From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756354AbcJWJdi (ORCPT ); Sun, 23 Oct 2016 05:33:38 -0400 Received: from kozue.soulik.info ([108.61.200.231]:50822 "EHLO kozue.soulik.info" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752845AbcJWJdh (ORCPT ); Sun, 23 Oct 2016 05:33:37 -0400 Subject: Re: [PATCH v9 1/2] usb: dwc2: assert phy reset when waking up in rk3288 platform To: John Youn , "linux-usb@vger.kernel.org" References: <1476988627-14953-1-git-send-email-ayaka@soulik.info> <1476988627-14953-2-git-send-email-ayaka@soulik.info> <4ee5817d-0076-9974-8b0e-182a0241e1f5@synopsys.com> Cc: "kishon@ti.com" , "felipe.balbi@linux.intel.com" , "mark.rutland@arm.com" , "devicetree@vger.kernel.org" , "heiko@sntech.de" , "gregkh@linuxfoundation.org" , "linux-kernel@vger.kernel.org" , "robh+dt@kernel.org" , "randy.li@rock-chips.com" From: ayaka Message-ID: <922d7f12-a520-7c07-bc4c-887321177c6d@soulik.info> Date: Sun, 23 Oct 2016 17:33:23 +0800 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.3.0 MIME-Version: 1.0 In-Reply-To: <4ee5817d-0076-9974-8b0e-182a0241e1f5@synopsys.com> Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 10/22/2016 03:27 AM, John Youn wrote: > On 10/20/2016 11:38 AM, Randy Li wrote: >> On the rk3288 USB host-only port (the one that's not the OTG-enabled >> port) the PHY can get into a bad state when a wakeup is asserted (not >> just a wakeup from full system suspend but also a wakeup from >> autosuspend). >> >> We can get the PHY out of its bad state by asserting its "port reset", >> but unfortunately that seems to assert a reset onto the USB bus so it >> could confuse things if we don't actually deenumerate / reenumerate the >> device. >> >> We can also get the PHY out of its bad state by fully resetting it using >> the reset from the CRU (clock reset unit) in chip, which does a more full >> reset. The CRU-based reset appears to actually cause devices on the bus >> to be removed and reinserted, which fixes the problem (albeit in a hacky >> way). >> >> It's unfortunate that we need to do a full re-enumeration of devices at >> wakeup time, but this is better than alternative of letting the bus get >> wedged. >> >> Signed-off-by: Randy Li >> --- >> drivers/usb/dwc2/core.h | 1 + >> drivers/usb/dwc2/core_intr.c | 11 +++++++++++ >> drivers/usb/dwc2/platform.c | 9 +++++++++ >> 3 files changed, 21 insertions(+) >> >> diff --git a/drivers/usb/dwc2/core.h b/drivers/usb/dwc2/core.h >> index 2a21a04..e91ddbc 100644 >> --- a/drivers/usb/dwc2/core.h >> +++ b/drivers/usb/dwc2/core.h >> @@ -859,6 +859,7 @@ struct dwc2_hsotg { >> unsigned int ll_hw_enabled:1; >> >> struct phy *phy; >> + struct work_struct phy_rst_work; >> struct usb_phy *uphy; >> struct dwc2_hsotg_plat *plat; >> struct regulator_bulk_data supplies[ARRAY_SIZE(dwc2_hsotg_supply_names)]; >> diff --git a/drivers/usb/dwc2/core_intr.c b/drivers/usb/dwc2/core_intr.c >> index d85c5c9..c3d2168 100644 >> --- a/drivers/usb/dwc2/core_intr.c >> +++ b/drivers/usb/dwc2/core_intr.c >> @@ -345,6 +345,7 @@ static void dwc2_handle_session_req_intr(struct dwc2_hsotg *hsotg) >> static void dwc2_handle_wakeup_detected_intr(struct dwc2_hsotg *hsotg) >> { >> int ret; >> + struct device_node *np = hsotg->dev->of_node; >> >> /* Clear interrupt */ >> dwc2_writel(GINTSTS_WKUPINT, hsotg->regs + GINTSTS); >> @@ -379,6 +380,16 @@ static void dwc2_handle_wakeup_detected_intr(struct dwc2_hsotg *hsotg) >> /* Restart the Phy Clock */ >> pcgcctl &= ~PCGCTL_STOPPCLK; >> dwc2_writel(pcgcctl, hsotg->regs + PCGCTL); >> + >> + /* >> + * It is a quirk in Rockchip RK3288, causing by >> + * a hardware bug. This will propagate out and >> + * eventually we'll re-enumerate the device. >> + * Not great but the best we can do >> + */ >> + if (of_device_is_compatible(np, "rockchip,rk3288-usb")) >> + schedule_work(&hsotg->phy_rst_work); >> + >> mod_timer(&hsotg->wkp_timer, >> jiffies + msecs_to_jiffies(71)); >> } else { >> diff --git a/drivers/usb/dwc2/platform.c b/drivers/usb/dwc2/platform.c >> index 8e1728b..65953cf 100644 >> --- a/drivers/usb/dwc2/platform.c >> +++ b/drivers/usb/dwc2/platform.c >> @@ -366,6 +366,14 @@ int dwc2_lowlevel_hw_disable(struct dwc2_hsotg *hsotg) >> return ret; >> } >> >> +/* Only used to reset usb phy at interrupter runtime */ >> +static void dwc2_reset_phy_work(struct work_struct *data) >> +{ >> + struct dwc2_hsotg *hsotg = container_of(data, struct dwc2_hsotg, >> + phy_rst_work); >> + phy_reset(hsotg->phy); >> +} >> + >> static int dwc2_lowlevel_hw_init(struct dwc2_hsotg *hsotg) >> { >> int i, ret; >> @@ -410,6 +418,7 @@ static int dwc2_lowlevel_hw_init(struct dwc2_hsotg *hsotg) >> return ret; >> } >> } >> + INIT_WORK(&hsotg->phy_rst_work, dwc2_reset_phy_work); >> >> if (!hsotg->phy) { >> hsotg->uphy = devm_usb_get_phy(hsotg->dev, USB_PHY_TYPE_USB2); >> > Hi Randy, > > This fails compile if CONFIG_GENERIC_PHY is disabled. I think you need > to make a fix to your phy_reset patch first. In the last time, cac18ecb6f44b11bc303d7afbae3887b27938fa4 have not been merged, I though the [PATCH v8 1/3] phy: Add reset callback for not generic phy have been merged before that. when the rebase abandon it. Should re-send that patch? As the mainline have not been affected, could you arrange a squash for it? > > Regards, > John From mboxrd@z Thu Jan 1 00:00:00 1970 From: ayaka Subject: Re: [PATCH v9 1/2] usb: dwc2: assert phy reset when waking up in rk3288 platform Date: Sun, 23 Oct 2016 17:33:23 +0800 Message-ID: <922d7f12-a520-7c07-bc4c-887321177c6d@soulik.info> References: <1476988627-14953-1-git-send-email-ayaka@soulik.info> <1476988627-14953-2-git-send-email-ayaka@soulik.info> <4ee5817d-0076-9974-8b0e-182a0241e1f5@synopsys.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <4ee5817d-0076-9974-8b0e-182a0241e1f5-HKixBCOQz3hWk0Htik3J/w@public.gmane.org> Sender: devicetree-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: John Youn , "linux-usb-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" Cc: "kishon-l0cyMroinI0@public.gmane.org" , "felipe.balbi-VuQAYsv1563Yd54FQh9/CA@public.gmane.org" , "mark.rutland-5wv7dgnIgG8@public.gmane.org" , "devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "heiko-4mtYJXux2i+zQB+pC5nmwQ@public.gmane.org" , "gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org" , "linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org" , "randy.li-TNX95d0MmH7DzftRWevZcw@public.gmane.org" List-Id: devicetree@vger.kernel.org On 10/22/2016 03:27 AM, John Youn wrote: > On 10/20/2016 11:38 AM, Randy Li wrote: >> On the rk3288 USB host-only port (the one that's not the OTG-enabled >> port) the PHY can get into a bad state when a wakeup is asserted (not >> just a wakeup from full system suspend but also a wakeup from >> autosuspend). >> >> We can get the PHY out of its bad state by asserting its "port reset", >> but unfortunately that seems to assert a reset onto the USB bus so it >> could confuse things if we don't actually deenumerate / reenumerate the >> device. >> >> We can also get the PHY out of its bad state by fully resetting it using >> the reset from the CRU (clock reset unit) in chip, which does a more full >> reset. The CRU-based reset appears to actually cause devices on the bus >> to be removed and reinserted, which fixes the problem (albeit in a hacky >> way). >> >> It's unfortunate that we need to do a full re-enumeration of devices at >> wakeup time, but this is better than alternative of letting the bus get >> wedged. >> >> Signed-off-by: Randy Li >> --- >> drivers/usb/dwc2/core.h | 1 + >> drivers/usb/dwc2/core_intr.c | 11 +++++++++++ >> drivers/usb/dwc2/platform.c | 9 +++++++++ >> 3 files changed, 21 insertions(+) >> >> diff --git a/drivers/usb/dwc2/core.h b/drivers/usb/dwc2/core.h >> index 2a21a04..e91ddbc 100644 >> --- a/drivers/usb/dwc2/core.h >> +++ b/drivers/usb/dwc2/core.h >> @@ -859,6 +859,7 @@ struct dwc2_hsotg { >> unsigned int ll_hw_enabled:1; >> >> struct phy *phy; >> + struct work_struct phy_rst_work; >> struct usb_phy *uphy; >> struct dwc2_hsotg_plat *plat; >> struct regulator_bulk_data supplies[ARRAY_SIZE(dwc2_hsotg_supply_names)]; >> diff --git a/drivers/usb/dwc2/core_intr.c b/drivers/usb/dwc2/core_intr.c >> index d85c5c9..c3d2168 100644 >> --- a/drivers/usb/dwc2/core_intr.c >> +++ b/drivers/usb/dwc2/core_intr.c >> @@ -345,6 +345,7 @@ static void dwc2_handle_session_req_intr(struct dwc2_hsotg *hsotg) >> static void dwc2_handle_wakeup_detected_intr(struct dwc2_hsotg *hsotg) >> { >> int ret; >> + struct device_node *np = hsotg->dev->of_node; >> >> /* Clear interrupt */ >> dwc2_writel(GINTSTS_WKUPINT, hsotg->regs + GINTSTS); >> @@ -379,6 +380,16 @@ static void dwc2_handle_wakeup_detected_intr(struct dwc2_hsotg *hsotg) >> /* Restart the Phy Clock */ >> pcgcctl &= ~PCGCTL_STOPPCLK; >> dwc2_writel(pcgcctl, hsotg->regs + PCGCTL); >> + >> + /* >> + * It is a quirk in Rockchip RK3288, causing by >> + * a hardware bug. This will propagate out and >> + * eventually we'll re-enumerate the device. >> + * Not great but the best we can do >> + */ >> + if (of_device_is_compatible(np, "rockchip,rk3288-usb")) >> + schedule_work(&hsotg->phy_rst_work); >> + >> mod_timer(&hsotg->wkp_timer, >> jiffies + msecs_to_jiffies(71)); >> } else { >> diff --git a/drivers/usb/dwc2/platform.c b/drivers/usb/dwc2/platform.c >> index 8e1728b..65953cf 100644 >> --- a/drivers/usb/dwc2/platform.c >> +++ b/drivers/usb/dwc2/platform.c >> @@ -366,6 +366,14 @@ int dwc2_lowlevel_hw_disable(struct dwc2_hsotg *hsotg) >> return ret; >> } >> >> +/* Only used to reset usb phy at interrupter runtime */ >> +static void dwc2_reset_phy_work(struct work_struct *data) >> +{ >> + struct dwc2_hsotg *hsotg = container_of(data, struct dwc2_hsotg, >> + phy_rst_work); >> + phy_reset(hsotg->phy); >> +} >> + >> static int dwc2_lowlevel_hw_init(struct dwc2_hsotg *hsotg) >> { >> int i, ret; >> @@ -410,6 +418,7 @@ static int dwc2_lowlevel_hw_init(struct dwc2_hsotg *hsotg) >> return ret; >> } >> } >> + INIT_WORK(&hsotg->phy_rst_work, dwc2_reset_phy_work); >> >> if (!hsotg->phy) { >> hsotg->uphy = devm_usb_get_phy(hsotg->dev, USB_PHY_TYPE_USB2); >> > Hi Randy, > > This fails compile if CONFIG_GENERIC_PHY is disabled. I think you need > to make a fix to your phy_reset patch first. In the last time, cac18ecb6f44b11bc303d7afbae3887b27938fa4 have not been merged, I though the [PATCH v8 1/3] phy: Add reset callback for not generic phy have been merged before that. when the rebase abandon it. Should re-send that patch? As the mainline have not been affected, could you arrange a squash for it? > > Regards, > John -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html