[PATCH v3 0/5] usb: musb: da8xx: Add DMA support

[PATCH v3 0/5] usb: musb: da8xx: Add DMA support

[Alexandre Bailon]

This series update MUSB driver to add DMA support to DA8xx.
It should be applied on top of
"[PATCH v3 0/3] usb: musb: cppi41: Add a way to manage DMA irq" but
"[PATCH v3 0/3] dmaengine: cppi41: Add dma support to da8xx" and
"[PATCH] arm: davinci: Make the usb20 clock available to PM runtime"
are required to make it work.

Changes in v3:
- Remove PM runtime callbacks.
  I have update arch/arm/mach-davinci/pm_domain.c to let PM runtime control
  the usb20 clock.
- Only use PM runtime sync operation.

Changes in v2:
- Clock and IRQ management has been moved to MUSB DA8xx glue
  (was in CPPI 4.1 driver)
- I have added a partial support PM runtime. The goal was to use PM
  runtime to manage clock of MUSB and CPPI 4.1 (they use the same clock).
- CPPI 4.1 is now achild of MUSB DA8xx glue.

Alexandre Bailon (5):
  usb: musb: da8xx: Remove CPPI 3.0 quirk and methods
  usb: musb: Use shared irq
  usb: musb: Add support of CPPI 4.1 DMA controller to DA8xx
  usb: musb: musb_cppi41: Workaround DMA stall issue during teardown
  usb: musb: da8xx: Add a primary support of PM runtime

 drivers/usb/musb/Kconfig       |  4 +--
 drivers/usb/musb/da8xx.c       | 60 +++++++++++++++++++++++++++---------------
 drivers/usb/musb/musb_core.c   |  2 +-
 drivers/usb/musb/musb_core.h   |  1 +
 drivers/usb/musb/musb_cppi41.c |  4 +++
 5 files changed, 47 insertions(+), 24 deletions(-)

-- 
2.10.2

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH v3 1/5] usb: musb: da8xx: Remove CPPI 3.0 quirk and methods

[Alexandre Bailon]

DA8xx driver is registering and using the CPPI 3.0 DMA controller but
actually, the DA8xx has a CPPI 4.1 DMA controller.
Remove the CPPI 3.0 quirk and methods.

Fixes: f8e9f34f80a2 ("usb: musb: Fix up DMA related macros")
Fixes: 7f6283ed6fe8 ("usb: musb: Set up function pointers for DMA")
Signed-off-by: Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>
Acked-by: Sergei Shtylyov <sergei.shtylyov-M4DtvfQ/ZS1MRgGoP+s0PdBPR1lH4CV8@public.gmane.org>
Acked-by: Tony Lindgren <tony-4v6yS6AI5VpBDgjK7y7TUQ@public.gmane.org>
---
 drivers/usb/musb/da8xx.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c
index e89708d..cd3d763 100644
--- a/drivers/usb/musb/da8xx.c
+++ b/drivers/usb/musb/da8xx.c
@@ -458,15 +458,11 @@ static inline u8 get_vbus_power(struct device *dev)
 }
 
 static const struct musb_platform_ops da8xx_ops = {
-	.quirks		= MUSB_DMA_CPPI | MUSB_INDEXED_EP,
+	.quirks		= MUSB_INDEXED_EP,
 	.init		= da8xx_musb_init,
 	.exit		= da8xx_musb_exit,
 
 	.fifo_mode	= 2,
-#ifdef CONFIG_USB_TI_CPPI_DMA
-	.dma_init	= cppi_dma_controller_create,
-	.dma_exit	= cppi_dma_controller_destroy,
-#endif
 	.enable		= da8xx_musb_enable,
 	.disable	= da8xx_musb_disable,
 
-- 
2.10.2

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH v3 2/5] usb: musb: Use shared irq

[Alexandre Bailon]

In the DA8xx, USB and CPPI 4.1 are sharing the same interrupt line.
Update the driver to request a shared irq.

Signed-off-by: Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>
---
 drivers/usb/musb/musb_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/usb/musb/musb_core.c b/drivers/usb/musb/musb_core.c
index fca288bb..cf40adf 100644
--- a/drivers/usb/musb/musb_core.c
+++ b/drivers/usb/musb/musb_core.c
@@ -2329,7 +2329,7 @@ musb_init_controller(struct device *dev, int nIrq, void __iomem *ctrl)
 	setup_timer(&musb->otg_timer, musb_otg_timer_func, (unsigned long) musb);
 
 	/* attach to the IRQ */
-	if (request_irq(nIrq, musb->isr, 0, dev_name(dev), musb)) {
+	if (request_irq(nIrq, musb->isr, IRQF_SHARED, dev_name(dev), musb)) {
 		dev_err(dev, "request_irq %d failed!\n", nIrq);
 		status = -ENODEV;
 		goto fail3;
-- 
2.10.2

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH v3 3/5] usb: musb: Add support of CPPI 4.1 DMA controller to DA8xx

[Alexandre Bailon]

Currently, only the PIO mode is supported.
This add support of CPPI 4.1 to DA8xx.
As the In DA8xx the CPPI 4.1 DMA is a part of the USB.
Create the CPPI 4.1 device as a child of USB.

Signed-off-by: Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>
---
 drivers/usb/musb/Kconfig |  4 ++--
 drivers/usb/musb/da8xx.c | 35 ++++++++++++++++++++++++++++++++++-
 2 files changed, 36 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/musb/Kconfig b/drivers/usb/musb/Kconfig
index 72a2a50..5506a9c 100644
--- a/drivers/usb/musb/Kconfig
+++ b/drivers/usb/musb/Kconfig
@@ -160,8 +160,8 @@ config USB_TI_CPPI_DMA
 	  Enable DMA transfers when TI CPPI DMA is available.
 
 config USB_TI_CPPI41_DMA
-	bool 'TI CPPI 4.1 (AM335x)'
-	depends on ARCH_OMAP && DMADEVICES
+	bool 'TI CPPI 4.1'
+	depends on (ARCH_OMAP || ARCH_DAVINCI_DA8XX) && DMADEVICES
 	select TI_CPPI41
 
 config USB_TUSB_OMAP_DMA
diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c
index cd3d763..5f50a78 100644
--- a/drivers/usb/musb/da8xx.c
+++ b/drivers/usb/musb/da8xx.c
@@ -33,6 +33,7 @@
 #include <linux/clk.h>
 #include <linux/err.h>
 #include <linux/io.h>
+#include <linux/of_platform.h>
 #include <linux/phy/phy.h>
 #include <linux/platform_device.h>
 #include <linux/dma-mapping.h>
@@ -457,12 +458,40 @@ static inline u8 get_vbus_power(struct device *dev)
 	return current_uA / 1000 / 2;
 }
 
+#ifdef CONFIG_USB_TI_CPPI41_DM
+static void da8xx_dma_controller_callback(struct dma_controller *c)
+{
+	struct musb *musb = c->musb;
+	void __iomem *reg_base = musb->ctrl_base;
+
+	musb_writel(reg_base, DA8XX_USB_END_OF_INTR_REG, 0);
+}
+
+static struct dma_controller *
+da8xx_dma_controller_create(struct musb *musb, void __iomem *base)
+{
+	struct dma_controller *controller;
+
+	controller = cppi41_dma_controller_create(musb, base);
+	if (IS_ERR_OR_NULL(controller))
+		return controller;
+
+	controller->dma_callback = da8xx_dma_controller_callback;
+
+	return controller;
+}
+#endif
+
 static const struct musb_platform_ops da8xx_ops = {
-	.quirks		= MUSB_INDEXED_EP,
+	.quirks		= MUSB_INDEXED_EP | MUSB_DMA_CPPI41,
 	.init		= da8xx_musb_init,
 	.exit		= da8xx_musb_exit,
 
 	.fifo_mode	= 2,
+#ifdef CONFIG_USB_TI_CPPI41_DMA
+	.dma_init	= da8xx_dma_controller_create,
+	.dma_exit	= cppi41_dma_controller_destroy,
+#endif
 	.enable		= da8xx_musb_enable,
 	.disable	= da8xx_musb_disable,
 
@@ -534,6 +563,10 @@ static int da8xx_probe(struct platform_device *pdev)
 	}
 	platform_set_drvdata(pdev, glue);
 
+	ret = of_platform_populate(pdev->dev.of_node, NULL, NULL, &pdev->dev);
+	if (ret)
+		return ret;
+
 	memset(musb_resources, 0x00, sizeof(*musb_resources) *
 			ARRAY_SIZE(musb_resources));
 
-- 
2.10.2

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH v3 3/5] usb: musb: Add support of CPPI 4.1 DMA controller to DA8xx

[Alexandre Bailon]

On 01/19/2017 03:08 PM, Alexandre Bailon wrote:
> Currently, only the PIO mode is supported.
> This add support of CPPI 4.1 to DA8xx.
> As the In DA8xx the CPPI 4.1 DMA is a part of the USB.
> Create the CPPI 4.1 device as a child of USB.
> 
> Signed-off-by: Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>
> ---
>  drivers/usb/musb/Kconfig |  4 ++--
>  drivers/usb/musb/da8xx.c | 35 ++++++++++++++++++++++++++++++++++-
>  2 files changed, 36 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/usb/musb/Kconfig b/drivers/usb/musb/Kconfig
> index 72a2a50..5506a9c 100644
> --- a/drivers/usb/musb/Kconfig
> +++ b/drivers/usb/musb/Kconfig
> @@ -160,8 +160,8 @@ config USB_TI_CPPI_DMA
>  	  Enable DMA transfers when TI CPPI DMA is available.
>  
>  config USB_TI_CPPI41_DMA
> -	bool 'TI CPPI 4.1 (AM335x)'
> -	depends on ARCH_OMAP && DMADEVICES
> +	bool 'TI CPPI 4.1'
> +	depends on (ARCH_OMAP || ARCH_DAVINCI_DA8XX) && DMADEVICES
>  	select TI_CPPI41
>  
>  config USB_TUSB_OMAP_DMA
> diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c
> index cd3d763..5f50a78 100644
> --- a/drivers/usb/musb/da8xx.c
> +++ b/drivers/usb/musb/da8xx.c
> @@ -33,6 +33,7 @@
>  #include <linux/clk.h>
>  #include <linux/err.h>
>  #include <linux/io.h>
> +#include <linux/of_platform.h>
>  #include <linux/phy/phy.h>
>  #include <linux/platform_device.h>
>  #include <linux/dma-mapping.h>
> @@ -457,12 +458,40 @@ static inline u8 get_vbus_power(struct device *dev)
>  	return current_uA / 1000 / 2;
>  }
>  
> +#ifdef CONFIG_USB_TI_CPPI41_DM
There is a typo here...
> +static void da8xx_dma_controller_callback(struct dma_controller *c)
> +{
> +	struct musb *musb = c->musb;
> +	void __iomem *reg_base = musb->ctrl_base;
> +
> +	musb_writel(reg_base, DA8XX_USB_END_OF_INTR_REG, 0);
> +}
> +
> +static struct dma_controller *
> +da8xx_dma_controller_create(struct musb *musb, void __iomem *base)
> +{
> +	struct dma_controller *controller;
> +
> +	controller = cppi41_dma_controller_create(musb, base);
> +	if (IS_ERR_OR_NULL(controller))
> +		return controller;
> +
> +	controller->dma_callback = da8xx_dma_controller_callback;
> +
> +	return controller;
> +}
> +#endif
> +
>  static const struct musb_platform_ops da8xx_ops = {
> -	.quirks		= MUSB_INDEXED_EP,
> +	.quirks		= MUSB_INDEXED_EP | MUSB_DMA_CPPI41,
>  	.init		= da8xx_musb_init,
>  	.exit		= da8xx_musb_exit,
>  
>  	.fifo_mode	= 2,
> +#ifdef CONFIG_USB_TI_CPPI41_DMA
> +	.dma_init	= da8xx_dma_controller_create,
> +	.dma_exit	= cppi41_dma_controller_destroy,
> +#endif
>  	.enable		= da8xx_musb_enable,
>  	.disable	= da8xx_musb_disable,
>  
> @@ -534,6 +563,10 @@ static int da8xx_probe(struct platform_device *pdev)
>  	}
>  	platform_set_drvdata(pdev, glue);
>  
> +	ret = of_platform_populate(pdev->dev.of_node, NULL, NULL, &pdev->dev);
> +	if (ret)
> +		return ret;
> +
>  	memset(musb_resources, 0x00, sizeof(*musb_resources) *
>  			ARRAY_SIZE(musb_resources));
>  
> 

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH v3 3/5] usb: musb: Add support of CPPI 4.1 DMA controller to DA8xx

[Sergei Shtylyov]

On 01/19/2017 05:08 PM, Alexandre Bailon wrote:

> Currently, only the PIO mode is supported.
> This add support of CPPI 4.1 to DA8xx.
> As the In DA8xx the CPPI 4.1 DMA is a part of the USB.
> Create the CPPI 4.1 device as a child of USB.
>
> Signed-off-by: Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>
> ---
>  drivers/usb/musb/Kconfig |  4 ++--
>  drivers/usb/musb/da8xx.c | 35 ++++++++++++++++++++++++++++++++++-
>  2 files changed, 36 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/usb/musb/Kconfig b/drivers/usb/musb/Kconfig
> index 72a2a50..5506a9c 100644
> --- a/drivers/usb/musb/Kconfig
> +++ b/drivers/usb/musb/Kconfig
> @@ -160,8 +160,8 @@ config USB_TI_CPPI_DMA
>  	  Enable DMA transfers when TI CPPI DMA is available.
>
>  config USB_TI_CPPI41_DMA
> -	bool 'TI CPPI 4.1 (AM335x)'
> -	depends on ARCH_OMAP && DMADEVICES
> +	bool 'TI CPPI 4.1'
> +	depends on (ARCH_OMAP || ARCH_DAVINCI_DA8XX) && DMADEVICES
>  	select TI_CPPI41
>
>  config USB_TUSB_OMAP_DMA
> diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c
> index cd3d763..5f50a78 100644
> --- a/drivers/usb/musb/da8xx.c
> +++ b/drivers/usb/musb/da8xx.c
[...]
> @@ -457,12 +458,40 @@ static inline u8 get_vbus_power(struct device *dev)
>  	return current_uA / 1000 / 2;
>  }
>
> +#ifdef CONFIG_USB_TI_CPPI41_DM

    CONFIG_USB_TI_CPPI41_DMA! :-)

[...]

MBR, Sergei

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH v3 4/5] usb: musb: musb_cppi41: Workaround DMA stall issue during teardown

[Alexandre Bailon]

The DMA may hung up if a teardown is initiated while an endpoint is still
active (Advisory 2.3.27 of DA8xx errata).
To workaround this issue, add a delay before to initiate the teardown.

Signed-off-by: Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>
---
 drivers/usb/musb/da8xx.c       | 2 +-
 drivers/usb/musb/musb_core.h   | 1 +
 drivers/usb/musb/musb_cppi41.c | 4 ++++
 3 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c
index 5f50a78..8c9850c 100644
--- a/drivers/usb/musb/da8xx.c
+++ b/drivers/usb/musb/da8xx.c
@@ -483,7 +483,7 @@ da8xx_dma_controller_create(struct musb *musb, void __iomem *base)
 #endif
 
 static const struct musb_platform_ops da8xx_ops = {
-	.quirks		= MUSB_INDEXED_EP | MUSB_DMA_CPPI41,
+	.quirks		= MUSB_INDEXED_EP | MUSB_DMA_CPPI41 | MUSB_DA8XX,
 	.init		= da8xx_musb_init,
 	.exit		= da8xx_musb_exit,
 
diff --git a/drivers/usb/musb/musb_core.h b/drivers/usb/musb/musb_core.h
index ade902e..d129278 100644
--- a/drivers/usb/musb/musb_core.h
+++ b/drivers/usb/musb/musb_core.h
@@ -172,6 +172,7 @@ struct musb_io;
  */
 struct musb_platform_ops {
 
+#define MUSB_DA8XX		BIT(7)
 #define MUSB_DMA_UX500		BIT(6)
 #define MUSB_DMA_CPPI41		BIT(5)
 #define MUSB_DMA_CPPI		BIT(4)
diff --git a/drivers/usb/musb/musb_cppi41.c b/drivers/usb/musb/musb_cppi41.c
index 1fe7eae..d371d05 100644
--- a/drivers/usb/musb/musb_cppi41.c
+++ b/drivers/usb/musb/musb_cppi41.c
@@ -554,6 +554,10 @@ static int cppi41_dma_channel_abort(struct dma_channel *channel)
 		}
 	}
 
+	/* DA8xx Advisory 2.3.27: wait 250 ms before to start the teardown */
+	if (musb->io.quirks & MUSB_DA8XX)
+		mdelay(250);
+
 	tdbit = 1 << cppi41_channel->port_num;
 	if (is_tx)
 		tdbit <<= 16;
-- 
2.10.2

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH v3 4/5] usb: musb: musb_cppi41: Workaround DMA stall issue during teardown

[Tony Lindgren]

* Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org> [170119 06:09]:
> The DMA may hung up if a teardown is initiated while an endpoint is still
> active (Advisory 2.3.27 of DA8xx errata).
> To workaround this issue, add a delay before to initiate the teardown.
> 
> Signed-off-by: Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>
> ---
>  drivers/usb/musb/da8xx.c       | 2 +-
>  drivers/usb/musb/musb_core.h   | 1 +
>  drivers/usb/musb/musb_cppi41.c | 4 ++++
>  3 files changed, 6 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c
> index 5f50a78..8c9850c 100644
> --- a/drivers/usb/musb/da8xx.c
> +++ b/drivers/usb/musb/da8xx.c
> @@ -483,7 +483,7 @@ da8xx_dma_controller_create(struct musb *musb, void __iomem *base)
>  #endif
>  
>  static const struct musb_platform_ops da8xx_ops = {
> -	.quirks		= MUSB_INDEXED_EP | MUSB_DMA_CPPI41,
> +	.quirks		= MUSB_INDEXED_EP | MUSB_DMA_CPPI41 | MUSB_DA8XX,
>  	.init		= da8xx_musb_init,
>  	.exit		= da8xx_musb_exit,
>  
> diff --git a/drivers/usb/musb/musb_core.h b/drivers/usb/musb/musb_core.h
> index ade902e..d129278 100644
> --- a/drivers/usb/musb/musb_core.h
> +++ b/drivers/usb/musb/musb_core.h
> @@ -172,6 +172,7 @@ struct musb_io;
>   */
>  struct musb_platform_ops {
>  
> +#define MUSB_DA8XX		BIT(7)
>  #define MUSB_DMA_UX500		BIT(6)
>  #define MUSB_DMA_CPPI41		BIT(5)
>  #define MUSB_DMA_CPPI		BIT(4)
> diff --git a/drivers/usb/musb/musb_cppi41.c b/drivers/usb/musb/musb_cppi41.c
> index 1fe7eae..d371d05 100644
> --- a/drivers/usb/musb/musb_cppi41.c
> +++ b/drivers/usb/musb/musb_cppi41.c
> @@ -554,6 +554,10 @@ static int cppi41_dma_channel_abort(struct dma_channel *channel)
>  		}
>  	}
>  
> +	/* DA8xx Advisory 2.3.27: wait 250 ms before to start the teardown */
> +	if (musb->io.quirks & MUSB_DA8XX)
> +		mdelay(250);
> +
>  	tdbit = 1 << cppi41_channel->port_num;
>  	if (is_tx)
>  		tdbit <<= 16;

How about replace the do while loop calling  dmaengine_terminate_all() with
deferred work doing that?

That way it's more generic as it seems there's no hurry doing that
after the musb registers are cleared. That would also allow getting rid of
the two current udelay() calls there that also seems nasty.

Regards,

Tony
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH v3 4/5] usb: musb: musb_cppi41: Workaround DMA stall issue during teardown

[Alexandre Bailon]

On 01/19/2017 05:56 PM, Tony Lindgren wrote:
> * Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org> [170119 06:09]:
>> The DMA may hung up if a teardown is initiated while an endpoint is still
>> active (Advisory 2.3.27 of DA8xx errata).
>> To workaround this issue, add a delay before to initiate the teardown.
>>
>> Signed-off-by: Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>
>> ---
>>  drivers/usb/musb/da8xx.c       | 2 +-
>>  drivers/usb/musb/musb_core.h   | 1 +
>>  drivers/usb/musb/musb_cppi41.c | 4 ++++
>>  3 files changed, 6 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c
>> index 5f50a78..8c9850c 100644
>> --- a/drivers/usb/musb/da8xx.c
>> +++ b/drivers/usb/musb/da8xx.c
>> @@ -483,7 +483,7 @@ da8xx_dma_controller_create(struct musb *musb, void __iomem *base)
>>  #endif
>>  
>>  static const struct musb_platform_ops da8xx_ops = {
>> -	.quirks		= MUSB_INDEXED_EP | MUSB_DMA_CPPI41,
>> +	.quirks		= MUSB_INDEXED_EP | MUSB_DMA_CPPI41 | MUSB_DA8XX,
>>  	.init		= da8xx_musb_init,
>>  	.exit		= da8xx_musb_exit,
>>  
>> diff --git a/drivers/usb/musb/musb_core.h b/drivers/usb/musb/musb_core.h
>> index ade902e..d129278 100644
>> --- a/drivers/usb/musb/musb_core.h
>> +++ b/drivers/usb/musb/musb_core.h
>> @@ -172,6 +172,7 @@ struct musb_io;
>>   */
>>  struct musb_platform_ops {
>>  
>> +#define MUSB_DA8XX		BIT(7)
>>  #define MUSB_DMA_UX500		BIT(6)
>>  #define MUSB_DMA_CPPI41		BIT(5)
>>  #define MUSB_DMA_CPPI		BIT(4)
>> diff --git a/drivers/usb/musb/musb_cppi41.c b/drivers/usb/musb/musb_cppi41.c
>> index 1fe7eae..d371d05 100644
>> --- a/drivers/usb/musb/musb_cppi41.c
>> +++ b/drivers/usb/musb/musb_cppi41.c
>> @@ -554,6 +554,10 @@ static int cppi41_dma_channel_abort(struct dma_channel *channel)
>>  		}
>>  	}
>>  
>> +	/* DA8xx Advisory 2.3.27: wait 250 ms before to start the teardown */
>> +	if (musb->io.quirks & MUSB_DA8XX)
>> +		mdelay(250);
>> +
>>  	tdbit = 1 << cppi41_channel->port_num;
>>  	if (is_tx)
>>  		tdbit <<= 16;
> 
> How about replace the do while loop calling  dmaengine_terminate_all() with
> deferred work doing that?
That would be great but I don't know how hard to would be to do it.
Right after the teardown, we may have to queue a new request.
But we must be sure that the teardown has completed before to start
a new transfer. And because it happen in atomic section, we must also
defer it.
> 
> That way it's more generic as it seems there's no hurry doing that
> after the musb registers are cleared. That would also allow getting rid of
> the two current udelay() calls there that also seems nasty.
> 
> Regards,
> 
> Tony
> 
Regards,
Alexandre
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH v3 4/5] usb: musb: musb_cppi41: Workaround DMA stall issue during teardown

[Bin Liu]

On Thu, Jan 19, 2017 at 08:56:46AM -0800, Tony Lindgren wrote:
> * Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org> [170119 06:09]:
> > The DMA may hung up if a teardown is initiated while an endpoint is still
> > active (Advisory 2.3.27 of DA8xx errata).
> > To workaround this issue, add a delay before to initiate the teardown.
> > 
> > Signed-off-by: Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>
> > ---
> >  drivers/usb/musb/da8xx.c       | 2 +-
> >  drivers/usb/musb/musb_core.h   | 1 +
> >  drivers/usb/musb/musb_cppi41.c | 4 ++++
> >  3 files changed, 6 insertions(+), 1 deletion(-)
> > 
> > diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c
> > index 5f50a78..8c9850c 100644
> > --- a/drivers/usb/musb/da8xx.c
> > +++ b/drivers/usb/musb/da8xx.c
> > @@ -483,7 +483,7 @@ da8xx_dma_controller_create(struct musb *musb, void __iomem *base)
> >  #endif
> >  
> >  static const struct musb_platform_ops da8xx_ops = {
> > -	.quirks		= MUSB_INDEXED_EP | MUSB_DMA_CPPI41,
> > +	.quirks		= MUSB_INDEXED_EP | MUSB_DMA_CPPI41 | MUSB_DA8XX,
> >  	.init		= da8xx_musb_init,
> >  	.exit		= da8xx_musb_exit,
> >  
> > diff --git a/drivers/usb/musb/musb_core.h b/drivers/usb/musb/musb_core.h
> > index ade902e..d129278 100644
> > --- a/drivers/usb/musb/musb_core.h
> > +++ b/drivers/usb/musb/musb_core.h
> > @@ -172,6 +172,7 @@ struct musb_io;
> >   */
> >  struct musb_platform_ops {
> >  
> > +#define MUSB_DA8XX		BIT(7)
> >  #define MUSB_DMA_UX500		BIT(6)
> >  #define MUSB_DMA_CPPI41		BIT(5)
> >  #define MUSB_DMA_CPPI		BIT(4)
> > diff --git a/drivers/usb/musb/musb_cppi41.c b/drivers/usb/musb/musb_cppi41.c
> > index 1fe7eae..d371d05 100644
> > --- a/drivers/usb/musb/musb_cppi41.c
> > +++ b/drivers/usb/musb/musb_cppi41.c
> > @@ -554,6 +554,10 @@ static int cppi41_dma_channel_abort(struct dma_channel *channel)
> >  		}
> >  	}
> >  
> > +	/* DA8xx Advisory 2.3.27: wait 250 ms before to start the teardown */
> > +	if (musb->io.quirks & MUSB_DA8XX)
> > +		mdelay(250);
> > +
> >  	tdbit = 1 << cppi41_channel->port_num;
> >  	if (is_tx)
> >  		tdbit <<= 16;
> 
> How about replace the do while loop calling  dmaengine_terminate_all() with
> deferred work doing that?

It would be great, but it seems a little hard to implement. If the
_abort is initiated by application due to transfer timeout, the
application might immediately retry the transfer, then the driver would
have to ensure the deferred work is done and the channel is free by that
time.

I guess we have to evaluate the impact of the do-while loop and the
performance impact of the deferred work to know if it is worth it.

> 
> That way it's more generic as it seems there's no hurry doing that
> after the musb registers are cleared. That would also allow getting rid of
> the two current udelay() calls there that also seems nasty.

I don't think replacing the do-while loop will allow to get rid of the
two udelay() calls, which are there as barriers for the register changes
- clearing MUSB_RXCSR_DMAENAB and flushing FIFO.

Regards,
-Bin.
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH v3 4/5] usb: musb: musb_cppi41: Workaround DMA stall issue during teardown

[Tony Lindgren]

* Bin Liu <b-liu-l0cyMroinI0@public.gmane.org> [170120 09:24]:
> On Thu, Jan 19, 2017 at 08:56:46AM -0800, Tony Lindgren wrote:
> > * Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org> [170119 06:09]:
> > > The DMA may hung up if a teardown is initiated while an endpoint is still
> > > active (Advisory 2.3.27 of DA8xx errata).
> > > To workaround this issue, add a delay before to initiate the teardown.
> > > 
> > > Signed-off-by: Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>
> > > ---
> > >  drivers/usb/musb/da8xx.c       | 2 +-
> > >  drivers/usb/musb/musb_core.h   | 1 +
> > >  drivers/usb/musb/musb_cppi41.c | 4 ++++
> > >  3 files changed, 6 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c
> > > index 5f50a78..8c9850c 100644
> > > --- a/drivers/usb/musb/da8xx.c
> > > +++ b/drivers/usb/musb/da8xx.c
> > > @@ -483,7 +483,7 @@ da8xx_dma_controller_create(struct musb *musb, void __iomem *base)
> > >  #endif
> > >  
> > >  static const struct musb_platform_ops da8xx_ops = {
> > > -	.quirks		= MUSB_INDEXED_EP | MUSB_DMA_CPPI41,
> > > +	.quirks		= MUSB_INDEXED_EP | MUSB_DMA_CPPI41 | MUSB_DA8XX,
> > >  	.init		= da8xx_musb_init,
> > >  	.exit		= da8xx_musb_exit,
> > >  
> > > diff --git a/drivers/usb/musb/musb_core.h b/drivers/usb/musb/musb_core.h
> > > index ade902e..d129278 100644
> > > --- a/drivers/usb/musb/musb_core.h
> > > +++ b/drivers/usb/musb/musb_core.h
> > > @@ -172,6 +172,7 @@ struct musb_io;
> > >   */
> > >  struct musb_platform_ops {
> > >  
> > > +#define MUSB_DA8XX		BIT(7)
> > >  #define MUSB_DMA_UX500		BIT(6)
> > >  #define MUSB_DMA_CPPI41		BIT(5)
> > >  #define MUSB_DMA_CPPI		BIT(4)
> > > diff --git a/drivers/usb/musb/musb_cppi41.c b/drivers/usb/musb/musb_cppi41.c
> > > index 1fe7eae..d371d05 100644
> > > --- a/drivers/usb/musb/musb_cppi41.c
> > > +++ b/drivers/usb/musb/musb_cppi41.c
> > > @@ -554,6 +554,10 @@ static int cppi41_dma_channel_abort(struct dma_channel *channel)
> > >  		}
> > >  	}
> > >  
> > > +	/* DA8xx Advisory 2.3.27: wait 250 ms before to start the teardown */
> > > +	if (musb->io.quirks & MUSB_DA8XX)
> > > +		mdelay(250);
> > > +
> > >  	tdbit = 1 << cppi41_channel->port_num;
> > >  	if (is_tx)
> > >  		tdbit <<= 16;
> > 
> > How about replace the do while loop calling  dmaengine_terminate_all() with
> > deferred work doing that?
> 
> It would be great, but it seems a little hard to implement. If the
> _abort is initiated by application due to transfer timeout, the
> application might immediately retry the transfer, then the driver would
> have to ensure the deferred work is done and the channel is free by that
> time.
> 
> I guess we have to evaluate the impact of the do-while loop and the
> performance impact of the deferred work to know if it is worth it.
> 
> > 
> > That way it's more generic as it seems there's no hurry doing that
> > after the musb registers are cleared. That would also allow getting rid of
> > the two current udelay() calls there that also seems nasty.
> 
> I don't think replacing the do-while loop will allow to get rid of the
> two udelay() calls, which are there as barriers for the register changes
> - clearing MUSB_RXCSR_DMAENAB and flushing FIFO.

OK fine with me.

Tony
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH v3 5/5] usb: musb: da8xx: Add a primary support of PM runtime

[Alexandre Bailon]

Currently, DA8xx doesn't support PM runtime.
In addition, the glue driver is managing the clock itself.
But the CPPI DMA needs to manage this clock too.
Add support to PM runtime and use the callback to enable / disable
the clock. And because the CPPI 4.1 is a child of Da8xx USB,
it will be able to enable / disable the clock by using PM runtime.

Signed-off-by: Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>
---
 drivers/usb/musb/da8xx.c | 23 ++++++-----------------
 1 file changed, 6 insertions(+), 17 deletions(-)

diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c
index 8c9850c..c4478c7 100644
--- a/drivers/usb/musb/da8xx.c
+++ b/drivers/usb/musb/da8xx.c
@@ -30,7 +30,6 @@
  */
 
 #include <linux/module.h>
-#include <linux/clk.h>
 #include <linux/err.h>
 #include <linux/io.h>
 #include <linux/of_platform.h>
@@ -86,7 +85,6 @@ struct da8xx_glue {
 	struct device		*dev;
 	struct platform_device	*musb;
 	struct platform_device	*usb_phy;
-	struct clk		*clk;
 	struct phy		*phy;
 };
 
@@ -377,11 +375,7 @@ static int da8xx_musb_init(struct musb *musb)
 
 	musb->mregs += DA8XX_MENTOR_CORE_OFFSET;
 
-	ret = clk_prepare_enable(glue->clk);
-	if (ret) {
-		dev_err(glue->dev, "failed to enable clock\n");
-		return ret;
-	}
+	pm_runtime_get_sync(musb->controller->parent);
 
 	/* Returns zero if e.g. not clocked */
 	rev = musb_readl(reg_base, DA8XX_USB_REVISION_REG);
@@ -424,7 +418,7 @@ static int da8xx_musb_init(struct musb *musb)
 err_phy_power_on:
 	phy_exit(glue->phy);
 fail:
-	clk_disable_unprepare(glue->clk);
+	pm_runtime_put_sync(musb->controller->parent);
 	return ret;
 }
 
@@ -436,7 +430,7 @@ static int da8xx_musb_exit(struct musb *musb)
 
 	phy_power_off(glue->phy);
 	phy_exit(glue->phy);
-	clk_disable_unprepare(glue->clk);
+	pm_runtime_put_sync(musb->controller->parent);
 
 	usb_put_phy(musb->xceiv);
 
@@ -519,7 +513,6 @@ static int da8xx_probe(struct platform_device *pdev)
 	struct musb_hdrc_platform_data	*pdata = dev_get_platdata(&pdev->dev);
 	struct da8xx_glue		*glue;
 	struct platform_device_info	pinfo;
-	struct clk			*clk;
 	struct device_node		*np = pdev->dev.of_node;
 	int				ret;
 
@@ -527,12 +520,6 @@ static int da8xx_probe(struct platform_device *pdev)
 	if (!glue)
 		return -ENOMEM;
 
-	clk = devm_clk_get(&pdev->dev, "usb20");
-	if (IS_ERR(clk)) {
-		dev_err(&pdev->dev, "failed to get clock\n");
-		return PTR_ERR(clk);
-	}
-
 	glue->phy = devm_phy_get(&pdev->dev, "usb-phy");
 	if (IS_ERR(glue->phy)) {
 		if (PTR_ERR(glue->phy) != -EPROBE_DEFER)
@@ -541,7 +528,6 @@ static int da8xx_probe(struct platform_device *pdev)
 	}
 
 	glue->dev			= &pdev->dev;
-	glue->clk			= clk;
 
 	if (IS_ENABLED(CONFIG_OF) && np) {
 		pdata = devm_kzalloc(&pdev->dev, sizeof(*pdata), GFP_KERNEL);
@@ -587,6 +573,8 @@ static int da8xx_probe(struct platform_device *pdev)
 	pinfo.data = pdata;
 	pinfo.size_data = sizeof(*pdata);
 
+	pm_runtime_enable(&pdev->dev);
+
 	glue->musb = platform_device_register_full(&pinfo);
 	ret = PTR_ERR_OR_ZERO(glue->musb);
 	if (ret) {
@@ -603,6 +591,7 @@ static int da8xx_remove(struct platform_device *pdev)
 
 	platform_device_unregister(glue->musb);
 	usb_phy_generic_unregister(glue->usb_phy);
+	pm_runtime_disable(&pdev->dev);
 
 	return 0;
 }
-- 
2.10.2

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH v3 5/5] usb: musb: da8xx: Add a primary support of PM runtime

[Sergei Shtylyov]

On 01/19/2017 05:08 PM, Alexandre Bailon wrote:

> Currently, DA8xx doesn't support PM runtime.
> In addition, the glue driver is managing the clock itself.
> But the CPPI DMA needs to manage this clock too.
> Add support to PM runtime and use the callback to enable / disable
> the clock.

    I think this sentence is stale now....

> And because the CPPI 4.1 is a child of Da8xx USB,
> it will be able to enable / disable the clock by using PM runtime.
>
> Signed-off-by: Alexandre Bailon <abailon-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>
> ---
>  drivers/usb/musb/da8xx.c | 23 ++++++-----------------
>  1 file changed, 6 insertions(+), 17 deletions(-)
>
> diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c
> index 8c9850c..c4478c7 100644
> --- a/drivers/usb/musb/da8xx.c
> +++ b/drivers/usb/musb/da8xx.c
[...]
> @@ -377,11 +375,7 @@ static int da8xx_musb_init(struct musb *musb)
>
>  	musb->mregs += DA8XX_MENTOR_CORE_OFFSET;
>
> -	ret = clk_prepare_enable(glue->clk);
> -	if (ret) {
> -		dev_err(glue->dev, "failed to enable clock\n");
> -		return ret;
> -	}
> +	pm_runtime_get_sync(musb->controller->parent);
>
>  	/* Returns zero if e.g. not clocked */
>  	rev = musb_readl(reg_base, DA8XX_USB_REVISION_REG);
> @@ -424,7 +418,7 @@ static int da8xx_musb_init(struct musb *musb)
>  err_phy_power_on:
>  	phy_exit(glue->phy);
>  fail:
> -	clk_disable_unprepare(glue->clk);
> +	pm_runtime_put_sync(musb->controller->parent);

    Not sure we need put_sync() here and below -- usually a simple put() is 
used...

[...]

MBR, Sergei

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH v3 5/5] usb: musb: da8xx: Add a primary support of PM runtime

[Sekhar Nori]

On Thursday 19 January 2017 10:12 PM, Sergei Shtylyov wrote:
> On 01/19/2017 05:08 PM, Alexandre Bailon wrote:
> 
>> Currently, DA8xx doesn't support PM runtime.
>> In addition, the glue driver is managing the clock itself.
>> But the CPPI DMA needs to manage this clock too.
>> Add support to PM runtime and use the callback to enable / disable
>> the clock.
> 
>    I think this sentence is stale now....

And please clarify what you mean by "DA8xx doesn't support PM runtime."
because we do have drivers using PM runtime being used on DA8xx.

Thanks,
Sekhar
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html