msimpson has quit [Remote host closed the connection]
msimpson has joined #linux-sunxi
msevo has joined #linux-sunxi
msevo has quit [Quit: Leaving]
SopaXorzTaker has quit [Quit: Leaving]
codekipper29 has joined #linux-sunxi
codekipper29 has quit [Remote host closed the connection]
codekipper97 has joined #linux-sunxi
<codekipper97>
z3ntu: Looking at the manuals it looks like you need to set up the I2S interface for the codec. This is done via AT+QDAI=
<z3ntu>
codekipper97: I have the modem set to AT+QDAI=1,1,0,4,0 (io: PCM output (1), mode: slave mode (1), fsync: primary mode (short sync) (0), clock: 2048K (4), format: 16-bit linear (0))
<codekipper97>
I think it should be AT+QDAI=1,1,1,4,0
<codekipper97>
we don't support the PCM format yet...although it can be hacked in
<codekipper97>
so you need to set the long sync
<z3ntu>
ah
<z3ntu>
But in auxiliary mode (long sync), "PCM interface operates with a 128kHz PCM_CLK and an 8kHz, 50% duty cycle PCM_SYNC only."
<z3ntu>
Is that okay?
<codekipper97>
mmm...this is a mono channel?
<codekipper97>
now looking at it..that doesn't look right
<codekipper97>
I've only ever looked at multi-channel pcm
matthias_bgg has quit [Read error: Connection reset by peer]
<KotCzarny>
kinda useful with li-ion, although it might be already in recent kernels
warpme_ has joined #linux-sunxi
<PaddleStroke>
wens btw I heard you know very well AXP subject. I think I found an issue with uboot AXP209 driver. When doing shutdown it overwrite some settings such a battery detection. In particular if you boot back with PEK button, then AXP register are not reset, so the overwrite is still there and battery detection is off.
<KotCzarny>
but as i've said most likely it's already included in some form
<PaddleStroke>
Thanks
<PaddleStroke>
this is the patch that Armbian use too.
arete74 has quit [Ping timeout: 248 seconds]
arete74 has joined #linux-sunxi
<PaddleStroke>
what is the sysfs to read battery capacity? I tried /sys/class/power_supply/qxp_pmu/capacity but it's not working
<wens>
PaddleStroke: send a fix for it?
<PaddleStroke>
wens to tell the truth I'm bit of a noob, I started to dig inside linux kernel and uboot only 2 weeks ago. I tried to send an email but I am not sure it went through
DrFrankensteinUK has quit [Ping timeout: 248 seconds]
DrFrankensteinUK has joined #linux-sunxi
warpme_ has quit [Quit: warpme_]
dev1990_ has quit [Remote host closed the connection]
<wens>
for u-boot, if you aren't subscribed, it would be put in holding
<codekipper97>
z3ntu: it's do-able but I won't be able to look until next week. Won't be mainline-able without a lot more effort though....I know that some other members are doing similar things but nothing has been shared.
<z3ntu>
I mostly care about it working for now - mainline can come later. Can you maybe give me some pointers what has to be done?
* mru
gives z3ntu a null pointer
<codekipper97>
you need to get the registers in I2S/PCM_FMT0 and I2S/PCM_FMT1 set up correctly. So that the signalling is what you see in "Figure 7-12. Timing Diagram for PCM/TDM-PCM mode"
vagrantc has joined #linux-sunxi
warpme_ has joined #linux-sunxi
msimpson has quit [Remote host closed the connection]
<qschulz>
PaddleStroke: well the thing is, we need to set this bit in the axp20x_battery driver in the probe function so we don't rely on the bootloader to set it up for us
<qschulz>
so fixing it in U-Boot, why not, it's always good to have. But it's important to have it in the kernel
<PaddleStroke>
it's same it writes bit(7) into reg32
<qschulz>
this one's arguable I think, we're anyway shutting down the board so I'm not sure it makes much of a difference but would be nice to fix I guess
warpme_ has quit [Quit: warpme_]
msimpson has joined #linux-sunxi
<PaddleStroke>
the issue is when we power on from PEK button. That's when the axp setting is not overwriten. The u-boot error is the same, it's when power-off that it overwrite the reg32.
msimpson has quit [Remote host closed the connection]
<PaddleStroke>
if you reboot with AC plug in, this bug don't have any effect because regs are reset. But when booting on PEK press, then the regs are not reset and the battery detection is off
<qschulz>
PaddleStroke: it's supposed to be in /sys/class/power_supply/smth/capacity for the capacity of the battery in mainline yes
<PaddleStroke>
(note my first message is wrong. I say "That's when the axp setting is not overwriten" I meant "that's when axp regs are not edited"
<qschulz>
Calibration is not supported yet though
<PaddleStroke>
"that's when axp regs are not reset" sorry double mistake.
<qschulz>
PaddleStroke: The point I was trying to make is that I'm not sure we actually care about monitoring the battery when we're shutting down the board.
<qschulz>
But the fix in the battery driver is definitely needed yes
<PaddleStroke>
Yes. What I meant is that apparantly after this shutdown, when you boot on PEK press, apparantly the AXP reg are not set back to default values. So on this second boot the reg32 is bit(7) which implies that battery detection is deactivate on this new boot.
<PaddleStroke>
but if reg32 value is set on boot by battery driver than it does not matter much
megi has quit [Ping timeout: 244 seconds]
warpme_ has joined #linux-sunxi
selfbg has quit [Remote host closed the connection]
<libv>
ElBarto: csi of allwinner is actually a parallel bus
<libv>
the adafruit module is to drive lcd panels
<libv>
but we don't care about what it is officially for
<libv>
we were only using this board to test the bandwidth
<libv>
and to verify our lcd and vga connection
<ElBarto>
libv: ok, nice
<libv>
this is stage 1 for the fosdem videobox
<libv>
stage 2 has us produce the first adv7611 based board with lcd and vga
<libv>
now that we know that it can do fullhd all day long, we can proceed