminding you :D
newbie has joined #arm-netbook
newbie is now known as Guest21858
<mnemoc>
oliv3r: oh
<mnemoc>
oliv3r: yes. thank you
hp__ has quit [Ping timeout: 276 seconds]
<oliv3r>
:p
<mnemoc>
i was stuck debugging an scheduler and didn't see the clock
<shivansps>
ok im a uart noob here, for uart only is needed to connect gnd, tx and rx?
<mnemoc>
right
Fab_FR has joined #arm-netbook
<Fab_FR>
Hello
<mnemoc>
oliv3r: maybe fixed
<mnemoc>
s/maybe/may be/
<ibot>
mnemoc meant: oliv3r: may be fixed
<mnemoc>
ibot: tired?
<ibot>
tired is probably a state of consciousness between soda and coffee.
<mnemoc>
oliv3r: it's really weird, on preview it's working
cat_x301 has quit [Read error: Connection reset by peer]
<mnemoc>
?action=purge did the trick
cat_n9 has quit [Ping timeout: 246 seconds]
cat_n9 has joined #arm-netbook
cat_n9 has quit [Ping timeout: 252 seconds]
<shivansps>
ok i have the UART working on the jz4770 right now
<Fab_FR>
Excuse me, I'm looking for some advice, maybe some of you could share ideas with me? I got an A13 tablet with HDMI, I know it sounds weird on A13, but I would like to find a way to make it work...
<ZaEarl>
doesn't sound possible
<Fab_FR>
That's strange, why they did put a HDMI port on an A13 board then? Just for decorative? :D
<ZaEarl>
re-using parts, such as a shell
<ZaEarl>
or it really is an a10
<Fab_FR>
Well I opened it, I even took pics of PCB, it is an A13, and board is labelled "A0709-4GB512M-HDSI"
<ZaEarl>
and you see an hdmi port on the motherboard?
<Fab_FR>
Yeah, Mini HDMI-out plug
<ZaEarl>
crazy
<Fab_FR>
I inspected the board for finding eventual separate IC which would manage HDMI, but I think I didn't see one...
<ZaEarl>
so probably it's not wired up
<mnemoc>
lkcl said once that the A13 can do hdmi but they decided to disable it...
<mnemoc>
Fab_FR: in the A13 chip, what's the first letter of the number under the A13 text?
<mnemoc>
but the C means it's the same rev of mine
<mnemoc>
very very very strange
<mnemoc>
i thought it could be an older chip with hdmi not-disabled
<mnemoc>
Fab_FR: you can make a page at linux-sunxi.org for your tablet if you want
<mnemoc>
with the pictures and your findings
<Fab_FR>
Okay, I will do that
<Fab_FR>
I hope it will help many people there
<Fab_FR>
Even if this tablet seems very uncommon...
<Fab_FR>
I did put the pic on imgur.com/ToLdc
<mnemoc>
Fab_FR: even if uncommon there are at least a couple of thousand units
<mnemoc>
maybe on the back?
<Fab_FR>
Pictures of back of PCB?
<mnemoc>
the hdmi thing
<mnemoc>
i see the A13, ddr, nand, axp (pmu) ... what is that tiny chip between the ddr and nand? on the top of the picturre?
shivansps has quit [Ping timeout: 245 seconds]
<Fab_FR>
The two little chips on the top are clock and calendar ic, and gsensor dmtard06
<Fab_FR>
The side with connectors: imgur.com/c6bDT if you want to see the connectors on the PCB itself let me open the tablet again
<mnemoc>
Fab_FR: nah, don't worry
<mnemoc>
it would be interesting to see your script.fex
<Fab_FR>
Sure
<mnemoc>
if the A13 is indeed controlling that hdmi the script file will show it
<Fab_FR>
Pastebin is ok for the file?
ZaEarl has quit [Quit: Ex-Chat]
<mnemoc>
yes
ZaEarl has joined #arm-netbook
<Fab_FR>
Ok here we go! pastebin.com/ESdVFiK2
<RaYmAn>
tv out dac_used = 1 but hdmi_used = 0
<Fab_FR>
Should I switch tv_out to 0 and hdmi to 1 then reflash and test?
vgrade has quit [Ping timeout: 252 seconds]
vgrade has joined #arm-netbook
<mnemoc>
and only lcd0_para
<ZaEarl>
dmard06 gsensor. does anyone have the source for that one yet?
<Fab_FR>
It's supposed to be a lcd1_para for HDMI too?
<mnemoc>
not necesarily, but defines a second lcd output.... which could be turned maybe into dvi
<mnemoc>
but i don't see anything that could point to hdmi
<mnemoc>
(or dvi)
<mnemoc>
ZaEarl: doubtful
<ZaEarl>
all right, I'll ask around see if I can get a copy
<Fab_FR>
So even if I switch the hdmi_used to 1, it won't change anything? Or if the HDMI is really managed by the board, is there a way to find which port it uses and write the corresponding section in the script.fex ?
<mnemoc>
ZaEarl: awesome
<ZaEarl>
I saw that part in a lot of sample tablets last week.
<mnemoc>
Fab_FR: the script.bin would need to have info about the involved pins
<Fab_FR>
Ok, is there a way I can find these pins?
<mnemoc>
the A13 has all it's legs exposed :)
tuliom has joined #arm-netbook
<mnemoc>
in the datasheet you can see a diagram showing them all, so you can skip those that obviusly aren't going there. like nand or ddr
<Fab_FR>
So if the A13 really manages HDMI, it should be on unassigned pins?
<mnemoc>
each pin supports several features
<mnemoc>
we don't know if some of the undefined values actually have meaning in the A13, like to enable hdmi support
cat_x301 has joined #arm-netbook
<mnemoc>
sure there is no IC on the back of the board?
<mnemoc>
it makes no sense to add an hdmi connector on a board that can only be used with an A13
<Fab_FR>
Nope, except the connectors, I think I did not see, but I'll check again to be sure
<ManoftheSea>
Well, if you're a huckster, you can truthfully claim it "plugs into HDMI"
<ManoftheSea>
But that would be dishonest.
<Fab_FR>
Yeah.... Haha, no I'm not like that :D
<ZaEarl>
does the hdmi work in android?
<mnemoc>
going to sleep. good night
<Fab_FR>
Good night :)
<Fab_FR>
I'll go soon too, 1.35am here
<bsdfox_>
Error: selected processor does not support ARM mode `ldfs f0,[sp],#4'
<bsdfox_>
I get that trying to cross compile gcc target arm-linux-gnueabihf on amd64. anyone got ideas?