[Pipo-Project] M7Pro kernel

[raxy]

Raxy, I am currently testing with rileyrom 3.4.2. Would you like me to test with Randy's 4.4.4 rom?

yes please!

for that 4.4.4. I have the source, but not for the rrileyrom 3.4.2,
so our milage is very different.

[rrileypm]

I tested with my modified version of the ROM you sent me. I posted the link above. I just flashed D33's kernel.

[tgray]

I tested with my modified version of the ROM you sent me. I posted the link above. I just flashed D33's kernel.

Hi Randy, I am using the Pipo_M7_Pro_FreaktabROM_1.0. Is this the latest as the GPS doesn't work.

[raxy]

I cannot activate GPS nor Bluetooth nor auto brightness in my current combo of ke4nel and android.,
but usb audio is OK (even if I can hear statics depending of audio files)
USB DACs working:
- aureon dual usb
- hifimediy with sabre ESS 9023 : qute an achievement ! exceptional for such an old kernel !!!!!
- turtle beach audio advantage micro 2

[tgray]

I cannot activate GPS nor Bluetooth nor auto brightness in my current combo of ke4nel and android.,
but usb audio is OK (even if I can hear statics depending of audio files)

Hi Raxy, Same here.

[raxy]

tgray, you tested OK those 3 things when the android was the one in rrileyrom 3.4.2, and my latest kernel ,
right ?

[tgray]

Raxy, Yes, I tested those three things on rileyrom 3.4.2 and your latest kernel. They worked okay. I have just reinstalled rileyrom 3.4.2 with your latest kernel and I am going to test them again.

[raxy]

OK, thanks again all,
now I really need to have a rest and sleep,
it is almost 3 am here, bye!

[tgray]

Hi Raxy, I have tested your kernel with rileyrom 3.4.2 again and the Bluetooth, GPS and auto brightness works. There is a problem with the host USB port though. It works for awhile after reflashing but then stops responding. I tested it with a USB flash drive. I have checked this 3 times to be sure.

[raxy]

ok, so flash drive on usb host misbehaves after a while, and that seems to be kernel related;
I'm going to have same behavior on my hardware and look for error message for hints on why it is so... could be tricky to fix !

[TriA]

I'm joining the test
Flashed the last kernel (kernelLowBrightnessFixedRaxy) for use with Riley ROM 3.4.2.
I've overclocked the kernel with overclockomatic_kitkat2.pl script, having the same settings I use with Riley ROM (CPU 1800, RAM 792, GPU 768) - have no stability issues so far.
First impression - I like the lowest screen brightness level, it is much more comfortable in a dark room.
However, only 2/3 of brightness range is useful on my tablet - that is, no difference in brightness level between 2/3 and full brightness settings.
Because of that, the middle level brightness is too bright

[raxy]

...
First impression - I like the lowest screen brightness level, it is much more comfortable in a dark room.
However, only 2/3 of brightness range is useful on my tablet - that is, no difference in brightness level between 2/3 and full brightness settings.
Because of that, the middle level brightness is too bright

hello and thanks for joining in !

like you, I appreciated the lowest brightness setup, i have tried a few value, I was in complete darkness and the screen is OK in that context.
but, yes, the high range is less perfect, highest brightness is not that high, and the perceived linearity of the control is so so.

[d33]

What are you amending Raxy? The indexing in the us5151 source or the range in the SoC source?

[raxy]

what I've changed was
.min_brightness and .max_brightness and .brightness_mode in board-rk3188-ds1006h.c as I've seen one of your commits was on this file

static struct rk29_bl_info rk29_bl_info = {
        .min_brightness = 60, //20,//10
        .max_brightness = 150,//255
        .brightness_mode =BRIGHTNESS_MODE_LINE, //BRIGHTNESS_MODE_CONIC,
    .pre_div = 40 * 1000,  // pwm output clk: 40k;
    .pwm_id = PWM_ID,
    .bl_ref = PWM_EFFECT_VALUE,
    .io_init = rk29_backlight_io_init,
    .io_deinit = rk29_backlight_io_deinit,
    .pwm_suspend = rk29_backlight_pwm_suspend,
    .pwm_resume = rk29_backlight_pwm_resume,
};

[raxy]

There are very numerous differences for this file between your (multi tablet) copy and the one from Pipo; and they are not just ifdefine here and there, so I wonder if I should not try to just inject in your kernel tree Pipo's M7Pro file (which is supposed to be fully OK and production quality)... any comment on that ?