<mmind00>
sjoerd: are you getting any messages? There is a dma issue still unsolved in mainline, but the emmc should use the dw_mmc internal dma, so I'm not sure
<sjoerd>
i do see some dma error/issues but that seems tobe related to serial
<sjoerd>
e.g. 1.340248] dmaengine: __dma_request_channel: fail ((null))
<sjoerd>
[ 1.340252] ttyS2 - failed to request DMA
<sjoerd>
hmm, looks like the mmc bits aren't using dma
<sjoerd>
[ 0.550357] dwmmc_rockchip ff0c0000.dwmmc: Using PIO mode.
<sjoerd>
[ 0.629319] dwmmc_rockchip ff0d0000.dwmmc: Using PIO mode.
<sjoerd>
[ 0.713565] dwmmc_rockchip ff0f0000.dwmmc: Using PIO mode.
<sjoerd>
hmm looks like i'd need to get Shawn Lin's internal dma patches to solve that at least, which may fix that then
<sjoerd>
mmind00: Another random question you might know, how do rockchip devices tend to store their mac address. I'm used to it being in the eFuse on e.g. i.mx6, but at least in the documentation i couldn't find much about that on rockchip
<sjoerd>
e.g. no standard fusemap documentation
levd has quit [Remote host closed the connection]
levd has joined #linux-rockchip
<mmind00>
sjoerd: re dw_mac dma ... are you on rk3188 or 3288 now? because the 3288 should use idma already
<mmind00>
sjoerd: the patches from Shawn enable the external dma ... for older socs like the 3188 I think
<mmind00>
sjoerd: but just to make sure ... check if the idma kconfig option is enabled ;-)
<mmind00>
as for the mac ... I actually don't know ... while there is a fuse driver now, I think they story only other stuff in there
<sjoerd>
mmind00: RK3288, i didn't see this issue on RK3188
<sjoerd>
Oh irght i misparsed his patch subject
<sjoerd>
coffee is still sinking in
<sjoerd>
# CONFIG_MMC_DW_IDMAC is not set
<sjoerd>
lalala
<mmind00>
:-D
honx has left #linux-rockchip [#linux-rockchip]
<mmind00>
a spooky ghost changed your .config
<sjoerd>
and this is why we need a rockchip defconfig tbh :p
<sjoerd>
well that or mare sure multiv7 has it all
<mmind00>
sjoerd: no need ... with Shawn's patch that is going away .... thankfully
<sjoerd>
well that specific bit is, but not the general problem of having to hunt down the right configuration :)
<mmind00>
yeah
premoboss has joined #linux-rockchip
<sjoerd>
mmind00: the IDMAC option solved it btw, thanks for the tip
<mmind00>
great
<sjoerd>
mmind00: btw i had a quick look at the io-domains last night for the sdio bit but setting it up for 1.8v didn't change the behaviour
<sjoerd>
something to look at later
levd1 has joined #linux-rockchip
naobsd has joined #linux-rockchip
levd has quit [Ping timeout: 244 seconds]
sunilmohan has joined #linux-rockchip
levd has joined #linux-rockchip
dlezcano_ has quit [Remote host closed the connection]
dlezcano has quit [Remote host closed the connection]
dlezcano has joined #linux-rockchip
levd1 has quit [Ping timeout: 245 seconds]
levd1 has joined #linux-rockchip
levd has quit [Ping timeout: 260 seconds]
levd has joined #linux-rockchip
levd1 has quit [Ping timeout: 250 seconds]
levd1 has joined #linux-rockchip
levd has quit [Ping timeout: 256 seconds]
levd1 has quit [Ping timeout: 272 seconds]
pietrushnic`away has joined #linux-rockchip
pietrushnic`away is now known as pietrushnic
sunilmohan has quit [Ping timeout: 252 seconds]
JohnDoe_71Rus has joined #linux-rockchip
markm has quit [Ping timeout: 252 seconds]
nighty^ has joined #linux-rockchip
markm has joined #linux-rockchip
cosm has joined #linux-rockchip
tlwoerner has quit [Quit: Leaving]
cosm has quit [Ping timeout: 260 seconds]
leowt has joined #linux-rockchip
<leowt>
Hi!
<leowt>
is there any kernel source for RK3168?
<leowt>
any content at all?
cosm has joined #linux-rockchip
leowt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
cosm has quit [Read error: Connection reset by peer]
leowt has joined #linux-rockchip
wb|biketron has joined #linux-rockchip
mrueg has quit [Remote host closed the connection]
mrueg has joined #linux-rockchip
<wb|biketron>
Hi, i got an RK3188 board here and i'm running Mainline (4.2). Currently i struggle with the USB Host controller (Host/OTG1) and the dwc2 driver. Seems that now an phy entry is required like the 3288 has. Did anybody got usb host running?
<wb|biketron>
My next step would be to try the phy bindings from 3288 but im missing the correct offset (can't find in TRM /maybe i'm blind)
<mmind00>
pacopad: depends on what kernel you use ... mainline does not yet support it, but sjoerd is working on that
mrueg has joined #linux-rockchip
<sjoerd>
mmind00: btw i've got some odd issue still with the sdpif output which seem to be clocking related, some audio rates just come out wrong
<sjoerd>
mmind00: almost as if the input clock doesn't end up being setup right even though the clk_summary says it is
<mmind00>
sjoerd: that may very well be the fault of the fraction clock
<sjoerd>
That's the one i suspect
<mmind00>
sjoerd: the downstream mux must point to the fraction clock for it to actually accept the rate
<mmind00>
(from the register to the actual clock)
<sjoerd>
mmind00: Oh so you have to first swtich the mux and *then* change the rate?
<mmind00>
sjoerd: yep, or at least point the mux to it when changing the rate
<sjoerd>
hmm
<mmind00>
give me a second, I'll send you some patches, I did back in december
<sjoerd>
mmind00: hah yes that seems be it
<sjoerd>
mmind00: switching rates backend forward seeminly fixes it
<sjoerd>
(it only chagnes the parent clock of the mux the first time i start playing)
<sjoerd>
mmind00: great thank you
<mmind00>
sjoerd: yep, it's one of the things on my gah-I-need-to-fix-that-list ;-)
<mmind00>
sjoerd: in the mail you can find my take on that ... never got around to test/finish that ... not even sure if it works in practice
<sjoerd>
awesome
markm has joined #linux-rockchip
cosm has joined #linux-rockchip
<mmind00>
sjoerd: although, now that I think about that again, it might be simplified, by putting the notifier on the mux and just writing the fractional value again when the mux switches to the fractional divider
mrueg has quit [Remote host closed the connection]
mrueg has joined #linux-rockchip
<sjoerd>
mmind00: Right rather then blipping it on and of to program it, just program it when required
<sjoerd>
mmind00: btw was this an issue on 3188 as well? I never hit the issue there
<mmind00>
sjoerd: not sure ... you might've gotten lucky ... from what I remember that was supposed to be present in all fractional dividers
<mmind00>
sjoerd: but I guess nobody actually double-checked the older ones at the time
mrueg has quit [Remote host closed the connection]
mrueg has joined #linux-rockchip
<mmind00>
sjoerd: I guess I'll simply get me toslink -> cinch converter later, give your spdif patches a spin and then finally fix that divider issue ;-)
gb_master has joined #linux-rockchip
<sjoerd>
that would be great
pietrushnic has quit [Read error: Connection reset by peer]
<sjoerd>
mmind00: ooi what board do you have with toslink?
gb_master has quit [Remote host closed the connection]
mrueg has quit [Remote host closed the connection]
mrueg has joined #linux-rockchip
<mmind00>
sjoerd: firdfly, radxa rock, rayeager popmetal
<mmind00>
oh and a Netxeon R89, that is currently serving as raidbox
<wb|biketron>
Okey new issue.... is there some extra kernel config for suspend to mem? As i thuought till now, CONFIG_SUSPEND=y should be enough... got only disk support "working" *LoL*
<sjoerd>
mmind00: guess teh firefly will be good to compare against the rock2
<sjoerd>
mmind00: currently doing something else, i'll push my wip patchset out later so you can pick the rk3288 stuff from that
<mmind00>
sjoerd: great
<sjoerd>
it's still hacking (didn't do the grf bits properly yet) but should get you started easily
<mmind00>
wb|biketron: suspend should work ... resume may be flaky, and on boards with the act8846 + syr* nobody tested it yet
<sjoerd>
*hackish
<wb|biketron>
mmind00: yeah this is what i expected and indeed i do have an act8846 so i could test it... but before this i have to figure out the standby issue
<mmind00>
wb|biketron: what doesn't work? aka does it suspend, but doesn't wake or something else?
<wb|biketron>
mmind00: It does not even list "mem" as supported value in /sys/power/state
<sjoerd>
mmind00: which also reminds me, do you have the wifi/sdio working on firefly?
<wb|biketron>
mmind00: got an idea kernel prompts PMU not found during early boot.... my investigation came to the point that may the DT biniding is not correct (need compatible rk3288-pmu to enable suspend ops in mack-rockchip/pm.c)
<wb|biketron>
btw: is there an fully featured uboot out there presenting an prompt not this rockchip fastboot/parameter stuff?
<mmind00>
wb|biketron: :-O what devicetree are you using?
<mmind00>
the pmu is a standard node in the mainline devicetree
<wb|biketron>
mmind00: The one in 4.2 ...at this point it is the same inside linux-rockchip ;)
<mmind00>
sjoerd: I did play around with the firefly's wifi a bit, but the driver got stuck at some point after loading the firmware
<wb|biketron>
mmind00: When i read this file (pm.c) correctly it probes only for rk3288-pmu and if not -> fail no pm ops
<mmind00>
wb|biketron: ah, on rk3188 you don't get suspend at all, at the moment
<mmind00>
the current suspend code does not support those yet ... especially as the suspend handling there is a bit more complex
<wb|biketron>
mmind00: Ahhh okey ...yeah figured out some more errors using the 3288 code at this point.. sadly i do not found an 3288 TRM to compare the diffs
<wb|biketron>
mmind00: Is someone working on this issue?
<mmind00>
wb|biketron: nope ... the whole putting code into the sram to power down the regular ram is largely unexplored
<wb|biketron>
mmind00: Yeah but for suspend to mem the regular ram is still powered up IIRC?
<wb|biketron>
so the only open point would be to set memory in SelfRefreshh
<mmind00>
wb|biketron: yes, selfrefresh is also missing on the 3288 right now ... called "deep" suspend
<wb|biketron>
mmind00: Just for background: i currently working on my Thesis about PowerManagement on ARM-Platforms (Linux-based) an want to use the RK for measurements of the Cortex Power Domains.. suspend to mem would have been an great additional point
<mmind00>
wb|biketron: sounds interesting :-) ... sadly I don't think much will happen in the rk3188 area in the near future
JohnDoe_71Rus has joined #linux-rockchip
cnxsoft has quit [Remote host closed the connection]
<wb|biketron>
mmind00: Yeah, so if i'll find some time i may patch this into the kernel... shouldn't be that complicated for simple suspend to mem
<wb|biketron>
Does anybody know if there is an fully featured uboot for rk3 1 88(not doing this fastboot parameter stuff) presenting an prompt out there?
naobsd has quit [Quit: naobsd]
<wb|biketron>
I am to foul to plug the SDC every time *shame
mrueg_ has joined #linux-rockchip
mrueg has quit [Read error: Connection reset by peer]
<sjoerd>
mmind00: appologise for it still being messy, but that should have all the bits you need, for spidf the sumitted patches + the last two of the top 3 are relevant
<sjoerd>
other bits are just the rock2 dts wip
<mmind00>
sjoerd: ok ... I'm not yet sure, when I'll have time anyway ... this weekend is planned for personal stuff ;-)
<mmind00>
sjoerd: for wifi, try increasing the drive strength? ... see veyron
<sjoerd>
mmind00: I did that did't help either
<sjoerd>
but didn't combine that yet with the io domain
<sjoerd>
need to run myself, later!
<sjoerd>
but if you wnat a play you have the bits needed :p
<mmind00>
sjoerd: :-P
cosm has quit [Ping timeout: 255 seconds]
mrueg_ is now known as mrueg
gb_master has joined #linux-rockchip
pizthewiz has joined #linux-rockchip
nighty^ has quit [Quit: Disappears in a puff of smoke]
sunilmohan has quit [Ping timeout: 244 seconds]
sunilmohan has joined #linux-rockchip
sunilmohan has joined #linux-rockchip
<wb|biketron>
As i see there is no thermal implementation on rk3188?
nighty^ has joined #linux-rockchip
<wb|biketron>
OK last question before weekend: does anybody has an online source for RK3188 PMU handbook?
markm has quit [Ping timeout: 244 seconds]
akaizen_ has joined #linux-rockchip
akaizen has quit [Read error: Connection reset by peer]
<mmind00>
wb|biketron: correct, rk3188 does not do thermal
<mmind00>
wb|biketron: looks like radxa is leaking at least parts
pizthewiz has quit [Quit:
pizthewiz has joined #linux-rockchip
akaizen_ has quit [Read error: Connection reset by peer]
akaizen has joined #linux-rockchip
akaizen has quit [Ping timeout: 246 seconds]
akaizen has joined #linux-rockchip
akaizen has quit [Remote host closed the connection]
gb_master has quit [Remote host closed the connection]
sunilmohan has quit [Ping timeout: 260 seconds]
nighty^ has quit [Read error: Connection reset by peer]