Hi,
Thanks for your reply.
On 09/09/16 09:42, Peter Chen wrote:
On Thu, Sep 08, 2016 at 10:00:22AM +0200, Fabien Lahoudere wrote:
Hi,
I have an imx53 based boards with an ULPI phy (smsc 3315) connected to usbh2
and usbh3 and encounter problem with patch
"usb: chipidea: coordinate usb phy initialization for different phy type".
In drivers/usb/chipidea/core.c:ci_usb_phy_init, if I call
hw_phymode_configure() before usb_phy_init(), the system freeze. If I call
usb_phy_init() before hw_phymode_configure(), everything works fine.
So I wonder if someone test ULPI phy? and if the following piece of patch
(that works in my case) could be acceptable :
Do you use upstream kernel? I wonder if current drivers/usb/phy/phy-ulpi.c
code can work well with DT?
I use next-20160905 and my patch works fine if I call usb_phy_init()
before hw_phymode_configure(). So my issue with the current kernel
appear before calling phy-ulpi functions.
Stephen Boyd posted ULPI support for chipidea recently, would you please
try if it works for you:
usb: chipidea: Add support for ULPI PHY bus
I try the patch and it did not works in my case.
When I call _ci_usb_phy_init() before hw_phymode_configure(). I have a
timeout returned by ulpi_write() in function ulpi_register(). So the
result is worst.
If I call _ci_usb_phy_init() after hw_phymode_configure(). The result is
still the same.
The system freeze in hw_phymode_configure() when doing :
hw_write(ci, OP_PORTSC, PORTSC_PTS(7) | PORTSC_PTW, portsc);
So I think that the register can be written if the phy is not powered
but don 't understand why.
If you want I can submit my patch to discuss about it.
Thanks
Fabien
--
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
