diff mbox series

[1/2] dt-bindings: usb: dwc3: Add snps,host-vbus-glitches avoiding vbus glitch

Message ID 20240119213130.3147517-1-Frank.Li@nxp.com
State Superseded
Headers show
Series [1/2] dt-bindings: usb: dwc3: Add snps,host-vbus-glitches avoiding vbus glitch | expand

Commit Message

Frank Li Jan. 19, 2024, 9:31 p.m. UTC
From: Ran Wang <ran.wang_1@nxp.com>

When DWC3 is set to host mode by programming register DWC3_GCTL, VBUS
(or its control signal) will turn on immediately on related Root Hub
ports. Then the VBUS will be de-asserted for a little while during xhci
reset (conducted by xhci driver) for a little while and back to normal.

This VBUS glitch might cause some USB devices emuration fail if kernel
boot with them connected. One SW workaround which can fix this is to
program all PORTSC[PP] to 0 to turn off VBUS immediately after setting
host mode in DWC3 driver(per signal measurement result, it will be too
late to do it in xhci-plat.c or xhci.c).

Signed-off-by: Ran Wang <ran.wang_1@nxp.com>
Reviewed-by: Peter Chen <peter.chen@nxp.com>
Signed-off-by: Frank Li <Frank.Li@nxp.com>
---
 Documentation/devicetree/bindings/usb/snps,dwc3.yaml | 7 +++++++
 1 file changed, 7 insertions(+)

Comments

Thinh Nguyen Jan. 20, 2024, 12:51 a.m. UTC | #1
On Fri, Jan 19, 2024, Frank Li wrote:
> From: Ran Wang <ran.wang_1@nxp.com>
> 
> When DWC3 is set to host mode by programming register DWC3_GCTL, VBUS
> (or its control signal) will be turned on immediately on related Root Hub
> ports. Then, the VBUS is turned off for a little while(15us) when do xhci
> reset (conducted by xhci driver) and back to normal finally, we can
> observe a negative glitch of related signal happen.
> 
> This VBUS glitch might cause some USB devices enumeration fail if kernel
> boot with them connected. Such as LS1012AFWRY/LS1043ARDB/LX2160AQDS
> /LS1088ARDB with Kingston 16GB USB2.0/Kingston USB3.0/JetFlash Transcend
> 4GB USB2.0 drives. The fail cases include enumerated as full-speed device
> or report wrong device descriptor, etc.
> 
> One SW workaround which can fix this is by programing all xhci PORTSC[PP]
> to 0 to turn off VBUS immediately after setting host mode in DWC3 driver
> (per signal measurement result, it will be too late to do it in
> xhci-plat.c or xhci.c). Then, after xhci reset complete in xhci driver,
> PORTSC[PP]s' value will back to 1 automatically and VBUS on at that time,
> no glitch happen and normal enumeration process has no impact.
> 
> Signed-off-by: Ran Wang <ran.wang_1@nxp.com>
> Reviewed-by: Peter Chen <peter.chen@nxp.com>
> Signed-off-by: Frank Li <Frank.Li@nxp.com>
> ---
> 
> Notes:
>     Last review at June 06, 2019. Fixed all review comments and sent again.
>     
>     https://urldefense.com/v3/__https://lore.kernel.org/linux-kernel/AM5PR0402MB2865979E26017BC5937DBBA5F1170@AM5PR0402MB2865.eurprd04.prod.outlook.com/__;!!A4F2R9G_pg!bdutJWi1Tcz8SYscB7Mr96SWYMKIo8ElUKgEILFJfK3_60EbECQHXPBmJYAMMhNwQ4YrjxqMZWHdokXhHB6a$ 
> 
>  drivers/usb/dwc3/core.c |  2 ++
>  drivers/usb/dwc3/core.h |  2 ++
>  drivers/usb/dwc3/host.c | 46 +++++++++++++++++++++++++++++++++++++++++
>  3 files changed, 50 insertions(+)
> 
> diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
> index 3e55838c00014..a57adf0c11dd1 100644
> --- a/drivers/usb/dwc3/core.c
> +++ b/drivers/usb/dwc3/core.c
> @@ -1626,6 +1626,8 @@ static void dwc3_get_properties(struct dwc3 *dwc)
>  	dwc->dis_split_quirk = device_property_read_bool(dev,
>  				"snps,dis-split-quirk");
>  
> +	dwc->host_vbus_glitches = device_property_read_bool(dev, "snps,host-vbus-glitches");

This is a quirk. The property should be named with "quirk" subfix.
But do we need a new quirk? How many different platforms are affected?
If it's just 1 or 2, then just use compatible string.

> +
>  	dwc->lpm_nyet_threshold = lpm_nyet_threshold;
>  	dwc->tx_de_emphasis = tx_de_emphasis;
>  
> diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h
> index e3eea965e57bf..0269bacbbf6bd 100644
> --- a/drivers/usb/dwc3/core.h
> +++ b/drivers/usb/dwc3/core.h
> @@ -1135,6 +1135,7 @@ struct dwc3_scratchpad_array {
>   * @dis_split_quirk: set to disable split boundary.
>   * @wakeup_configured: set if the device is configured for remote wakeup.
>   * @suspended: set to track suspend event due to U3/L2.
> + * @host_vbus_glitches: set to avoid vbus glitch during xhci reset.

This is only applicable to the first xhci reset in its initialization
and not every xhci reset right? If so, please reword.

Also, please place it correspond to the order of the field.

>   * @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.
> @@ -1353,6 +1354,7 @@ struct dwc3 {
>  	unsigned		tx_de_emphasis:2;
>  
>  	unsigned		dis_metastability_quirk:1;
> +	unsigned		host_vbus_glitches:1;
>  
>  	unsigned		dis_split_quirk:1;
>  	unsigned		async_callbacks:1;
> diff --git a/drivers/usb/dwc3/host.c b/drivers/usb/dwc3/host.c
> index 61f57fe5bb783..af8903ee37c20 100644
> --- a/drivers/usb/dwc3/host.c
> +++ b/drivers/usb/dwc3/host.c
> @@ -11,6 +11,7 @@
>  #include <linux/of.h>
>  #include <linux/platform_device.h>
>  
> +#include "../host/xhci.h"

Let's not import the entire xhci.h. If we're going to share some macros
from xhci.h, please split them from xhci.h and perhaps create
xhci-ports.h for dwc3 to share.

>  #include "core.h"
>  
>  static void dwc3_host_fill_xhci_irq_res(struct dwc3 *dwc,
> @@ -28,6 +29,44 @@ static void dwc3_host_fill_xhci_irq_res(struct dwc3 *dwc,
>  		dwc->xhci_resources[1].name = name;
>  }
>  
> +#define XHCI_HCSPARAMS1		0x4
> +#define XHCI_PORTSC_BASE	0x400
> +
> +/*
> + * dwc3_power_off_all_roothub_ports - Power off all Root hub ports
> + * @dwc3: Pointer to our controller context structure
> + */
> +static void dwc3_power_off_all_roothub_ports(struct dwc3 *dwc)
> +{
> +	int i, port_num;
> +	u32 reg, op_regs_base, offset;
> +	void __iomem *xhci_regs;
> +
> +	/* xhci regs is not mapped yet, do it temperary here */
> +	if (dwc->xhci_resources[0].start) {
> +		xhci_regs = ioremap(dwc->xhci_resources[0].start,
> +				DWC3_XHCI_REGS_END);
> +		if (IS_ERR(xhci_regs)) {
> +			dev_err(dwc->dev, "Failed to ioremap xhci_regs\n");
> +			return;
> +		}
> +
> +		op_regs_base = HC_LENGTH(readl(xhci_regs));
> +		reg = readl(xhci_regs + XHCI_HCSPARAMS1);
> +		port_num = HCS_MAX_PORTS(reg);
> +
> +		for (i = 1; i <= port_num; i++) {
> +			offset = op_regs_base + XHCI_PORTSC_BASE + 0x10*(i-1);
> +			reg = readl(xhci_regs + offset);
> +			reg &= ~PORT_POWER;
> +			writel(reg, xhci_regs + offset);
> +		}
> +
> +		iounmap(xhci_regs);
> +	} else
> +		dev_err(dwc->dev, "xhci base reg invalid\n");
> +}
> +
>  static int dwc3_host_get_irq(struct dwc3 *dwc)
>  {
>  	struct platform_device	*dwc3_pdev = to_platform_device(dwc->dev);
> @@ -66,6 +105,13 @@ int dwc3_host_init(struct dwc3 *dwc)
>  	int			ret, irq;
>  	int			prop_idx = 0;
>  
> +	/*
> +	 * We have to power off all Root hub ports immediately after DWC3 set
> +	 * to host mode to avoid VBUS glitch happen when xhci get reset later.
> +	 */
> +	if (dwc->host_vbus_glitches)
> +		dwc3_power_off_all_roothub_ports(dwc);
> +

It's part of the dwc3_host_init(), but don't do this in
dwc3_host_get_irq(). Place it where it makes sense.

>  	irq = dwc3_host_get_irq(dwc);
>  	if (irq < 0)
>  		return irq;
> -- 
> 2.34.1
> 

Please run checkpatch.pl and fix them next time. Regarding
PARENTHESIS_ALIGNMENT or line continuation, it can be two indentations
or parenthesis aligned, whichever one makes it easier to read.


From checkpatch:

CHECK:PARENTHESIS_ALIGNMENT: Alignment should match open parenthesis
#119: FILE: drivers/usb/dwc3/host.c:48:
+		xhci_regs = ioremap(dwc->xhci_resources[0].start,
+				DWC3_XHCI_REGS_END);

CHECK:SPACING: spaces preferred around that '*' (ctx:VxV)
#130: FILE: drivers/usb/dwc3/host.c:59:
+			offset = op_regs_base + XHCI_PORTSC_BASE + 0x10*(i-1);
 			                                               ^

CHECK:SPACING: spaces preferred around that '-' (ctx:VxV)
#130: FILE: drivers/usb/dwc3/host.c:59:
+			offset = op_regs_base + XHCI_PORTSC_BASE + 0x10*(i-1);
 			                                                  ^

CHECK:BRACES: Unbalanced braces around else statement
#137: FILE: drivers/usb/dwc3/host.c:66:
+	} else

total: 0 errors, 0 warnings, 4 checks, 86 lines checked


Thanks,
Thinh
Thinh Nguyen Jan. 20, 2024, 12:55 a.m. UTC | #2
On Sat, Jan 20, 2024, Thinh Nguyen wrote:
> > +
> >  static int dwc3_host_get_irq(struct dwc3 *dwc)
> >  {
> >  	struct platform_device	*dwc3_pdev = to_platform_device(dwc->dev);
> > @@ -66,6 +105,13 @@ int dwc3_host_init(struct dwc3 *dwc)
> >  	int			ret, irq;
> >  	int			prop_idx = 0;
> >  
> > +	/*
> > +	 * We have to power off all Root hub ports immediately after DWC3 set
> > +	 * to host mode to avoid VBUS glitch happen when xhci get reset later.
> > +	 */
> > +	if (dwc->host_vbus_glitches)
> > +		dwc3_power_off_all_roothub_ports(dwc);
> > +
> 
> It's part of the dwc3_host_init(), but don't do this in
> dwc3_host_get_irq(). Place it where it makes sense.

Ignore this comment, I thought it's called from dwc3_host_get_irq(), but
it's not.

BR,
Thinh

> 
> >  	irq = dwc3_host_get_irq(dwc);
> >  	if (irq < 0)
> >  		return irq;
> > -- 
> > 2.34.1
> > 
>
Frank Li Jan. 20, 2024, 2 a.m. UTC | #3
On Sat, Jan 20, 2024 at 12:51:07AM +0000, Thinh Nguyen wrote:
> On Fri, Jan 19, 2024, Frank Li wrote:
> > From: Ran Wang <ran.wang_1@nxp.com>
> > 
> > When DWC3 is set to host mode by programming register DWC3_GCTL, VBUS
> > (or its control signal) will be turned on immediately on related Root Hub
> > ports. Then, the VBUS is turned off for a little while(15us) when do xhci
> > reset (conducted by xhci driver) and back to normal finally, we can
> > observe a negative glitch of related signal happen.
> > 
> > This VBUS glitch might cause some USB devices enumeration fail if kernel
> > boot with them connected. Such as LS1012AFWRY/LS1043ARDB/LX2160AQDS
> > /LS1088ARDB with Kingston 16GB USB2.0/Kingston USB3.0/JetFlash Transcend
> > 4GB USB2.0 drives. The fail cases include enumerated as full-speed device
> > or report wrong device descriptor, etc.
> > 
> > One SW workaround which can fix this is by programing all xhci PORTSC[PP]
> > to 0 to turn off VBUS immediately after setting host mode in DWC3 driver
> > (per signal measurement result, it will be too late to do it in
> > xhci-plat.c or xhci.c). Then, after xhci reset complete in xhci driver,
> > PORTSC[PP]s' value will back to 1 automatically and VBUS on at that time,
> > no glitch happen and normal enumeration process has no impact.
> > 
> > Signed-off-by: Ran Wang <ran.wang_1@nxp.com>
> > Reviewed-by: Peter Chen <peter.chen@nxp.com>
> > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > ---
> > 
> > Notes:
> >     Last review at June 06, 2019. Fixed all review comments and sent again.
> >     
> >     https://urldefense.com/v3/__https://lore.kernel.org/linux-kernel/AM5PR0402MB2865979E26017BC5937DBBA5F1170@AM5PR0402MB2865.eurprd04.prod.outlook.com/__;!!A4F2R9G_pg!bdutJWi1Tcz8SYscB7Mr96SWYMKIo8ElUKgEILFJfK3_60EbECQHXPBmJYAMMhNwQ4YrjxqMZWHdokXhHB6a$ 
> > 
> >  drivers/usb/dwc3/core.c |  2 ++
> >  drivers/usb/dwc3/core.h |  2 ++
> >  drivers/usb/dwc3/host.c | 46 +++++++++++++++++++++++++++++++++++++++++
> >  3 files changed, 50 insertions(+)
> > 
> > diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
> > index 3e55838c00014..a57adf0c11dd1 100644
> > --- a/drivers/usb/dwc3/core.c
> > +++ b/drivers/usb/dwc3/core.c
> > @@ -1626,6 +1626,8 @@ static void dwc3_get_properties(struct dwc3 *dwc)
> >  	dwc->dis_split_quirk = device_property_read_bool(dev,
> >  				"snps,dis-split-quirk");
> >  
> > +	dwc->host_vbus_glitches = device_property_read_bool(dev, "snps,host-vbus-glitches");
> 
> This is a quirk. The property should be named with "quirk" subfix.
> But do we need a new quirk? How many different platforms are affected?
> If it's just 1 or 2, then just use compatible string.

more than 2 platform. I think quirk is more flexible.

Frank
> 
> > +
> >  	dwc->lpm_nyet_threshold = lpm_nyet_threshold;
> >  	dwc->tx_de_emphasis = tx_de_emphasis;
> >  
> > diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h
> > index e3eea965e57bf..0269bacbbf6bd 100644
> > --- a/drivers/usb/dwc3/core.h
> > +++ b/drivers/usb/dwc3/core.h
> > @@ -1135,6 +1135,7 @@ struct dwc3_scratchpad_array {
> >   * @dis_split_quirk: set to disable split boundary.
> >   * @wakeup_configured: set if the device is configured for remote wakeup.
> >   * @suspended: set to track suspend event due to U3/L2.
> > + * @host_vbus_glitches: set to avoid vbus glitch during xhci reset.
> 
> This is only applicable to the first xhci reset in its initialization
> and not every xhci reset right? If so, please reword.
> 
> Also, please place it correspond to the order of the field.
> 
> >   * @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.
> > @@ -1353,6 +1354,7 @@ struct dwc3 {
> >  	unsigned		tx_de_emphasis:2;
> >  
> >  	unsigned		dis_metastability_quirk:1;
> > +	unsigned		host_vbus_glitches:1;
> >  
> >  	unsigned		dis_split_quirk:1;
> >  	unsigned		async_callbacks:1;
> > diff --git a/drivers/usb/dwc3/host.c b/drivers/usb/dwc3/host.c
> > index 61f57fe5bb783..af8903ee37c20 100644
> > --- a/drivers/usb/dwc3/host.c
> > +++ b/drivers/usb/dwc3/host.c
> > @@ -11,6 +11,7 @@
> >  #include <linux/of.h>
> >  #include <linux/platform_device.h>
> >  
> > +#include "../host/xhci.h"
> 
> Let's not import the entire xhci.h. If we're going to share some macros
> from xhci.h, please split them from xhci.h and perhaps create
> xhci-ports.h for dwc3 to share.
> 
> >  #include "core.h"
> >  
> >  static void dwc3_host_fill_xhci_irq_res(struct dwc3 *dwc,
> > @@ -28,6 +29,44 @@ static void dwc3_host_fill_xhci_irq_res(struct dwc3 *dwc,
> >  		dwc->xhci_resources[1].name = name;
> >  }
> >  
> > +#define XHCI_HCSPARAMS1		0x4
> > +#define XHCI_PORTSC_BASE	0x400
> > +
> > +/*
> > + * dwc3_power_off_all_roothub_ports - Power off all Root hub ports
> > + * @dwc3: Pointer to our controller context structure
> > + */
> > +static void dwc3_power_off_all_roothub_ports(struct dwc3 *dwc)
> > +{
> > +	int i, port_num;
> > +	u32 reg, op_regs_base, offset;
> > +	void __iomem *xhci_regs;
> > +
> > +	/* xhci regs is not mapped yet, do it temperary here */
> > +	if (dwc->xhci_resources[0].start) {
> > +		xhci_regs = ioremap(dwc->xhci_resources[0].start,
> > +				DWC3_XHCI_REGS_END);
> > +		if (IS_ERR(xhci_regs)) {
> > +			dev_err(dwc->dev, "Failed to ioremap xhci_regs\n");
> > +			return;
> > +		}
> > +
> > +		op_regs_base = HC_LENGTH(readl(xhci_regs));
> > +		reg = readl(xhci_regs + XHCI_HCSPARAMS1);
> > +		port_num = HCS_MAX_PORTS(reg);
> > +
> > +		for (i = 1; i <= port_num; i++) {
> > +			offset = op_regs_base + XHCI_PORTSC_BASE + 0x10*(i-1);
> > +			reg = readl(xhci_regs + offset);
> > +			reg &= ~PORT_POWER;
> > +			writel(reg, xhci_regs + offset);
> > +		}
> > +
> > +		iounmap(xhci_regs);
> > +	} else
> > +		dev_err(dwc->dev, "xhci base reg invalid\n");
> > +}
> > +
> >  static int dwc3_host_get_irq(struct dwc3 *dwc)
> >  {
> >  	struct platform_device	*dwc3_pdev = to_platform_device(dwc->dev);
> > @@ -66,6 +105,13 @@ int dwc3_host_init(struct dwc3 *dwc)
> >  	int			ret, irq;
> >  	int			prop_idx = 0;
> >  
> > +	/*
> > +	 * We have to power off all Root hub ports immediately after DWC3 set
> > +	 * to host mode to avoid VBUS glitch happen when xhci get reset later.
> > +	 */
> > +	if (dwc->host_vbus_glitches)
> > +		dwc3_power_off_all_roothub_ports(dwc);
> > +
> 
> It's part of the dwc3_host_init(), but don't do this in
> dwc3_host_get_irq(). Place it where it makes sense.
> 
> >  	irq = dwc3_host_get_irq(dwc);
> >  	if (irq < 0)
> >  		return irq;
> > -- 
> > 2.34.1
> > 
> 
> Please run checkpatch.pl and fix them next time. Regarding
> PARENTHESIS_ALIGNMENT or line continuation, it can be two indentations
> or parenthesis aligned, whichever one makes it easier to read.
> 
> 
> From checkpatch:
> 
> CHECK:PARENTHESIS_ALIGNMENT: Alignment should match open parenthesis
> #119: FILE: drivers/usb/dwc3/host.c:48:
> +		xhci_regs = ioremap(dwc->xhci_resources[0].start,
> +				DWC3_XHCI_REGS_END);
> 
> CHECK:SPACING: spaces preferred around that '*' (ctx:VxV)
> #130: FILE: drivers/usb/dwc3/host.c:59:
> +			offset = op_regs_base + XHCI_PORTSC_BASE + 0x10*(i-1);
>  			                                               ^
> 
> CHECK:SPACING: spaces preferred around that '-' (ctx:VxV)
> #130: FILE: drivers/usb/dwc3/host.c:59:
> +			offset = op_regs_base + XHCI_PORTSC_BASE + 0x10*(i-1);
>  			                                                  ^
> 
> CHECK:BRACES: Unbalanced braces around else statement
> #137: FILE: drivers/usb/dwc3/host.c:66:
> +	} else
> 

Sorry, I too trust original patch author. Will fix next version

Frank

> total: 0 errors, 0 warnings, 4 checks, 86 lines checked
> 
> 
> Thanks,
> Thinh
Frank Li Jan. 26, 2024, 4:06 a.m. UTC | #4
On Thu, Jan 25, 2024 at 05:43:22PM +0000, Conor Dooley wrote:
> On Wed, Jan 24, 2024 at 02:17:22PM -0500, Frank Li wrote:
> > On Wed, Jan 24, 2024 at 05:59:00PM +0000, Conor Dooley wrote:
> > > On Wed, Jan 24, 2024 at 12:47:14PM -0500, Frank Li wrote:
> > > > On Wed, Jan 24, 2024 at 05:36:42PM +0000, Conor Dooley wrote:
> > > > > On Fri, Jan 19, 2024 at 04:31:28PM -0500, Frank Li wrote:
> > > > > > From: Ran Wang <ran.wang_1@nxp.com>
> > > > > > 
> > > > > > When DWC3 is set to host mode by programming register DWC3_GCTL, VBUS
> > > > > > (or its control signal) will turn on immediately on related Root Hub
> > > > > > ports. Then the VBUS will be de-asserted for a little while during xhci
> > > > > > reset (conducted by xhci driver) for a little while and back to normal.
> > > > > > 
> > > > > > This VBUS glitch might cause some USB devices emuration fail if kernel
> > > > > > boot with them connected. One SW workaround which can fix this is to
> > > > > > program all PORTSC[PP] to 0 to turn off VBUS immediately after setting
> > > > > > host mode in DWC3 driver(per signal measurement result, it will be too
> > > > > > late to do it in xhci-plat.c or xhci.c).
> > > > > > 
> > > > > > Signed-off-by: Ran Wang <ran.wang_1@nxp.com>
> > > > > > Reviewed-by: Peter Chen <peter.chen@nxp.com>
> > > > > > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > > > > > ---
> > > > > >  Documentation/devicetree/bindings/usb/snps,dwc3.yaml | 7 +++++++
> > > > > >  1 file changed, 7 insertions(+)
> > > > > > 
> > > > > > diff --git a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
> > > > > > index 203a1eb66691f..dbf272b76e0b5 100644
> > > > > > --- a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
> > > > > > +++ b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
> > > > > > @@ -273,6 +273,13 @@ properties:
> > > > > >        with an external supply.
> > > > > >      type: boolean
> > > > > >  
> > > > > > +  snps,host-vbus-glitches:
> > > > > > +    description:
> > > > > > +      When set, power off all Root Hub ports immediately after
> > > > > > +      setting host mode to avoid vbus (negative) glitch happen in later
> > > > > > +      xhci reset. And the vbus will back to 5V automatically when reset done.
> > > 
> > > nit: "will return to"
> > > 
> > > > > > +    type: boolean
> > > > > 
> > > > > Why do we want to have a property for this at all? The commit message
> > > > > seems to describe a problem that's limited to specific configurations
> > > > > and appears to be somethng the driver should do unconditionally.
> > > > > 
> > > > > Could you explain why this cannot be done unconditionally please?
> > > > 
> > > > It depends on board design, not all system vbus can be controller by root
> > > > hub port. If it is always on, it will not trigger this issue.
> > > 
> > > Okay, that seems reasonable to have a property for. Can you add that
> > > info to the commit message please?
> > 
> > By the way, I sent v4 at
> > https://lore.kernel.org/imx/20240124152525.3910311-1-Frank.Li@nxp.com/T/#t
> 
> I see.
> 
> > How about add below sentence?
> > 
> > This was only happen when PORTSC[PP} can control vbus. Needn't set it if
> > vbus is always on.
> 
> "This can only happen when ... controls vbus, if vbus is always on, omit
> this property".
> 
> Just a wee grammatical nitpicking.
> 
> > > On another note, I like it when the property name explains why you would
> > > add it, rather than the thing it is trying to solve.
> > > Named after the disease, rather than the symptoms, if you get me. I
> > > tried to come up with a name here, but could not really suggest
> > > something good. If you can think of something, that'd be good, but don't
> > > stress it.
> > 
> > snps,host-vbus-glitches change to snps,host-vbus-glitches-quirk.
> 
> I don't think adding "quirk" moves the needle.

I think "quirk" is reasonable because it is workaround.

Frank

> 
> > How about use below description:
> > 
> > When set, power off all Root Hub ports immediately after
> > setting host mode to avoid vbus (negative) glitch happen in later
> > xhci reset. That may cause some USB devices emuration fail when kernel boot
> > with device connected and PORTSC[PP] control vbus in board desgin.
> 
> "When set, all root hub ports should be powered off immediately after
> enabling host mode, to avoid Vbus (negative) glitches that may happen
> during xHCI reset. These glitches can cause enumeration of some USB
> devices to fail when PORTSC[PP] controls Vbus. If Vbus is always on,
> omit this property."
> 
> How's that?
Rob Herring Jan. 30, 2024, 6:13 p.m. UTC | #5
On Wed, Jan 24, 2024 at 05:59:00PM +0000, Conor Dooley wrote:
> On Wed, Jan 24, 2024 at 12:47:14PM -0500, Frank Li wrote:
> > On Wed, Jan 24, 2024 at 05:36:42PM +0000, Conor Dooley wrote:
> > > On Fri, Jan 19, 2024 at 04:31:28PM -0500, Frank Li wrote:
> > > > From: Ran Wang <ran.wang_1@nxp.com>
> > > > 
> > > > When DWC3 is set to host mode by programming register DWC3_GCTL, VBUS
> > > > (or its control signal) will turn on immediately on related Root Hub
> > > > ports. Then the VBUS will be de-asserted for a little while during xhci
> > > > reset (conducted by xhci driver) for a little while and back to normal.
> > > > 
> > > > This VBUS glitch might cause some USB devices emuration fail if kernel
> > > > boot with them connected. One SW workaround which can fix this is to
> > > > program all PORTSC[PP] to 0 to turn off VBUS immediately after setting
> > > > host mode in DWC3 driver(per signal measurement result, it will be too
> > > > late to do it in xhci-plat.c or xhci.c).
> > > > 
> > > > Signed-off-by: Ran Wang <ran.wang_1@nxp.com>
> > > > Reviewed-by: Peter Chen <peter.chen@nxp.com>
> > > > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > > > ---
> > > >  Documentation/devicetree/bindings/usb/snps,dwc3.yaml | 7 +++++++
> > > >  1 file changed, 7 insertions(+)
> > > > 
> > > > diff --git a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
> > > > index 203a1eb66691f..dbf272b76e0b5 100644
> > > > --- a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
> > > > +++ b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
> > > > @@ -273,6 +273,13 @@ properties:
> > > >        with an external supply.
> > > >      type: boolean
> > > >  
> > > > +  snps,host-vbus-glitches:
> > > > +    description:
> > > > +      When set, power off all Root Hub ports immediately after
> > > > +      setting host mode to avoid vbus (negative) glitch happen in later
> > > > +      xhci reset. And the vbus will back to 5V automatically when reset done.
> 
> nit: "will return to"
> 
> > > > +    type: boolean
> > > 
> > > Why do we want to have a property for this at all? The commit message
> > > seems to describe a problem that's limited to specific configurations
> > > and appears to be somethng the driver should do unconditionally.
> > > 
> > > Could you explain why this cannot be done unconditionally please?
> > 
> > It depends on board design, not all system vbus can be controller by root
> > hub port. If it is always on, it will not trigger this issue.
> 
> Okay, that seems reasonable to have a property for. Can you add that
> info to the commit message please?

But if vbus is always on, then applying the work-around would be a NOP, 
right? So you could just apply this unconditionally.

Rob
Frank Li Jan. 30, 2024, 7:29 p.m. UTC | #6
On Tue, Jan 30, 2024 at 12:13:22PM -0600, Rob Herring wrote:
> On Wed, Jan 24, 2024 at 05:59:00PM +0000, Conor Dooley wrote:
> > On Wed, Jan 24, 2024 at 12:47:14PM -0500, Frank Li wrote:
> > > On Wed, Jan 24, 2024 at 05:36:42PM +0000, Conor Dooley wrote:
> > > > On Fri, Jan 19, 2024 at 04:31:28PM -0500, Frank Li wrote:
> > > > > From: Ran Wang <ran.wang_1@nxp.com>
> > > > > 
> > > > > When DWC3 is set to host mode by programming register DWC3_GCTL, VBUS
> > > > > (or its control signal) will turn on immediately on related Root Hub
> > > > > ports. Then the VBUS will be de-asserted for a little while during xhci
> > > > > reset (conducted by xhci driver) for a little while and back to normal.
> > > > > 
> > > > > This VBUS glitch might cause some USB devices emuration fail if kernel
> > > > > boot with them connected. One SW workaround which can fix this is to
> > > > > program all PORTSC[PP] to 0 to turn off VBUS immediately after setting
> > > > > host mode in DWC3 driver(per signal measurement result, it will be too
> > > > > late to do it in xhci-plat.c or xhci.c).
> > > > > 
> > > > > Signed-off-by: Ran Wang <ran.wang_1@nxp.com>
> > > > > Reviewed-by: Peter Chen <peter.chen@nxp.com>
> > > > > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > > > > ---
> > > > >  Documentation/devicetree/bindings/usb/snps,dwc3.yaml | 7 +++++++
> > > > >  1 file changed, 7 insertions(+)
> > > > > 
> > > > > diff --git a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
> > > > > index 203a1eb66691f..dbf272b76e0b5 100644
> > > > > --- a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
> > > > > +++ b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
> > > > > @@ -273,6 +273,13 @@ properties:
> > > > >        with an external supply.
> > > > >      type: boolean
> > > > >  
> > > > > +  snps,host-vbus-glitches:
> > > > > +    description:
> > > > > +      When set, power off all Root Hub ports immediately after
> > > > > +      setting host mode to avoid vbus (negative) glitch happen in later
> > > > > +      xhci reset. And the vbus will back to 5V automatically when reset done.
> > 
> > nit: "will return to"
> > 
> > > > > +    type: boolean
> > > > 
> > > > Why do we want to have a property for this at all? The commit message
> > > > seems to describe a problem that's limited to specific configurations
> > > > and appears to be somethng the driver should do unconditionally.
> > > > 
> > > > Could you explain why this cannot be done unconditionally please?
> > > 
> > > It depends on board design, not all system vbus can be controller by root
> > > hub port. If it is always on, it will not trigger this issue.
> > 
> > Okay, that seems reasonable to have a property for. Can you add that
> > info to the commit message please?
> 
> But if vbus is always on, then applying the work-around would be a NOP, 
> right? So you could just apply this unconditionally.

Supposed yes. But I have not confidence to apply this unconditionaly.
There are too much difference SOC and dwc3 version. Not sure if it brokes
something. I think it should apply workround as less as possible.

Frank

> 
> Rob
diff mbox series

Patch

diff --git a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
index 203a1eb66691f..dbf272b76e0b5 100644
--- a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
+++ b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
@@ -273,6 +273,13 @@  properties:
       with an external supply.
     type: boolean
 
+  snps,host-vbus-glitches:
+    description:
+      When set, power off all Root Hub ports immediately after
+      setting host mode to avoid vbus (negative) glitch happen in later
+      xhci reset. And the vbus will back to 5V automatically when reset done.
+    type: boolean
+
   snps,is-utmi-l1-suspend:
     description:
       True when DWC3 asserts output signal utmi_l1_suspend_n, false when