00:24 <amstan> steev: another kali fan: https://www.reddit.com/r/chromeos/comments/3giybu/finding_the_right_chromebook_for_running_linux_on/

00:25 <steev> *shrug*

00:30 <steev> i actually just finished pushing up the final images to the server, release is tomorrow

07:17 <sjoerd> cool, my rock2 should arrive tomorrow

13:06 <nfstry> hello

13:07 <nfstry> looking for help on rk3288

13:07 <nfstry> anybody here?

13:08 <greggypoo> nfstry, you'll have best luck asking your question and then idling until (if) it attracts interest

13:10 <nfstry> i'm trying to use the TS input on the rk3288 but without success the pti does not seem to start or produce any data in the circular buffer even if i'm setting up the control registers correctly

13:10 <nfstry> my guess is that there is a clock problem or some kind of similar issue

13:11 <nfstry> anybody has been able to use GPIO bank 2 port A+B as Ts input?

13:18 <greggypoo> i don't know anything about it but i will say there is a lot of talk about clocking in here :)

13:19 <nfstry> the only way to get data in the circular buffer was to completely bypass the pti1

13:19 <nfstry> i was hoping to get any pointer on why pti1 did not process the data

14:09 <tlwoerner> naobsd: any thoughts on why the mainline can't mount the eMMC "partitions"?

14:12 <naobsd> tlwoerner: please define what is "partitions". Linux supports common partitioning method such as MBR, GPT.

14:14 <naobsd> today is the first day of vacation... I'm getting some time for my hobby, but it's not enough to complete all my todo items ;)

14:26 <tlwoerner> naobsd: enjoy your vacation, i'll give you a poke when you're back :-)

14:35 <naobsd> I'm busy when not in vacation. I can get some time only in vacation...

16:13 <rperier> tlwoerner: I was thinking about your issue... you don't see your eMMC partitions at all or you're unable to mount of them ?

16:13 <rperier> one of them*

16:15 <rperier> classic questions: 1)° is the emmc support enabled in your kernel config ? 2)° does u-boot export correctly your dtb to your kernel ? (check the model name or machine name to be sure) 3)° do you use the right partition table type in your kernel ? (I had the same issue few days ago on my chomebook, UFI partition support was missing)

16:15 <rperier> have you a log or something ?

16:19 <tlwoerner> rperier: please see https://groups.google.com/forum/#!topic/linux-rockchip/5ww6VckbYnQ

16:20 <tlwoerner> the device nodes at /dev/block/mtd/by-name aren't created

16:22 <tlwoerner> i believe the support is enabled in the kernel (although i haven't explicitly verified this) because the instruction at http://wiki.t-firefly.com/index.php/Firefly-RK3288/Mainline_Kernel/en#Kernel say to "make rk3288_firefly_defconfig" and later state that 'support of command line parameter "mtdparts=" is added'

16:23 <tlwoerner> i could push an experimental branch to git.yoctoproject.org that anyone could use to build exactly what i'm building ;-)

18:59 <rperier> tlwoerner: yes you could

19:22 <naobsd> tlwoerner: are you sure that mainline doesn't support partition info in parameter?

19:25 <naobsd> well

19:25 <naobsd> oh, 4AM now...

19:26 <naobsd> tlwoerner: even if kernel supports partition in parameter, /dev/block/mtd/by-name/* is not populated by kernel. you have to prepare it in userland.

19:29 <mrjay> Is there a way deal with high number of interrupts of dwc2 usb ? i'm getting ~12000/s

19:33 <greggypoo> mrjay, when nothing is happening over it?

19:35 <mrjay> i have connected usb stick to this port with system on it. but diode is not flashing. isn't a little to much?

19:43 <greggypoo> mrjay, i don't know if you're interested in this much work, but i would personally do printk debugging... drivers/usb/dwc2/hcd_intr.c dwc2_handle_hcd_intr()

19:43 <greggypoo> i would add a printk there to see which kind of interrutps it's getting

19:45 <greggypoo> and if they're a bunch of GINTSTS_RXFLVL, then at the end of dwc2_rx_fifo_level_intr() i'd add a printk of readl(hsotg->regs + GRXSTSP);, to see if maybe bcnt is non-zero even after the packet is read

19:45 <greggypoo> but i don't actually know if that's possible on that chip

19:45 <greggypoo> my point is, you can find out exactly what is happening and then you will at lesat know :)

19:49 <mrjay> greggypoo: ok i can try ... never done that before but ok.

20:06 <mrjay> greggypoo: strange ... after poweroff and disconnecting cables it stopped. Maybe something is wrong with the usb stick

20:07 <greggypoo> that was my guess, that something silly was generating unnecessary traffic

20:10 <mrjay> greggypoo: good guess :).

22:09 <steev> mmind00: nice!

22:09 <steev> just saw the closing of the issue re: libGL

22:09 <steev> if i can find some free time, i'll play with it myself

22:11 <mmind00> steev: hehe ... at least with this, it is in line with general debian practices regarding vendor graphics drivers, and you can switch between mali and mesa on the fly

22:11 <steev> yep