Hi,
On Tue, Dec 08, 2009 at 06:31:56PM +0100, ext Greg KH wrote:
with e.g. RX-51 board, which uses a ulpi-connected transceiver
(no i2c control bus) and no irq line to cpu, we don't have
information on when to kick the charger detection. Moreover, we
have to keep e.g. musb (in RX-51) alive because it's the only
way to have access to the ulpi bus. Kicking the charger
detection also breaks usb communication during the time the
detection is ongoing.
How to tackle those problems ? Should we leave that to userland ?
Should we implement a "musb_power_supply.c" driver ?
How does the hardware designers of the latter issue expect it to work?
I can't imagine that loosing communication is a correct thing to have
happen here, can't they fix their design?
that's in usb battery charging spec [1]
when detecting charger, a small voltage is sourced on d- (or d+, can't
remember and out of office now). If it's a usb dedicated charger or
host/hub charger, the voltage sourced on data line is looped back on the
other data line.
It seems like a driver is going to be needed, as either way you need to
detect it from within the kernel, right?
the problem is when you have transceivers without a control bus other
than ulpi. You have to keep the link powered, which means the link will
see vbus and try to connect to the bus. But that shouldn't happen right
now because we are still trying to detect the charger. On RX-51 case, we
get vbus irq from twl4030-usb but the actual usb communication (data) is
routed through isp1707 connected to musb via ulpi. I know, it's a
horrible hw design but we couldn't get that changed because we needed
usb charger detection which isn't support by twl4030 or twl5030. Only
now in twl5031 we have support for usb bc 1.0 (only) on the usb
transceiver available on that pmic.
2. Preventing usb gadget controller from connecting to the usb bus while
usb charger detection is ongoing.
Just do the detection first before binding the driver to the device?
I wish it was that simple. And maybe it is, I'll try tomorrow and see
what happens if I disable data pullups and try to detect charger...
Hopefully all will work, but then we still have one problem:
on RX-51, we get vbus irq from pmic which doesn't reach musb driver at
all, so we wouldn't have information about when to kick the charger
detection unless we export some symbol from musb to twl4030-usb to be
called when vbus irq comes. That doesn't sound good, right ? Any ideas
for this case ?
[1] http://www.usb.org/developers/devclass_docs/batt_charging_1_1.zip
--
balbi
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
