tllim has quit [Read error: Connection reset by peer]
tllim has joined #linux-sunxi
pmpp has quit [Ping timeout: 268 seconds]
Gerwin_J has quit [Quit: Gerwin_J]
cnxsoft has joined #linux-sunxi
popolon has quit [Quit: WeeChat 2.0.1]
return0e has joined #linux-sunxi
soderstrom has quit [Ping timeout: 268 seconds]
chewitt has joined #linux-sunxi
dddddd has quit [Remote host closed the connection]
BenG83_ has joined #linux-sunxi
huawei has quit [Quit: Leaving]
BenG83 has quit [Ping timeout: 240 seconds]
tl_lim has joined #linux-sunxi
tllim has quit [Ping timeout: 265 seconds]
uwe_ has quit [Ping timeout: 240 seconds]
uwe_ has joined #linux-sunxi
Hao has joined #linux-sunxi
cnxsoft1 has joined #linux-sunxi
cnxsoft has quit [Ping timeout: 252 seconds]
cnxsoft1 is now known as cnxsoft
reinforce has joined #linux-sunxi
Hao has quit [Ping timeout: 240 seconds]
vagrantc has joined #linux-sunxi
TheSeven has quit [Ping timeout: 265 seconds]
mpmc has quit [Ping timeout: 248 seconds]
JohnDoe_71Rus has joined #linux-sunxi
SP7RT has quit [Quit: Lost terminal]
lurchi_ has joined #linux-sunxi
lurchi__ has quit [Ping timeout: 264 seconds]
TheSeven has joined #linux-sunxi
tllim has joined #linux-sunxi
<wens>
didn't realize debconf was a week long... O.o
tl_lim has quit [Ping timeout: 250 seconds]
<vagrantc>
wens: and preceeded by a 5+ day debcamp! :)
TheSeven has quit [Ping timeout: 272 seconds]
IgorPec has joined #linux-sunxi
<wens>
uh... wow :)
<wens>
dead of summer, it's going to be really hot
<vagrantc>
yup
mpmc has joined #linux-sunxi
cnxsoft has quit [Ping timeout: 248 seconds]
vagrantc has quit [Quit: leaving]
cnxsoft has joined #linux-sunxi
TheSeven has joined #linux-sunxi
iamfrankenstein has joined #linux-sunxi
nuuuciano has quit [Ping timeout: 240 seconds]
f0xx has joined #linux-sunxi
VargaD has quit [Ping timeout: 240 seconds]
BenG83 has joined #linux-sunxi
VargaD has joined #linux-sunxi
BenG83_ has quit [Ping timeout: 256 seconds]
vpeter has quit [Remote host closed the connection]
chewitt has quit [Quit: Adios!]
vpeter has joined #linux-sunxi
Andy-D_ has joined #linux-sunxi
<kilobyte>
I don't know if I'll make it to your side of the planet, but if I'll do, you're responsible for dragging the mainland gang to Taiwan :)
<kilobyte>
debconf there is an occasion that would be a shame to waste
Hao has joined #linux-sunxi
Hao has quit [Ping timeout: 265 seconds]
<wens>
afaik it's a bit harder for them to come over, as you need permits from _both_ sides
tl_lim has joined #linux-sunxi
tllim has quit [Ping timeout: 240 seconds]
Hao has joined #linux-sunxi
VargaD has quit [Ping timeout: 255 seconds]
Hao has quit [Ping timeout: 248 seconds]
DullTube has joined #linux-sunxi
VargaD has joined #linux-sunxi
fkluknav has joined #linux-sunxi
Hao has joined #linux-sunxi
junnie__ has joined #linux-sunxi
lemonzest has joined #linux-sunxi
tl_lim has quit [Read error: Connection reset by peer]
Hao has quit [Ping timeout: 250 seconds]
<montjoie>
BenG83: I have started to use distcc, I have some siginifiant perf boost for linux compile
<sunxi_fan>
hi, i've beetn testing a "dual head" setup with mainline DRM/KMX sunxi-next kernel tree and i have an issue when testing two "kmstest" processes, each one on a connector/display
<wens>
sunxi_fan: IIRC you were the one that reported the error in the device tree?
<wens>
anyone getting the "451-Could not complete sender verify callout" messages from infradead.org?
<sunxi_fan>
wens: yeah right.. then it was just that i didn't enable the "simple panel" module in the kernel tree..
<wens>
the error likely prevents you from using both pipelines at the same time
<sunxi_fan>
wens: when i had enabled the "simple panel" module, a pretty simple DTS was able to show the testcards on both "heads/connectors"..
Hao has joined #linux-sunxi
<wens>
sunxi_fan: right, but you couldn't get two pipelines to output together
<wens>
due to the error, the driver might not be putting the pipeline together correctly
<wens>
i.e. the two pipelines might be trying to use the same component
fkluknav has quit [Ping timeout: 255 seconds]
<junnie__>
icenowy[m]: Did you make OTG works in u-boot on H3? I just realize that FEL did some work in my previous test. Without booting u-boot with FEL, ums does not work.
<wens>
ok, and with that fixed and simple-panel enabled, you still can't get two pipelines to work together
afaerber has quit [Quit: Leaving]
<sunxi_fan>
yes: both displays work with a single kmstest process, but dont work with two different kmstest, each one with its own connector.. the second process fails when allocating the resource.
<wens>
sunxi_fan: Net147 recently posted some patches for the hdmi driver ( drm/sun4i: hdmi: Fix sun4i_tmds_determine_rate )
<wens>
sunxi_fan: you might want to look at those
msimpson has joined #linux-sunxi
<sunxi_fan>
yeah, i'll try those too.. anyway i was just wondering if there's a more "deterministic" process to find out WHAT really goes wrong when allocating the "kernel resources". this DRM thing is pretty huge and deep!! :-|
<wens>
the process would likely be adding more DRM_DEBUG_DRIVER calls
<sunxi_fan>
yeah good ol' prints all over source, that could help someway.. anyway let me tell you that since a number of years, i've been soooo happy with a gdbserver/gdb-multiarch setup for network cross debugging on ARM SBC (and i like a lot the TUI mode - text user interface... feels like "vim" for debuggers!!) that resorting back to prints make me feel a lot younger!! :-)
<sunxi_fan>
indeed i still didn't find the best way to debug kernel mode.. kgdb is not that clear to me..
afaerber has joined #linux-sunxi
<KotCzarny>
printk all the way +1
<KotCzarny>
unless you write an uart console interface that could access arbitrary data
<wens>
sunxi_fan: can I add your name/email for a Reported-by tag on my patch?
<sunxi_fan>
i suppose that for this DRM issue here, it's not really a matter to trap "wrong data into HW registers" but it's a matter to find which "internal state" of the DRM FSM is a bit messed up!
<sunxi_fan>
wens: of course you can add that tag!
<wens>
it's probably one of the atomic check callbacks in sun4i-drm failing
<wens>
the fact that it doesn't produce any usable error/debug messages doesn't help
<sunxi_fan>
-13 is not that meaningful, but i suppose it's a ENUM somewhere! :-)
<wens>
#define EACCES 13 /* Permission denied */
<wens>
include/uapi/asm-generic/errno-base.h
<sunxi_fan>
and let me tell you: this ATOMIC stuff in DRM setting is completely out of my current context. i suppose i should study more about it to be more proficient..
freemangordon has quit [Quit: Leaving.]
<wens>
don't know where that error came from :/
<sunxi_fan>
curently i have a "drm.debug=15 " as kernel param..
<wens>
it is one of the largest subsystems in the kernel (the others being ALSA and media), no worries
freemangordon has joined #linux-sunxi
<sunxi_fan>
i'm wondering if a different setting can lead to some kind of "stack trace"!
<wens>
afaik there's no stack trace option
<sunxi_fan>
last time i worked on ALSA for the I2S module, and indeed it was a nightmare too there! :-)
<sunxi_fan>
i'm going to look more about the kgdb stuff, anyway.
<sunxi_fan>
i need to find a generic way to deal with this kind of kernel code, not too timing sensitive..
<wens>
the really noisy version is drm.debug=0x3f
<wens>
see include/drm/drm_print.h for the options
elros has joined #linux-sunxi
<sunxi_fan>
oho. let's try this setting too.. the
fkluknav has joined #linux-sunxi
lurchi_ is now known as lurchi__
romanis has quit [Quit: Leaving]
Hao has quit [Ping timeout: 256 seconds]
d_ed is now known as d__ed
<KotCzarny>
hdmi tx phy s40 spec v0.1.pdf on wiki
yann has quit [Ping timeout: 240 seconds]
junnie__ has quit [Ping timeout: 268 seconds]
kloczek has joined #linux-sunxi
lkcl has joined #linux-sunxi
cnxsoft has quit [Remote host closed the connection]
cnxsoft has joined #linux-sunxi
WInk[AW] has joined #linux-sunxi
fdcx has joined #linux-sunxi
WInk[AW] has quit [Quit: Page closed]
IgorPec has quit [Ping timeout: 268 seconds]
<Net147>
sunxi_fan: I haven't tried with kmstest. when testing dual pipeline I used 2 instances of modetest with -d flag.
<sunxi_fan>
Neti147: interesting pointer to DRM-Master.. thanx
enrico_ has joined #linux-sunxi
<Net147>
sunxi_fan: so you probably need to add a drmDropMaster() call in kmstest somewhere after setting the resolution so that the second instance doesn't get access denied when trying to set resolution of the second output
<Net147>
sunxi_fan: okay try running both "modetest -M sun4i-drm -d -s 32@30:800x480" and "modetest -M sun4i-drm -d -s 37@35:1920x1080" at same time
Pe3ucTop has joined #linux-sunxi
IgorPec has joined #linux-sunxi
elros has quit [Quit: IRC for Sailfish 0.9]
elros_ has joined #linux-sunxi
<Pe3ucTop>
Hi to all, question about Single Image capture, do anyboy could provide comment or point for correct way to do it ? Not capture frame from video stream, but realy request only one frame capture?
<Pe3ucTop>
Would like to implement it in applilcation and update CSI driver as well...
<sunxi_fan>
Net147: this way , they both works works!! :-) i can see the two testcards, then check the last "dmesg" on 1920x1080 setup here: https://paste.debian.net/1003388/
<sunxi_fan>
there's indeed the "[94188.090862] [drm:drm_ioctl] pid=959, dev=0xe200, auth=1, DRM_IOCTL_DROP_MASTER"
<Net147>
sunxi_fan: good
paulk-gagarine has quit [Ping timeout: 268 seconds]
nobe has quit [Read error: Connection reset by peer]
nobe has joined #linux-sunxi
kaspter has quit [Remote host closed the connection]
clemens3 has joined #linux-sunxi
<wens>
you might want to verify the dotclock (which affects refresh rate)
dddddd has joined #linux-sunxi
<KotCzarny>
nice. my h3fakeoff also works on h5 (just powered off my opi0+2h5)
kaspter has joined #linux-sunxi
Hao has joined #linux-sunxi
paulk-gagarine has joined #linux-sunxi
IlyaM has joined #linux-sunxi
IlyaM has quit [Quit: IlyaM]
IlyaM has joined #linux-sunxi
IlyaM has quit [Client Quit]
IlyaM has joined #linux-sunxi
IlyaM has quit [Client Quit]
paulk-gagarine has quit [Ping timeout: 268 seconds]
IlyaM has joined #linux-sunxi
IlyaM has quit [Client Quit]
IlyaM has joined #linux-sunxi
IlyaM has quit [Client Quit]
IlyaM has joined #linux-sunxi
IlyaM has quit [Client Quit]
IlyaM has joined #linux-sunxi
pmpp has joined #linux-sunxi
IlyaM has quit [Client Quit]
IlyaM has joined #linux-sunxi
IlyaM has quit [Client Quit]
IlyaM has joined #linux-sunxi
IlyaM has quit [Client Quit]
skiboy has joined #linux-sunxi
paulk-gagarine has joined #linux-sunxi
nuuuciano has joined #linux-sunxi
VargaD has quit [Ping timeout: 240 seconds]
reinforce has quit [Quit: Leaving.]
IlyaM has joined #linux-sunxi
gfa has quit [Quit: WeeChat 1.0.1]
gfa has joined #linux-sunxi
VargaD has joined #linux-sunxi
DullTube has quit [Quit: Leaving]
junnie__ has joined #linux-sunxi
Worf has joined #linux-sunxi
Pe3ucTop has quit [Ping timeout: 240 seconds]
dddddd has quit [Ping timeout: 248 seconds]
Pe3ucTop has joined #linux-sunxi
dddddd has joined #linux-sunxi
dddddd has quit [Ping timeout: 240 seconds]
dddddd has joined #linux-sunxi
Pe3ucTop has quit [Ping timeout: 240 seconds]
IlyaM has quit [Quit: IlyaM]
IlyaM has joined #linux-sunxi
Pe3ucTop has joined #linux-sunxi
Hao has quit [Ping timeout: 256 seconds]
gumblex_ has joined #linux-sunxi
msimpson has quit [Read error: Connection reset by peer]
gumblex has quit [Ping timeout: 256 seconds]
gumblex_ is now known as gumblex
msimpson has joined #linux-sunxi
cnxsoft has quit [Quit: cnxsoft]
reinforce has joined #linux-sunxi
Hao has joined #linux-sunxi
dave0x6d has quit [Quit: Connection closed for inactivity]
xyntrix has joined #linux-sunxi
Ntemis has joined #linux-sunxi
Hao has quit [Quit: Leaving]
Hao has joined #linux-sunxi
tom_nov has quit [Quit: Leaving]
IlyaM has quit [Quit: IlyaM]
pgreco has joined #linux-sunxi
<BenG83>
montjoie, do you use distcc with some load balancing setup?
<BenG83>
I have 7 nodes in my distcc cluster, but even with some tweaks the workload usually is pretty uneven
<BenG83>
I followed some of Willy Tarreau's hints from the build farm talk, but have not tried using haproxy yet...
<montjoie>
BenG83: i use basic distcc. just added node in priority order in DISTCC_XXX=""
kloczek has quit [Quit: kloczek]
kloczek has joined #linux-sunxi
<BenG83>
time for a kernel build goes down from 24 minutes to 7 minutes for me (1 vs 6 nodes)
<BenG83>
(on Pine64 cluster board)
junnie__ has quit [Ping timeout: 248 seconds]
phipli has joined #linux-sunxi
chewitt has joined #linux-sunxi
<montjoie>
notbad
<montjoie>
If I remember I go for 11(main pc) to 8min with (opipc and a83t(1core) and main PC)
diego71 has quit [Ping timeout: 256 seconds]
Worf has quit [Quit: Konversation terminated!]
<BenG83>
my laptop is still faster at 4.5 minutes ;/
<BenG83>
but I think the 7 minutes can be improved upon at lest 20%
<BenG83>
with load balancing
<BenG83>
A64's A53 cores are slow :(
IlyaM has joined #linux-sunxi
<BenG83>
but hey, maybe my i7 will be 30% slower with 4.15.x :P
<KotCzarny>
wont be if you dont care about security
<KotCzarny>
:P
<montjoie>
BenG83: distcc loadbalance by default, you just need to have -j = numcores
<kevans91>
wens: Assuming the a83t-sid patch is acceptable, when would it likely get merged? Just curious- not too familiar with how things work in Linux world =)