mnemoc changed the topic of #linux-sunxi to: Allwinner/sunxi development discussion - did you try looking at our wiki? https://linux-sunxi.org - Don't ask to ask. Just ask and wait! - https://github.com/linux-sunxi/ - Logs at http://irclog.whitequark.org/linux-sunxi
CaptHindsight has joined #linux-sunxi
ganbold has joined #linux-sunxi
ecelis has quit [Quit: leaving]
maksimlin has joined #linux-sunxi
ecelis has joined #linux-sunxi
popolon has quit [Quit: Quitte]
mmarker1 has joined #linux-sunxi
mmarker has quit [Ping timeout: 252 seconds]
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
mmarker1 has quit [Remote host closed the connection]
mmarker has joined #linux-sunxi
kuldeepdhaka has quit [Quit: Leaving]
maksimlin has quit [Quit: ChatZilla 0.9.90.1 [Firefox 29.0/20140428193838]]
montjoie1home] has quit [Ping timeout: 252 seconds]
montjoie[home] has joined #linux-sunxi
hramrach has quit [Remote host closed the connection]
tomboy65 has quit [Write error: Connection reset by peer]
<Skaag> is there a board supported by this project that has a full OpenGL implementation GPU?
<Skaag> or are they all OpenGL ES?
TheSeven has quit [Ping timeout: 264 seconds]
TheSeven has joined #linux-sunxi
tomboy65 has joined #linux-sunxi
hramrach has joined #linux-sunxi
jebba has quit [Quit: Leaving.]
FreezingCold has quit [Ping timeout: 240 seconds]
nicksydney has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
amitk has joined #linux-sunxi
MasterChief10 has joined #linux-sunxi
hramrach has quit [Remote host closed the connection]
hramrach has joined #linux-sunxi
MasterChief10 has quit [Quit: Leaving.]
MasterChief10 has joined #linux-sunxi
MadSpark has joined #linux-sunxi
rz2k has joined #linux-sunxi
maksimlin has joined #linux-sunxi
Black_Horseman has quit [Quit: Zwi se logou mou!!!]
ZetaNeta has joined #linux-sunxi
MasterChief10 has quit [Quit: Leaving.]
MasterChief10 has joined #linux-sunxi
Tsvetan has quit [Read error: Connection reset by peer]
libcg has joined #linux-sunxi
bbrezillon has joined #linux-sunxi
tomboy65 has quit [Ping timeout: 272 seconds]
tomboy65 has joined #linux-sunxi
sehraf has joined #linux-sunxi
popolon has joined #linux-sunxi
<mnemoc> moin
issueat has joined #linux-sunxi
deasy has joined #linux-sunxi
fredy has quit [Excess Flood]
bgal has joined #linux-sunxi
fredy has joined #linux-sunxi
ZetaNeta has quit [Quit: Leaving]
notmart has joined #linux-sunxi
tomboy65 has quit [Remote host closed the connection]
tomboy65 has joined #linux-sunxi
gzamboni has quit [Ping timeout: 260 seconds]
<marcin_> Hi. I'm trying to use g2d for streatchblt operation on a buffer taken from UMP. I have a problem with memory corruption after run a g2d now. I would like to ask if someone understand line addr = (__u64)(buffer_addr-0x40000000)*8 + (__u64)((stride * y + x)*bpp);//bits (https://github.com/linux-sunxi/linux-sunxi/blob/sunxi-3.4/drivers/char/sunxi_g2d/g2d_bsp.c, line 535). From "A10 User manual. V1.20" pages 462-463, I know that address (I
<marcin_> assume physical) is 40-bits, but why we should substract 0x40000000 and next multiply by 8?
hramrach has quit [Remote host closed the connection]
hramrach has joined #linux-sunxi
FDCX_ has quit [Ping timeout: 265 seconds]
paulk-aldrin has joined #linux-sunxi
prahal_ has joined #linux-sunxi
<ccaione> moin
issueat has quit [Remote host closed the connection]
<ssvb> marcin_: 0x40000000 is the difference between physical addresses (how the CPU addresses memory) and bus addresses (how the G2D addresses memory)
<ssvb> marcin_: and the comment there says "// bits", this explains multiplication by 8
<ssvb> marcin_: what kind of UMP buffer are you trying to use? how did you allocate it?
<ssvb> marcin_: my guess is that you are somehow messing up virtual addresses vs. physical addresses translation, or maybe even using a non-physically contiguous UMP buffer
<ssvb> marcin_: G2D can only work with physically contiguous memory
<marcin_> ssvb: Thank you for explanation about memory address calculation.
<marcin_> ssvb: I trying to use UMP buffer created in maliModifyPixmapHeader(), which is allocated by ump_ref_drv_allocate( size, UMP_REF_DRV_CONSTRAINT_PHYSICALLY_LINEAR ...). I modified g2d kernel driver to accept ump_secure_id and offset.
<marcin_> ssvb: BTW It is first time, when I can see passing DMA address in "bit" unit...
<marcin_> ssvb: I take physical address for UMP buffer with functions: ump_dd_handle_create_from_secure_id and ump_dd_phys_block_get.
<ssvb> marcin_: if you check the sources, you can find that UMP_REF_DRV_CONSTRAINT_PHYSICALLY_LINEAR has no effect
<ssvb> marcin_: Mali400 has MMU and is allocating memory one page at a time, so the memory is very likely fragmented
prahal_ has quit [Ping timeout: 252 seconds]
<marcin_> ssvb: I understand (it's not OK). Thank you a lot.
<ssvb> marcin_: the only really physically contiguous UMP buffer is the one which wraps the framebuffer
<ssvb> marcin_: it looks like you are using xf86-video-mali ddx, is there a good reason for doing this?
<ssvb> marcin_: also can't you just do image scaling unsing Mali400 instead of trying to use G2D for this?
<marcin_> ssvb: One more question. In general I would like to convert frame from I420 to ARGB. I thought that I use G2D's stretchblt operation. Now I also see that input format can be YUV422 only (it looks like planar YUV format are reserved for output format - three channels are required).
<marcin_> ssvb: by Mali400 you mean use of EGL/OpenGL in DDX?
FDCX has joined #linux-sunxi
<ssvb> marcin_: I have not tried to do YUV <-> RGB conversion with G2D yet, so don't know
<linkmauve1> Skaag, the only mobile GPU I know of that has full OpenGL support is the Nvidia K1, you should go to the Nouveau project for that.
<ssvb> marcin_: by the use of UMP, I was just assuming that you are trying to scale the picture produced by OpenGL ES
<rz2k> linkmauve1: freedreno supports opengl 2.x, vivante GC has unofficial glx support via translation to egl (shaders are a big problem there, many segfaults, but you can run simple games on pure glx).
<linkmauve1> Ah, indeed.
<ssvb> marcin_: G2D has 3 input channels, so planar YUV might work as an input too, this just needs to be carefully checked (in the documentation and by trying some test code)
<linkmauve1> Although the K1 has hardware for up to OpenGL 4.4, while even the latest Adreno GPUs target only GLES.
<rz2k> I'm trying to convince my boss to buy a K1 devkit :p
<linkmauve1> ^^
FDCX has quit [Ping timeout: 265 seconds]
<linkmauve1> rz2k, why would anyone want to use GLX instead of EGL, again? :p
<linkmauve1> You can create a non-ES OpenGL context just fine on EGL.
<rz2k> vivante has DRI1(one) Xorg drivers
FDCX has joined #linux-sunxi
<rz2k> so they kinda have problems with egl
Black_Horseman has joined #linux-sunxi
MasterChief10 has quit [Read error: Connection reset by peer]
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
maksimlin has quit [Ping timeout: 265 seconds]
deasy has joined #linux-sunxi
<Montjoie> If I create two driver with the same exact of_match_table, only one will be loaded right ?
Seppoz_nowork has joined #linux-sunxi
zumbi_ is now known as zumbi
Seppoz has quit [Ping timeout: 252 seconds]
libcg has quit [Ping timeout: 276 seconds]
<rz2k> Montjoie: check what type is of_match_table, if it is then put to a linked list or something - you wont be able to make two with the same name. I bet guys who designed OF already thought it out.
<Montjoie> you speak about which name ? the vamue of .compatible ="" ?
<Montjoie> I want to load 2 plaftorm_driver with the same .compatible = "" name
<Montjoie> it seems to not work
<mripard> Montjoie: you can't
<Montjoie> I explain more my problem: for the seucrity system I made a core platform_driver who own the device
<Montjoie> for each crypto algo I want a platform_driver who "attach" to the core module
<Montjoie> for the moment each algo are in a simple module
<Montjoie> does I need to stay like that or can I use a platfrom_driver
wickwire has joined #linux-sunxi
libcg has joined #linux-sunxi
<mripard> I guess you can do it either by using an mfd, or by not doing it at all.
<Montjoie> "not doing it", you think trying to create a module for each algo is "too much work for nothing" ?
<mripard> yep.
<Montjoie> I want to give choice, but yes I have pain for trying that
blsd has joined #linux-sunxi
Black_Horseman has quit [Remote host closed the connection]
paulk-aldrin has quit [Remote host closed the connection]
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
avsm1 has joined #linux-sunxi
amitk has quit [Quit: leaving]
kivutar has joined #linux-sunxi
vindolin has joined #linux-sunxi
issueat has joined #linux-sunxi
FDCX has quit [Remote host closed the connection]
FDCX has joined #linux-sunxi
avsm has joined #linux-sunxi
rm has quit [Quit: ZNC - http://znc.sourceforge.net]
rm has joined #linux-sunxi
avsm1 has quit [Ping timeout: 245 seconds]
jebba has joined #linux-sunxi
vindolin has quit [Quit: Leaving]
FR^2 has joined #linux-sunxi
ganbold has quit [Ping timeout: 252 seconds]
BMK has joined #linux-sunxi
BMK has quit [Client Quit]
<hramrach> Montjoie: if you really want choice make it possible to unregister the algo without actually loading/unloading anything
<hramrach> so you can pick which algo is handled by the module
<hramrach> hmm, but unregistering without unloading may cause issues too
<hramrach> meh
<Montjoie> I will propose the actual state to linux-crypto, if they do not like it, I will make a monolithic one
<Skaag> linkmauve1: thank you!
nedko has left #linux-sunxi [#linux-sunxi]
libcg has quit [Remote host closed the connection]
avsm has quit [Quit: Leaving.]
avsm has joined #linux-sunxi
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
avsm has quit [Quit: Leaving.]
avsm has joined #linux-sunxi
libcg has joined #linux-sunxi
CaptHindsight has quit [Quit: gone]
avsm has quit [Quit: Leaving.]
CaptHindsight has joined #linux-sunxi
bbrezillon has quit [Ping timeout: 265 seconds]
bgal has quit [Ping timeout: 245 seconds]
Netlynx has joined #linux-sunxi
FR^2 has quit [Quit: Connection reset by peer]
<Turl> Montjoie: you were interested on DMA on mainline, weren't you?
rz2k has quit []
ZetaNeta has joined #linux-sunxi
xavia has joined #linux-sunxi
Undertasker has joined #linux-sunxi
libcg has quit [Quit: dogecoin DDNPK9LJwEzXJK1f7hoMwjvGGGHF8gkBmw]
nove has joined #linux-sunxi
bertrik has joined #linux-sunxi
<nove> in this weekly monday blob news: after spending more than enough time already, just consider libvdecoder.so(new api) as defect and move on
<nove> and did the jpegencoder with the old api, at least now with this one the quality can be set
<nove> the good news is that this blobs aren't so useless as they appear, the encoder is setting new bits and registers
<nove> this gives some hits about the thumbnail talk around the ISP subengine
libcg has joined #linux-sunxi
<nove> 0xa70, 0xa74 is output buffers from ISP, that at first look appears to have a scaled down raw picture data
hansg has joined #linux-sunxi
imcsk8 has quit [Quit: Reconnecting]
imcsk8 has joined #linux-sunxi
<nove> this could mean that ISP is capable and that maybe can be used as raw picture scaler, crop and colorspace conversion
<nove> maybe
wickwire has quit [Remote host closed the connection]
<nove> for this, we have to guess how to use ISP alone
bbrezillon has joined #linux-sunxi
ninolein has quit [Ping timeout: 240 seconds]
ninolein has joined #linux-sunxi
notmart has quit [Quit: notmart terminated!]
FDCX has quit [Remote host closed the connection]
hansg has quit [Quit: Leaving]
FDCX has joined #linux-sunxi
avsm has joined #linux-sunxi
libcg has quit [Quit: dogecoin DDNPK9LJwEzXJK1f7hoMwjvGGGHF8gkBmw]
libcg has joined #linux-sunxi
libcg has quit [Remote host closed the connection]
libcg has joined #linux-sunxi
libcg has quit [Read error: Connection reset by peer]
libcg_ has joined #linux-sunxi
^hawk^ has joined #linux-sunxi
libcg_ has quit [Read error: Connection reset by peer]
libcg has joined #linux-sunxi
[hawk] has quit [Ping timeout: 252 seconds]
bonbons has quit [Quit: Leaving]
avsm has quit [Quit: Leaving.]
<nove> 2 Mbytes patch
bonbons has joined #linux-sunxi
<libv> iirc, there was a difference between 1 and 2
<libv> and since this guy is using a13, he triggered different modes.
shineworld has joined #linux-sunxi
paulk-aldrin has joined #linux-sunxi
libcg has quit [Remote host closed the connection]
libcg has joined #linux-sunxi
jinzo has joined #linux-sunxi
libcg has quit [Remote host closed the connection]
libcg has joined #linux-sunxi
libcg has quit [Remote host closed the connection]
Netlynx has quit [Quit: Leaving]
libcg has joined #linux-sunxi
libcg has quit [Quit: dogecoin DDNPK9LJwEzXJK1f7hoMwjvGGGHF8gkBmw]
MasterChief10 has joined #linux-sunxi
MasterChief10 has quit [Client Quit]
libcg has joined #linux-sunxi
Undertasker has left #linux-sunxi [#linux-sunxi]
protoCall7 has joined #linux-sunxi
libcg_ has joined #linux-sunxi
libcg_ has quit [Client Quit]
libcg_ has joined #linux-sunxi
libcg has quit [Ping timeout: 276 seconds]
libcg_ has quit [Client Quit]
libcg has joined #linux-sunxi
libcg has quit [Quit: dogecoin DDNPK9LJwEzXJK1f7hoMwjvGGGHF8gkBmw]
libcg has joined #linux-sunxi
<mturquette> Turl: -rc8 is very late to take that pull request, and in fact the merge window is already open
<mturquette> Turl: so I think it's best to take your pull request for 3.17
<mturquette> Turl: it will be the first thing I pull once -rc1 drops.
BMK has joined #linux-sunxi
bonbons has quit [Quit: Leaving]
<BMK> Has anybodey ever seen chinese characters in the kernel messages at boot?
<BMK> On the UART
<mnemoc> garbage that decodes as chinese glyphs or real chinese?
<BMK> I presumed garbage - but the weird thing is that it is only where english text would be timestamps etc were readable e.g [1.2345], as were hex values of things.
Black_Horseman has joined #linux-sunxi
issueat has quit [Remote host closed the connection]
shineworld has quit [Quit: Leaving]
paulk-aldrin has quit [Remote host closed the connection]
libcg has quit [Quit: dogecoin DDNPK9LJwEzXJK1f7hoMwjvGGGHF8gkBmw]
libcg has joined #linux-sunxi
<BMK> Sorry for the noise- It seems that somehow Alpha characters are getting mangled but numeric / symbols are not. like my terminal has been tricked into using chinese charset?
<mturquette> you've been hacked by the Chinese army elite hacking unit.
nove has quit [Quit: nove]
FreezingCold has joined #linux-sunxi
sehraf has quit [Quit: www.miranda-fusion.de ... be part of it...]
maksimlin has joined #linux-sunxi
libcg has quit [Quit: dogecoin DDNPK9LJwEzXJK1f7hoMwjvGGGHF8gkBmw]
libcg has joined #linux-sunxi
kivutar has quit [Quit: Ex-Chat]
tomboy65 has quit [Ping timeout: 272 seconds]
bertrik has quit [Remote host closed the connection]
BMK has quit [Ping timeout: 240 seconds]
tomboy65 has joined #linux-sunxi
jinzo has quit [Quit: Leaving]
libcg has quit [Ping timeout: 252 seconds]
ZetaNeta has quit [Quit: Leaving]
libcg has joined #linux-sunxi
libcg has quit [Remote host closed the connection]
Skaag has quit [Ping timeout: 240 seconds]
<atsampson> if BMK comes back, his terminal's been switched into one of the alternative charsets -- restart it or run "reset"...
VargaD has quit [Ping timeout: 240 seconds]
kieppie has joined #linux-sunxi
<kieppie> hi
<kieppie> I'm trying to do some work on my CubieBoard2 running ArkOS (based on Arch) & need some guidance, please. I'm unable to boot (either ArchOnArm refs) or ArkOS, so I'm trying to set the uEnv.txt kernel options to serve me serial terminal over the mini-USB. I'm looking for where/how the interface is served up - i.e. /dev/ttyX. Can anyone please help me out here?
<maksimlin> kieppie: doesnt the cubieboad2 have a real uart ?
<kieppie> maksimlin: it does, but I simply don't have the cable atm
<maksimlin> kieppie: ah, fair enough
<kieppie> besides - it's serving up & talks to the serial (slave) interface anyway
<kieppie> might as use what I've got - provided I can address it
<kieppie> I'm looking at this page, but honestly a little lost : http://linux-sunxi.org/UEnv.txt
<maksimlin> yeah sorry I haven't come across UEnv.txt before either but looks like its a diff way to pass params to the kernel
<kieppie> yip
<mnemoc> linux-sunxi.org/Display is more detailed
<TheSeven> mnemoc: just had another mmc failure, with 90-r1
<TheSeven> so this seems to happen somewhat randomly, not directly related to the kernel version
<mnemoc> pll maybe?
<TheSeven> ...or just a bad card. I need to try another one later
<mnemoc> :)
<TheSeven> I don't see how a PLL could throw this protocol off, other than just general system instability, which would likely cause other problems as well
<mnemoc> ack
<TheSeven> for reference, this is an old SanDisk Mobile Ultra 4GB Class 4 MicroSDHC card on an OLinuXino A20 MICRO board's microSD slot
<TheSeven> no signs of HW instability with rootfs on SATA, with the same kernel
kieppie1 has joined #linux-sunxi
kieppie has quit [Ping timeout: 252 seconds]
ninolein has quit [Remote host closed the connection]
Turl has quit [*.net *.split]
Andy-D has joined #linux-sunxi
Turl has joined #linux-sunxi
xavia has quit [Quit: Leaving.]