Jump to content

Igor

Administrators
 View Profile  See their activity

  • Posts

    13648

  • Joined

  • Last visited

 Content Type 

Posts posted by Igor

    FTDI FT232 on Cubietruck

    in Allwinner sunxi

    Posted

    Jessie with 4.0.4 / v3.8, v3.9 is reported to have problems but works for me. Odd. I just installed Jessie with 4.0.5 on Orange Pi and started with long term stress test. It's re-booting fine, hard drive is working, CPU freq scaling, battery charge / discharge, ... running some stress test periodically and running some services.

     

    Current uptime is only few days so it's a bit too soon to make conclusions but it looks it's stable.

     

    BTW. Distribution upgrades should work but they are not bullet proof.

    0

    Change ramlog to direct log writing in kernel 3.4.107

    in Allwinner sunxi

    Posted

    How Do i downgrade the kernel to 3.4.106 together with all needed components like uboot and such? Is there an howto? Or is it the same as in FAQ to upgrade kernel?

     

    Manual is in the FAQ but not sure if it's 100% valid for this older kernel. U-boot is not packed with kernel but separate. 

     

    http://mirror.igorpecovnik.com/kernel/3.4.106-cubieboard-default.tar

     

    I can't find the exact u-boot, but try with this one.

    http://mirror.igorpecovnik.com/u-boot/linux-u-boot-3.4.107-cubietruck_3.6_armhf.deb

    0

    R1 - wifi issue

    in Allwinner sunxi

    Posted

    There is some problem with latest 3.4.107 build. Currently not sure where - it just hangs ... after days. I experience this first on Orange, than on Banana ... so I thing it affect all A20 boards.

     

    I suspect u-boot but am currently very tight with time for serious debugging.

     

    A good start would be to debug. :D We need to find out what's causing this. B)

     

    Try also with kernel 4.0.5 (u-boot is also a bit different).

    0

    compiling the external module

    in Allwinner sunxi

    Posted

    Use this kernel. FTDI should work:

     

    http://mirror.igorpecovnik.com/kernel/4.0.5-lamobo-r1-next.tar

     

    But I got wrong architecture after compiling sources from kernel.org:

     

    Because you need to define this in the compilation process - make ARCH=arm or whatever target.

     

    Kernel compilation is / can be a nightmare ... I made a script to minimize risk, to simplify process, to save time but even if you fully stick to defaults, problems might occur. If you want do compile in any other but theoretically proper environment, native included, it might fail :mellow:  :huh:

    0

    Change ramlog to direct log writing in kernel 3.4.107

    in Allwinner sunxi

    Posted

    Yes, I encounter this problem in recent build, but haven't got a solution yet. I thought it's related to hardware fail on some particular board. The problem could also be in U-boot.

     

    100% working combination is kernel 3.4.106 with matched u-boot or older releases. I have zero problems with this, uptime several months.

     

    To disable ramlog:

    service ramlog disable
reboot

    I am also working on to catch the bug, but it's a slow progress. Sometimes it works for weeks without a crash :(

    0

    Unable to build working bananapi-jessie-next image

    in Advanced users - Development

    Posted

    Yes, working normally ... with some extra default kernel debugging in the log  :P

     

    SD size you define in your initial call script.

    SDSIZE="2500"                               # SD image size in MB

    https://github.com/igorpecovnik/lib/blob/next/README.md

     

    This is the actual build script. You can build with old, not so recent, except in this scenario where is the latest kernel (and u-boot) hardencoded in / as an example. Some note should be there to clarify, I agree.

     

     

     

    in compile.sh I left "DEST_LANG="en_US.UTF-8", but it was built in Ubuntu with Czech language (so host/building pc uses czech language). Could that be a problem? If yes, we should mention this in Documentation. 

    No, it shouldn't. It should build properly regarding of the host LANG / locales settings. This needs to be fixed.

     

    I just build another image with Jessie. Runs o.k. - with XFCE desktop, installed manually ... need to fix that post auto install.

     

    Thanks for trying and finding problems  :D I know that they are and I am doing my best to knock them down.

     

    If you haven alter anything in scripts under /lib than scripts are not updating anymore - at least your changes files. This is not yet under FORCE="yes" yet. Workaround - delete whole lib/ - save your customization elsewhere be4 ;)

    0

    Unable to build working bananapi-jessie-next image

    in Advanced users - Development

    Posted

    Buggy development kernel with desktop on Jessie with some language customization is far worst possible scenario ...  but I manage to build an image on first try with default language settings. Only desktop building failed for no apparent reason. I am going to fix that today if possible.

    • Kernel 4.0.5 is latest stable, stick to this. I manage to build 4.1 RC
    • Jessie has problems with or without my interventions.
    • Desktop is here experimental - I don't deal with it.

    I hope I manage to find time and write a proper manual on this. I should  :o  :huh:  Holidays are coming  :P

     

    Enlarge default SD size a little bit, leave 4.0.5, leave default language, choose Wheezy / Trusty. Much better for start. Than tweak gradually. Most of things is cached, so build time is getting down to 2 minutes in best case.

    0

    Battery

    in Advanced users - Development

    Posted · Edited by Igor
    update

    I came out with this: 

    ######################################################################################
#
# Battery info
#


i2cset -y -f 0 0x34 0x82 0xC3
# read power OPERATING MODE register @01h
POWER_OP_MODE=$(i2cget -y -f 0 0x34 0x01)
BAT_EXIST=$(($(($POWER_OP_MODE&0x20))/32))  # divide by 32 is like shifting rigth 5 times
CHARG_IND=$(($(($POWER_OP_MODE&0x40))/64))  # divide by 64 is like shifting rigth 6 times

########################################################################################
#read battery voltage   79h, 78h        0 mV -> 000h,   1.1 mV/bit      FFFh -> 4.5045 V
BAT_VOLT_LSB=$(i2cget -y -f 0 0x34 0x79)
BAT_VOLT_MSB=$(i2cget -y -f 0 0x34 0x78)

BAT_BIN=$(( $(($BAT_VOLT_MSB << 4)) | $(($(($BAT_VOLT_LSB & 0xF0)) >> 4)) ))
BAT_VOLT=$(echo "($BAT_BIN*1.1)"|bc)

# store maximum battery voltage to compare to
if [ -f "/etc/default/battery" ]; then
        source "/etc/default/battery"
else
        echo "MAX=$BAT_VOLT" > /etc/default/battery
	echo "MIN=3484" >> /etc/default/battery
fi

# integer is enough, cut down the decimals
MAX=${MAX%.*}
BAT_VOLT=${BAT_VOLT%.*}

# mainline kernel shows battery existence even if not exists. this is walkaround 
if [ "$BAT_VOLT" -ge "3200" ]; then


# if we have new max value, alter defaults
if [ "$BAT_VOLT" -gt "$MAX" ]; then 
MAX=$BAT_VOLT
sed -e 's/MAX=.*/MAX='$BAT_VOLT'/g' -i /etc/default/battery
fi

# if we have new min value, alter defaults
if [ "$BAT_VOLT" -lt "$MIN" ]; then
MIN=$BAT_VOLT
#sed -e 's/MIN=.*/MIN='$BAT_VOLT'/g' -i /etc/default/battery
fi

# calculate percentage
percent=$(echo "($BAT_VOLT-$MIN)*100/($MAX-$MIN)"|bc)

# colorize output under certain percentage
if [ $percent -le 15 ]
then
  color="31"
fi

if [ "$BAT_EXIST" == "1" ]; then
BATT=" - Batt: "
# dispay charging / percentage
if [ "$CHARG_IND" == "1" ]; then
BATT=$BATT"charging $percent%"
else
BATT=$BATT"\e["$color"m$percent%\x1B[0m"
fi
fi
fi
OUT="${OUT}${BATT}"


echo ""
echo -e ${OUT}
echo ""

    Load: 1.64, 1.14, 1.00 - Board: 36.2°C - Ambient: 27.0°C - Drive: 43°C / 2.5Tb - Memory: 934Mb - Batt: 7%

     

    Working on both kernels.

    1

    SPI on Allwinner / Banana / Cubietruck ...

    in Advanced users - Development

    Posted

    I am having the exact same hw configuration as described here.

     

    banana-spi-display.jpg

     

    I am using latest (stable) mainline kernel 4.0.5, SPI is enabled in DTB like this:

    		spi0: spi@01c05000 {
			pinctrl-names = "default";
			pinctrl-0 = <&spi0_pins_a>;
			status = "okay";
			spi0_0 {
				#address-cells = <1>;
				#size-cells = <0>;
				compatible = "spidev";
				reg = <0>;
				spi-max-frequency = <50000000>;
			};
		};

    I am loading the driver like this and the displays back light goes on - so far is o.k. 

    modprobe fbtft_device name=adafruit22a rotate=90 speed=4800000 fps=30 gpios=reset:272,led:226,dc:245 busnum=32766

    This way, with different GPIOS numbers and without busnum parameter works perfectly well on kernel 3.4.x

     

    Kernel output when loading the driver on 4.0.5

    [27880.540870] fbtft: module is from the staging directory, the quality is unknown, you have been warned.
[27880.549004] fbtft_device: module is from the staging directory, the quality is unknown, you have been warned.
[27880.551333] fbtft_device:  SPI devices registered:
[27880.551386] fbtft_device:      spidev spi32766.0 50000kHz 8 bits mode=0x00
[27880.551406] fbtft_device:  'fb' Platform devices registered:
[27880.551570] fbtft_device: Deleting spi32766.0
[27880.553696] fbtft_device:  GPIOS used by 'adafruit22a':
[27880.553735] fbtft_device:    'reset' = GPIO272
[27880.553756] fbtft_device:    'led' = GPIO226
[27880.553775] fbtft_device:    'dc' = GPIO245
[27880.553793] fbtft_device:  SPI devices registered:
[27880.553820] fbtft_device:      fb_ili9340 spi32766.0 4800kHz 8 bits mode=0x00
[27880.575688] fb_ili9340: module is from the staging directory, the quality is unknown, you have been warned.
[27880.819789] fb_ili9340 spi32766.0: SPI transfer failed: -22
[27880.819830] spi_master spi32766: failed to transfer one message from queue
[27880.819850] fb_ili9340 spi32766.0: fbtft_update_display: write_vmem failed to update display buffer
[27880.821093] graphics fb1: fb_ili9340 frame buffer, 320x240, 150 KiB video memory, 4 KiB DMA buffer memory, fps=33, spi32766.0 at 4 MHz

    When I write some picture to the display I got this:

    [28727.125092] fb_ili9340 spi32766.0: SPI transfer failed: -22
[28727.125141] spi_master spi32766: failed to transfer one message from queue
[28727.125162] fb_ili9340 spi32766.0: fbtft_update_display: write_vmem failed to update display buffer
[28727.154895] fb_ili9340 spi32766.0: SPI transfer failed: -22
[28727.154937] spi_master spi32766: failed to transfer one message from queue
[28727.154957] fb_ili9340 spi32766.0: fbtft_update_display: write_vmem failed to update display buffer

    I haven't done the famous spi_test.c test and I also doesn't have any equipment to debug further. Is there any stupid thing that I am doing wrong or is this simply not done yet? Anyone got further on this?

    0

    CPU voltage control on 3.4 kernels and som OC

    in Beginners

    Posted

     

     

    In Bananian with kernel 3,4 (that means "NOT from Igor") I could change freq as I wanted and I did not respected 48MHz steps and it worked. But I am not sure if things were not slower...

     

    My version of kernel 3.4 is not that much different (at least in this part) from the one in Bananian so I think this feature works equal. I am not 100% sure since I don't pay much attention - I pay attention to the main community Linux:

    https://github.com/linux-sunxi/linux-sunxi

     

    2 & 3

    Status of mainline:

    http://linux-sunxi.org/Linux_mainlining_effort

     

    I don't use any extra configuration script except NAND / SATA / USB installer. The rest is either Debian/Ubuntu stock (keyboard, timezone, ...) or automatic (partition resize, SWAP) or done via .deb install (kernel switching, upgrade). I tend to have images as clean as possible.

    0
×
×
  • Create New...

Important Information

Terms of Use - Privacy Policy - Guidelines

  I accept