linux-i2c.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] i2c: mediatek: fix transfer error handling
@ 2015-07-28  3:38 Eddie Huang
       [not found] ` <1438054685-25214-1-git-send-email-eddie.huang-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
  0 siblings, 1 reply; 3+ messages in thread
From: Eddie Huang @ 2015-07-28  3:38 UTC (permalink / raw)
  To: Wolfram Sang
  Cc: srv_heupstream-NuS5LvNUpcJWk0Htik3J/w, Matthias Brugger,
	Liguo Zhang, Xudong Chen, Sascha Hauer,
	linux-i2c-u79uwXL29TY76Z2rM5mHXA,
	linux-kernel-u79uwXL29TY76Z2rM5mHXA,
	linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
	linux-mediatek-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r, Eddie Huang

From: Liguo Zhang <liguo.zhang-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>

Reset i2c dma engine in hw init function.
When occur i2c ack error, mtk_i2c_irq may is twice,
first is the ack error interrupt, then the complete interrupt,
so i2c->irq_stat need keep the two interrupt value, and only
call complete() for the complete interrupt.

Signed-off-by: Liguo Zhang <liguo.zhang-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
Signed-off-by: Eddie Huang <eddie.huang-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
---
 drivers/i2c/busses/i2c-mt65xx.c |   25 ++++++++++++++++++-------
 1 file changed, 18 insertions(+), 7 deletions(-)

diff --git a/drivers/i2c/busses/i2c-mt65xx.c b/drivers/i2c/busses/i2c-mt65xx.c
index 9920eef..57d11b7 100644
--- a/drivers/i2c/busses/i2c-mt65xx.c
+++ b/drivers/i2c/busses/i2c-mt65xx.c
@@ -59,6 +59,7 @@
 #define I2C_DMA_START_EN		0x0001
 #define I2C_DMA_INT_FLAG_NONE		0x0000
 #define I2C_DMA_CLR_FLAG		0x0000
+#define I2C_DMA_HARD_RST		0x0002
 
 #define I2C_DEFAULT_SPEED		100000	/* hz */
 #define MAX_FS_MODE_SPEED		400000
@@ -81,6 +82,7 @@ enum DMA_REGS_OFFSET {
 	OFFSET_INT_FLAG = 0x0,
 	OFFSET_INT_EN = 0x04,
 	OFFSET_EN = 0x08,
+	OFFSET_RST = 0x0c,
 	OFFSET_CON = 0x18,
 	OFFSET_TX_MEM_ADDR = 0x1c,
 	OFFSET_RX_MEM_ADDR = 0x20,
@@ -262,6 +264,10 @@ static void mtk_i2c_init_hw(struct mtk_i2c *i2c)
 		      I2C_CONTROL_CLK_EXT_EN | I2C_CONTROL_DMA_EN;
 	writew(control_reg, i2c->base + OFFSET_CONTROL);
 	writew(I2C_DELAY_LEN, i2c->base + OFFSET_DELAY_LEN);
+
+	writel(I2C_DMA_HARD_RST, i2c->pdmabase + OFFSET_RST);
+	udelay(50);
+	writel(I2C_DMA_CLR_FLAG, i2c->pdmabase + OFFSET_RST);
 }
 
 /*
@@ -550,16 +556,20 @@ err_exit:
 static irqreturn_t mtk_i2c_irq(int irqno, void *dev_id)
 {
 	struct mtk_i2c *i2c = dev_id;
-	u16 restart_flag = 0;
+	u16 intr_stat = 0;
 
-	if (i2c->dev_comp->auto_restart)
-		restart_flag = I2C_RS_TRANSFER;
+	intr_stat = readw(i2c->base + OFFSET_INTR_STAT);
+	writew(intr_stat, i2c->base + OFFSET_INTR_STAT);
 
-	i2c->irq_stat = readw(i2c->base + OFFSET_INTR_STAT);
-	writew(restart_flag | I2C_HS_NACKERR | I2C_ACKERR
-		| I2C_TRANSAC_COMP, i2c->base + OFFSET_INTR_STAT);
+	/*
+	 * when occurs i2c ack error, mtk_i2c_irq is called twice,
+	 * first is the ack error interrupt, then the complete interrupt,
+	 * i2c->irq_stat need keep the two interrupt value.
+	 */
+	i2c->irq_stat |= intr_stat;
 
-	complete(&i2c->msg_complete);
+	if (i2c->irq_stat & I2C_TRANSAC_COMP)
+		complete(&i2c->msg_complete);
 
 	return IRQ_HANDLED;
 }
@@ -729,3 +739,4 @@ module_platform_driver(mtk_i2c_driver);
 MODULE_LICENSE("GPL v2");
 MODULE_DESCRIPTION("MediaTek I2C Bus Driver");
 MODULE_AUTHOR("Xudong Chen <xudong.chen-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>");
+MODULE_AUTHOR("Liguo Zhang <liguo.zhang-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>");
-- 
1.7.9.5

^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH] i2c: mediatek: fix transfer error handling
       [not found] ` <1438054685-25214-1-git-send-email-eddie.huang-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
@ 2015-07-31 11:00   ` Wolfram Sang
  2015-08-04  6:13     ` Eddie Huang
  0 siblings, 1 reply; 3+ messages in thread
From: Wolfram Sang @ 2015-07-31 11:00 UTC (permalink / raw)
  To: Eddie Huang
  Cc: srv_heupstream-NuS5LvNUpcJWk0Htik3J/w, Matthias Brugger,
	Liguo Zhang, Xudong Chen, Sascha Hauer,
	linux-i2c-u79uwXL29TY76Z2rM5mHXA,
	linux-kernel-u79uwXL29TY76Z2rM5mHXA,
	linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
	linux-mediatek-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r

[-- Attachment #1: Type: text/plain, Size: 3244 bytes --]

On Tue, Jul 28, 2015 at 11:38:05AM +0800, Eddie Huang wrote:
> From: Liguo Zhang <liguo.zhang-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
> 
> Reset i2c dma engine in hw init function.
> When occur i2c ack error, mtk_i2c_irq may is twice,
> first is the ack error interrupt, then the complete interrupt,
> so i2c->irq_stat need keep the two interrupt value, and only
> call complete() for the complete interrupt.
> 
> Signed-off-by: Liguo Zhang <liguo.zhang-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
> Signed-off-by: Eddie Huang <eddie.huang-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>

Looks to me this patch needs to be split up into one patch per issue?
And doesn't it kill the auto_restart functionality? Sascha?

> ---
>  drivers/i2c/busses/i2c-mt65xx.c |   25 ++++++++++++++++++-------
>  1 file changed, 18 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/i2c/busses/i2c-mt65xx.c b/drivers/i2c/busses/i2c-mt65xx.c
> index 9920eef..57d11b7 100644
> --- a/drivers/i2c/busses/i2c-mt65xx.c
> +++ b/drivers/i2c/busses/i2c-mt65xx.c
> @@ -59,6 +59,7 @@
>  #define I2C_DMA_START_EN		0x0001
>  #define I2C_DMA_INT_FLAG_NONE		0x0000
>  #define I2C_DMA_CLR_FLAG		0x0000
> +#define I2C_DMA_HARD_RST		0x0002
>  
>  #define I2C_DEFAULT_SPEED		100000	/* hz */
>  #define MAX_FS_MODE_SPEED		400000
> @@ -81,6 +82,7 @@ enum DMA_REGS_OFFSET {
>  	OFFSET_INT_FLAG = 0x0,
>  	OFFSET_INT_EN = 0x04,
>  	OFFSET_EN = 0x08,
> +	OFFSET_RST = 0x0c,
>  	OFFSET_CON = 0x18,
>  	OFFSET_TX_MEM_ADDR = 0x1c,
>  	OFFSET_RX_MEM_ADDR = 0x20,
> @@ -262,6 +264,10 @@ static void mtk_i2c_init_hw(struct mtk_i2c *i2c)
>  		      I2C_CONTROL_CLK_EXT_EN | I2C_CONTROL_DMA_EN;
>  	writew(control_reg, i2c->base + OFFSET_CONTROL);
>  	writew(I2C_DELAY_LEN, i2c->base + OFFSET_DELAY_LEN);
> +
> +	writel(I2C_DMA_HARD_RST, i2c->pdmabase + OFFSET_RST);
> +	udelay(50);
> +	writel(I2C_DMA_CLR_FLAG, i2c->pdmabase + OFFSET_RST);
>  }
>  
>  /*
> @@ -550,16 +556,20 @@ err_exit:
>  static irqreturn_t mtk_i2c_irq(int irqno, void *dev_id)
>  {
>  	struct mtk_i2c *i2c = dev_id;
> -	u16 restart_flag = 0;
> +	u16 intr_stat = 0;
>  
> -	if (i2c->dev_comp->auto_restart)
> -		restart_flag = I2C_RS_TRANSFER;
> +	intr_stat = readw(i2c->base + OFFSET_INTR_STAT);
> +	writew(intr_stat, i2c->base + OFFSET_INTR_STAT);
>  
> -	i2c->irq_stat = readw(i2c->base + OFFSET_INTR_STAT);
> -	writew(restart_flag | I2C_HS_NACKERR | I2C_ACKERR
> -		| I2C_TRANSAC_COMP, i2c->base + OFFSET_INTR_STAT);
> +	/*
> +	 * when occurs i2c ack error, mtk_i2c_irq is called twice,
> +	 * first is the ack error interrupt, then the complete interrupt,
> +	 * i2c->irq_stat need keep the two interrupt value.
> +	 */
> +	i2c->irq_stat |= intr_stat;
>  
> -	complete(&i2c->msg_complete);
> +	if (i2c->irq_stat & I2C_TRANSAC_COMP)
> +		complete(&i2c->msg_complete);
>  
>  	return IRQ_HANDLED;
>  }
> @@ -729,3 +739,4 @@ module_platform_driver(mtk_i2c_driver);
>  MODULE_LICENSE("GPL v2");
>  MODULE_DESCRIPTION("MediaTek I2C Bus Driver");
>  MODULE_AUTHOR("Xudong Chen <xudong.chen-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>");
> +MODULE_AUTHOR("Liguo Zhang <liguo.zhang-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>");
> -- 
> 1.7.9.5
> 

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH] i2c: mediatek: fix transfer error handling
  2015-07-31 11:00   ` Wolfram Sang
@ 2015-08-04  6:13     ` Eddie Huang
  0 siblings, 0 replies; 3+ messages in thread
From: Eddie Huang @ 2015-08-04  6:13 UTC (permalink / raw)
  To: Wolfram Sang
  Cc: srv_heupstream, Matthias Brugger, Liguo Zhang, Xudong Chen,
	Sascha Hauer, linux-i2c, linux-kernel, linux-arm-kernel,
	linux-mediatek

On Fri, 2015-07-31 at 13:00 +0200, Wolfram Sang wrote:
> On Tue, Jul 28, 2015 at 11:38:05AM +0800, Eddie Huang wrote:
> > From: Liguo Zhang <liguo.zhang@mediatek.com>
> > 
> > Reset i2c dma engine in hw init function.
> > When occur i2c ack error, mtk_i2c_irq may is twice,
> > first is the ack error interrupt, then the complete interrupt,
> > so i2c->irq_stat need keep the two interrupt value, and only
> > call complete() for the complete interrupt.
> > 
> > Signed-off-by: Liguo Zhang <liguo.zhang@mediatek.com>
> > Signed-off-by: Eddie Huang <eddie.huang@mediatek.com>
> 
> Looks to me this patch needs to be split up into one patch per issue?
OK, I can split 

> And doesn't it kill the auto_restart functionality? Sascha?

No. restart_flag already set in mtk_i2c_do_transfer() function.It is not
necessary check restart_flag again in mtk_i2c_irq(). It is simpler to
just read status bit and write back to clear interrupt status.

Eddie
Thanks

> 
> > ---
> >  drivers/i2c/busses/i2c-mt65xx.c |   25 ++++++++++++++++++-------
> >  1 file changed, 18 insertions(+), 7 deletions(-)
> > 
> > diff --git a/drivers/i2c/busses/i2c-mt65xx.c b/drivers/i2c/busses/i2c-mt65xx.c
> > index 9920eef..57d11b7 100644
> > --- a/drivers/i2c/busses/i2c-mt65xx.c
> > +++ b/drivers/i2c/busses/i2c-mt65xx.c
> > @@ -59,6 +59,7 @@
> >  #define I2C_DMA_START_EN		0x0001
> >  #define I2C_DMA_INT_FLAG_NONE		0x0000
> >  #define I2C_DMA_CLR_FLAG		0x0000
> > +#define I2C_DMA_HARD_RST		0x0002
> >  
> >  #define I2C_DEFAULT_SPEED		100000	/* hz */
> >  #define MAX_FS_MODE_SPEED		400000
> > @@ -81,6 +82,7 @@ enum DMA_REGS_OFFSET {
> >  	OFFSET_INT_FLAG = 0x0,
> >  	OFFSET_INT_EN = 0x04,
> >  	OFFSET_EN = 0x08,
> > +	OFFSET_RST = 0x0c,
> >  	OFFSET_CON = 0x18,
> >  	OFFSET_TX_MEM_ADDR = 0x1c,
> >  	OFFSET_RX_MEM_ADDR = 0x20,
> > @@ -262,6 +264,10 @@ static void mtk_i2c_init_hw(struct mtk_i2c *i2c)
> >  		      I2C_CONTROL_CLK_EXT_EN | I2C_CONTROL_DMA_EN;
> >  	writew(control_reg, i2c->base + OFFSET_CONTROL);
> >  	writew(I2C_DELAY_LEN, i2c->base + OFFSET_DELAY_LEN);
> > +
> > +	writel(I2C_DMA_HARD_RST, i2c->pdmabase + OFFSET_RST);
> > +	udelay(50);
> > +	writel(I2C_DMA_CLR_FLAG, i2c->pdmabase + OFFSET_RST);
> >  }
> >  
> >  /*
> > @@ -550,16 +556,20 @@ err_exit:
> >  static irqreturn_t mtk_i2c_irq(int irqno, void *dev_id)
> >  {
> >  	struct mtk_i2c *i2c = dev_id;
> > -	u16 restart_flag = 0;
> > +	u16 intr_stat = 0;
> >  
> > -	if (i2c->dev_comp->auto_restart)
> > -		restart_flag = I2C_RS_TRANSFER;
> > +	intr_stat = readw(i2c->base + OFFSET_INTR_STAT);
> > +	writew(intr_stat, i2c->base + OFFSET_INTR_STAT);
> >  
> > -	i2c->irq_stat = readw(i2c->base + OFFSET_INTR_STAT);
> > -	writew(restart_flag | I2C_HS_NACKERR | I2C_ACKERR
> > -		| I2C_TRANSAC_COMP, i2c->base + OFFSET_INTR_STAT);
> > +	/*
> > +	 * when occurs i2c ack error, mtk_i2c_irq is called twice,
> > +	 * first is the ack error interrupt, then the complete interrupt,
> > +	 * i2c->irq_stat need keep the two interrupt value.
> > +	 */
> > +	i2c->irq_stat |= intr_stat;
> >  
> > -	complete(&i2c->msg_complete);
> > +	if (i2c->irq_stat & I2C_TRANSAC_COMP)
> > +		complete(&i2c->msg_complete);
> >  
> >  	return IRQ_HANDLED;
> >  }
> > @@ -729,3 +739,4 @@ module_platform_driver(mtk_i2c_driver);
> >  MODULE_LICENSE("GPL v2");
> >  MODULE_DESCRIPTION("MediaTek I2C Bus Driver");
> >  MODULE_AUTHOR("Xudong Chen <xudong.chen@mediatek.com>");
> > +MODULE_AUTHOR("Liguo Zhang <liguo.zhang@mediatek.com>");
> > -- 
> > 1.7.9.5
> > 

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2015-08-04  6:13 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-07-28  3:38 [PATCH] i2c: mediatek: fix transfer error handling Eddie Huang
     [not found] ` <1438054685-25214-1-git-send-email-eddie.huang-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
2015-07-31 11:00   ` Wolfram Sang
2015-08-04  6:13     ` Eddie Huang

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).