On Mon, Aug 18, 2014 at 5:36 PM, wrote:
> On 17/08/14 21:18, jonsm...@gmail.com wrote:
>> I think I found the bug, my sgtl5000 is happy now at lower clock
>> speeds. I was missing a 'L' and a value got truncated to 32 bits.
>
> Sorry for the late reply, I guess we're in different timezones :)
>
>
On 17/08/14 21:18, jonsm...@gmail.com wrote:
> I think I found the bug, my sgtl5000 is happy now at lower clock
> speeds. I was missing a 'L' and a value got truncated to 32 bits.
Sorry for the late reply, I guess we're in different timezones :)
Yes, that seems to work here too, giving expected o
I think I found the bug, my sgtl5000 is happy now at lower clock
speeds. I was missing a 'L' and a value got truncated to 32 bits.
On Sun, Aug 17, 2014 at 4:04 PM, jonsm...@gmail.com wrote:
> On Sun, Aug 17, 2014 at 3:22 PM, wrote:
>> I should probably add that I'm applying your patch onto a va
On Sun, Aug 17, 2014 at 3:22 PM, wrote:
> I should probably add that I'm applying your patch onto a vanilla 3.16
> and I'm using a cubietruck.
>
> Let me know if you'd like me to start with a different base kernel or
> to try it on different hardware, I have a couple of other A20 boards
> I can t
I should probably add that I'm applying your patch onto a vanilla 3.16
and I'm using a cubietruck.
Let me know if you'd like me to start with a different base kernel or
to try it on different hardware, I have a couple of other A20 boards
I can try if needed
--
You received this message beca
On 17/08/14 16:03, jonsm...@gmail.com wrote:
> Debug output from my test. Don't know if the pin is ticking.
debug output looks identical here
If I leave enable at 0 the output stays high, set enable to 1
and the output goes low and stays there regardless of any other
changes I make.
Even in th
On 17/08/14 16:02, jonsm...@gmail.com wrote:
> Thanks for testing. That gave me some things to fix. New version
> revmoved the SGTL clutter. It still has a lot of debug in it so that I
> can see what is going one.
>
> Note that the sysfs parameter 'duty_cycle' is not a percent
> duty_cycle. It wan
On Sun, Aug 17, 2014 at 11:16 AM, Chen-Yu Tsai wrote:
> Curious, why did you make a PWM driver when there is one in the works?
>
> See: https://lkml.org/lkml/2014/5/19/568
I suspect this one is going to have trouble with generating high
frequency clocks. I had to do the math in picoseconds to kee
On Sun, Aug 17, 2014 at 11:16 AM, Chen-Yu Tsai wrote:
> Curious, why did you make a PWM driver when there is one in the works?
Because I didn't know about it.
>
> See: https://lkml.org/lkml/2014/5/19/568
>
> On Sun, Aug 17, 2014 at 4:16 AM, jonsm...@gmail.com
> wrote:
>> I attached a PWM drive
Curious, why did you make a PWM driver when there is one in the works?
See: https://lkml.org/lkml/2014/5/19/568
On Sun, Aug 17, 2014 at 4:16 AM, jonsm...@gmail.com wrote:
> I attached a PWM driver for the mainline kernel. I think it is working
> but I don't have access to a scope at my current l
Debug output from my test. Don't know if the pin is ticking.
# ls
duty_cycle enable period polaritypower uevent
# echo 5 > period
[ 41.678063] JDS - sunxi_pwm_config duty 0 period 5
# echo 50 > duty_cycle
[ 48.753739] JDS - sunxi_pwm_config duty 50 period 5
[
Thanks for testing. That gave me some things to fix. New version
revmoved the SGTL clutter. It still has a lot of debug in it so that I
can see what is going one.
Note that the sysfs parameter 'duty_cycle' is not a percent
duty_cycle. It wants the duty_cycle in nanoseconds like the period.
That's
On 16/08/14 21:16, jonsm...@gmail.com wrote:
> I attached a PWM driver for the mainline kernel. I think it is working
> but I don't have access to a scope at my current location. I know it
> is making pulses, but without a scope I have no idea if they are the
> correct pulses.
>
> I tried this and
I attached a PWM driver for the mainline kernel. I think it is working
but I don't have access to a scope at my current location. I know it
is making pulses, but without a scope I have no idea if they are the
correct pulses.
I tried this and the audio chip wasn't happy with the pulses. It
should h
14 matches
Mail list logo
