<mmind00>
c0d3z3r0: no idea, I'm not aware of specific changes there
<c0d3z3r0>
mmind00: ok, thanks anyways
<mmind00>
c0d3z3r0: in your reply to the dwc2 patch, why were you talking about sd-cards? :-)
<c0d3z3r0>
let me think...
<c0d3z3r0>
I have no clue…
<c0d3z3r0>
\o/
<c0d3z3r0>
one second
<mmind00>
:-)
paulk-collins has quit [Ping timeout: 260 seconds]
<c0d3z3r0>
mmind00: corrected. Maybe I should think before typing haha
<mmind00>
:-D
<dianders>
mrjay: Do patchwork.kernel.org/patch/8506901 and patchwork.kernel.org/patch/8506921 solve your USB problems too?
<c0d3z3r0>
dianders: msleep(50) is there in next-20160302 but not in *03 and *04 even though it has not been reverted, yet. that's why I don't see it but you do o.O
<dianders>
c0d3z3r0: Ah, maybe there was a merge conflict? Let's see.
<c0d3z3r0>
that's what I thought but I couldn't find any
<dianders>
c0d3z3r0: Weird. Removal shows in "git diff 62718e304aa6..next-20160304 -- drivers/usb/dwc2" but not in "git log --patch 62718e304aa6..next-20160304 -- drivers/usb/dwc2"
<c0d3z3r0>
interesting… just like I said - "magic reverting" :D
<dianders>
:) Always figure there are lurkers around following things, so nice to drop breadcrumbs. Sometimes I'm the lurker and I like breadcrumbs... ;)
naobsd has joined #linux-rockchip
cnxsoft has joined #linux-rockchip
nighty has joined #linux-rockchip
c0d3z3r0_ has joined #linux-rockchip
c0d3z3r0 has quit [Ping timeout: 276 seconds]
c0d3z3r0_ is now known as c0d3z3r0
c0d3z3r0 is now known as Guest38640
cnxsoft1 has joined #linux-rockchip
cnxsoft has quit [Ping timeout: 240 seconds]
cnxsoft1 is now known as cnxsoft
_vaibhav_ has joined #linux-rockchip
JohnDoe_71Rus has joined #linux-rockchip
cosm has quit [Ping timeout: 252 seconds]
sunilmohan has quit [Ping timeout: 264 seconds]
sunilmohan has joined #linux-rockchip
sunilmohan has quit [Changing host]
sunilmohan has joined #linux-rockchip
dlezcano has quit [Ping timeout: 246 seconds]
_vaibhav_ has quit [Ping timeout: 276 seconds]
mrjay has joined #linux-rockchip
<mrjay>
dianders: msleep(50); patch did the trick ... i can test this two patches ... but should i revert msleep patch from my tree?
paulk-collins has joined #linux-rockchip
pjob has joined #linux-rockchip
pjob has quit [Quit: Page closed]
VargaD has quit [Ping timeout: 244 seconds]
VargaD has joined #linux-rockchip
paulk-collins has quit [Remote host closed the connection]
<mrjay>
dianders: i think 50 - 55ms is a safe value (20 reboots ok) ...weird
Ueno_Otoko has quit [Remote host closed the connection]
gb_master has joined #linux-rockchip
gb_master has quit [Quit: Leaving]
JohnDoe0 has joined #linux-rockchip
<dianders>
mrjay: Thanks for testing. The problem is that they delay doesn't really make sense where it is. It probably belongs somewhere else.
JohnDoe_71Rus has quit [Ping timeout: 250 seconds]
<dianders>
mrjay: Actually, just for kicks, can you try reverting commit 97e463886b87 ("usb: dwc2: Reduce delay when forcing mode in reset')
<dianders>
Instead of 50ms delay.
<dianders>
So to summarize: take the two patches I posted but _don't_ take the 50ms delay patch. ...then revert 97e463886b87. Sort of a shot in the dark, but ...
paulk-collins has joined #linux-rockchip
<Guest38640>
mmind00: I found the cause of my voltage problem… looks like barebox sets max-voltage since my last config change and then the kernel isn't able to override that value
<Guest38640>
hm.. im a guest.. interesting
Guest38640 is now known as c0d3z3r0
gb_master has joined #linux-rockchip
<mmind00>
c0d3z3r0: where does barebox set the voltage? *confused*
gb_master has quit [Remote host closed the connection]
gb_master has joined #linux-rockchip
<c0d3z3r0>
mmind00: I did not look into the code so that was just a guess. Disabling CONFIG_REGULATOR in barebox config solves the problem
<mmind00>
c0d3z3r0: is that barebox support even moving forward? last I know there was a tiny bit of rk3188 support in there 1 or 2 years ago
<c0d3z3r0>
mmind00: barebox works fine for me :)
<c0d3z3r0>
mind00: last commit for rk3188 was in september 2014
_vaibhav_ has joined #linux-rockchip
eli198 has joined #linux-rockchip
eli198 is now known as eli1981
<eli1981>
hey all, anyone knows where I can get original firmware/ROM of my Beelink R68 device ? (Need it as a backup as I'm planning to try Wasser ROM)
<mrjay>
c0d3z3r0: for me with dianders patches it worked once for 3-5 reboots
<dianders>
mrjay, c0d3z3r0: presumably my two patches + the 50 ms _does_ work at least. AKA my patch properly allows us to revert "usb: dwc2: Fix probe problem on bcm2835"?
<c0d3z3r0>
mrjay: the 50ms patch alone does not work for you?
<dianders>
c0d3z3r0: Earlier he said it did.
<dianders>
AKA: ToT + 50ms (which is mysteriously missing from ToT) seems to fix everyone. I'm hoping at least that ToT + 50ms + my two patches _also_ fixes people since it's good to revert the "bcm2835" patch and fix it correctly.
<dianders>
...and at some point I think we should find a more correct patch for the "50 ms" one...
<dianders>
Do the extra two patches break things for you?
<mrjay>
dianders: this combination work ok for me ;)
<dianders>
mrjay: OK, great! c0d3z3r0: You too?
<c0d3z3r0>
dianders: same here
<dianders>
mrjay: If you happen to want to do one more test. Skip 50ms (https://patchwork.kernel.org/patch/8319851/), take patchwork.kernel.org/patch/8506901 and patchwork.kernel.org/patch/8506921, and change msleep(25) in dwc2_force_mode() and dwc2_clear_force_mode() to msleep(75).
premoboss has joined #linux-rockchip
<mrjay>
dianders: usb usb2-port1: connect-debounce failed ... :(
<dianders>
Sigh. OK I'll have to spend more time thinking about it.
<mmind00>
mrjay: although I remember seeing that on my rk3036 recently as well (without dianders patches)
<mrjay>
dianders: thanks anyway ...
<mrjay>
dianders: did anyone bisect when this dwc2 issue started?
<mrjay>
mmind00: good to know that i and c0d3z3r0 are not the only one
<dianders>
mrjay: I think these problems are related to John's patches to decrease some of the delays.
<dianders>
Of course, the connect-debounce could be something different? I haven't seen that.
<mrjay>
dianders: when usb doesn't work debounce failure apears for me
<mrjay>
dianders: one more thing ... only host mode doesn't work ... gadget mode works fine