00:58 <naobsd> hm, why rk3368 reference/product board doesn't use PMIC

03:15 <naobsd> ir remote on R68 should be connected to pwm

06:26 <lautriv_> naobsd, mayve they fot the glue to use a GPIO for the sake of possible lirc compatibility ?

06:33 <naobsd> lautriv_: well?

06:35 <lautriv_> naobsd, afair there is a in-kernel driver for gpio using the RC-tables and bridge to lirc or translate to evdev.

06:38 <naobsd> it will work

06:43 <lautriv_> naobsd, i guess you got me a bit wrong, i was about your "<naobsd> ir remote on R68 should be connected to pwm" approx 3h ago ;)

06:44 <naobsd> well

06:46 <naobsd> I have no problem for now

06:47 <naobsd> well

06:47 <naobsd> that was not problem report

06:48 <lautriv_> i took that not as a problem, just a guess why they changed away from PWM.

08:42 <tony_> hi, how to install & start android-tools-adbd NOT adb.

08:42 <tony_> ubuntu on rk3188.

08:42 <tony_> who can help me ?

09:01 <lautriv> tony_, if you run linux, why do you need adb ?

09:10 <tony_> lautriv, adb is so easy. ;P

09:10 <tony_> I can install android-tools-adbd.

09:10 <tony_> I can also start it by run "adbd"

09:10 <tony_> but I can not get any port in PC.

09:10 <tony_> by lsusb.

09:14 <tony_> lautriv, do you think so ?

09:17 <lautriv> ssh is the tool.

09:17 <tony_> lautriv, okay, cool.

09:17 <tony_> lautriv, thank you.

12:14 <rperier> technically speaking... if you design an IP block inside an fpga which is mapped at 0x40000000. This physical address seems to overlap the physical address range of the kernel, in fact some functions in arch/arm/kernel/setup.c seems to be located at this address (mmu is not activated yet at this early stage). do we agree that this is a problem ? I mean even if you map later this physical address into the kernel linear memory via ioremap,

12:14 <rperier> it cannot work right ?

12:14 <rperier> that's a low arm related question

12:16 <rperier> perhaps that... I should probably ask on #armlinux :/

12:44 <mrutland> rperier: I'm not sure I follow

12:44 <mrutland> rperier: Where's your memory located?

12:45 <mrutland> and what split are you using? 1G user / 3G kernel?

12:52 <sjoerd> rperier: there are a whole bunch of arm boards that don't even have normal memory at that physical location, so i'm pretty sure that's not a hardcoded physical address (afaik linux itself is relocatable anyway)

13:00 <mrutland> The MMU is turned on much earlier than setup.c -- page tables get created in __create_page_tables, then we go to__enable_mmu and __turn_mmu_on enable the mmu, then we go to __mmap_switched, then onwards to start_kernel

13:01 <mrutland> everything before __mmap_switched should be position independent; the VA is irrelevant

13:08 <rperier> mrutland: indeed, it seems to be already enabled...