Nokia CBUS code

Nokia CBUS code

[Michael Büsch]

Hi there,

I was wondering if somebody was working on mainline integration of the
cbus code.
That probably is a major effort, as it needs a rewrite in some parts,
but I'm certainly interested in getting it mainlined, because I'm
currently developing a driver on top of it.
It's a battery management driver. If somebody is interested, it can
be found here:
https://dev.openwrt.org/browser/trunk/target/linux/omap24xx/patches-2.6.37/900-n810-battery-management.patch
It currently doesn't support charging, however, almost all information
for the charging hardware is available. Registers and bits are known.
The only thing that's missing is the interpreter of the
calibration data. That's going to be some major effort, though,
because bme uses software floating point for that. That's currently
screwing my brains, but it should be doable.

-- 
Greetings Michael.

Re: Nokia CBUS code

[Felipe Balbi]

Hi,

On Tue, Feb 22, 2011 at 11:48:12AM +0100, Michael Büsch wrote:
> I was wondering if somebody was working on mainline integration of the
> cbus code.

just look at the cbus from on linux-omap tree:

Alexander Shishkin (1):
      cbus: fix comilation breakage

Felipe Balbi (85):
      cbus: tahvo-usb: make it build again
      cbus: switch to kzalloc
      cbus: move EXPORT_SYMBOL close to its exported symbols
      cbus: separate into several MODULE_AUTHOR entries
      cbus: move to platform_driver
      cbus: add platform_data to pass gpios
      cbus: convert u32 base to void __iomem *base
      cbus: NULL global variable on exit
      cbus: checkpatch.pl fix on cbus.c
      cbus: don't export the global cbus_host variable
      cbus: rely on gpiolib
      cbus: no ternary on return
      cbus: add read/write flag to cbus_transfer
      cbus: don't type case when issuing read/write
      cbus: fix a resource leakage
      cbus: move cbus_host definition to C source
      cbus: handle possible errors on cbus_send/receive_bit
      cbus: introduce cbus_send/receive_data wrappers
      cbus: add kerneldoc
      cbus: retu-wdt: fix compile breakage
      cbus: retu: fix compile breakage on retu-headset
      cbus: retu: split one MODULE_AUTHOR into several
      cbus: retu: don't assing ret inside the if ()
      cbus: retu: convert printk to dev_*
      cbus: retu: convert to a platform_driver
      cbus: tahvo: split MODULE_AUTHOR into several entries
      cbus: tahvo: move EXPORT_SYMBOL macros closer to functions
      cbus: tahvo: don't assign ret inside if ()
      cbus: tahvo: convert printk into dev_*
      cbus: tahvo: convert to platform_driver
      cbus: retu: allocate platform_device for Retu's children
      cbus: retu-pwrbutton: convert to platform_driver
      cbus: retu-headset: simplify module_init
      cbus: retu-rtc: remove platform_device code
      cbus: retu-rtc: convert to platform_driver
      cbus: retu-rtc: split MODULE_AUTHOR into several entries
      cbus: retu-rtc: get rid of globals
      cbus: retu-rtc: move retu_rtc_barrier up on source code
      cbus: retu-rtc: make checkpatch.pl happy
      cbus: retu-rtc: switch to rtc class device
      cbus: retu-wdt: remove the platform_device
      cbus: retu-wtd: convert to platform_driver
      cbus: retu-wdt: misc cleanup on retu-wdt driver
      cbus: fix compilation with current mainline
      cbus: switch kmalloc() + memset() to kzalloc()
      cbus: remove device_release completion
      cbus: retu: pass irq number via struct resource
      cbus: retu: avoid section mismatch
      cbus: tahvo: usb: convert to platform_driver
      cbus: tahvo: remove device_release
      cbus: tahvo: pass irq via struct resource
      cbus: tahvo: avoid section mismatch
      cbus: remove unneded includes
      cbus: retu: move platform_device to board file
      cbus: tahvo: move platform_device to board file
      cbus: tahvo-usb: move platform_device to board file
      cbus: retu: get rid of retu-user.c
      cbus: retu: give it a context structure
      cbus: retu: move module_* close to the matching symbol
      cbus: retu: cleanup error path
      cbus: retu: move to threaded IRQ and GENIRQ
      cbus: retu: headset: convert to threaded_irq
      cbus: retu-pwrbutton: convert to threaded irq
      cbus: retu-rtc: move to threaded irq
      cbus: retu-rtc: drop the reset_occurred flag
      cbus: Makefile: re-enable retu-wdt
      cbus: tahvo: drop tahvo-user
      cbus: retu: pass IRQ via struct resource
      cbus: retu: headset: grab IRQ via struct resource
      cbus: retu: pwrbutton: grab IRQ via struct resource
      cbus: retu: rtc: grab IRQ via struct resource
      cbus: retu: drop retu_get_status
      cbus: retu: replace BUG_ON with WARN
      cbus: retu: drop the unnecessary spinlock_t
      cbus: retu: drop unused PFX macro
      cbus: retu: use the devid from platform_data
      cbus: retu: introduce internal read/write functions
      cbus: retu: search and replace
      cbus: retu: pwrbutton: save device pointer on our structure
      cbus: retu: wdt: save dev in retu_wdt_dev
      cbus: retu: pass the child device pointer to all retu functions
      cbus: retu: headset: don't save pdev pointer
      cbus: retu: replace EXPORT_SYMBOL with EXPORT_SYMBOL_GPL
      cbus: retu: tabify retu initialization
      cbus: retu: set pm_power_off to NULL when removing retu

Francisco Alecrim (1):
      fix compile for tahvo-usb.c

Jarkko Nikula (1):
      cbus: Fix compile by converting ioctl calls to unlocked_ioctlcalls

Juha Yrjola (1):
      omap: Add drivers/cbus support

Tony Lindgren (11):
      omap: Search and replace headers to use plat
      cbus: Fix compile and don't try to use tag
      cbus: Make retu watchdog behave like a standard Linux watchdog
      cbus: Fix init on boards with no cbus
      cbus: Fix tahvo init without cbus
      cbus: Fix compile for 770
      cbus: Fix retu init order
      cbus: Disable retu_rtc_do_reset for now
      Revert "cbus: Disable retu_rtc_do_reset for now"
      cbus: Fix retu_rtc_do_reset
      cbus: Fix nested interrupts for retu


> That probably is a major effort, as it needs a rewrite in some parts,
> but I'm certainly interested in getting it mainlined, because I'm
> currently developing a driver on top of it.

We're almost there actually. Retu is in great shape and tahvo is a lot
simpler, but it still needs help. Probably porting the changes I made on
retu would be the way to go.

A small TODO would be to use dev_pm_ops on all driver's suspend/resume
functions, then move retu-headset.c to Jack framework and port all
changes to tahvo as well.

> It's a battery management driver. If somebody is interested, it can
> be found here:
> https://dev.openwrt.org/browser/trunk/target/linux/omap24xx/patches-2.6.37/900-n810-battery-management.patch

cool.

> It currently doesn't support charging, however, almost all information
> for the charging hardware is available. Registers and bits are known.
> The only thing that's missing is the interpreter of the
> calibration data. That's going to be some major effort, though,
> because bme uses software floating point for that. That's currently
> screwing my brains, but it should be doable.

heh, good luck and thanks for the initiative :-)

-- 
balbi
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Nokia CBUS code

[Michael Büsch]

On Tue, 2011-02-22 at 13:05 +0200, Felipe Balbi wrote: 
> > That probably is a major effort, as it needs a rewrite in some parts,
> > but I'm certainly interested in getting it mainlined, because I'm
> > currently developing a driver on top of it.
> 
> We're almost there actually. Retu is in great shape and tahvo is a lot
> simpler, but it still needs help. Probably porting the changes I made on
> retu would be the way to go.

Ok, that's pretty cool. I'll try to port this to OpenWRT, soon,
to check how it works.


On another issue: Did you try tahvo-usb recently?
I was trying it with the OpenWRT cbus codebase (which is some months
old) and I couldn't get it to work. It would always crash with
some weird NULL pointer accesses and stuff like this. I was wondering
if the omap-tree tahvo code would actually work on the USB device.

-- 
Greetings Michael.

Re: Nokia CBUS code

[Felipe Balbi]

On Tue, Feb 22, 2011 at 12:12:07PM +0100, Michael Büsch wrote:
> On Tue, 2011-02-22 at 13:05 +0200, Felipe Balbi wrote: 
> > > That probably is a major effort, as it needs a rewrite in some parts,
> > > but I'm certainly interested in getting it mainlined, because I'm
> > > currently developing a driver on top of it.
> > 
> > We're almost there actually. Retu is in great shape and tahvo is a lot
> > simpler, but it still needs help. Probably porting the changes I made on
> > retu would be the way to go.
> 
> Ok, that's pretty cool. I'll try to port this to OpenWRT, soon,
> to check how it works.
> 
> 
> On another issue: Did you try tahvo-usb recently?

Nope :-(

> I was trying it with the OpenWRT cbus codebase (which is some months
> old) and I couldn't get it to work. It would always crash with
> some weird NULL pointer accesses and stuff like this. I was wondering
> if the omap-tree tahvo code would actually work on the USB device.

It's been a long time since I tested that part of the code. My n810
actually died :-( so Tony has been testing my patches.

But hey, if you get a NULL pointer with linux-omap's code, then please
let us know about it. Give us whatever information you can give us and
we will work together on fixing it ;-)

-- 
balbi
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html