[PATCH] usb: dwc3: core: Fix system suspend on TI AM62 platforms

[Roger Quadros]

On 08/10/2024 23:57, Thinh Nguyen wrote:
> Hi,
> 
> On Mon, Oct 07, 2024, Roger Quadros wrote:
>> Hi,
>>
>> On 05/10/2024 04:04, Thinh Nguyen wrote:
>>> Hi,
>>>
>>> On Tue, Oct 01, 2024, Roger Quadros wrote:
>>>> Since commit 6d735722063a ("usb: dwc3: core: Prevent phy suspend during init"),
>>>> system suspend is broken on AM62 TI platforms.
>>>>
>>>> Before that commit, both DWC3_GUSB3PIPECTL_SUSPHY and DWC3_GUSB2PHYCFG_SUSPHY
>>>> bits (hence forth called 2 SUSPHY bits) were being set during core
>>>> initialization and even during core re-initialization after a system
>>>> suspend/resume.
>>>>
>>>> These bits are required to be set for system suspend/resume to work correctly
>>>> on AM62 platforms.
>>>>
>>>> Since that commit, the 2 SUSPHY bits are not set for DEVICE/OTG mode if gadget
>>>> driver is not loaded and started.
>>>> For Host mode, the 2 SUSPHY bits are set before the first system suspend but
>>>> get cleared at system resume during core re-init and are never set again.
>>>>
>>>> This patch resovles these two issues by ensuring the 2 SUSPHY bits are set
>>>> before system suspend and restored to the original state during system resume.
>>>>
>>>> Cc: stable at vger.kernel.org # v6.9+
>>>> Fixes: 6d735722063a ("usb: dwc3: core: Prevent phy suspend during init")
>>>> Link: https://urldefense.com/v3/__https://lore.kernel.org/all/1519dbe7-73b6-4afc-bfe3-23f4f75d772f@kernel.org/__;!!A4F2R9G_pg!ahChm4MaKd6VGYqbnM4X1_pY_jqavYDv5HvPFbmicKuhvFsBwlEFi1xO5itGuHmfjbRuUSzReJISf5-1gXpr$ 
>>>> Signed-off-by: Roger Quadros <rogerq at kernel.org>
>>>> ---
>>>>  drivers/usb/dwc3/core.c | 16 ++++++++++++++++
>>>>  drivers/usb/dwc3/core.h |  2 ++
>>>>  2 files changed, 18 insertions(+)
>>>>
>>>> diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
>>>> index 9eb085f359ce..1233922d4d54 100644
>>>> --- a/drivers/usb/dwc3/core.c
>>>> +++ b/drivers/usb/dwc3/core.c
>>>> @@ -2336,6 +2336,9 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>>>  	u32 reg;
>>>>  	int i;
>>>>  
>>>> +	dwc->susphy_state = !!(dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0)) &
>>>> +			    DWC3_GUSB2PHYCFG_SUSPHY);
>>>> +
>>>>  	switch (dwc->current_dr_role) {
>>>>  	case DWC3_GCTL_PRTCAP_DEVICE:
>>>>  		if (pm_runtime_suspended(dwc->dev))
>>>> @@ -2387,6 +2390,11 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>>>  		break;
>>>>  	}
>>>>  
>>>> +	if (!PMSG_IS_AUTO(msg)) {
>>>> +		if (!dwc->susphy_state)
>>>> +			dwc3_enable_susphy(dwc, true);
>>>> +	}
>>>> +
>>>>  	return 0;
>>>>  }
>>>>  
>>>> @@ -2454,6 +2462,14 @@ static int dwc3_resume_common(struct dwc3 *dwc, pm_message_t msg)
>>>>  		break;
>>>>  	}
>>>>  
>>>> +	if (!PMSG_IS_AUTO(msg)) {
>>>> +		/* dwc3_core_init_for_resume() disables SUSPHY so just handle
>>>> +		 * the enable case
>>>> +		 */
>>>
>>> Can we note that this is a particular behavior needed for AM62 here?
>>> And can we use this comment style:
>>>
>>> /*
>>>  * xxxxx
>>>  * xxxxx
>>>  */
>>
>> OK.
>>
>>>
>>>
>>>> +		if (dwc->susphy_state)
>>>
>>> Shouldn't we check for if (!dwc->susphy_state) and clear the susphy
>>> bits?
>>
>> In that case it would have already been cleared so no need to check
>> and clear again.
>>
>>>
>>>> +			dwc3_enable_susphy(dwc, true);
>>>
>>> The dwc3_enable_susphy() set and clear both GUSB3PIPECTL_SUSPHY and
>>> GUSB2PHYCFG_SUSPHY, perhaps we should split that function out so we can
>>> only need to set for GUSB2PHYCFG_SUSPHY since you only track for that.
>>
>> As  dwc3_enable_susphy() sets and clears both GUSB3PIPECTL_SUSPHY and
>> GUSB2PHYCFG_SUSPHY together it doesn't really help to track both
>> separately, but just complicates things.
> 
> Then we should check if either GUSB2PHYCFG_SUSPHY or GUSB3PIPECTL_SUSPHY
> is set, then apply this.

Yes. I will do this.
> 
>>
>>>
>>>> +	}
>>>> +
>>>>  	return 0;
>>>>  }
>>>>  
>>>> diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h
>>>> index c71240e8f7c7..b2ed5aba4c72 100644
>>>> --- a/drivers/usb/dwc3/core.h
>>>> +++ b/drivers/usb/dwc3/core.h
>>>> @@ -1150,6 +1150,7 @@ struct dwc3_scratchpad_array {
>>>>   * @sys_wakeup: set if the device may do system wakeup.
>>>>   * @wakeup_configured: set if the device is configured for remote wakeup.
>>>>   * @suspended: set to track suspend event due to U3/L2.
>>>> + * @susphy_state: state of DWC3_GUSB2PHYCFG_SUSPHY before PM suspend.
>>>>   * @imod_interval: set the interrupt moderation interval in 250ns
>>>>   *			increments or 0 to disable.
>>>>   * @max_cfg_eps: current max number of IN eps used across all USB configs.
>>>> @@ -1382,6 +1383,7 @@ struct dwc3 {
>>>>  	unsigned		sys_wakeup:1;
>>>>  	unsigned		wakeup_configured:1;
>>>>  	unsigned		suspended:1;
>>>> +	unsigned		susphy_state:1;
>>>>  
>>>>  	u16			imod_interval;
>>>>  
>>>>
>>>> ---
>>>> base-commit: 9852d85ec9d492ebef56dc5f229416c925758edc
>>>> change-id: 20240923-am62-lpm-usb-f420917bd707
>>>>
>>>> Best regards,
>>>> -- 
>>>> Roger Quadros <rogerq at kernel.org>
>>>>
>>>
>>> <rant/>
>>> While reviewing your change, I see that we misuse the
>>> dis_u2_susphy_quirk to make this property a conditional thing during
>>> suspend and resume for certain platform. That bugs me because we can't
>>> easily change it without the reported hardware to test.
>>> </rant>
>>
>> No it is not conditional. if dis_u2_susphy_quirk or dis_u3_susphy_quirk
>> is set then we never enable the respective U2/U3 SUSPHY bit.
>>
> 
> I'm not referring to your change. I was referring to this in particular:
> bcb128777af5 ("usb: dwc3: core: Suspend PHYs on runtime suspend in host mode")

I get it now.

-- 
cheers,
-roger