Hi Martin,
On Mon, 4 Feb 2019 at 19:36, Martin Blumenstingl
<martin.blumenstingl@xxxxxxxxxxxxxx> wrote:
>
> Hi Anand,
>
> On Tue, Jan 29, 2019 at 11:25 AM Anand Moon <linux.amoon@xxxxxxxxx> wrote:
> >
> > Override the dr_mode from "host" to "otg" for USB_HOST controller (usb0)
> > changes help usb_phy0->usb0 initialization of OTG bus port.
> >
> > Changes fix the below warning.
> >
> > [ 1.183681] dwc2 c9000000.usb: c9000000.usb supply vusb_d not found, using dummy regulator
> > [ 1.190627] dwc2 c9000000.usb: Linked as a consumer to regulator.0
> > [ 1.196696] dwc2 c9000000.usb: c9000000.usb supply vusb_a not found, using dummy regulator
> > [ 1.206094] phy phy-c0000000.phy.0: USB ID detect failed!
> > [ 1.210243] phy phy-c0000000.phy.0: phy poweron failed --> -22
> I wonder if the warning below are a side-effect of "phy poweron failed"...
As pointed by you we need the fix this issue earlier email.
unless we enable the phy and set the correct dr_mode to tune the phy
we cannot the tune and power on phy bus properly.
After this patch we get phy and power correctly registered.
# cat /sys/kernel/debug/regulator/regulator_summary
regulator use open bypass opmode voltage
current min max
---------------------------------------------------------------------------------------
regulator-dummy 5 4 0 unknown 0mV
0mA 0mV 0mV
c9100000.usb 1
0mA 0mV 0mV
c9100000.usb 1
0mA 0mV 0mV
c9000000.usb 1
0mA 0mV 0mV
c9000000.usb 1
0mA 0mV 0mV
P5V0 1 1 0 unknown 5000mV
0mA 5000mV 5000mV
USB_OTG_PWR 4 2 0 unknown 5000mV
0mA 5000mV 5000mV
phy-c0000020.phy.1 2
0mA 0mV 0mV
phy-c0000000.phy.0 2
0mA 0mV 0mV
>
> > [ 1.216072] WARNING: CPU: 0 PID: 78 at drivers/regulator/core.c:2054 _regulator_put.part.11+0xf0/0xf8
> > [ 1.225147] Modules linked in:
> > [ 1.228169] CPU: 0 PID: 78 Comm: kworker/0:1 Not tainted 5.0.0-rc4-xc2ml #1
> > [ 1.235066] Hardware name: Hardkernel ODROID-C2 (DT)
> > [ 1.239987] Workqueue: events deferred_probe_work_func
> > [ 1.245072] pstate: 80000005 (Nzcv daif -PAN -UAO)
> > [ 1.249816] pc : _regulator_put.part.11+0xf0/0xf8
> > [ 1.254473] lr : regulator_put+0x34/0x48
> > [ 1.258353] sp : ffff0000117b3b00
> > [ 1.261630] x29: ffff0000117b3b00 x28: 0000000000000000
> > [ 1.266891] x27: ffff80007ec1a7b8 x26: ffff000010eec280
> > [ 1.272153] x25: 0000000000000000 x24: 0000000000000009
> > [ 1.277414] x23: ffff0000117b3bc8 x22: ffff80007ecd3c00
> > [ 1.282675] x21: ffff0000111ad6c8 x20: ffff80007ecd3e00
> > [ 1.287936] x19: ffff80007ecd3e00 x18: 0000000000000124
> > [ 1.293198] x17: 0000000000000000 x16: 0000000000000000
> > [ 1.298459] x15: 0000000000000400 x14: 0000000000000400
> > [ 1.303720] x13: ff00000000000000 x12: ffffffffffffffff
> > [ 1.308981] x11: 0000000000000038 x10: 0000000000000040
> > [ 1.314243] x9 : ffff0000111c58b0 x8 : ffff0000111c58a8
> > [ 1.319504] x7 : ffff80006b055200 x6 : ffff80007edd2880
> > [ 1.324765] x5 : ffff80007f400248 x4 : ffff80007fb92b00
> > [ 1.330026] x3 : ffff0000112c9d58 x2 : ffff80007edd2880
> > [ 1.335288] x1 : 0000000000000000 x0 : 0000000000000001
> > [ 1.340549] Call trace:
> > [ 1.342966] _regulator_put.part.11+0xf0/0xf8
> > [ 1.347278] regulator_put+0x34/0x48
> > [ 1.350814] regulator_bulk_free+0x30/0x50
> > [ 1.354868] devm_regulator_bulk_release+0x18/0x20
> > [ 1.359612] release_nodes+0x1b0/0x220
> > [ 1.363320] devres_release_all+0x34/0x50
> > [ 1.367288] really_probe+0xec/0x290
> > [ 1.370824] driver_probe_device+0x54/0xe8
> > [ 1.374878] __device_attach_driver+0xb8/0xe8
> > [ 1.379190] bus_for_each_drv+0x78/0xc8
> > [ 1.382985] __device_attach+0xd4/0x130
> > [ 1.386780] device_initial_probe+0x10/0x18
> > [ 1.390920] bus_probe_device+0x90/0x98
> > [ 1.394715] deferred_probe_work_func+0x6c/0xa0
> > [ 1.399203] process_one_work+0x1e0/0x318
> > [ 1.403168] worker_thread+0x228/0x428
> > [ 1.406877] kthread+0x124/0x128
> > [ 1.410070] ret_from_fork+0x10/0x18
> > [ 1.413603] ---[ end trace 498aca4cb07e5377 ]---
> phy_power_on fails above with -EINVAL and then something else breaks
> (it seems like you're hitting the WARN_ON in drivers/regulator/core.c
> below the comment which states: "Docs say you must disable before
> calling regulator_put()")
Ok will look into this later.
> so this looks like a bug in either the dwc2 driver or the PHY
> framework. can you please report that to the dwc2 and PHY maintainers
> so it can be fixed independently?
>
>
> Regards
> Martin
Best Regards
-Anand
