GTA02 sysfs

From Openmoko

(Difference between revisions)
Jump to: navigation, search
(/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/chgmode)
m (new: /sys/class/i2c-adapter/i2c-0/0-0073/lis302dl.{1|2}/...: accelerometers location)
 
(18 intermediate revisions by 13 users not shown)
Line 1: Line 1:
 +
'''NOTE: These only apply to Linux kernel 2.6.24, 2.6.28 has different paths''' (see below)
 +
 
==<font color=75d806>GTA02 Kernel sysfs highlights</font>==
 
==<font color=75d806>GTA02 Kernel sysfs highlights</font>==
 
sysfs is a filesystem that is mounted on /sys which contains various fake "files" that are actually filled by a variety of drivers and other kernel subsystems.  You can often also write stuff into these "files" using
 
sysfs is a filesystem that is mounted on /sys which contains various fake "files" that are actually filled by a variety of drivers and other kernel subsystems.  You can often also write stuff into these "files" using
Line 5: Line 7:
  
 
to change settings and behaviours of the kernel and drivers dynamically.  (When doing this, take care you have a space before the redirection operator &gt; or the shell will do something completely different with your command.)
 
to change settings and behaviours of the kernel and drivers dynamically.  (When doing this, take care you have a space before the redirection operator &gt; or the shell will do something completely different with your command.)
 +
However[http://permalink.gmane.org/gmane.comp.hardware.smartphones.userland/165], one should use [[Dbus device API]]s!
  
 
===USB Host / Device===
 
===USB Host / Device===
Line 14: Line 17:
 
* Set to "host" to logically configure the peripheral to act like a host for other USB devices to be plugged into it
 
* Set to "host" to logically configure the peripheral to act like a host for other USB devices to be plugged into it
  
Normally you set these two guys at the same time for the same mode, but there is a trick possible to select logical host mode but leave the power arrangements as if it was in device mode and apply external power.  You need to externally add 15K pulldowns on D+ and D- if you do this, because we did not enable them the normal way.  With this, you can have a USB device attached to GTA02, and use external power and charge the battery at the same time.
+
Normally you set these two guys at the same time for the same mode, but there is a trick possible to select logical host mode but leave the power arrangements as if it was in device mode and apply external power.  You need to [[Specialized USB cables|externally add]] 15K pulldowns on D+ and D- if you do this, because we did not enable them the normal way.  With this, you can have a USB device attached to GTA02, and use external power and charge the battery at the same time.
  
 
===GSM Subsystem===
 
===GSM Subsystem===
Line 33: Line 36:
 
===MEMCONFIG===
 
===MEMCONFIG===
 
====/sys/devices/platform/neo1973-memconfig.0/BANKCON0 .. 7====
 
====/sys/devices/platform/neo1973-memconfig.0/BANKCON0 .. 7====
For extreme meddlers (I salute you), these let you control the wait states and other characteristics of the memory regions of the Freerunner.  Information and warnings about how to use these are here:
+
For extreme meddlers (I salute you), these let you control the wait states and other characteristics of the memory regions of the FreeRunner.  Information and warnings about how to use these are here:
  
 
http://git.openmoko.org/?p=kernel.git;a=commitdiff;h=fd25f90517322fc6c07bfb8d34752d3cb41cb4b6
 
http://git.openmoko.org/?p=kernel.git;a=commitdiff;h=fd25f90517322fc6c07bfb8d34752d3cb41cb4b6
Line 47: Line 50:
 
===power_supply battery information===
 
===power_supply battery information===
 
Most of these are coming from the coulomb counter over HDQ dynamically, but the bq27000 does not update a lot of its registers any more often than once per 4 seconds.  Therefore a lot of this info can be a bit stale or averaged over that period.
 
Most of these are coming from the coulomb counter over HDQ dynamically, but the bq27000 does not update a lot of its registers any more often than once per 4 seconds.  Therefore a lot of this info can be a bit stale or averaged over that period.
 +
====/sys/devices/platform/bq27000-battery.0/power_supply/bat/uevent====
 +
All informations about battery.
 
====/sys/devices/platform/bq27000-battery.0/power_supply/bat/type====
 
====/sys/devices/platform/bq27000-battery.0/power_supply/bat/type====
 
Just says "Battery"
 
Just says "Battery"
Line 60: Line 65:
 
Current being drawn from battery (+ve) or pushed into battery during charging (-ve) in uA
 
Current being drawn from battery (+ve) or pushed into battery during charging (-ve) in uA
 
====/sys/devices/platform/bq27000-battery.0/power_supply/bat/charge_full====
 
====/sys/devices/platform/bq27000-battery.0/power_supply/bat/charge_full====
Coulomb Counter's estimate of the capacity of the battery measured in uA/h.  It roughly means that if full, for one hour it could supply current at this rate.  I saw 1197913 reported on mine, so it estimates it can provide 1.2A for an hour.
+
Coulomb Counter's estimate of the capacity of the battery measured in uAh.  It roughly means that if full, for one hour it could supply current at this rate.  I saw 1197913 reported on mine, so it estimates it can provide 1.2A for an hour.
 +
 
 
====/sys/devices/platform/bq27000-battery.0/power_supply/bat/temp====
 
====/sys/devices/platform/bq27000-battery.0/power_supply/bat/temp====
 
This is the battery temperature reported by the Coulomb Counter chip that is part of the physical battery.  It is in Celcius * 10, so 251 means 25.1 degrees.
 
This is the battery temperature reported by the Coulomb Counter chip that is part of the physical battery.  It is in Celcius * 10, so 251 means 25.1 degrees.
Line 86: Line 92:
  
 
====/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/charger_type====
 
====/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/charger_type====
Shows the detected charger type followed by the mode it is running in.  So "charger 500mA mode 100mA" would be the case for non-enumerated USB connection, "charger 500mA mode 500mA" for after enumeration.
+
Shows the detected charger type followed by the mode it is running in.  So "charger 500mA mode 100mA" would be the case for non-enumerated USB connection, "charger 500mA mode 500mA" for after enumeration (or "host/500mA usb mode 500mA" as here [[User:OlivierBerger|OlivierBerger]]).
 +
 
 
====/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/force_usb_limit_dangerous====
 
====/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/force_usb_limit_dangerous====
 
This is added to allow the PMU to be told to take more current than the default rules allow.  Use it only if you know your charger source can handle it.  For example
 
This is added to allow the PMU to be told to take more current than the default rules allow.  Use it only if you know your charger source can handle it.  For example
  
echo 500 > /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/force_usb_limit_dangerous
+
echo 500 > /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/force_usb_limit_dangerous
  
 
will allow PMU to draw 500mA even though you are connected to a dumb charger where the limit is normally held at 100mA, since it didn't enumerate the GTA02.  You can give 0, 100, 500 or 1000 here, but make sure your charger source can handle what you give here, otherwise it can exceed what is safe for your charging source to provide.
 
will allow PMU to draw 500mA even though you are connected to a dumb charger where the limit is normally held at 100mA, since it didn't enumerate the GTA02.  You can give 0, 100, 500 or 1000 here, but make sure your charger source can handle what you give here, otherwise it can exceed what is safe for your charging source to provide.
 +
 +
More details at [[Forcing_fast_charge_mode]].
 +
 
====/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/dump_regs====
 
====/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/dump_regs====
 
cat this to get a dump of the I2C registers for the PMU, useful for debugging power problems.
 
cat this to get a dump of the I2C registers for the PMU, useful for debugging power problems.
Line 119: Line 129:
 
cat this to get a dump of selected Glamo regs.  You can write the regs by echoing "<glamo reg index in decimal> <value in decimal>" here.
 
cat this to get a dump of selected Glamo regs.  You can write the regs by echoing "<glamo reg index in decimal> <value in decimal>" here.
  
[[Category:Flashing Openmoko]]
+
===LED & Vibrator===
 +
(per: Jeff Tickle)
 +
====/sys/devices/platform/neo1973-vibrator.0/leds/neo1973:vibrator/brightness====
 +
 
 +
For the Vibration, my very limited testing indicates it will accept a
 +
range of 0-255, with 0 being off and 255 being full force.  So, you can
 +
actually set the strength of vibration.
 +
 
 +
Keep in mind that the vibration will positively eat your battery, so
 +
use with caution!
 +
 
 +
====/sys/devices/platform/gta02-led.0/leds/gta02-aux:red/brightness====
 +
====/sys/devices/platform/gta02-led.0/leds/gta02-power:blue/brightness====
 +
====/sys/devices/platform/gta02-led.0/leds/gta02-power:orange/brightness====
 +
 
 +
For the LEDs, you can echo 1 for on or 0 for off to any of these files
 +
to control the state of the LEDs.
 +
 
 +
===Accelerometers===
 +
====/sys/devices/platform/spi_s3c24xx_gpio.1/spi0.0/power/wakeup====
 +
====/sys/devices/platform/spi_s3c24xx_gpio.1/spi0.1/power/wakeup====
 +
Turns on accelerometers.
 +
 
 +
====/sys/devices/platform/spi_s3c24xx_gpio.1/spi0.0/dump====
 +
====/sys/devices/platform/spi_s3c24xx_gpio.1/spi0.1/dump====
 +
Dumps accelerometers raw contents.
 +
 
 +
====/sys/devices/platform/spi_s3c24xx_gpio.1/spi0.0/full_scale====
 +
====/sys/devices/platform/spi_s3c24xx_gpio.1/spi0.1/full_scale====
 +
Switches the accelerometers between 2G and 8G mode.  Echo "9.2" into it for 8G, any other value for 2G (default)[http://lists.openmoko.org/pipermail/openmoko-kernel/2008-July/004029.html]
 +
 
 +
 
 +
 
 +
==<font color=75d806>GTA02 Kernel sysfs highlights for kernel 2.6.28</font>==
 +
(info from: [http://lists.openmoko.org/nabble.html#nabble-td1569863 http://lists.openmoko.org/nabble.html#nabble-td1569863]
 +
 
 +
Warning: this is still a work in progress, and things might still change in the future.
 +
 
 +
 
 +
Here are the current /sys mappings for andy-tracking (which has Balaji's
 +
regulator PMU stuff).  The reason virtually everything has changed is
 +
mainly due to the device tree changes to fix suspend - resume, many more
 +
things have a child relationship to the PMU now.
 +
 
 +
 
 +
====old: /sys/devices/platform/neo1973-pm-host.0/hostmode ====
 +
====new: /sys/class/i2c-adapter/i2c-0/0-0073/neo1973-pm-host.0/hostmode ====
 +
 
 +
====same: /sys/devices/platform/s3c2410-ohci/usb_mode ====
 +
(from 901d73fe51f33032b34b2ae5612eb863ec90532a moved to /sys/devices/platform/s3c-ohci/usb_mode)
 +
 
 +
====old: /sys/devices/platform/neo1973-pm-gsm.0/power_on|reset|download ====
 +
====new: /sys/class/i2c-adapter/i2c-0/0-0073/neo1973-pm-gsm.0/... ====
 +
 
 +
====old: /sys/devices/platform/neo1973-resume.0/resume_reason ====
 +
====new: /sys/class/i2c-adapter/i2c-0/0-0073/neo1973-resume.0/resume_reason ====
 +
    /sys/class/i2c-adapter/i2c-0/0-0073/resume_reason
 +
    (split into CPU wake interrupt and secondly PMU-specific reason)
 +
 
 +
====old: /sys/devices/platform/sc32440_fiq.0/fiq/count|dump|write ====
 +
====new: /sys/devices/platform/sc32440_fiq.0/gta02-hdq.0/hdq/... ====
 +
 
 +
====old: /sys/devices/platform/bq27000-battery.0/power_supply/bat/ ====
 +
    uevent|type|status|voltage_now|current_now|charge_full|temp|
 +
    technology|present|time_to_empty_now|time_to_full_now|capacity|
 +
    online
 +
====new: /sys/class/power_supply/battery/... ====
 +
 
 +
====old: /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/ ====
 +
    chgmode|charger_type|force_usb_limit_dangerous|dump_regs
 +
====new: /sys/class/i2c-adapter/i2c-0/0-0073/pcf50633-mbc/... ====
 +
 
 +
====old: /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/neo1973-pm-bt.0/ ====
 +
power_on|reset
 +
====new: /sys/class/i2c-adapter/i2c-0/0-0073/pcf50633-regltr.6/neo1973-pm-bt.0/... ====
 +
 
 +
====old: /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/neo1973-pm-gps.0/pwron ====
 +
====new: /sys/class/i2c-adapter/i2c-0/0-0073/pcf50633-regltr.7/neo1973-pm-gps.0/power_on ====
 +
 
 +
====same: /sys/devices/platform/soc-audio/codec_reg|codec_reg_write ====
 +
 
 +
====old: /sys/devices/platform/glamo3362.0/regs ====
 +
====new: /sys/class/i2c-adapter/i2c-0/0-0073/pcf50633-regltr.9/glamo3362.0/regs ====
 +
 
 +
====old: /sys/devices/platform/neo1973-vibrator.0/leds/neo1973:vibrator/brightness ====
 +
====new: /sys/class/leds/neo1973:vibrator/brightness ====
 +
 
 +
====old: /sys/devices/platform/gta02-led.0/leds/gta02-aux:red/brightness ====
 +
====new: /sys/class/leds/gta02-aux:red/brightness ====
 +
 
 +
====old: /sys/devices/platform/gta02-led.0/leds/gta02-power:blue/brightness ====
 +
====new: /sys/class/leds/gta02-power:blue/brightness ====
 +
 
 +
====old: /sys/devices/platform/gta02-led.0/leds/gta02-power:orange/brightness ====
 +
====new: /sys/class/leds/gta02-power:orange/brightness ====
 +
 
 +
====old: /sys/devices/platform/spi_s3c24xx_gpio.1/spi0.{0|1}/ ====
 +
power/wakeup|dump|full_scale...
 +
====new: /sys/class/i2c-adapter/i2c-0/0-0073/spi_s3c24xx_gpio.0/spi3.{0|1}/ ====
 +
 
 +
 
 +
 
 +
[[Category:Kernel]]

Latest revision as of 21:53, 27 October 2009

NOTE: These only apply to Linux kernel 2.6.24, 2.6.28 has different paths (see below)

Contents

[edit] GTA02 Kernel sysfs highlights

sysfs is a filesystem that is mounted on /sys which contains various fake "files" that are actually filled by a variety of drivers and other kernel subsystems. You can often also write stuff into these "files" using

echo 1 > /sys/blah...

to change settings and behaviours of the kernel and drivers dynamically. (When doing this, take care you have a space before the redirection operator > or the shell will do something completely different with your command.) However[1], one should use Dbus device APIs!

[edit] USB Host / Device

[edit] /sys/devices/platform/neo1973-pm-host.0/hostmode

  • Defaults to 0, GTA02 USB hardware is configured to be a device, no power is generated for USB, charging is enabled and host 15K pulldowns are removed from D+ and D-
  • Set to 1 to put the GTA02 USB hardware into host mode, so it starts to generate 5V power, disables charging from USB, and applies 15K pulldowns to USB D+ and D-

[edit] /sys/devices/platform/s3c2410-ohci/usb_mode

  • Defaults to "device", it means the USB peripheral in the CPU is logically configured for device mode
  • Set to "host" to logically configure the peripheral to act like a host for other USB devices to be plugged into it

Normally you set these two guys at the same time for the same mode, but there is a trick possible to select logical host mode but leave the power arrangements as if it was in device mode and apply external power. You need to externally add 15K pulldowns on D+ and D- if you do this, because we did not enable them the normal way. With this, you can have a USB device attached to GTA02, and use external power and charge the battery at the same time.

[edit] GSM Subsystem

[edit] /sys/devices/platform/neo1973-pm-gsm.0/power_on

  • Defaults to "0", GSM power OFF
  • Set to "1" to enable power to GSM logic

[edit] /sys/devices/platform/neo1973-pm-gsm.0/reset

  • Defaults to "0", no reset
  • Set to "1" briefly after power applied to reset GSM logic

[edit] /sys/devices/platform/neo1973-pm-gsm.0/download

  • Defaults to "0", no insane clicking sound
  • Set to "1" to drive yourself crazy and enable serial access to GSM chips down headphone socket / once a second loud clicks / chirps

[edit] Resume Reason

[edit] /sys/devices/platform/neo1973-resume.0/resume_reason

If you cat this you get a list of possible resume sources in text with one or more * at the left of the active source that woke us. You need a recent U-Boot to get the * set.

[edit] MEMCONFIG

[edit] /sys/devices/platform/neo1973-memconfig.0/BANKCON0 .. 7

For extreme meddlers (I salute you), these let you control the wait states and other characteristics of the memory regions of the FreeRunner. Information and warnings about how to use these are here:

http://git.openmoko.org/?p=kernel.git;a=commitdiff;h=fd25f90517322fc6c07bfb8d34752d3cb41cb4b6

[edit] FIQ and HDQ

[edit] /sys/devices/platform/sc32440_fiq.0/fiq/count

FIQ is used to provide precision interrupt service in order to implement the HDQ protocol used to the battery couloumb counter. This shows how many FIQ interrupts have been run since the kernel started. It's turned off unless it is required.

[edit] /sys/devices/platform/gta02-hdq.0/hdq/dump

If you cat this, you will see the raw contents of HDQ device register space. This is the bq27000 couloumb counter in our case.

[edit] /sys/devices/platform/gta02-hdq.0/hdq/write

This allows you to write to raw HDQ registers, using "<decimal register index> <decimal value>"

[edit] power_supply battery information

Most of these are coming from the coulomb counter over HDQ dynamically, but the bq27000 does not update a lot of its registers any more often than once per 4 seconds. Therefore a lot of this info can be a bit stale or averaged over that period.

[edit] /sys/devices/platform/bq27000-battery.0/power_supply/bat/uevent

All informations about battery.

[edit] /sys/devices/platform/bq27000-battery.0/power_supply/bat/type

Just says "Battery"

[edit] /sys/devices/platform/bq27000-battery.0/power_supply/bat/status

This will summarize what goes on generally with the battery

  • "Not Charging" - there's a charger in, but right now it isn't charging us. Normal if you are fully charged and still powered.
  • "Charging" - there's a charger in and it is charging the battery
  • "Discharging" - we are running on battery
  • "Full" - not normally given as a result (Not Charging used instead)

[edit] /sys/devices/platform/bq27000-battery.0/power_supply/bat/voltage_now

Battery voltage in uV, averaged. Obviously this depends on load and place on discharge curve, etc.

[edit] /sys/devices/platform/bq27000-battery.0/power_supply/bat/current_now

Current being drawn from battery (+ve) or pushed into battery during charging (-ve) in uA

[edit] /sys/devices/platform/bq27000-battery.0/power_supply/bat/charge_full

Coulomb Counter's estimate of the capacity of the battery measured in uAh. It roughly means that if full, for one hour it could supply current at this rate. I saw 1197913 reported on mine, so it estimates it can provide 1.2A for an hour.

[edit] /sys/devices/platform/bq27000-battery.0/power_supply/bat/temp

This is the battery temperature reported by the Coulomb Counter chip that is part of the physical battery. It is in Celcius * 10, so 251 means 25.1 degrees.

[edit] /sys/devices/platform/bq27000-battery.0/power_supply/bat/technology

"Li-ion"

[edit] /sys/devices/platform/bq27000-battery.0/power_supply/bat/present

  • "0" the battery is absent, or not one with a Coulomb Counter
  • "1" the smart battery is present

[edit] /sys/devices/platform/bq27000-battery.0/power_supply/bat/time_to_empty_now

At current rate of discharge, estimate of how long we can run for. If battery is not discharging, it won't make an estimate and will return a magic value "3932100" meaning "no estimate". The coulomb counter averages the load and adjusts this value slowly to be its estimate of when we will blow chunks.

[edit] /sys/devices/platform/bq27000-battery.0/power_supply/bat/time_to_full_now

This estimates how long until we are fully charged, at current rate of charging, in seconds. If we are not charging, it gives the magic value "3932100" meaning "no estimate".

[edit] /sys/devices/platform/bq27000-battery.0/power_supply/bat/capacity

This is the remaining capacity of charge in the battery in percent. This is probably the most useful figure here.

[edit] /sys/devices/platform/bq27000-battery.0/power_supply/bat/online

  • "0" no charger is present
  • "1" A powered charger is present

[edit] PMU info

[edit] /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/chgmode

  • "enabled" means the charger is powered and willing to charge, although if you're full it might not be right now
  • "disabled" means there was no power or not enough power supplied to charge

Looks like it may also report "fast" here. OlivierBerger 07:47, 25 July 2008 (UTC)

[edit] /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/charger_type

Shows the detected charger type followed by the mode it is running in. So "charger 500mA mode 100mA" would be the case for non-enumerated USB connection, "charger 500mA mode 500mA" for after enumeration (or "host/500mA usb mode 500mA" as here OlivierBerger).

[edit] /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/force_usb_limit_dangerous

This is added to allow the PMU to be told to take more current than the default rules allow. Use it only if you know your charger source can handle it. For example

echo 500 > /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/force_usb_limit_dangerous

will allow PMU to draw 500mA even though you are connected to a dumb charger where the limit is normally held at 100mA, since it didn't enumerate the GTA02. You can give 0, 100, 500 or 1000 here, but make sure your charger source can handle what you give here, otherwise it can exceed what is safe for your charging source to provide.

More details at Forcing_fast_charge_mode.

[edit] /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/dump_regs

cat this to get a dump of the I2C registers for the PMU, useful for debugging power problems.

[edit] Bluetooth

[edit] /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/neo1973-pm-bt.0/power_on

  • Default "0" no power to Bluetooth device
  • set to "1" to enable power to Bluetooth

[edit] /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/neo1973-pm-bt.0/reset

  • Default to "0", no reset
  • set to "1" to reset BT (needed after powerup)

[edit] GPS

[edit] /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/neo1973-pm-gps.0/pwron

  • Default "0" GPS unpowered
  • set to "1" to power GPS section

[edit] Audio codec

[edit] /sys/devices/platform/soc-audio/codec_reg

cat this to get a dump of codec registers. Note they are 9-bit wide.

[edit] /sys/devices/platform/soc-audio/codec_reg_write

You can write to any codec register by echoing "<codec reg index in hex> <value in hex>" to here

[edit] Glamo

[edit] /sys/devices/platform/glamo3362.0/regs

cat this to get a dump of selected Glamo regs. You can write the regs by echoing "<glamo reg index in decimal> <value in decimal>" here.

[edit] LED & Vibrator

(per: Jeff Tickle)

[edit] /sys/devices/platform/neo1973-vibrator.0/leds/neo1973:vibrator/brightness

For the Vibration, my very limited testing indicates it will accept a range of 0-255, with 0 being off and 255 being full force. So, you can actually set the strength of vibration.

Keep in mind that the vibration will positively eat your battery, so use with caution!

[edit] /sys/devices/platform/gta02-led.0/leds/gta02-aux:red/brightness

[edit] /sys/devices/platform/gta02-led.0/leds/gta02-power:blue/brightness

[edit] /sys/devices/platform/gta02-led.0/leds/gta02-power:orange/brightness

For the LEDs, you can echo 1 for on or 0 for off to any of these files to control the state of the LEDs.

[edit] Accelerometers

[edit] /sys/devices/platform/spi_s3c24xx_gpio.1/spi0.0/power/wakeup

[edit] /sys/devices/platform/spi_s3c24xx_gpio.1/spi0.1/power/wakeup

Turns on accelerometers.

[edit] /sys/devices/platform/spi_s3c24xx_gpio.1/spi0.0/dump

[edit] /sys/devices/platform/spi_s3c24xx_gpio.1/spi0.1/dump

Dumps accelerometers raw contents.

[edit] /sys/devices/platform/spi_s3c24xx_gpio.1/spi0.0/full_scale

[edit] /sys/devices/platform/spi_s3c24xx_gpio.1/spi0.1/full_scale

Switches the accelerometers between 2G and 8G mode. Echo "9.2" into it for 8G, any other value for 2G (default)[2]


[edit] GTA02 Kernel sysfs highlights for kernel 2.6.28

(info from: http://lists.openmoko.org/nabble.html#nabble-td1569863

Warning: this is still a work in progress, and things might still change in the future.


Here are the current /sys mappings for andy-tracking (which has Balaji's regulator PMU stuff). The reason virtually everything has changed is mainly due to the device tree changes to fix suspend - resume, many more things have a child relationship to the PMU now.


[edit] old: /sys/devices/platform/neo1973-pm-host.0/hostmode

[edit] new: /sys/class/i2c-adapter/i2c-0/0-0073/neo1973-pm-host.0/hostmode

[edit] same: /sys/devices/platform/s3c2410-ohci/usb_mode

(from 901d73fe51f33032b34b2ae5612eb863ec90532a moved to /sys/devices/platform/s3c-ohci/usb_mode)

[edit] old: /sys/devices/platform/neo1973-pm-gsm.0/power_on|reset|download

[edit] new: /sys/class/i2c-adapter/i2c-0/0-0073/neo1973-pm-gsm.0/...

[edit] old: /sys/devices/platform/neo1973-resume.0/resume_reason

[edit] new: /sys/class/i2c-adapter/i2c-0/0-0073/neo1973-resume.0/resume_reason

    /sys/class/i2c-adapter/i2c-0/0-0073/resume_reason 
    (split into CPU wake interrupt and secondly PMU-specific reason) 

[edit] old: /sys/devices/platform/sc32440_fiq.0/fiq/count|dump|write

[edit] new: /sys/devices/platform/sc32440_fiq.0/gta02-hdq.0/hdq/...

[edit] old: /sys/devices/platform/bq27000-battery.0/power_supply/bat/

    uevent|type|status|voltage_now|current_now|charge_full|temp| 
    technology|present|time_to_empty_now|time_to_full_now|capacity| 
    online 

[edit] new: /sys/class/power_supply/battery/...

[edit] old: /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/

    chgmode|charger_type|force_usb_limit_dangerous|dump_regs 

[edit] new: /sys/class/i2c-adapter/i2c-0/0-0073/pcf50633-mbc/...

[edit] old: /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/neo1973-pm-bt.0/

power_on|reset

[edit] new: /sys/class/i2c-adapter/i2c-0/0-0073/pcf50633-regltr.6/neo1973-pm-bt.0/...

[edit] old: /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/neo1973-pm-gps.0/pwron

[edit] new: /sys/class/i2c-adapter/i2c-0/0-0073/pcf50633-regltr.7/neo1973-pm-gps.0/power_on

[edit] same: /sys/devices/platform/soc-audio/codec_reg|codec_reg_write

[edit] old: /sys/devices/platform/glamo3362.0/regs

[edit] new: /sys/class/i2c-adapter/i2c-0/0-0073/pcf50633-regltr.9/glamo3362.0/regs

[edit] old: /sys/devices/platform/neo1973-vibrator.0/leds/neo1973:vibrator/brightness

[edit] new: /sys/class/leds/neo1973:vibrator/brightness

[edit] old: /sys/devices/platform/gta02-led.0/leds/gta02-aux:red/brightness

[edit] new: /sys/class/leds/gta02-aux:red/brightness

[edit] old: /sys/devices/platform/gta02-led.0/leds/gta02-power:blue/brightness

[edit] new: /sys/class/leds/gta02-power:blue/brightness

[edit] old: /sys/devices/platform/gta02-led.0/leds/gta02-power:orange/brightness

[edit] new: /sys/class/leds/gta02-power:orange/brightness

[edit] old: /sys/devices/platform/spi_s3c24xx_gpio.1/spi0.{0|1}/

power/wakeup|dump|full_scale...

[edit] new: /sys/class/i2c-adapter/i2c-0/0-0073/spi_s3c24xx_gpio.0/spi3.{0|1}/

Personal tools

GTA02 Kernel sysfs highlights

sysfs is a filesystem that is mounted on /sys which contains various fake "files" that are actually filled by a variety of drivers and other kernel subsystems. You can often also write stuff into these "files" using

echo 1 > /sys/blah...

to change settings and behaviours of the kernel and drivers dynamically. (When doing this, take care you have a space before the redirection operator > or the shell will do something completely different with your command.)

USB Host / Device

/sys/devices/platform/neo1973-pm-host.0/hostmode

  • Defaults to 0, GTA02 USB hardware is configured to be a device, no power is generated for USB, charging is enabled and host 15K pulldowns are removed from D+ and D-
  • Set to 1 to put the GTA02 USB hardware into host mode, so it starts to generate 5V power, disables charging from USB, and applies 15K pulldowns to USB D+ and D-

/sys/devices/platform/s3c2410-ohci/usb_mode

  • Defaults to "device", it means the USB peripheral in the CPU is logically configured for device mode
  • Set to "host" to logically configure the peripheral to act like a host for other USB devices to be plugged into it

Normally you set these two guys at the same time for the same mode, but there is a trick possible to select logical host mode but leave the power arrangements as if it was in device mode and apply external power. You need to externally add 15K pulldowns on D+ and D- if you do this, because we did not enable them the normal way. With this, you can have a USB device attached to GTA02, and use external power and charge the battery at the same time.

GSM Subsystem

/sys/devices/platform/neo1973-pm-gsm.0/power_on

  • Defaults to "0", GSM power OFF
  • Set to "1" to enable power to GSM logic

/sys/devices/platform/neo1973-pm-gsm.0/reset

  • Defaults to "0", no reset
  • Set to "1" briefly after power applied to reset GSM logic

/sys/devices/platform/neo1973-pm-gsm.0/download

  • Defaults to "0", no insane clicking sound
  • Set to "1" to drive yourself crazy and enable serial access to GSM chips down headphone socket / once a second loud clicks / chirps

Resume Reason

/sys/devices/platform/neo1973-resume.0/resume_reason

If you cat this you get a list of possible resume sources in text with one or more * at the left of the active source that woke us. You need a recent U-Boot to get the * set.

MEMCONFIG

/sys/devices/platform/neo1973-memconfig.0/BANKCON0 .. 7

For extreme meddlers (I salute you), these let you control the wait states and other characteristics of the memory regions of the Freerunner. Information and warnings about how to use these are here:

http://git.openmoko.org/?p=kernel.git;a=commitdiff;h=fd25f90517322fc6c07bfb8d34752d3cb41cb4b6

FIQ and HDQ

/sys/devices/platform/sc32440_fiq.0/fiq/count

FIQ is used to provide precision interrupt service in order to implement the HDQ protocol used to the battery couloumb counter. This shows how many FIQ interrupts have been run since the kernel started. It's turned off unless it is required.

/sys/devices/platform/gta02-hdq.0/hdq/dump

If you cat this, you will see the raw contents of HDQ device register space. This is the bq27000 couloumb counter in our case.

/sys/devices/platform/gta02-hdq.0/hdq/write

This allows you to write to raw HDQ registers, using "<decimal register index> <decimal value>"

power_supply battery information

Most of these are coming from the coulomb counter over HDQ dynamically, but the bq27000 does not update a lot of its registers any more often than once per 4 seconds. Therefore a lot of this info can be a bit stale or averaged over that period.

/sys/devices/platform/bq27000-battery.0/power_supply/bat/type

Just says "Battery"

/sys/devices/platform/bq27000-battery.0/power_supply/bat/status

This will summarize what goes on generally with the battery

  • "Not Charging" - there's a charger in, but right now it isn't charging us. Normal if you are fully charged and still powered.
  • "Charging" - there's a charger in and it is charging the battery
  • "Discharging" - we are running on battery
  • "Full" - not normally given as a result (Not Charging used instead)

/sys/devices/platform/bq27000-battery.0/power_supply/bat/voltage_now

Battery voltage in uV, averaged. Obviously this depends on load and place on discharge curve, etc.

/sys/devices/platform/bq27000-battery.0/power_supply/bat/current_now

Current being drawn from battery (+ve) or pushed into battery during charging (-ve) in uA

/sys/devices/platform/bq27000-battery.0/power_supply/bat/charge_full

Coulomb Counter's estimate of the capacity of the battery measured in uA/h. It roughly means that if full, for one hour it could supply current at this rate. I saw 1197913 reported on mine, so it estimates it can provide 1.2A for an hour.

/sys/devices/platform/bq27000-battery.0/power_supply/bat/temp

This is the battery temperature reported by the Coulomb Counter chip that is part of the physical battery. It is in Celcius * 10, so 251 means 25.1 degrees.

/sys/devices/platform/bq27000-battery.0/power_supply/bat/technology

"Li-ion"

/sys/devices/platform/bq27000-battery.0/power_supply/bat/present

  • "0" the battery is absent, or not one with a Coulomb Counter
  • "1" the smart battery is present

/sys/devices/platform/bq27000-battery.0/power_supply/bat/time_to_empty_now

At current rate of discharge, estimate of how long we can run for. If battery is not discharging, it won't make an estimate and will return a magic value "3932100" meaning "no estimate". The coulomb counter averages the load and adjusts this value slowly to be its estimate of when we will blow chunks.

/sys/devices/platform/bq27000-battery.0/power_supply/bat/time_to_full_now

This estimates how long until we are fully charged, at current rate of charging, in seconds. If we are not charging, it gives the magic value "3932100" meaning "no estimate".

/sys/devices/platform/bq27000-battery.0/power_supply/bat/capacity

This is the remaining capacity of charge in the battery in percent. This is probably the most useful figure here.

/sys/devices/platform/bq27000-battery.0/power_supply/bat/online

  • "0" no charger is present
  • "1" A powered charger is present

PMU info

/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/chgmode

  • "enabled" means the charger is powered and willing to charge, although if you're full it might not be right now
  • "disabled" means there was no power or not enough power supplied to charge

Looks like it may also report "fast" here. OlivierBerger 07:47, 25 July 2008 (UTC)

/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/charger_type

Shows the detected charger type followed by the mode it is running in. So "charger 500mA mode 100mA" would be the case for non-enumerated USB connection, "charger 500mA mode 500mA" for after enumeration.

/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/force_usb_limit_dangerous

This is added to allow the PMU to be told to take more current than the default rules allow. Use it only if you know your charger source can handle it. For example

echo 500 > /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/force_usb_limit_dangerous

will allow PMU to draw 500mA even though you are connected to a dumb charger where the limit is normally held at 100mA, since it didn't enumerate the GTA02. You can give 0, 100, 500 or 1000 here, but make sure your charger source can handle what you give here, otherwise it can exceed what is safe for your charging source to provide.

/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/dump_regs

cat this to get a dump of the I2C registers for the PMU, useful for debugging power problems.

Bluetooth

/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/neo1973-pm-bt.0/power_on

  • Default "0" no power to Bluetooth device
  • set to "1" to enable power to Bluetooth

/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/neo1973-pm-bt.0/reset

  • Default to "0", no reset
  • set to "1" to reset BT (needed after powerup)

GPS

/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/neo1973-pm-gps.0/pwron

  • Default "0" GPS unpowered
  • set to "1" to power GPS section

Audio codec

/sys/devices/platform/soc-audio/codec_reg

cat this to get a dump of codec registers. Note they are 9-bit wide.

/sys/devices/platform/soc-audio/codec_reg_write

You can write to any codec register by echoing "<codec reg index in hex> <value in hex>" to here

Glamo

/sys/devices/platform/glamo3362.0/regs

cat this to get a dump of selected Glamo regs. You can write the regs by echoing "<glamo reg index in decimal> <value in decimal>" here.