LanDi has quit [Read error: Connection reset by peer]
amitk has joined #linux-exynos
<steev>
Wizzup: torvald's tree
<steev>
and using the 5800-peach-pi
<steev>
Wizzup: oh blah... i didn't pay attention, didn't have DRM_EXYNOS* enabled
<steev>
probably need to send a patch similar to my last one i had to send to enable usb on exynos
aballier has joined #linux-exynos
aballier has joined #linux-exynos
zombah has joined #linux-exynos
amitk has quit [Ping timeout: 250 seconds]
amitk has joined #linux-exynos
<Wizzup>
steev: usb should be enabled by default in defconfig I think
<Wizzup>
oh, right
<Wizzup>
hurr
<steev>
Wizzup: yeah, i submitted that patch :P
<Wizzup>
any luck with the drm now?
<steev>
neg
<steev>
i'm giving up
<javier__>
steev: are you using exynos_defconfig? the EXYNOS_DRM* options should be enabled in v3.19 afaict
<steev>
javier__: yes
<steev>
javier__: or rather - i switched to exynos_defconfig after multi_v7 didn't work
<steev>
still didn't work
<javier__>
steev: yeah, multi_v7 doesn't have all the options enabled and also builds a huge kernel binary so is only useful for build / boot test farms
<javier__>
steev: I thought v3.19 would work...
<sjoerd>
javier__: don't forget distributions ;)
<javier__>
sjoerd: well on its current form (with everything built-in) is not useful for distributions either
<javier__>
if things are moved to be built as a module then yes
<sjoerd>
it's useful as a reference
<Wizzup>
steev: hmmm, I'll give it a shot after my exams and I'll let you know
<javier__>
sjoerd: that's true
<sjoerd>
Wizzup: good luck with the exams ;)
<Wizzup>
only a few more to go :)
<sjoerd>
javier__: Which reminds me, should write a script at some point to diff/add all the options from e.g. exynos defcofig to multiv7 and see what blows up :P
<javier__>
sjoerd: yeah, I've had it on my TODO list for a long time but just with low priority
<sjoerd>
I think the iommu's still make everything explode
<javier__>
sjoerd: now the exynos_drm driver has a check to fail gracefully instead of crashing the system if IOMMU is enabled
<javier__>
but yeah, no display and who knows what other problems
<sjoerd>
javier__: is disabled you mean
<Wizzup>
I don't think so :p
<Wizzup>
gracefully fail when it is enabled -> instead of just crashing
<javier__>
sjoerd: no, it is working without iommu. The problem is when iommu is enabled. Used to crash now it only fails to probe
<javier__>
which reminds me that I've to post my patch to revert Andrzej's that added the DISP1 power domain for exynos5250
<sjoerd>
javier__: oh right hdmi didn't work if iommu wasn't on and crashed if it was
* sjoerd
got the combinations confused
<javier__>
sjoerd: nod
<javier__>
but the problem turned out to be ot iommu but the fact that iommu enabled the correct power domais and clocks
<javier__>
so now hdmi is also working without iommu
<javier__>
*not to be iommu
<sjoerd>
cool
<sjoerd>
now we just need it to work again with iommu and everything should be happy
Tenkawa has joined #linux-exynos
Tenkawa has joined #linux-exynos
<javier__>
yeah
<javier__>
steev: I can try 3.19 later but I suggest you to use 4.0-rc4 or linux-next
<steev>
javier__: i was using 4.0-rcX - whatever's in torvald's tree as of a few hours ago
<steev>
*shrug* i don't have any more free time to try it til at least thursday
<javier__>
steev: ok, for 4.0-rcX you need to disable IOMMU support because of what we were just talking about ^
<javier__>
the patch made it to linux-next commit: 6562f3bd396a ARM: exynos_defconfig: Disable IOMMU support
<javier__>
but unfortantely is still not in v4.0-rcX even when I asked the samsung maintainer to add it several times...
<steev>
ah
<steev>
can grab that commit i suppose
<javier__>
I usually test that a feature / peripheral is working, post the patches and then move on
<javier__>
is hard to keep track of all the regressions that were introduced on each kernel release :-/
<steev>
oh i know it ain't easy, believe me
<sjoerd>
javier__: *cought*write those test*cought* :)
<javier__>
sjoerd: haha, I knew you would say that
<sjoerd>
:)
<javier__>
sjoerd: I'm writing a test for the cros_ec user-space driver as we speak :)
Tenkawa has quit [Ping timeout: 250 seconds]
<Tenkawa_>
javier__: so if I read this correctly the cb support should be "working" again
<Tenkawa_>
?
<javier__>
Tenkawa_: cb 1 or 2?
<sjoerd>
javier__: \o/
<Tenkawa_>
1
<javier__>
Tenkawa_: yes, I'm about to post a patch to revert the commit that added the regression
<Tenkawa_>
sweet
<Tenkawa_>
that will help
<javier__>
zombah: I don't know about CPUFreq for Snow but the config symbol mentioned on that thread CONFIG_I2C_S3C2410 is enabled in both exynos_defconfig and multi_v7
<zombah>
javier__: yes it is enabled, but exynos-cpufreq spam log with failed to set cpu voltage even with it enabled
Tenkawa_ is now known as Tenkawa
Tenkawa has quit [Changing host]
Tenkawa has joined #linux-exynos
<Tenkawa>
hmm
<Tenkawa>
I have been messing with rc4 however I'm still getting blank screen on boot
<Tenkawa>
what is it I'm missihg again?
<Tenkawa>
I know I am missing something small i think
<Tenkawa>
starts up fine and is running.. just gets no output
<Tenkawa>
(on snow)
<javier__>
Tenkawa: no, all the patches didn't make it for -rc4
<javier__>
you need linux-next
<Tenkawa>
ahhh
<Tenkawa>
I misread
<Tenkawa>
yeah I'm currently running next-0222
<Tenkawa>
let me grab the newer build
<Tenkawa>
and start compiling
<Tenkawa>
heheheh
<javier__>
Tenkawa: but you need to revert a commit
<javier__>
2d2c9a8d0a4f90e298315d2f4a282d8bd5d45e5c ("ARM: dts: add display power domain for exynos5250")
<Tenkawa>
oh that one still?
<Tenkawa>
meh I'll wait
<Tenkawa>
I'm mostly working on a x86 box atm so I can wait till its all sorted out
<Tenkawa>
that and my rpi2's which are already caught up