ganbold_ has quit [Remote host closed the connection]
geecko has quit [Quit: bye bye]
cajg has joined #linux-sunxi
FreezingCold has joined #linux-sunxi
deffrag has quit [Ping timeout: 240 seconds]
eagles0513875 has joined #linux-sunxi
deffrag has joined #linux-sunxi
Seppoz has quit [Remote host closed the connection]
Seppoz has joined #linux-sunxi
errorex has joined #linux-sunxi
errorex has quit [Client Quit]
FreezingCold has quit [Ping timeout: 240 seconds]
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
FreezingCold has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
wingrime has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
[7] has quit [Disconnected by services]
TheSeven has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
<memleak>
someone pinged me
<memleak>
why do i always get pinged by someone in here, i cant find out who it was in the logs, and i never get pinged again until the next time i leave for a few days.
akaizen has quit [Remote host closed the connection]
<Turl>
memleak: it was mrnuke
<memleak>
ah
<memleak>
what did he say?
code-ninja has joined #linux-sunxi
<code-ninja>
hi :). The Wifi page on linux Sunxi site says that for Realtek USB Wifi ICs, we have to edit the scrip.bin. My tablet uses 8188eu. How can I know to which "usbc_num" my wireless chip is attached to?
<rellla>
the only 2 timesone, that sets the finish_flag = 1 is g2d_handle_irq or the , and this one is only called
<rellla>
sry
<rellla>
the only 2 times, finish_flag = 1 is set, is when reaching timeout and in g2d_handle_irq, and this one is only called in g2d_probe. is this all correct?
wingrime has joined #linux-sunxi
kuldeepdhaka has joined #linux-sunxi
t3st3r has joined #linux-sunxi
deasy has joined #linux-sunxi
<rellla>
something has to check bit 8 FINISHIRQ_FLAG from 0x4 register and set g2d_ext_hd.finish_flag ?
wingrime1 has joined #linux-sunxi
wingrime has quit [Read error: Connection reset by peer]
wingrime has joined #linux-sunxi
wingrime1 has quit [Ping timeout: 265 seconds]
wingrime1 has joined #linux-sunxi
wingrime has quit [Read error: Connection reset by peer]
leviathanch2 has quit [Ping timeout: 240 seconds]
leviathanch2 has joined #linux-sunxi
wingrime has joined #linux-sunxi
wingrime1 has quit [Ping timeout: 240 seconds]
wingrime1 has joined #linux-sunxi
wingrime has quit [Read error: Connection reset by peer]
wingrime has joined #linux-sunxi
wingrime2 has joined #linux-sunxi
wingrime has quit [Read error: Connection reset by peer]
wingrime1 has quit [Ping timeout: 265 seconds]
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
leviathanch2 has quit [Ping timeout: 241 seconds]
wingrime2 has quit [Read error: Connection reset by peer]
wingrime has joined #linux-sunxi
wingrime1 has joined #linux-sunxi
fredy has quit [Excess Flood]
wingrime has quit [Ping timeout: 240 seconds]
fredy has joined #linux-sunxi
wingrime has joined #linux-sunxi
wingrime1 has quit [Ping timeout: 240 seconds]
wingrime1 has joined #linux-sunxi
pwhalen has quit [Ping timeout: 240 seconds]
wingrime has quit [Ping timeout: 265 seconds]
nicksydney has quit [Remote host closed the connection]
t3st3r has quit [Ping timeout: 265 seconds]
nicksydney has joined #linux-sunxi
wingrime has joined #linux-sunxi
wingrime1 has quit [Ping timeout: 252 seconds]
nicksydney has quit [Remote host closed the connection]
t3st3r has joined #linux-sunxi
nicksydney has joined #linux-sunxi
wingrime1 has joined #linux-sunxi
wingrime has quit [Ping timeout: 265 seconds]
t3st3r has quit [Disconnected by services]
t3st3r has joined #linux-sunxi
Black_Horseman has quit [Ping timeout: 265 seconds]
ykchavan has quit [Read error: No route to host]
nicksydney has quit [Remote host closed the connection]
tkoskine has quit [Quit: leaving]
nicksydney has joined #linux-sunxi
nicksydney has quit [Remote host closed the connection]
pwhalen has joined #linux-sunxi
rz2k has quit []
F1skr has quit [Quit: WeeChat 0.4.3]
t3st3r has quit [Excess Flood]
t3st3r has joined #linux-sunxi
diego_r has quit [Quit: Konversation terminated!]
diego_r has joined #linux-sunxi
kivutar has joined #linux-sunxi
nicksydney has joined #linux-sunxi
kivutar has quit [Quit: Ex-Chat]
Guest458207 has joined #linux-sunxi
kivutar has joined #linux-sunxi
Guest458207 has quit [Client Quit]
ferro has joined #linux-sunxi
t3st3r has quit [Ping timeout: 265 seconds]
t3st3r has joined #linux-sunxi
t3st3r has quit [Ping timeout: 265 seconds]
jemk has joined #linux-sunxi
t3st3r has joined #linux-sunxi
nicksydney has quit [Remote host closed the connection]
nicksydney has joined #linux-sunxi
<jemk>
rellla: can you apply this patch and paste its output when g2d crashes: http://sprunge.us/AeCe
kivutar has quit [Ping timeout: 240 seconds]
ferro has quit [Quit: Page closed]
jemk has quit [Ping timeout: 240 seconds]
t3st3r has quit [Ping timeout: 265 seconds]
paulk-collins has joined #linux-sunxi
<shineworld>
spl: not an uImage at 1600 - this mean I have to place uImage at 1600 ?
pwhalen has quit [Ping timeout: 240 seconds]
kivutar has joined #linux-sunxi
paulk-collins_ has joined #linux-sunxi
paulk-collins has quit [Ping timeout: 246 seconds]
pwhalen has joined #linux-sunxi
pwhalen has quit [Changing host]
pwhalen has joined #linux-sunxi
<ccaione>
weird memory location 1600
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
deffrag has quit [Remote host closed the connection]
theskilledworker has quit [Quit: Leaving]
deasy has joined #linux-sunxi
leviathanch2 has joined #linux-sunxi
Guest723471 has joined #linux-sunxi
Guest723471 has quit [Client Quit]
enrico_ has quit [Quit: Bye]
pwhalen has quit [Ping timeout: 265 seconds]
rz2k has joined #linux-sunxi
paulk-collins_ has quit [Remote host closed the connection]
pwhalen has joined #linux-sunxi
pwhalen has quit [Changing host]
pwhalen has joined #linux-sunxi
bertrik has joined #linux-sunxi
bertrik has joined #linux-sunxi
bertrik has quit [Changing host]
lkcl_ has quit [Ping timeout: 240 seconds]
rz2k has quit [Ping timeout: 246 seconds]
Skaag_ has quit [Ping timeout: 264 seconds]
pwhalen has quit [Ping timeout: 265 seconds]
pwhalen has joined #linux-sunxi
deffrag has joined #linux-sunxi
shineworld has quit [Quit: Leaving]
tomboy64 has quit [Read error: Operation timed out]
Dodger78 has joined #linux-sunxi
tomboy64 has joined #linux-sunxi
<Dodger78>
does the uboot git from hansg also boot the older 3.4 kernels on cubietruck ? to me it seems it doesnt
nabblet has quit [Quit: leaving]
kuldeepdhaka has joined #linux-sunxi
kuldeepdhaka has quit [Max SendQ exceeded]
kuldeepdhaka has joined #linux-sunxi
popolon has quit [Quit: Quitte]
netlynx has joined #linux-sunxi
netlynx has quit [Changing host]
netlynx has joined #linux-sunxi
_massi_ has quit [Quit: Leaving]
ynezz has quit [Ping timeout: 264 seconds]
ynezz has joined #linux-sunxi
rings_IIV has quit []
rings_IIV has joined #linux-sunxi
HeHoPMaJIeH has quit [Read error: Connection reset by peer]
Skaag has joined #linux-sunxi
<oliv3r>
shineworld: that uImage at 1600 is a known bogus message
leviathanch2 has quit [Ping timeout: 240 seconds]
Jachy has quit [Read error: Operation timed out]
Seppoz has quit [Remote host closed the connection]
Seppoz has joined #linux-sunxi
notmart has quit [Excess Flood]
notmart_ has joined #linux-sunxi
<memleak>
is there a (big) difference between Guinness extra stout and Guinness foreign extra stout? (most likely intended for mrnuke) i'm drinking one now (not foreign) and i love it.
IrcDroidClient has joined #linux-sunxi
IrcDroidClient is now known as pirea
<pirea>
hi
<memleak>
hello!
<memleak>
like beer? xD
<pirea>
vdpau-libs can't be used without x?
<pirea>
yes:)
<memleak>
vdpau is a GPU-based video decoding acceleration API so i don't see how you COULD use it without X
<pirea>
i had realise something... libvdpau-sunxi is fast, but x server is slow on sunxi based hardware
<libv>
hrm, codekipper his mk808c claims on its case to have a dual core a9 inside
<memleak>
perhaps an incredibly fancy GPU-based framebuffer with full 3D rendering support, like super low-level purely GPU-driven fbcon...
<memleak>
futuristic stuff that you'd never actually see in Linux but an incredibly awesome idea.
<memleak>
@ pirea btw^
<memleak>
that'd be the only way I see to use vdpau without X.
<libv>
memleak: ever heard of directfb?
rz2k has joined #linux-sunxi
<memleak>
you can't use vdpau with directfb, can you?
<pirea>
libv in what state is disp driver for sunxi?
<libv>
disp? it's horrid, but it somewhat works
<memleak>
on the kernel side of things, the code is pretty ugly.
<pirea>
i saw that
<memleak>
XD
<libv>
memleak: i have no idea what media decoding support there is in directfb
<memleak>
it's that obvious eh?
<libv>
but it is an incredibly fancy gpu-based framebuffer with full 3d rendering support
<pirea>
openmax?
Jachy has joined #linux-sunxi
<memleak>
libv can directfb work without any X server / wayland?
<libv>
memleak: hence the direct in its name
<memleak>
why isn't it popular then?
<pirea>
memleak the name tells you that :))
<pirea>
DIRECT...
<libv>
memleak: it actually is fairly popular, but just not for the desktop use case
<memleak>
why not for desktops? X sucks and everyone on the planet knows it, even the developers of it know it's terrible.
<libv>
memleak: why does X suck?
<libv>
memleak: and what developers are stating so?
<libv>
keithp, daniels and ajax
<libv>
but this is just so people think "gor, those guys are soo great, they work on this antique codebase and they apparently do not enjoy it, but they push on through, for our benefit, and are truly great chaps"
<memleak>
because it still uses that 1980s vibe (X server keyword: SERVER and stuff and wayland was supposed to (and hopefully eventually stabilizes enough to be a drop-in replacement))
<libv>
memleak: how long has wayland been claimed to absolve X?
<libv>
5 years already.
<libv>
and they are still busy reinventing half of the xserver
<memleak>
true it's been awhile but desktop managers such as LXDM still need to work with it.
<pirea>
libv mir will succed?
<memleak>
i think GDM / KDM (gnome / KDE) still need further support as well
<ssvb>
memleak: the linux kernel is being developed using decades old programming language, so it must suck by definition, right?
<libv>
pirea: i don't know
<libv>
pirea: it would actually suck quite badly if it doesn't
<libv>
pirea: read my mir blog entry from a few months back on why that is
<libv>
+so
<memleak>
ssvb, when referring to the SCSI stack and networking stack, yes but some of it is actually modernized/
<pirea>
i ill do that libv
<libv>
libv.livejournal.com/25325.html
akaizen has joined #linux-sunxi
<libv>
one sentence from it: "What exactly gives these people the sole monopoly on reinvention?"
<hramrach>
ssvb: does it not suck?
akaizen has quit [Read error: Connection reset by peer]
<ssvb>
memleak: I'm talking about the C language, which has been also challenged many times by wannabe replacements during these decades :)
<hramrach>
it is using decades old obsolete design, too
<memleak>
ssvb, no I love C.
akaizen has joined #linux-sunxi
<hramrach>
it's not that C does not suck. There just is no replacement
<memleak>
libv, don't you think it's a bit silly to call something a reinvention when the original invention is now FUBAR?
<ssvb>
hramrach: the alternatives just happen to suck more
<libv>
memleak: that's a very stupid and shortsighted statement that you just made.
<libv>
memleak: i would've suggested you to read that blog entry as well, but now i would like to discourage you from doing so
<memleak>
when a project is FUBAR, for example, take X, what can you do it to it besides re-write the official tree? or do you just not like multiple trees of the same thing?
leviathanch2 has joined #linux-sunxi
<libv>
you wouldn't get the point anyway, and you would just be shouting out further mantras without thinking
<hramrach>
the alternatives for application programming actually provide different tradeoffs compared to C and for majority of applications C is not the language of choice. The laternatives for system programming do not exist.
uRandomMM has joined #linux-sunxi
<memleak>
what you call re-inventions i call replacements. some things are terrible to where developers don't want to touch it anymore and rather in fact decide, let's "re-invent" it!
<libv>
and the really cool thing is that you are actively being proven quite wrong
<libv>
but the noisemakers of the world haven't noticed what's really going on yet
<memleak>
so whats wayland for then? honestly.
<pirea>
will be blood
<libv>
memleak: wayland is krh his own little playground, first and foremost
<hramrach>
laughing
<hramrach>
or what is it for?
<libv>
and quite some people happen to be running along with it
<hramrach>
can wayland handle user input?
<hramrach>
can wayland be used over network?
<hramrach>
can wayland be used securely?
<libv>
hramrach: the latter was first claimed to not be needed
<libv>
hramrach: and work to add it was done somewhere 2011 or so
<libv>
the network bit that is
<libv>
that's the hallmark of reinvention, right there.
kuldeepdhaka has quit [Ping timeout: 240 seconds]
<hramrach>
so it can be used over network but still has no security and user input
<hramrach>
sounds like another X, even worse
notmart_ has quit [Quit: notmart terminated!]
<libv>
no! it's better
<libv>
because it's newer!
<libv>
it was done all from scratch!
<libv>
therefor it must be better!
<hramrach>
you would think that people reinventing X will learn from its mistakes ...
<memleak>
well wait, i didnt say newer is always better, but the current state of X is a fucking mess.
<libv>
it fixes the things we learned in how graphics hw works today!
<memleak>
hence a replacement is in fact needed IMHO.
<libv>
and it throws away everything else X learned in the last 30 years
<hramrach>
well, I guess that pretty much sums it
<memleak>
i dont care if its an older yet cleaner, simplier variant of X, but X is a hodge podge, and someone like you libv, the graphics expert, all your work with lima and such, should know how much repair the X server needs.
<libv>
memleak: you really have no idea, and you are very actively prancing that lack of insight around here
<libv>
memleak: i write graphics drivers, always have and always will.
<memleak>
exactly.
<memleak>
so i'm surprised you're in bed with X so much...
<libv>
memleak: you really do not see it do you?
<libv>
memleak: i do not care about X, wayland or mir
<oliv3r>
memleak: a lot of set-top-boxes and entertainment systems etc, from old days a mostly, use(d) directfb
<libv>
i do care about people shortsightedly reinventing everything all the time and never ever getting anything decent working for a measurable amount of time
<memleak>
really? you've been defending it since this conversation started.
<oliv3r>
and xbmc on the raspberry pi uses the regular FB with openGL and hardware acceleration; so why wouldn't it be possible?
<libv>
memleak: unlike you, i have been in this game for more than a decade
<oliv3r>
pirea: to answer you question, vdpau at this moment, has only 1 output, and that is X
<libv>
and i have seen these things play out over a slightly longer amount of time
<mrnuke>
<libv> i do care about people shortsightedly reinventing everything all the time and never ever getting anything decent working for a measurable amount of time
<mrnuke>
true story, brother!
<libv>
and i never ever seen anything that worked satisfiedly for a measurable amount of time
kuldeepdhaka has joined #linux-sunxi
<libv>
things break, and people only seem to want to break things more
<pirea>
oliv3r ok
<memleak>
except.... ready for it.... X!
<libv>
nobody every goes and fixes anything
<libv>
people just go and reinvent
<libv>
memleak: you are being actively proven wrong
<memleak>
it's perfect, it works for everyone, its been around since the 1900s so it must be AWESOME
<libv>
memleak: and in the next two years everyone will see it
<ssvb>
memleak: are you currently using wayland on your desktop system?
<hramrach>
memleak: it's broken and ugly. But since the alternatives are no better ..
<memleak>
looks like i just flipped the table. everything old that has a much larger footprint and cruddy old crap is better and should never be changed.
<libv>
memleak: for now, only those who closely follow the games of the X world, and who have seen those games played for a longer period in time, they see what really is happening
<libv>
ssvb: nice one
<pirea>
hramrach :)) lol
<hramrach>
wayland has been designed by graphics people. They know how to do graphics and do not know or care about anything besides graphics
<libv>
memleak: another statement of yours that shows just how shortsighted you are in this whole story
<memleak>
we should all just go back to using Alpha architecture too because it's depreciated as shit
<libv>
hramrach: s/graphics people/a person who never wrote a graphics driver/
<libv>
hramrach: but that's ok
<libv>
that's not the bad thing about it
<libv>
the bad thing is this "let's go reinvent it from scratch" attitude that he has
<pirea>
good bye
pirea has quit [Quit: pirea]
<libv>
even though he was one of the people who killed XGL
<hramrach>
well, it's miles ahead of X in terms of doing graphics with it so the person who wrote it at least looked at a graphics I guess
<libv>
when he did AIGLX, and achieved similar effects with a standard X server
pirea has joined #linux-sunxi
pirea has quit [Client Quit]
<memleak>
libv, if i truly am shortsighted and everything must be new, why am i drinking Guinness that goes back to 1759? OH SNAP
<libv>
memleak: ah, right
<libv>
memleak: go to a pub, like normal people
<memleak>
pubs arent IRC friendly
<libv>
memleak: and drunkenly expell your alcoholized theories there.
<memleak>
snap again.
<mrnuke>
memleak: you can't win a drunken argument with libv. Google him
<memleak>
mrnuke, there you are!
<memleak>
and yes i know very well who libv is, i admire him greatly.
<memleak>
regardless of our disagreements he is a great guy, and i know i'm probably wrong but so be it.
<mrnuke>
s/probably//g
<ssvb>
memleak: you haven't replied to my question about your wayland usage experience, so I guess you don't really have much to say about it
<memleak>
ah harr harr harr
<memleak>
ssvb i've never used wayland because of it's stability, I use X and i hate it but i honestly should switch.
<memleak>
*instability
<ssvb>
memleak: why people like you think that destroying X is going to make your happy wayland future come sooner?
<memleak>
i didnt say destroy it.
diego_r has quit [Quit: Konversation terminated!]
<hramrach>
so open another Guinness and start switching :D
<memleak>
i just said it sucks and i wish it had a simplier better alternative.
<memleak>
wayland is getting closer than X ever will be, thats all.
<libv>
memleak: how old is wayland these days?
<memleak>
old.
<libv>
memleak: since it is supposedly all new and brilliant and clean
<memleak>
its not about age.
<memleak>
its about X server sucking.
<libv>
memleak: why is it not stable or useable yet?
<hramrach>
heh, they think that by destroying X people will have to switch :P
<memleak>
i never said kill X, i said i wish there was a newer alternative with a better code base that didnt suck so much fucking balls, thats all.
<hramrach>
but really they should make wayland so good that people will switch just because it's sooo better than X :p
<memleak>
now you all can enjoy putting words in my mouth but honestly i just hate X and wish something better came along and i have high hopes for wayland because of its simplier code base, thats all.
<hramrach>
sadly Wayland is not one
<memleak>
well what is, genius?
<memleak>
got one?
<memleak>
DO YA GOT ONE?
<hramrach>
no
<hramrach>
that's why I don't go around saying how horrible X is
<memleak>
but it is horrible.
<hramrach>
when there is nothing better it's the most beautiful windowing system there is :p
<memleak>
oh i see your point of view now.
<memleak>
if something isn't better, then it isn't bad, i see i see.
<memleak>
you're unable to look at a project and see how much it lacks because its the only one that got that far.
<ssvb>
memleak: what kind of graphics drivers and hardware do you have? the quality of drivers also matters a lot, not everything can be blamed on the X11 architecture
<libv>
memleak: come back with your arguments in march 2015
<libv>
memleak: let's see whether you still state the same things.
<memleak>
now see, regardless of a lack of better project, i can look at some source code and say "WOW THIS SHIT IS UGLY"
<memleak>
unlike.. anyone else here LOL
<libv>
memleak: i actually have done some work on X
<memleak>
i say X sucks because its filled with bloat, regardless of how far behind everything else is.
<memleak>
libv me too..
<libv>
memleak: but not on the protocol side
<memleak>
not as much tho
<libv>
and i was drinking just 16h ago with one of the authors of the parts i worked on
<libv>
and freshly re-appointed xorg board member as well
<libv>
it did a few things rather well
<libv>
things that got thrown out of the window with RandR1.2 and KMS
<libv>
reinvent kills the good bits as well.
<libv>
always.
<memleak>
libv are you saying that X is actually going to get better? if so, thats awesome and i can't wait, but honestly i have yet to see anything great happen to the project in the past while.
<libv>
and the end result is always worse.
<hramrach>
well, not always. But with something like a graphics driver nobody wants to redo all the debugging that went into the old code after starting afresh
<hramrach>
so the fresh codebase tends to be rather buggy as in working on very small subset of hardware it's supposed to
<ssvb>
memleak: again, you haven't said anything about your graphics driver, some of them really suck
<libv>
hramrach: people usually reinvent for the sole reason of not having to look at other peoples code
<libv>
hramrach: reinventing != rewriting
<libv>
rewriting is what people should be doing.
<libv>
failing that, they should first try to learn as much as possible about the existing implementation
<libv>
and actually take over the good bits
<libv>
and this is what never happens.
<libv>
the baby always gets thrown out with the bathwater
<memleak>
ssvb, about my hardware, being a professional gamer and all and previously sponsored by AMD: AMD FX-6300, two GIGABYTE R9 290s, 16GB DDR3 RAM @ 1866, Astro A40 System, Creative Labs Sound Blaster Z PCI Express sound card and before you talk about the drivers for my GPU being unstable and that being the reason I hate X, i've ran about 20 other cards with X as well, starting off with the radeon 9800 p
<memleak>
ro.
<libv>
memleak: how well is that r200 of yours still supported today?
<memleak>
Cooler Master 1000W Pro Gold PSU btw
<hramrach>
oooh the bugs in radeon 9200
<hramrach>
broken line acceleration that would lock up the card that never got fixed
<libv>
9200 was the rv200, right?
<libv>
heh
<hramrach>
they just turned it off
<memleak>
its about code base of X not GPU drivers im complaining
<memleak>
about^
<hramrach>
yeah, the cheap office one
<libv>
hramrach: the openchrome guys, who forked away from me because i didn't accept their VBE modesetting shortcut, did the same thing for a decade
<libv>
if for instance, the hw cursor didn't work, they disabled it
<libv>
instead of working towards fixing it.
<hramrach>
but you can probably still run X on rv200 without problems today given the installed base
<libv>
i am not too certain
<libv>
with radeonhd we wanted to start afresh, as to not trample all over r100-rs4xx code, and to give it a chance to stabilize
<ssvb>
memleak: how do you know that it is the code base of X at fault and not the bugs in the half-baked drivers for the bleeding edge hardware?
<libv>
but that sort of thing was not accepted
<hramrach>
well, there is the problem with AGP bridges locking up
drfu_ has joined #linux-sunxi
<memleak>
like i said, 9800 pro as well
<hramrach>
but in PCI mode it would most likely work
<memleak>
you're missing the other stuff im saying so i give up
<ssvb>
memleak: are you unhappy about your 3D games or what?
<memleak>
no
<memleak>
im happy
<memleak>
on phone bbl
<ssvb>
memleak: you are not telling what is your problem exactly
<libv>
hramrach: i never really dug out what was really failing there? was that a gart issue?
<libv>
or was that stuff like bus mastering and things?
<hramrach>
there was some mismatch between how the BIOS set up the AGP and how the driver thought it's set
<libv>
heh, surely that is fixable.
<ssvb>
memleak: and the source code being aesthetically displeasing when you looked at it is a laughable argument :)
<libv>
but then, agp days was when i was on my own in the graphics world with trying to be bios independent
<hramrach>
setting some random values of AGP speed until it matched what the BIOS set up (if it did set it up properly) 'fixed' the problem
<libv>
i see
<libv>
surely it wouldn't be overly hard to figure out what bits to poke for that on the northbridge?
<hramrach>
probably. but today AGP is obsolete so all i care it can run in PCI mode
<libv>
but one would have to do this for each northbridge family from scratch
<memleak>
ssvb its not so much the actual source code but how cludgey the interface is for X
<libv>
and do a partial bios disassembly (where you know what you are looking for) for new versions
<libv>
hramrach: true
<hramrach>
and accommodate buggy revisions, etc
<memleak>
like for an example, i use PAM for a lot of my applications and depending on how i start X, PAM may or may not work
<hramrach>
with PCIe being properly standardized this is moot
<ssvb>
memleak: interface for the drivers inside of X or for the applications?
<libv>
hramrach: yeah
<memleak>
application
<memleak>
i know what is a 3d driver issue and what is an X issue
<libv>
hramrach: a decade old already
<memleak>
but you're all just still going to be stuck on my r9 290 not thinking about the other 20 cards i used with X
<memleak>
so it doesnt matter really
<memleak>
im done
<libv>
memleak: finally.
<memleak>
lol and im short sighted
<libv>
memleak: good night
<memleak>
you guys can be real jerks
<memleak>
like honestly
<memleak>
libv im starting to not like you as much
<hramrach>
memleak: like X can be held responsible for you setting up PAM incorrectly :p
<memleak>
X sucks and it always will depending on video cards and on that note, fuck you.
memleak has left #linux-sunxi ["Leaving"]
<libv>
:)
memleak has joined #linux-sunxi
<memleak>
and btw that meant to say not depending on video cards
<memleak>
because X sucks regardless
memleak has left #linux-sunxi ["Leaving"]
<libv>
alcohol and irc really do not mix well.
<libv>
which reminds me, i was supposed to commit a patch while Turl was in a good mood about it
<ssvb>
I'm just afraid his dreams are going to be shattered after he moves to wayland and encounters his first wayland related bug :)
<Turl>
libv, hurry up, I'm on a bus and 20m away from a pc ;)
<libv>
Turl: hehe :)
<Turl>
irssinotifier ftw :P
<WarheadsSE>
the F was that.
<hramrach>
hello
<hramrach>
are you a second edition or what?
bonbons has joined #linux-sunxi
<WarheadsSE>
just caught that "pile" that memleak
<hramrach>
but SE
<WarheadsSE>
Well at least you didn't ask if I was from Sweden
<hramrach>
is asking about Sweden rude?
<WarheadsSE>
No, just a lot of people do
jemk has joined #linux-sunxi
<hramrach>
speaking of Sweden .. I heard Swedish movies are special but never cared enough to experience that myself
techn_ has joined #linux-sunxi
<oliv3r>
anybody able to help me with a brainblackout
<oliv3r>
if i have dividers of 1, 2, 4 and 8; and they correspond to bitvalues of 0, 1, 2 and 3
<oliv3r>
what's a quick and easy formula to convert between the two
<oliv3r>
other then sqrt()
<libv>
a loop
<oliv3r>
(n >> 1) -1
<oliv3r>
i think
<oliv3r>
no i think wrong; damnit i'm sure i've done it before with a define
hawi_ has joined #linux-sunxi
<libv>
oliv3r: there are assembler instructions for at least i386 which do that
<oliv3r>
i'm looking for pre-compiler macro's though
<libv>
oliv3r: but i do not know another option apart from a loop
<libv>
in C
<libv>
some code in the mali binary driver is pretty huge and nasty when you look at the assembly
<oliv3r>
hehe
<oliv3r>
this is just to setup some bitfields
<libv>
loads of weird masking
<oliv3r>
i'm pretty sure i've done it before; that's why it frustrating
<oliv3r>
Turl: the patch wasn't even done great;, it's a 2 line (copy paste patch mind you, it's from the SDK) and then plant your own name as copyright holder
<oliv3r>
Turl: is that even legally allowed? copy paste something from the SDK and then put your own name on it?
t3st3r has joined #linux-sunxi
<Turl>
oliv3r: -ENOPATCH != -ENOPATCH? o.O
<oliv3r>
Turl: english porvavor :p
mdp has joined #linux-sunxi
<oliv3r>
ijc: ^ since your doing most of the cleanups :p
<oliv3r>
i get that bit; just not how you mean it :p
diego71 has quit [Ping timeout: 240 seconds]
<Turl>
oliv3r: well you linked me to a patch with him as author :p
<Turl>
oliv3r: as for the 'is it legal?', IANAL
<oliv3r>
well if he wrote the code from scratch;s ure; but if it is copied from the SDK, is it still your copyright?
<oliv3r>
if you read the sdk, and rewrite it, sure I suppose
<Turl>
probably not on that case
<Turl>
althought the signed-off-by would be required
<oliv3r>
Turl: yeah but what i was after, it was a crappy 1 line patch; and then with his name as copyright holder in the top of the file
<oliv3r>
i can understand the AW devs there, they wrote it initially
<oliv3r>
but adding every person for each small patch ...
<ijc>
oliv3r: Not sure what you were trying to direct me to.
t3st3r has quit [Ping timeout: 265 seconds]
<oliv3r>
ijc: the copyright notice :p
<oliv3r>
ijc: and hi!
Black_Horseman has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
Black_Horseman has quit [Changing host]
<oliv3r>
ijc: also, i dunno if you checked your mail, i replied to the 2/9 cleanup patch
<ijc>
oliv3r: If you think something is wrong with the copyright notices can you send a patch to correct them please. I don't feel the need to dig into that particular pit.
<ijc>
I didn't see your mail, I suspect the sunxi list reply-to has struck again
<oliv3r>
ijc: i'll resend it to you; sec
<oliv3r>
ijc: as for the copyright, i guess i understand that that pit is an ugly one :p
<oliv3r>
i figured if it's going upstream; might aswell clean it a bit
<ijc>
Found your mail in the sunxi mbox
<oliv3r>
ijc: oops; you got it double then
<ijc>
The whole situation with the copyrights and signed off bys is such a mess its not worth dealin with IMHO. I listed everyone who touched the file inthe commit message, which is the best compromise I think
<oliv3r>
ijc: i agree
<ijc>
seocnd guessing whether someone was right or wrong to add their copyright header or whether they copied or rewrote the code etc is not a productive use of time as far as I'm concerned, I'm not going to touch that can of worms.
<oliv3r>
ijc: very fair enough :)
<oliv3r>
it is a nasty can, i give you that
deffrag has joined #linux-sunxi
<oliv3r>
ijc: anyhow, working on a patch for 2/9
<oliv3r>
i saw 3/9 addresses one issue
<ijc>
oliv3r: TBH I think those sorts of cleanups are just going to get in the way of the upstream process, they aren't necessary.
<oliv3r>
but it odesnt' help readability that much does it?
<ijc>
If you want to take a sweep at it after it has gone upstream then that would be great, but otherwise it is just creating needless churn for me.
<oliv3r>
you still have to look up everything in the manual
<Turl>
mripard: ping? :)
<oliv3r>
ijc: ok i don't wanna make it more difficult for you :)
<ijc>
readability> this sort of thing reads quite naturally to kernel and bootloader folks
<oliv3r>
i suppose so :)
t3st3r has joined #linux-sunxi
<oliv3r>
ijc: you are more expierenced in that regard :)
<oliv3r>
ijc: though cleanups like that have been done on the other half of the code (well i did some bits allready in dram.c etc ages ago)
<oliv3r>
won't it be confusing to have the two styles 'cross'?
<ijc>
I think the only reason to redo it another way would be if the u-boot maintainers want tit done differently
<oliv3r>
rgr!
<ijc>
oliv3r: If there are only two style of this stuff in u-boot I will eat my hat. and yours
<oliv3r>
ijc: :p
<oliv3r>
ijc: we should have a poule to see how many styles there will be eventually :)
diego71 has joined #linux-sunxi
<oliv3r>
anyway, i'll keep my patch somewhere safe for later then
<oliv3r>
was almost done :(
<ijc>
sounds good.
<ijc>
I'mhoping to send v2 in the next half hour or so. just need to check mmc booting
<oliv3r>
great!
<oliv3r>
if there's anything I can clean up on; do say so
<oliv3r>
ijc: what about the 2 GiB ram support bug; is it worth persuing that thing?
<ccaione>
/quit
<ccaione>
damn
<oliv3r>
nooo don't go!
<ijc>
oliv3r: 2GiB RAM support bug?
<ccaione>
oliv3r: I'm moving to the bed
<ccaione>
back in 5 min
<ccaione>
:)
<ijc>
Not sure what thatis, but it seems like it could be fixed later in mainline, things seem to work pretty well regardless
<oliv3r>
i think i submitted it to mainline twice :(
<oliv3r>
but it got semi-ignored both times aswell
<ijc>
Ah that, I'd been ignoring that for the first pass, itdidn't seem like a huge problem given I've only got 2GB of RAM on this board
<oliv3r>
i can dig up the whole thing if needed
<oliv3r>
the patch is against mainline mostly
<oliv3r>
and it only fixes a minor thing i guess
<oliv3r>
but it was slightly talked about; but nothing came from it :(
<ijc>
I htink it would be best to fix this against mainline, in principal it doesn't even need the sunxi stuff to be there, it's eithr an issue or it isn't
<oliv3r>
i only sent the mainline patches upstream :)
<oliv3r>
i kept the sunxi stuff here
<ijc>
it's not that bad if u-boot can only use 2gb of ram that is "plenty" for a bootloader. The only reason it would matter is if it ended up telling Linux there was only 2GB.
<oliv3r>
i thinkt he bug was that it was using long, instead of unsigned long
<oliv3r>
it was printing wrong memory sizes initially
<ijc>
u-boot on midway (an 8GB system) seems to work fine and Linux sees all the ram.
<oliv3r>
but that info also gets passed to linux aswell
<oliv3r>
it's only a minor bug i suppose as it gets type casted around a few times
netlynx has quit [Quit: Leaving]
<ijc>
I wouldn't be surprised if the real issues started at 4GB because the casting was hiding things...
<oliv3r>
yeah
<oliv3r>
it's a bug, just not a big bad one
<oliv3r>
and it probably won't bite often ;)
<oliv3r>
i do think i merged it into sunxi
jemk has quit [Quit: bye]
<ijc>
there's certainly some changes to unsigned long for the ram stuff in u-boot-sunxi.git
<ijc>
Anyway, there goes v2 of the series. Time to do something else for a bit!
<oliv3r>
ijc cool and enjoy!
<ijc>
thanks, good night
<oliv3r>
ijc: in case you read this later; i notice in clock.h for CCM_PLL5 i us CTRL, probably should have used CFG :p
ZetaNeta has joined #linux-sunxi
<ZetaNeta>
Hi
<oliv3r>
hey
<ZetaNeta>
I failed to find a allwinner smartphone.... Tried to do that several times already, and several times asked....
<ZetaNeta>
If not allwinner, what other "50 euros" smartphone i can take?
<ZetaNeta>
It should be GNU-Friendly. I wanna put Gentoo there :3
<ZetaNeta>
(RockChip for example. Not some qualcomms)
<tonikasch>
hi, sorry to bother here (i'm building on rockchip platform, but found something interesting on your site that touches rockchip case)
<skaczor>
Turl: Thanks for the tip, I will try that
<libv>
tonikasch: /topic says "Don't ask to ask. Just ask and wait!"
<tonikasch>
currently looking at the vpu elements and we have: @ vpu_service.ko kernel module (code is provided in some cases) @ libomxvpu.so and libomxvpu_enc.so that seem to be openmax il components (but require not only libvpu.so (rockchip specific), but many android libraries like libmedia.so and ohers like libstagefright.so)
<tonikasch>
here it goes
* Turl
assumes tonikasch is interested in libhybris
<libv>
or in REing
<libv>
but he hasn't asked his question yet :p
<tonikasch>
I know libhybris would be a way (but imho not a good one), but i'm overall interested in REing
<libv>
tonikasch: i am the guy who started the whole ARM graphics REing stuff
<tonikasch>
:D
<libv>
tonikasch: get the kernel source, find out how userspace communicates with the kernel
<Turl>
most of the Cedar RE people don't seem to be around now
<Turl>
they got some nice tools to trace stuff I hear
<libv>
tonikasch: then use libc wrappers (like the wrap/ directory in lima) or ptrace to capture all this communication