nexgen2 has quit [Remote host closed the connection]
tllim has quit [Read error: Connection reset by peer]
<smaeul>
anarsoul: it was originally in anx6345_bridge_enable -> drm_panel_enable -> ... -> sun4i_pwm_apply
nexgen2 has joined #linux-sunxi
<smaeul>
but I don't know what part of the chain is causing it to be atomic
<KotCzarny>
it looks more like some corrupted pointer?
<KotCzarny>
'unreadable address' ?
<smaeul>
KotCzarny: the actual error is [ 33.887413] BUG: scheduling while atomic: s6-linux-init-h/643/0x00000002
<smaeul>
which means you tried to sleep while holding spinlocks
fl_0 has quit [Quit: STRG + Q]
<wens>
a stack trace should have clues?
<smaeul>
wens: unfortunately I don't have the original stack trace; that was several hours ago. but I remember most of it from my vim search history: http://ix.io/1Tva
<anarsoul>
looks like a bug in panel simple to me?
<smaeul>
somehow I can't reproduce it now
<anarsoul>
:\
<smaeul>
I think it depends on how many i2c errors I get during anx initialization
<anarsoul>
sun4i_pwm_apply() does call msleep() or usleep_range()
<diego71>
KotCzarny: may be you want debug information, just not one byte per line :)
fl_0 has joined #linux-sunxi
tnovotny has joined #linux-sunxi
a|3x has joined #linux-sunxi
a|3x has left #linux-sunxi [#linux-sunxi]
ldevulder__ has quit [Quit: Leaving]
Mangy_Dog has joined #linux-sunxi
<MoeIcenowy>
smaeul: people reported 8723cs breaks even when s2idle
<MoeIcenowy>
smaeul: why is a new interrupt driver needed?
gnufan_home has quit [Ping timeout: 246 seconds]
aalm has quit [Quit: xyz 2.3]
Da_Coynul has joined #linux-sunxi
gnufan_home has joined #linux-sunxi
Alexis3D has joined #linux-sunxi
msevo has quit [Quit: Leaving]
Da_Coynul has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
cnxsoft has quit [Read error: Connection reset by peer]
cnxsoft1 has joined #linux-sunxi
Mangy_Dog has quit [Ping timeout: 246 seconds]
Da_Coynul has joined #linux-sunxi
gnufan_home has quit [Ping timeout: 246 seconds]
Da_Coynul has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
gnufan_home has joined #linux-sunxi
Alexis3D has quit [Ping timeout: 272 seconds]
no_- has quit [Ping timeout: 246 seconds]
xqdzn has joined #linux-sunxi
gamelaster has joined #linux-sunxi
nashpa has quit [Ping timeout: 268 seconds]
lurchi_ is now known as lurchi__
lurchi__ is now known as lurchi_
gamelaster has quit [Remote host closed the connection]
nashpa has joined #linux-sunxi
no_- has joined #linux-sunxi
lurchi_ is now known as lurchi__
gaston1980 has joined #linux-sunxi
nexgen2 has quit [Ping timeout: 272 seconds]
kaspter has quit [Read error: Connection reset by peer]
kaspter has joined #linux-sunxi
cnxsoft1 has quit [Quit: cnxsoft1]
dddddd has joined #linux-sunxi
no_- has quit [Ping timeout: 268 seconds]
Putti has joined #linux-sunxi
SopaXorzTaker has joined #linux-sunxi
<smaeul>
MoeIcenowy: because the GIC is turned off in suspend
<smaeul>
so for devices that can act as wakeup sources (those attached to R_INTC) we need to synchronize the GIC state with the R_INTC state
<smaeul>
there are 20+ of these "stacked" interrupt controller drivers in the Linux tree
<smaeul>
AW blob has a special memory area and protocol for configuring wakeup sources, but the simple way is "enabled IRQ during suspend == enabled wakeup source"
<smaeul>
since Linux already does the job of turning off non-wakeup IRQs
kaspter has quit [Quit: kaspter]
Alexis3D has joined #linux-sunxi
<MoeIcenowy>
smaeul: BTW I think you can consider to mainline it first
<MoeIcenowy>
current NMI driver doesn't support wakeup at all, even from s2idle
no_- has joined #linux-sunxi
random_yanek has quit [Ping timeout: 268 seconds]
JohnDoe_71Rus has joined #linux-sunxi
random_yanek has joined #linux-sunxi
DrFrankensteinUK has quit [Read error: Connection reset by peer]
selfbg has quit [Remote host closed the connection]
tllim has joined #linux-sunxi
sunshavi has quit [Remote host closed the connection]
sunshavi has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
no_- has quit [Ping timeout: 272 seconds]
<martinayotte>
tllim: My coupon for PBP received few days ago expired within 2 days ... Can it be extended or re-issued ? (I've said the same thing on pinebook@pine64.org, but no answer yet ...)
DrFrankensteinUK has joined #linux-sunxi
<gaston1980>
how i can exclude lichepi-zer-dock when cross-compiling dts?
<gaston1980>
im using linux kernel 5.2.9
<gaston1980>
I modify dts sun8i-v3s-licheepi-zero and sun8i-v3s.dtsi to enable ethernet
<gaston1980>
but got errors related to the licheepi-zero-dock
<gaston1980>
then I modify licheepi-zero-dock to enable ethernet but still got errors
<fALSO>
you probably added something to the general dtsi what is already on that zero-dock
<gaston1980>
sorry to ask
<gaston1980>
but could you take a look?
jstein_ has joined #linux-sunxi
jstein_ is now known as jstein
<fALSO>
just look at the error
<fALSO>
its probably specified there
jstein has quit [Client Quit]
<tllim>
@martinayotte, if till not getting reply from pinebook@pine64.org on 24 hours, just PM me and I will handle
<martinayotte>
tllim: Thanks ! I will be in touch ...
<martinayotte>
gaston1980: Since it shows "arch/arm/boot/dts/sun8i-v3s.dtsi:461.3-7 syntax error", please check line 461 of this DTSI ...
xqdzn has quit [Remote host closed the connection]
<fALSO>
if you changed sun8i-v3s.dtsi, you changed it badly
<fALSO>
check that line
<fALSO>
as martinayotte told
<mru>
and the surrounding lines
<mru>
the error could be something like a missing ; on the previous line
<gaston1980>
ok, thanks I solve it
<gaston1980>
missing ;
<gaston1980>
thanks guys
nexgen2 has joined #linux-sunxi
<fALSO>
=)
<jernej>
smaeul: From Crust README.md: "However, finishing the process and switching everyone over to use Crust will take a while" <- by that you don't mean that Crust will be mandatory, right?
<KotCzarny>
only if one would want to use suspend
<KotCzarny>
:)
<KotCzarny>
otherwise it's programmed to be optionl
<KotCzarny>
*optional
<jernej>
just checking :)
<KotCzarny>
i wouldnt mind it being mandatory tho
<KotCzarny>
;)
<jernej>
no, because there could be legit uses of AR100 for other purposes like runing real time tasks (offloaded)
<KotCzarny>
ah, that's what you mean
<jernej>
and I don't like having yet another FW if not really necessary
<KotCzarny>
btw. i still havent found a way to initialize arisc when linux already booted (without disabling all kinds of kernel protections)
<jernej>
I don't know that part well...
<KotCzarny>
once booted, registers are probably set in a way to prevent it
vagrantc has joined #linux-sunxi
<KotCzarny>
hmm
<KotCzarny>
funny idea
<KotCzarny>
i wonder what would be performance of aes on arisc
<KotCzarny>
:)
<KotCzarny>
assuming one could write it in 16kB
shfil has joined #linux-sunxi
yann has quit [Ping timeout: 272 seconds]
reinforce has joined #linux-sunxi
no_- has quit [Read error: Connection reset by peer]
no_- has joined #linux-sunxi
florian_kc has joined #linux-sunxi
<megi>
KotCzarny: I load it via /dev/mem :)
<megi>
I guess that fits into "disabling kernel protections"...
<KotCzarny>
:)
<KotCzarny>
yeah, /dev/mem is more of a hack
<KotCzarny>
but hmm
<megi>
you can even toggle the clock/reset directly in the iomem, for cpus
<megi>
you may need to toggle a few bits to make it work on other SoC
<KotCzarny>
although looking at it, its devmem2 based
<KotCzarny>
does it work on stock armbian or you compile own uboot/kernel?
<megi>
it should be fairly trivial to write a driver just for loading firmware into the CPUS over some ioctl with proper access controls on some file in /dev/*
<KotCzarny>
ahm. you didnt test for h3
<megi>
it should work anywhere, you just need to check the addresses
<KotCzarny>
addresses are ok because it works on legacy, for mainline i think it worked with boot_mode nonsec
<megi>
it doesn't do anything hacky like H3 firmware, which loads half of the firmware into SRAM and hlaf int DRAM
<megi>
yeah it expects working /dev/mem
<KotCzarny>
i mean it works for me with 'sec', with nonsec it doesnt
<megi>
what's that?
<KotCzarny>
setenv bootm_boot_mode nonsec
<KotCzarny>
nonsec is needed for psci and virtualization (basically mainline)
<megi>
i'm pretty sure I'm using it in nonsec mode then (that's the default, right?)
<KotCzarny>
for mainline, yes
<KotCzarny>
hmm, one thing, is that usleep(1000) necessary?
Mangy_Dog has joined #linux-sunxi
rexxster has quit [Remote host closed the connection]
<megi>
try removing it :)
<KotCzarny>
nah, i will rather try to add it to my code
<KotCzarny>
:)
quadjfet has joined #linux-sunxi
no_- has quit [Ping timeout: 245 seconds]
gsz has quit [Quit: Konversation terminated!]
tnovotny has quit [Ping timeout: 258 seconds]
tnovotny has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
rexxster has joined #linux-sunxi
no_- has joined #linux-sunxi
niceplace has quit [Ping timeout: 245 seconds]
yann has joined #linux-sunxi
shfil has quit [Quit: Connection closed for inactivity]