Skip to contentSkip to conversation

usb: dwc2: Force mode optimizations

PATCH New Felipe Balbi 1 participant 1 message Feb 15, 2018 USB

lore
Share & tools
lore
#0 lore raw

PATCH v2 2/49 usb: dwc2: Force mode optimizations

Quoted 24 lines · Grigor Tovmasyan

From: Vardan Mikayelyan <[email protected]>

If the dr_mode is USB_DR_MODE_OTG, forcing the mode is needed during driver probe to get the host and device specific HW parameters. Then we clear the force mode bits so that the core operates in OTG mode.

The force mode bits should not be touched at any other time during the driver lifetime and they should be preserved whenever the GUSBCFG register is written to. The force mode bit values will persist across soft resets of the core.

If the dr_mode is either USB_DR_MODE_HOST or USB_DR_MODE_PERIPHERAL, the force mode is set just once at probe to configure the core as either a host or peripheral.

Given the above, we no longer need any other reset delays, force delays, or any forced modes anywhere else in the driver. So replace all calls to dwc2_core_reset_and_force_dr_mode() with dwc2_core_reset() and remove all other unnecessary delays.

Also remove the dwc2_force_mode_if_needed() function since the "if needed" part is already taken care of by the polling in dwc2_force_mode().

Quoted 6 lines

Finally, remove all other calls to dwc2_clear_force_mode().

Only patch #1 applied to testing/next.

checking file drivers/usb/dwc2/core.c
checking file drivers/usb/dwc2/core.h
Hunk #1 succeeded at 1106 (offset 6 lines).
checking file drivers/usb/dwc2/hcd.c
Hunk #3 succeeded at 2261 (offset 5 lines).
checking file drivers/usb/dwc2/params.c
Hunk #1 succeeded at 579 (offset -1 lines).
Hunk #2 FAILED at 603.
1 out of 2 hunks FAILED
checking file drivers/usb/dwc2/platform.c
Hunk #1 succeeded at 425 (offset 10 lines).

please rebase