akaizen_ has quit [Remote host closed the connection]
akaizen has joined #linux-sunxi
arete74_ has joined #linux-sunxi
arete74 has quit [Read error: Connection reset by peer]
akaizen has quit [Ping timeout: 255 seconds]
kivutar has joined #linux-sunxi
quitte has quit [Read error: Connection reset by peer]
quitte has joined #linux-sunxi
sehraf has quit [Read error: Connection reset by peer]
<quitte>
does dma work in 3.4? if yes - what does it work with?
FR^2 has joined #linux-sunxi
notmart has joined #linux-sunxi
notmart has joined #linux-sunxi
sehraf has joined #linux-sunxi
physis has joined #linux-sunxi
sehraf has quit [Ping timeout: 250 seconds]
physis has quit [Ping timeout: 272 seconds]
paulk-aldrin has joined #linux-sunxi
akaizen has joined #linux-sunxi
programtruck has joined #linux-sunxi
<programtruck>
Question about Cedar X , is ther ant bosy here with succecful exprience in Gstreamer on A 20
<programtruck>
using hardware video decoding and encoding
physis has joined #linux-sunxi
linkmauve1 has joined #linux-sunxi
Renard has joined #linux-sunxi
linkmauve1 has left #linux-sunxi [#linux-sunxi]
physis has quit [Ping timeout: 250 seconds]
linkmauve1 has joined #linux-sunxi
konradoo77 has quit [Ping timeout: 255 seconds]
konradoo77 has joined #linux-sunxi
sehraf has joined #linux-sunxi
akaizen has quit [Remote host closed the connection]
Net147 has joined #linux-sunxi
akaizen has joined #linux-sunxi
akaizen has quit [Ping timeout: 240 seconds]
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
popolon has joined #linux-sunxi
sehraf has quit [Read error: Connection reset by peer]
huehner_ is now known as huehner
sehraf has joined #linux-sunxi
programtruck has quit [Ping timeout: 240 seconds]
_massi has joined #linux-sunxi
hramrach has quit [Remote host closed the connection]
programtruck has joined #linux-sunxi
hramrach has joined #linux-sunxi
programtruck has quit [Read error: No route to host]
MasterChief4942 has joined #linux-sunxi
physis has joined #linux-sunxi
physis has quit [Ping timeout: 255 seconds]
konradoo77 has quit [Ping timeout: 260 seconds]
MasterChief4942 has quit [Quit: Leaving.]
konradoo77 has joined #linux-sunxi
rds has joined #linux-sunxi
<rds>
programtruck: you read, the wiki. That is all about H/W encoding/decoding you get
<rds>
you won't hear the answer you like, once it does not exist
<rds>
if you H/W aided encode/decode, use Android or piss AllWinner for documentation
<rds>
the solutions that exist today, are documented in the Reverse Eng pages. It is as simple as that.
<rds>
All linux solutions for H/W oriented video encoded/decoded are experimental, and documented on the Cedrus pages. If you need beyond that, ask AW for docs and be prepared for long weekends, coding!
<rds>
it is as simple as that!
rds has quit [Quit: Page closed]
Gerwin_J has joined #linux-sunxi
Gerwin_J has quit [Client Quit]
akaizen has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
Gerwin_J_ has joined #linux-sunxi
Vanfanel has joined #linux-sunxi
Gerwin_J has quit [Ping timeout: 250 seconds]
Gerwin_J_ is now known as Gerwin_J
<Gerwin_J>
WITS send me some documents of A80 but it's all confidential :(
<mnemoc>
reference tablet design or some real documentation?
<Gerwin_J>
optimus board
<Gerwin_J>
it's small introduction to a80 optimus board and all the pinouts
<wens>
datasheet?
<mnemoc>
pinout of the board or the pinmuxes of the SoC?
physis has joined #linux-sunxi
<Gerwin_J>
GPIO, CAM, UART0 and JTAG pin out
<wens>
board pinouts... not so useful
cubear has quit [Quit: Leaving]
physis has quit [Ping timeout: 250 seconds]
<Gerwin_J>
also the talk about linux 3.4 source code
<mnemoc>
got any kernel/u-boot sources?
akaizen has quit [Remote host closed the connection]
akaizen has joined #linux-sunxi
akaizen has quit [Ping timeout: 260 seconds]
konradoo87 has joined #linux-sunxi
konradoo77 has quit [Ping timeout: 255 seconds]
konradoo87 has quit [Ping timeout: 250 seconds]
konradoo77 has joined #linux-sunxi
rinni has joined #linux-sunxi
<rinni>
Libv: Are you shure about the SMP support for sun7i via PSCI in u-boot v2014.10? I can't see the commit implementing it for sun7i
<deepe>
mripard_: yes, possible, but I am new in this, and sometimes I just do not know where I need to look even it blinks on front of my head
<deepe>
mripard_: SoC display engine means, that I will use display, which will be connected throught LCD_CON? on A10 olinuxino libe board?
<mripard_>
yes, or hdmi, or vga, for example
<deepe>
mripard_: thanks for explanation :), sorry for noob questions before Wiki read
HeHoPMaJIeH has joined #linux-sunxi
<mripard_>
it's fine, don't worry :)
<mripard_>
just remember that the wiki is there next time ;)
blsd1 has joined #linux-sunxi
blsd has quit [Read error: Connection reset by peer]
INdek has joined #linux-sunxi
physis has joined #linux-sunxi
physis has quit [Ping timeout: 244 seconds]
blsd1 is now known as blsd
leviathanch has quit [Ping timeout: 240 seconds]
konradoo77 has quit [Ping timeout: 250 seconds]
issue_ has joined #linux-sunxi
quitte_ is now known as quitte
konradoo77 has joined #linux-sunxi
<quitte>
it seems that for putting u-boot-spl in the nand a modified livesuit image still is the only way, right?
<quitte>
libv: ?
CaptHindsight has quit [Quit: gone]
CaptHindsight has joined #linux-sunxi
ddc has joined #linux-sunxi
<libv>
quitte: i have never dealt with livesuit
<ddc>
quitte: not quite right you can also do it in uboot or user space/ kernel
MasterChief4942 has joined #linux-sunxi
<MasterChief4942>
Any hint please how to run ANY 3d application on A13 Olinuxino Micro
<ddc>
quitte: have a look at yuq nand driver
<libv>
MasterChief4942: did you try looking at our wiki?
Vanfanel has quit [Quit: Lost terminal]
<quitte>
ddc: what distribution should I flash to get that u-boot without boot0/boot1 onto my ct?
<MasterChief4942>
libv: of course, but where is such article there?
<quitte>
preferably debian
HeHoPMaJIeH has quit [Quit: Konversation terminated!]
<MasterChief4942>
any hint please how to draw at least triangle with OpenGL ES 2.0 on A13
<ddc>
yuq nand driver has some serious issues with UBIFS. Currently Boris is working on a driver for the mainline. One it ready someone may take his work and implement it in uboot
rinni has quit [Quit: Page closed]
<ddc>
I'm afraid for the time being you are in the mercy of boo0/boot1
<quitte>
ddc: i want to use squashfs and jffs2 so ubifs wouldn't worry me
<quitte>
so the current state of executing u-boot is that it is read from a filesystem such as fat? except for yuq's, which is hard to get flashed?
<quitte>
ddc: boris is bbrellion? i didn't know he was working on u-boot,too
physis has joined #linux-sunxi
<quitte>
ah never mind
<ddc>
yuq uboot can boot from mmc so you should be able to write you uboot to the flash
<quitte>
okay. do you know if the wear-levelling for nand is activated on the "normal" kernels? does data in nand from a normal system look scrambled to the mtd kernels?
<quitte>
I guess I now have enough information to play around with. I'll force myself to put something in the wiki tomorrow ;)
avsm has joined #linux-sunxi
konradoo77 has quit [Ping timeout: 240 seconds]
ddc has quit [Quit: ddc]
MasterChief4942 has quit [Quit: Leaving.]
konradoo77 has joined #linux-sunxi
zeRez has joined #linux-sunxi
Skaag has quit [Ping timeout: 255 seconds]
libcg has quit [Quit: libcg]
xavia has joined #linux-sunxi
physis has joined #linux-sunxi
physis has quit [Ping timeout: 272 seconds]
Vanfanel has joined #linux-sunxi
konradoo77 has quit [Ping timeout: 255 seconds]
bonbons has joined #linux-sunxi
leviathanch has joined #linux-sunxi
techn has joined #linux-sunxi
ganbold_ has quit [Ping timeout: 255 seconds]
ganbold_ has joined #linux-sunxi
avsm has quit [Quit: Leaving.]
avsm has joined #linux-sunxi
codekipper has joined #linux-sunxi
<ssvb>
Vanfanel: g2d can only work with physically contiguous memory
<ssvb>
Vanfanel: you can't use it with any arbitrary memory buffers allocated in the userland
<ssvb>
Vanfanel: can't retroarch render directly to the (offscreen part of the) framebuffer?
<Vanfanel>
ssvb: what RetroArch does is to pass a "frame", which is a void* pointing to RGB565 array of pixels with the finished frame. Then I memcpy that frame to the sunxi_disp->framebuffer_addr
<Vanfanel>
do you mean I should re-allocate the frame void* to an area after the sunxi_disp->framebuffer_addr
<Vanfanel>
??
<Vanfanel>
To an area out of the visible screen, but with an offset after the initial framebuffer_addr
<Vanfanel>
right?
<ssvb>
Vanfanel: yes, if you can place this array of pixels somewhere inside of the framebuffer, then you can avoid this extra copy
<codekipper>
libv, Hi, just tried your uboot console patch
<codekipper>
fiat lux
<codekipper>
is a blank screen expected after "Starting Kernel"
<ssvb>
Vanfanel: just be warned that if retroarch ever wants to read back from the framebuffer with the CPU, then this is going to be slow
physis has joined #linux-sunxi
<codekipper>
mripard_, how would you want patches for new devices for mainline?....started with MK802 and MK808.
<Vanfanel>
ssvb: I don't think there's need for it except for screenshots, but I'm not sure I will implement that
<ikeeki>
hi guys, a couple of problems, the first one is: it is bad if dual-core a20 shows different speeds per core? For example 1819.52 for cpu0 and 1813.6 bogomips for cpu1
<libv>
codekipper: did you also try the other patches?
<codekipper>
looking at the cubietruck history it's a gradual process...would you want the same?
<codekipper>
I only saw the one with all the comments
<libv>
codekipper: go try the other patches.
<ssvb>
ikeeki: just ignore bogomips values and try cpufreq-info or some similar tool (or read from sysfs directly)
<codekipper>
shudder...I must of missed them
<libv>
codekipper: and also, simplefb does not exist on 3.4
<libv>
there were two mainline kernel patches to make the code provisionally work
<libv>
codekipper: this all seems to be news to you.
<ikeeki>
and the second one: does anybody experience stuttering when playing MKV files from sata (fbturbo, vdpau) but they goes perfect from USB hd (hdiparm shows good speeds for sata (80M/s) and for USB hdd (18 M/s)
Flashtek has left #linux-sunxi [#linux-sunxi]
<codekipper>
this is on mainline
<ikeeki>
ssvb: thanks
<libv>
codekipper: yes, but without simplefb and without fbconsole enabled
physis has quit [Ping timeout: 272 seconds]
<libv>
there were two mainline patches: one to keep the clocks up, one to alter the config to enable simplefb and fbconsole
<libv>
ah.
<libv>
did you remember to add console=tty0 to the kernel command line?
<codekipper>
I only see the uboot patch
<libv>
look again.
<libv>
linux-sunxi@
<codekipper>
sure thing boss
<ssvb>
ikeeki: afaik mplayer has some command line options to enable buffering of the files to mitigate such issues
<codekipper>
it's on my modified mk808
<ikeeki>
excuse me, I meant mpv, but I think the answer is valid too for mpv
<codekipper>
tty0 should be passed out through the serial as well
<libv>
codekipper: it is ok to pass multiple console= arguments to the kernel commandline
quitte has quit [Read error: Connection reset by peer]
<ikeeki>
the thing is... why do I have to enable those options for sata and not for usb?
<ikeeki>
And another little thing... Do you think that vlc will be able to use vdpau in next 2.2 version (so there will be mpv and vlc to choose between players)?
quitte has joined #linux-sunxi
<ssvb>
ikeeki: you don't have to, but you may try to see if that helps
<ikeeki>
ssvb: thanks again.
<ssvb>
ikeeki: I have no idea about vlc vdpau support plans, maybe it would be best to ask vlc people?
<mripard_>
codekipper: what do you mean?
<codekipper>
libv, found them. I'll have a play with this tomorrow. Ta, can't believe I missed those patches.
<codekipper>
mripard_, I've got the MK808C working with linux-next and was going to deliver the patches for the dts
<ikeeki>
but yes, I'll try to retrieve some info there...
<codekipper>
do I do a series of patches incrementing the funtionality or drop one.
<ikeeki>
It would be kind of birthday present to have accelerated vlc¡
<mripard_>
codekipper: send it as one patch please
<codekipper>
okey dokey
bertrik has joined #linux-sunxi
akaizen has joined #linux-sunxi
avsm has quit [Quit: Leaving.]
konradoo77 has joined #linux-sunxi
_massi has quit [Quit: Leaving]
techn has quit [Ping timeout: 245 seconds]
techn has joined #linux-sunxi
physis has joined #linux-sunxi
akaizen has quit [Remote host closed the connection]
physis has quit [Ping timeout: 272 seconds]
akaizen has joined #linux-sunxi
BorgCuba has joined #linux-sunxi
akaizen has quit [Ping timeout: 240 seconds]
<BorgCuba>
a short quote:
<BorgCuba>
9/13/2013 "...but i first want to fix the last few issues i am running into before i can start cleaning up the lima mesa driver for release." (Luv V.)
ikeeki has quit [Quit: Saliendo]
FR^2 has quit [Quit: Connection reset by peer]
konradoo77 has quit [Ping timeout: 260 seconds]
netlynx has joined #linux-sunxi
netlynx has joined #linux-sunxi
<Vanfanel>
ssvb: looking at how sunxi_disp_init() is implemented, I don't understand how can I access an off-screen area where graphics can be rendered before using sunxi_g2d_blt() to put them in place with the corrected blitting. Is an offscreen buffer also mmapped in sunxi_disp_init()? If that the case, how can I access it? sunxi_disp has a framebuffer_size member, but that's for the visible buffer, right?
RzR has joined #linux-sunxi
<RzR>
hi
<ssvb>
Vanfanel: the 'gfx_layer_size' is the size of the visible part
<ssvb>
Vanfanel: 'framebuffer_size' is the total size of the framebuffer, and it is already mmapped in your application
<Vanfanel>
ssvb: oh! so the difference between the gfx_layer_size and the framebuffer_size is where I am suppose to put "hidden things", right? :D
<RzR>
can anyone confirm the voltage of A20's SPI ?
<ssvb>
Vanfanel: yes
<Vanfanel>
ssvb: great!! Thanks! :)
<ssvb>
Vanfanel: 1920x1080 32bpp screen only needs 8294400 bytes, or less than 8 MiB
<Vanfanel>
ssvb: yes, well, I'm using a 640x480 mode for now, configured in that flex file, so I don't use that much :)
<Vanfanel>
for nes/snes/megadrive era hardware emulation, it's more than enough
<ssvb>
Vanfanel: the sunxi disp driver usually has at least 16MiB reserved, but you should always check at runtime
<Vanfanel>
and I'm in a 16bpp mode, too. RA uses RGB565 pixel format as preferred format
<Vanfanel>
yes, I do a check for memory, adapted for my bpp, as you do in the vsync examples
<ssvb>
Vanfanel: afaik, scaled layers do not support rgb565 format
<Vanfanel>
ssvb: yes, I had to disable scaler for fb0 in the flex file :(
<ssvb>
Vanfanel: but if you need to scale your buffer, then it can be a problem
<ssvb>
Vanfanel: but you should be able to use G2D hardware for this
<Vanfanel>
ssvb: the problem was that I didn't even know what a "scaler" was supposed to be. I mean: if I want to use hardware to scale a 256x240 image to, let's say, my 640x480 video mode, will I need the same "scaler" I disabled in the flex to be able to scale it using G2D?
drfu_ has joined #linux-sunxi
<Vanfanel>
is it a different "scaler"? can be enabled later on?
<Vanfanel>
so many questions, sorry
<ssvb>
Vanfanel: the disp scaler (which is a part of DEFE) and G2D (Mixer Processor) are different pieces of hardware
<ssvb>
Vanfanel: and they have different limitations
<ssvb>
Vanfanel: you can check the a20 user manual and kernel sources to find more information
<Vanfanel>
ssvb: so I can use a G2D "layer" to scale to fullscreen, even if I have disabled the fb0 "scaler", because they are different things, right?
<Vanfanel>
I mean, the G2D layer has it's own "scaler", according to sunxi_disp.c
<ssvb>
Vanfanel: G2D is not a layer, it is a 2D hardware accelerator for memory-to-memory operations
<Vanfanel>
ssvb: yes, but a sunxi_disp seems to have at least a layer. It can be made visible. You use it in the examples, too.
<Vanfanel>
and it's supposed to be used for fullscreen scaling, if I'm not confused
<ssvb>
Vanfanel: "layers" are the display controller things, and they allow up to 4 memory buffers to be composited together (with two of them scaled) on the flight when the data is sent to hdmi
<Vanfanel>
so if I have a frame with a lower resolution than the phisical one, I can make that layer's scanout buffer point to a framebuffer address and that layer will show what's in that address, and these contents can be scaled
<Vanfanel>
that's what I gathered from the examples, at least...
quitte has quit [Read error: Connection reset by peer]
<ssvb>
Vanfanel: yes, but with certain restrictions
<ssvb>
Vanfanel: you can't have a rgb565 layer directly
<ssvb>
Vanfanel: but you can have rgrb565 buffer, which gets converted to argb8888 by g2d in another buffer and then use this another buffer as a disp layer
<ssvb>
Vanfanel: or if you still have this rgb565 buffer in a normal malloc allocated buffer, then you can do the rgb565->argb8888 conversion using NEON instead of memcpy
physis has joined #linux-sunxi
<ssvb>
Vanfanel: this is not very nice and will use some percentage of CPU though (roughly the same as memcpy), so avoiding the copy is always preferable
<Vanfanel>
well, if I can have GD2 make the conversion, then it will be better I guess
<Vanfanel>
I'm trying to use sunxi_g2d_blt() to blit from "disp->framebuffer_addr + disp->gfx_layer_size" to "disp->framebuffer_addr", but it returns false. Both addresses are in the framebuffer adddress, so shouldn't it work?
<Vanfanel>
strangely, it works if I make it blit from, let's say, disp->framebuffer_addr+100 to disp->framebuffer_addr
physis has quit [Ping timeout: 272 seconds]
quitte has joined #linux-sunxi
<ssvb>
Vanfanel: you need to take the sources of sunxi_g2d_blt and modify them a bit
<quitte>
sweet. after disabling the BUG(); line in u-boot I get a more sensible error. ECC too many error at 0.
zeRez has quit []
<ssvb>
Vanfanel: just find which check is failing for you
<Vanfanel>
ssvb: I have them in front of me, so I can modify whatever it takes I guess..
<Vanfanel>
ok
<ssvb>
Vanfanel: what are the values of 'disp->gfx_layer_size' and 'disp->framebuffer_size' on your device?
konradoo77 has joined #linux-sunxi
drfu_ has left #linux-sunxi [#linux-sunxi]
akaizen has joined #linux-sunxi
notmart has quit [Quit: notmart terminated!]
techn has quit [Ping timeout: 250 seconds]
deasy has joined #linux-sunxi
leviathanch has quit [Ping timeout: 240 seconds]
physis has joined #linux-sunxi
codekipper has quit [Remote host closed the connection]
netlynx has quit [Quit: Leaving]
physis has quit [Ping timeout: 272 seconds]
amitk has quit [Quit: leaving]
Andy-D has joined #linux-sunxi
<Vanfanel>
ssvb: sorry, I had a loong phone call :( The tests are failing with this condition: if ((uint8_t *)src_bits >= disp->framebuffer_addr + disp->framebuffer_size)
<Vanfanel>
now I will check these values
techn has joined #linux-sunxi
pwhalen has joined #linux-sunxi
konradoo77 has quit [Ping timeout: 260 seconds]
<Vanfanel>
ssvb: framebuffer_size is 1384448 and gfx_layer_size is 691200
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
konradoo77 has joined #linux-sunxi
Zboonet has quit [Quit: Leaving]
Zboonet has joined #linux-sunxi
<Turl>
RzR: I think it's 3v3 like everything else, but don't take my word for it
bfree_ is now known as bfree
physis has joined #linux-sunxi
ninolein has joined #linux-sunxi
ninolein_ has quit [Ping timeout: 272 seconds]
avsm has joined #linux-sunxi
avsm has quit [Client Quit]
kivutar has quit [Quit: Ex-Chat]
T0mW has joined #linux-sunxi
bonbons has quit [Quit: Leaving]
leviathanch has joined #linux-sunxi
konradoo77 has quit [Ping timeout: 240 seconds]
konradoo77 has joined #linux-sunxi
leviathanch has quit [Ping timeout: 240 seconds]
physis has quit [Remote host closed the connection]
<T0mW>
on the sun7i (a20) how do you specify the pin number for gpio operations (such as gpio_to_int), there doesn't seem to be any header file macros to do this?
<T0mW>
e.g. GPIO_PINH(16)
akaizen has quit [Remote host closed the connection]
akaizen has joined #linux-sunxi
Andy-D has quit [Ping timeout: 272 seconds]
akaizen has quit [Ping timeout: 240 seconds]
akaizen has joined #linux-sunxi
deasy has joined #linux-sunxi
ricardocrudo has quit [Remote host closed the connection]