00:22 <anarsoul> I wonder if anyone is testing reboot and poweroff on mainline

07:48 <anarsoul> well, I fixed i2c driver in u-boot (will send a patch tomorrow) but unfortunately initializing cpu_b regulator doesn't fix reboot for me

07:49 <anarsoul> linux hangs on 2nd boot if I rebooted without pressing hw reset

09:15 <sigmaris> anarsoul: In my experience I also see hangs after about 75% of soft resets

09:15 <sigmaris> hw reset always works, and interrupting u-boot then using the u-boot "reset" command also works

09:16 <sigmaris> but rebooting straight from linux often hangs. in some cases rather than hangs I have seen this "Asynchronous SError Interrupt" panic: https://gist.github.com/sigmaris/c0e155c8cb0a325d84f549185f9a568c

10:06 <mmind00> anarsoul: I somewhat regulary do "reboots" when testing kernel changes, but haven't seen any hangs recently ... the only async serror interrupt I see is when the uboot mmc-driver tries to write stuff to sram (atf sram part) ... setting fifo-mode for the mmc controllers fixes that for the moment

12:51 <leming> ooooom tgyhhhhhhhhhhhhhhh[']

12:51 <hanetzer> cat?

14:52 <leming> he says hi

15:36 <anarsoul> sigmaris: yeah, same here :\

15:36 <anarsoul> mmind00: what board do you use?

15:39 <anarsoul> I get this after a while on unsuccessful boot: https://gist.github.com/anarsoul/35e51acfec2eb4e4e5ebb37a94dc164e

19:17 <anarsoul> mmind00: also, are you using mainline u-boot with mainline ATF?

19:52 <mrjay> i'm also having troubles with reboot/poweroff with current linux-next

20:01 <anarsoul> what board?

20:02 <mrjay> pipo p10 with rk818 pmic

20:13 <anarsoul> so rk3399

20:18 <stikonas> yeah, I'm having some reboot issues with rk3399 too...

20:18 <stikonas> anarsoul: I saw some patches from you? Do they fix it

20:19 <stikonas> I think it was pul lrequest for ATF

20:19 <anarsoul> it fixes immediate hang on reload so you can actually get back to u-boot

20:19 <anarsoul> but subsequent linux boot hangs with stuck task on CPU#4

20:20 <anarsoul> I though that it's somehow related to cpu_b regulator not being initalized (and linux indeed initializes it later than it boots big CPUs) but initializing it in u-boot didn't fix the issue

20:20 <anarsoul> stikonas: poweroff is still broken though

22:07 <sphalerite> hanetzer: coreboot won't build with the kernel uncompressed in the u-boot image, because it's too big

22:07 <sphalerite> I tried lzma'ing the vmlinux and specifying compression = "lzma" in the its file, but coreboot is refusing to load that because it doens't like the magic

22:09 <sphalerite> any suggestions?

22:12 <sphalerite> oh wait, it has to be in Image format

22:12 <sphalerite> not vmlinux

22:13 <sphalerite> trying that now

22:21 <hanetzer> ^this

22:25 <sphalerite> hooray! It loaded! It panicked, but it did load!

22:25 <sphalerite> Also, is it normal for flashing to take ~10min?

22:25 <anarsoul> :)

22:25 <hanetzer> \o/

22:25 <hanetzer> what's the kpanic?

22:27 <sphalerite> init no worky. Not sure exactly why, but I'm sure I'll manage to figure it out next time I look at this (and for now I'm going to sleep)

22:28 <sphalerite> but yeah, is ~10min normal for erase+write+verify flash? That'll make a painful testing cycle if so…

22:28 <hanetzer> sphalerite: what flashing device you using?

22:29 <sphalerite> suzyqable

22:29 <hanetzer> hmm. no idea about that one.

22:30 <sphalerite> alright. And does anyone here know a way to power the AP off from the cr50?

22:31 <hanetzer> nerp

22:33 <sphalerite> letting the battery run out it is, then…

22:34 <sphalerite> thanks for the help, gnight!

22:34 <hanetzer> sphalerite: wait

22:34 <hanetzer> sphalerite: basically you're trying to turn it off while kpanic'd?

22:35 <urjaman> i was gonna say i assume it'll power off with a long enough power button press

22:35 <hanetzer> yeh

22:38 <urjaman> and yeah i heard someone mention that the cr50 is slow at writing the SPI because tiny usb transfers

22:38 <hanetzer> yeh. I usually just clip :P