[PATCH 4/4] introduce-gta02-beachhead.patch

Andy Green andy at openmoko.com
Fri Feb 20 14:39:55 CET 2009


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Somebody in the thread at some point said:
| Hi Andy,
|
| just three comments:
|
| 1. Could you please hold this up three more days? I have some build
| system fixes that we need for mach-gta02, but I'm not able to send
| them before monday.

What does "build system fixes" mean, Kconfig stuff?  Don't forget we are
only talking about those four patches total, everything else we know and
love is not going anywhere yet.

| 2. Which revisions of GTA02 are actually in the wild? I'm asking
| because it would be nice to further the size of the header (and some
| switches). Btw. there are still some GTA02v1definitions in there...

I threw out everything that was commented as v1 or v2 only.  There are
apparently some people still using v3 and v4, but we only ever shipped v5+.

Personally, I only think we should take care about what we shipped.  But
we moved ahead by just nuking v1 and v2 so it's OK for now.  Practically
if v3 or v4 people had funny behaviour they would be on their own anyway.

| 3. What is the data flow when inserting USB? I'm asking because we
| startup the loader detection at two different places:
|
|> +static void gta02_pmu_event_callback(struct pcf50633 *pcf, int irq)
|> +{
|> +       if (irq == PCF50633_IRQ_USBINS) {
|> +               schedule_delayed_work(&gta02_charger_work,
              <<-HERE
|> +                               GTA02_CHARGER_CONFIGURE_TIMEOUT);
|> +               return;
|>  ...
|
| and
|
|> +static void gta02_udc_vbus_draw(unsigned int ma)
|> +{
|> ...
|> +       schedule_delayed_work(&gta02_charger_work,
             <<-HERE
|> +                               GTA02_CHARGER_CONFIGURE_TIMEOUT);
|> ...

These are very different animals, there's no confusion.

The first one comes when we accept PMU interrupt on noticing power has
arrived at USB socket.  Anything with appropriate voltage triggers this
and it means we can pull 100mA.

The second one occurs on logical enumeration of the USB device by the
host PC.  That happens after the dumb voltage is seen, and signals that
it is time to reassess how much current we can pull.  On enumeration, we
can pull 500mA.

- -Andy
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmesqsACgkQOjLpvpq7dMouVgCdGaE7gK0wE+0eJq2MGQ11Lf1e
C9wAn2BoOtGaFYkMUcVYFjrH0B2UUzuV
=5v0Y
-----END PGP SIGNATURE-----



More information about the openmoko-kernel mailing list