Ben NanoNote kernel, status of critical patches ?
apelete at seketeli.net
Sun Dec 15 13:07:50 EST 2013
On 25-Apr-13, Lars-Peter Clausen wrote:
> On 04/22/2013 03:36 PM, Werner Almesberger wrote:
> > Using the (nearly) mainline kernel (net-next, closely tracking 3.9),
> > one can almost build a working kernel for the Ben. "Working" here
> > means that it gets to a shell on the serial console.
> > In the OpenWRT build process, a fairly large number of kernel
> > patches gets applied. They live here:
> > http://projects.qi-hardware.com/index.php/p/openwrt-xburst/source/tree/master/target/linux/xburst/patches-3.3
> > I found that just one of them is essential for getting the system to
> > boot:
> > 0020-qi_lb60-NAND-add-data-partition.patch
> > I also used
> > 0008-qi_lb60-Don-t-use-3-wire-spi-mode-for-the-display-fo.patch
> > to get fewer complaints from the kernel.
> > To boot a mainline kernel, I also had to override the boot command
> > line and add init=/etc/preinit (OpenWRT traditionally had a kernel
> > hack that added this directly to init/main.c, but I don't know if
> > they still use this. In any case, our u-boot setup seems to expect
> > the kernel to find preinit on its own, which it doesn't.)
> > With this, I get to a shell and can play with ATBEN and such.
> > Since I'm using a "headless" board, I don't know whether the screen
> > would actually work. The absence of
> > 0007-Add-ili8960-lcd-driver.patch
> > in mainline suggests that it won't. Another thing that's missing
> > is the USB driver,
> > 0002-Add-jz4740-udc-driver.patch
> > The rest of those 3.3 patches are mainly about NAND optimizations,
> > audio, power saving, wireless, and assorted minor changes.
> > I wonder what the status of all those patches is. Have they been
> > submitted towards upstream ? If yes, have they been accepted by
> > the respective subsystem maintainers and just haven't propagated
> > into Linus' tree yet ? Is someone tracking them ?
> Some are upstream, some are not. Those which are not upstream are primarily not
> upstream because they aren't quite ready yet and need some more love.
> E.g. There is already a driver for the usb gadget core in upstream (musb). But
> it is used on a couple of SoCs and somebody needs to sit down and write the
> jz4740 glue code.
I've been working on an musb glue layer for the jz4740 for some time
now, and it's finally ready.
Using the glue layer code, the musb driver is a replacement for the
jz4740_udc usb gadget driver we've been using until now.
Ethernet over USB is working fine to connect the Ben NanoNote to the
Internet, so the glue layer seems to be working pretty well in its
The patches are already available in the jz-3.12 branch, along with
the other Qi-Hardware community patches.
I also submitted them upstream, hoping to contribute to having full
upstream support for the Ben NanoNote.
Thanks everyone for the outstanding guidance and support while doing
this, it's been a lot of fun :).
More information about the discussion