mbox series

[v5,0/3] usb: meson: fix shared reset control use

Message ID 20211212201844.114949-1-aouledameur@baylibre.com (mailing list archive)
Headers show
Series usb: meson: fix shared reset control use | expand

Message

Amjad Ouled-Ameur Dec. 12, 2021, 8:18 p.m. UTC
This patchset fixes a usb suspend warning seen on the libretech-cc by
using reset_control_rearm() call of the reset framework API. 
This call allows a reset consummer to release the reset line even when 
just triggered so that it may be triggered again by other reset
consummers.

reset_control_(de)assert() calls are called, in some meson usb drivers, 
on a shared reset line when reset_control_reset has been used. This is not
allowed by the reset framework.

Finally the meson usb drivers are updated to use this new call, which
solves the suspend issue addressed by the previous reverted 
commit 7a410953d1fb ("usb: dwc3: meson-g12a: fix shared reset control
use").

changes since v4:
- call reset_control_rearm() after clk_prepare_enable() fails

Amjad Ouled-Ameur (3):
  phy: amlogic: phy-meson-gxl-usb2: fix shared reset controller use
  phy: amlogic: meson8b-usb2: Use dev_err_probe()
  phy: amlogic: meson8b-usb2: fix shared reset control use

 drivers/phy/amlogic/phy-meson-gxl-usb2.c | 5 ++++-
 drivers/phy/amlogic/phy-meson8b-usb2.c   | 9 +++++++--
 2 files changed, 11 insertions(+), 3 deletions(-)

Comments

Neil Armstrong Dec. 13, 2021, 1:30 p.m. UTC | #1
Hi Amljad,

On 12/12/2021 21:18, Amjad Ouled-Ameur wrote:
> This patchset fixes a usb suspend warning seen on the libretech-cc by
> using reset_control_rearm() call of the reset framework API. 
> This call allows a reset consummer to release the reset line even when 
> just triggered so that it may be triggered again by other reset
> consummers.
> 
> reset_control_(de)assert() calls are called, in some meson usb drivers, 
> on a shared reset line when reset_control_reset has been used. This is not
> allowed by the reset framework.
> 
> Finally the meson usb drivers are updated to use this new call, which
> solves the suspend issue addressed by the previous reverted 
> commit 7a410953d1fb ("usb: dwc3: meson-g12a: fix shared reset control
> use").
> 
> changes since v4:
> - call reset_control_rearm() after clk_prepare_enable() fails
> 
> Amjad Ouled-Ameur (3):
>   phy: amlogic: phy-meson-gxl-usb2: fix shared reset controller use
>   phy: amlogic: meson8b-usb2: Use dev_err_probe()
>   phy: amlogic: meson8b-usb2: fix shared reset control use
> 
>  drivers/phy/amlogic/phy-meson-gxl-usb2.c | 5 ++++-
>  drivers/phy/amlogic/phy-meson8b-usb2.c   | 9 +++++++--
>  2 files changed, 11 insertions(+), 3 deletions(-)
> 

For whole serie:

Acked-by: Neil Armstrong <narmstrong@baylibre.com>

Thanks for fixing that !

Neil
Kevin Hilman Jan. 6, 2022, 7:06 p.m. UTC | #2
Hi Amjad,

Amjad Ouled-Ameur <aouledameur@baylibre.com> writes:

> This patchset fixes a usb suspend warning seen on the libretech-cc by
> using reset_control_rearm() call of the reset framework API. 
> This call allows a reset consummer to release the reset line even when 
> just triggered so that it may be triggered again by other reset
> consummers.
>
> reset_control_(de)assert() calls are called, in some meson usb drivers, 
> on a shared reset line when reset_control_reset has been used. This is not
> allowed by the reset framework.
>
> Finally the meson usb drivers are updated to use this new call, which
> solves the suspend issue addressed by the previous reverted 
> commit 7a410953d1fb ("usb: dwc3: meson-g12a: fix shared reset control
> use").
>
> changes since v4:
> - call reset_control_rearm() after clk_prepare_enable() fails
>
> Amjad Ouled-Ameur (3):
>   phy: amlogic: phy-meson-gxl-usb2: fix shared reset controller use
>   phy: amlogic: meson8b-usb2: Use dev_err_probe()
>   phy: amlogic: meson8b-usb2: fix shared reset control use
>
>  drivers/phy/amlogic/phy-meson-gxl-usb2.c | 5 ++++-
>  drivers/phy/amlogic/phy-meson8b-usb2.c   | 9 +++++++--
>  2 files changed, 11 insertions(+), 3 deletions(-)

The cover letter prefix is "usb: meson" but all the patches are for
drivers/phy.

Could you collect the reviewed-by etc tags and resend to the PHY
maintainers and linux-phy list?

Thanks,

Kevin