[PATCH 0/2] change PTR_ERR argument

[PATCH 0/2] change PTR_ERR argument

[Julia Lawall]

Apply PTR_ERR to the value that was recently assigned.

---

 var/julia/linuxcopy/drivers/spi/spi-ep93xx.c   |    3 ++-
 var/julia/linuxcopy/sound/soc/qcom/lpass-cpu.c |    3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

[PATCH 1/2] ASoC: qcom: change PTR_ERR argument

[Julia Lawall]

Apply PTR_ERR to the value that was recently assigned.

The semantic match that finds this problem is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@@
expression x,y;
@@

if (IS_ERR(x) || ...) {
  ... when any
      when != IS_ERR(...)
(
   PTR_ERR(x)
|
*  PTR_ERR(y)
)
  ... when any
}
// </smpl>

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>

---
 sound/soc/qcom/lpass-cpu.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/sound/soc/qcom/lpass-cpu.c b/sound/soc/qcom/lpass-cpu.c
index 23f3d59..94beb99 100644
--- a/sound/soc/qcom/lpass-cpu.c
+++ b/sound/soc/qcom/lpass-cpu.c
@@ -438,7 +438,8 @@ int asoc_qcom_lpass_cpu_platform_probe(struct platform_device *pdev)
 		if (IS_ERR(drvdata->mi2s_bit_clk[dai_id])) {
 			dev_err(&pdev->dev,
 				"%s() error getting mi2s-bit-clk: %ld\n",
-				__func__, PTR_ERR(drvdata->mi2s_bit_clk[i]));
+				__func__,
+				PTR_ERR(drvdata->mi2s_bit_clk[dai_id]));
 			return PTR_ERR(drvdata->mi2s_bit_clk[dai_id]);
 		}
 	}

[PATCH 2/2] spi: spi-ep93xx: change PTR_ERR argument

[Julia Lawall]

Apply PTR_ERR to the value that was recently assigned.

The semantic match that finds this problem is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@@
expression x,y;
@@

if (IS_ERR(x) || ...) {
  ... when any
      when != IS_ERR(...)
(
   PTR_ERR(x)
|
*  PTR_ERR(y)
)
  ... when any
}
// </smpl>

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>

---
 drivers/spi/spi-ep93xx.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/spi/spi-ep93xx.c b/drivers/spi/spi-ep93xx.c
index bb00be8..1a7696c 100644
--- a/drivers/spi/spi-ep93xx.c
+++ b/drivers/spi/spi-ep93xx.c
@@ -567,7 +567,8 @@ static void ep93xx_spi_dma_transfer(struct ep93xx_spi *espi)
 	txd = ep93xx_spi_dma_prepare(espi, DMA_MEM_TO_DEV);
 	if (IS_ERR(txd)) {
 		ep93xx_spi_dma_finish(espi, DMA_DEV_TO_MEM);
-		dev_err(&espi->pdev->dev, "DMA TX failed: %ld\n", PTR_ERR(rxd));
+		dev_err(&espi->pdev->dev, "DMA TX failed: %ld\n",
+			PTR_ERR(txd));
 		msg->status = PTR_ERR(txd);
 		return;
 	}

Re: [PATCH 2/2] spi: spi-ep93xx: change PTR_ERR argument

[walter harms]

Am 30.08.2015 20:05, schrieb Julia Lawall:
> Apply PTR_ERR to the value that was recently assigned.
> 
> The semantic match that finds this problem is as follows:
> (http://coccinelle.lip6.fr/)
> 
> // <smpl>
> @@
> expression x,y;
> @@
> 
> if (IS_ERR(x) || ...) {
>   ... when any
>       when != IS_ERR(...)
> (
>    PTR_ERR(x)
> |
> *  PTR_ERR(y)
> )
>   ... when any
> }
> // </smpl>
> 
> Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
> 
> ---
>  drivers/spi/spi-ep93xx.c |    3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/spi/spi-ep93xx.c b/drivers/spi/spi-ep93xx.c
> index bb00be8..1a7696c 100644
> --- a/drivers/spi/spi-ep93xx.c
> +++ b/drivers/spi/spi-ep93xx.c
> @@ -567,7 +567,8 @@ static void ep93xx_spi_dma_transfer(struct ep93xx_spi *espi)
>  	txd = ep93xx_spi_dma_prepare(espi, DMA_MEM_TO_DEV);
>  	if (IS_ERR(txd)) {
>  		ep93xx_spi_dma_finish(espi, DMA_DEV_TO_MEM);
> -		dev_err(&espi->pdev->dev, "DMA TX failed: %ld\n", PTR_ERR(rxd));
> +		dev_err(&espi->pdev->dev, "DMA TX failed: %ld\n",
> +			PTR_ERR(txd));
>  		msg->status = PTR_ERR(txd);
>  		return;
>  	}
>

I improve readability i would suggest:

ep93xx_spi_dma_finish(espi, DMA_DEV_TO_MEM);
msg->status = PTR_ERR(txd);
dev_err(&espi->pdev->dev, "DMA TX failed: %ld\n",msg->status):
return;


just my 2 cents,

re,
 wh
> --
> To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: [PATCH 1/2] ASoC: qcom: change PTR_ERR argument

[walter harms]

Am 30.08.2015 20:05, schrieb Julia Lawall:
> Apply PTR_ERR to the value that was recently assigned.
> 
> The semantic match that finds this problem is as follows:
> (http://coccinelle.lip6.fr/)
> 
> // <smpl>
> @@
> expression x,y;
> @@
> 
> if (IS_ERR(x) || ...) {
>   ... when any
>       when != IS_ERR(...)
> (
>    PTR_ERR(x)
> |
> *  PTR_ERR(y)
> )
>   ... when any
> }
> // </smpl>
> 
> Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
> 
> ---
>  sound/soc/qcom/lpass-cpu.c |    3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/sound/soc/qcom/lpass-cpu.c b/sound/soc/qcom/lpass-cpu.c
> index 23f3d59..94beb99 100644
> --- a/sound/soc/qcom/lpass-cpu.c
> +++ b/sound/soc/qcom/lpass-cpu.c
> @@ -438,7 +438,8 @@ int asoc_qcom_lpass_cpu_platform_probe(struct platform_device *pdev)
>  		if (IS_ERR(drvdata->mi2s_bit_clk[dai_id])) {
>  			dev_err(&pdev->dev,
>  				"%s() error getting mi2s-bit-clk: %ld\n",
> -				__func__, PTR_ERR(drvdata->mi2s_bit_clk[i]));
> +				__func__,
> +				PTR_ERR(drvdata->mi2s_bit_clk[dai_id]));
>  			return PTR_ERR(drvdata->mi2s_bit_clk[dai_id]);
>  		}
>  	}
> 

just a note:
using a shorter name instead of drvdata->mi2s_bit_clk[dai_id] whould help to make the code
more readable (yes, the other code is alike). something like:

struct clk *tmp =  devm_clk_get(&pdev->dev,clk_name);

if (IS_ERR(tmp)) {
	dev_err(&pdev->dev,"%s() error getting mi2s-bit-clk: %ld\n",__func__, PTR_ERR(tmp));
        return PTR_ERR(tmp);
}
drvdata->mi2s_bit_clk[dai_id]=tmp;


just one minor:
the dev_warn() just before says: " error getting mi2s-osr-clk" may be it should be "warnig ..."
That will make it more easy to rep for real error in a log.

re,
 wh

Re: [PATCH 1/2] ASoC: qcom: change PTR_ERR argument

[Julia Lawall]

On Sun, 30 Aug 2015, walter harms wrote:

> 
> 
> Am 30.08.2015 20:05, schrieb Julia Lawall:
> > Apply PTR_ERR to the value that was recently assigned.
> > 
> > The semantic match that finds this problem is as follows:
> > (http://coccinelle.lip6.fr/)
> > 
> > // <smpl>
> > @@
> > expression x,y;
> > @@
> > 
> > if (IS_ERR(x) || ...) {
> >   ... when any
> >       when != IS_ERR(...)
> > (
> >    PTR_ERR(x)
> > |
> > *  PTR_ERR(y)
> > )
> >   ... when any
> > }
> > // </smpl>
> > 
> > Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
> > 
> > ---
> >  sound/soc/qcom/lpass-cpu.c |    3 ++-
> >  1 file changed, 2 insertions(+), 1 deletion(-)
> > 
> > diff --git a/sound/soc/qcom/lpass-cpu.c b/sound/soc/qcom/lpass-cpu.c
> > index 23f3d59..94beb99 100644
> > --- a/sound/soc/qcom/lpass-cpu.c
> > +++ b/sound/soc/qcom/lpass-cpu.c
> > @@ -438,7 +438,8 @@ int asoc_qcom_lpass_cpu_platform_probe(struct platform_device *pdev)
> >  		if (IS_ERR(drvdata->mi2s_bit_clk[dai_id])) {
> >  			dev_err(&pdev->dev,
> >  				"%s() error getting mi2s-bit-clk: %ld\n",
> > -				__func__, PTR_ERR(drvdata->mi2s_bit_clk[i]));
> > +				__func__,
> > +				PTR_ERR(drvdata->mi2s_bit_clk[dai_id]));
> >  			return PTR_ERR(drvdata->mi2s_bit_clk[dai_id]);
> >  		}
> >  	}
> > 
> 
> just a note:
> using a shorter name instead of drvdata->mi2s_bit_clk[dai_id] whould help to make the code
> more readable (yes, the other code is alike). something like:
> 
> struct clk *tmp =  devm_clk_get(&pdev->dev,clk_name);

Where do you suggest to put this?

Maybe it would be reasonable to declare a variable struct clk *clk; at the 
top of the function, and then use that as a temporary variable for all 
three calls.

However, now I see that the first call, unlike the other two doesn't cause 
a return from the function.

                if (IS_ERR(drvdata->mi2s_osr_clk[dai_id])) {
                        dev_warn(&pdev->dev,
                                "%s() error getting mi2s-osr-clk: %ld\n",
				__func__,
		                PTR_ERR(drvdata->mi2s_osr_clk[dai_id]));
                }

Is that intentional?

thanks,
julia

> if (IS_ERR(tmp)) {
> 	dev_err(&pdev->dev,"%s() error getting mi2s-bit-clk: %ld\n",__func__, PTR_ERR(tmp));
>         return PTR_ERR(tmp);
> }
> drvdata->mi2s_bit_clk[dai_id]=tmp;
> 
> 
> just one minor:
> the dev_warn() just before says: " error getting mi2s-osr-clk" may be it should be "warnig ..."
> That will make it more easy to rep for real error in a log.
> 
> re,
>  wh
> 
> --
> To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

[PATCH 2/2 v2] spi: spi-ep93xx: fix PTR_ERR problem

[Julia Lawall]

Move initialization of msg->status up over the call to dev_err, in both
calls to ep93xx_spi_dma_prepare, and change the reference in the call to
dev_err to msg->status, to both fix the wrong argument to PTR_ERR in the
second case and to make the dev_err line a little shorter.  This required
furthermore replacing %ld by %d, since msg->status is an integer.

The semantic match that finds the PTR_ERR problem is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@@
expression x,y;
@@

if (IS_ERR(x) || ...) {
  ... when any
      when != IS_ERR(...)
(
   PTR_ERR(x)
|
*  PTR_ERR(y)
)
  ... when any
}
// </smpl>

Reorganizations at the suggestion of Walter Harms.

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>

---
v2: Reorganize the code, to solve the problem in a way that makes the
resulting code simpler.

 drivers/spi/spi-ep93xx.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/spi/spi-ep93xx.c b/drivers/spi/spi-ep93xx.c
index bb00be8..73d0df6 100644
--- a/drivers/spi/spi-ep93xx.c
+++ b/drivers/spi/spi-ep93xx.c
@@ -559,16 +559,16 @@ static void ep93xx_spi_dma_transfer(struct ep93xx_spi *espi)
 
 	rxd = ep93xx_spi_dma_prepare(espi, DMA_DEV_TO_MEM);
 	if (IS_ERR(rxd)) {
-		dev_err(&espi->pdev->dev, "DMA RX failed: %ld\n", PTR_ERR(rxd));
 		msg->status = PTR_ERR(rxd);
+		dev_err(&espi->pdev->dev, "DMA RX failed: %d\n", msg->status);
 		return;
 	}
 
 	txd = ep93xx_spi_dma_prepare(espi, DMA_MEM_TO_DEV);
 	if (IS_ERR(txd)) {
 		ep93xx_spi_dma_finish(espi, DMA_DEV_TO_MEM);
-		dev_err(&espi->pdev->dev, "DMA TX failed: %ld\n", PTR_ERR(rxd));
 		msg->status = PTR_ERR(txd);
+		dev_err(&espi->pdev->dev, "DMA TX failed: %d\n", msg->status);
 		return;
 	}
 

Re: [alsa-devel] [PATCH 1/2] ASoC: qcom: change PTR_ERR argument

[Kenneth Westfield]

On Sun, Aug 30, 2015 at 11:05:10AM -0700, Julia Lawall wrote:
> Apply PTR_ERR to the value that was recently assigned.
> 
> The semantic match that finds this problem is as follows:
> (http://coccinelle.lip6.fr/)
> 
> // <smpl>
> @@
> expression x,y;
> @@
> 
> if (IS_ERR(x) || ...) {
>   ... when any
>       when != IS_ERR(...)
> (
>    PTR_ERR(x)
> |
> *  PTR_ERR(y)
> )
>   ... when any
> }
> // </smpl>
> 
> Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
> 
> ---

The patch itself looks good.  Thanks.

Acked-by: Kenneth Westfield <kwestfie@codeaurora.org>

-- 
Kenneth Westfield
Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, 
a Linux Foundation Collaborative Project

Re: [alsa-devel] [PATCH 1/2] ASoC: qcom: change PTR_ERR argument

[Kenneth Westfield]

On Sun, Aug 30, 2015 at 11:54:33AM -0700, walter harms wrote:
> Am 30.08.2015 20:05, schrieb Julia Lawall:
> >  		if (IS_ERR(drvdata->mi2s_bit_clk[dai_id])) {
> >  			dev_err(&pdev->dev,
> >  				"%s() error getting mi2s-bit-clk: %ld\n",
> > -				__func__,
> PTR_ERR(drvdata->mi2s_bit_clk[i]));
> > +				__func__,
> > +				PTR_ERR(drvdata->mi2s_bit_clk[dai_id]));
> >  			return PTR_ERR(drvdata->mi2s_bit_clk[dai_id]);
> >  		}
> >  	}
> > 
> 
> just a note:
> using a shorter name instead of drvdata->mi2s_bit_clk[dai_id] whould help
> to make the code
> more readable (yes, the other code is alike). something like:
> 
> struct clk *tmp =  devm_clk_get(&pdev->dev,clk_name);
> 
> if (IS_ERR(tmp)) {
> 	dev_err(&pdev->dev,"%s() error getting mi2s-bit-clk:
> %ld\n",__func__, PTR_ERR(tmp));
>         return PTR_ERR(tmp);
> }
> drvdata->mi2s_bit_clk[dai_id]=tmp;

Yes, it would make the code more readable.

> 
> 
> just one minor:
> the dev_warn() just before says: " error getting mi2s-osr-clk" may be it
> should be "warnig ..."
> That will make it more easy to rep for real error in a log.

"error [gs]etting" could be re-phrased to "could not [gs]et".  The term
"error" was not meant to indicate the log level, but evidently, can
cause some confusion for someone reading the logs.

-- 
Kenneth Westfield
Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, 
a Linux Foundation Collaborative Project

Re: [alsa-devel] [PATCH 1/2] ASoC: qcom: change PTR_ERR argument

[Kenneth Westfield]

On Sun, Aug 30, 2015 at 12:54:15PM -0700, Julia Lawall wrote:
> On Sun, 30 Aug 2015, walter harms wrote:
> > Am 30.08.2015 20:05, schrieb Julia Lawall:
> > >  		if (IS_ERR(drvdata->mi2s_bit_clk[dai_id])) {
> > >  			dev_err(&pdev->dev,
> > >  				"%s() error getting mi2s-bit-clk: %ld\n",
> > > -				__func__,
> PTR_ERR(drvdata->mi2s_bit_clk[i]));
> > > +				__func__,
> > > +				PTR_ERR(drvdata->mi2s_bit_clk[dai_id]));
> > >  			return PTR_ERR(drvdata->mi2s_bit_clk[dai_id]);
> > >  		}
> > >  	}
> > > 
> > 
> > just a note:
> > using a shorter name instead of drvdata->mi2s_bit_clk[dai_id] whould
> help to make the code
> > more readable (yes, the other code is alike). something like:
> > 
> > struct clk *tmp =  devm_clk_get(&pdev->dev,clk_name);
> 
> Where do you suggest to put this?
> 
> Maybe it would be reasonable to declare a variable struct clk *clk; at the
> 
> top of the function, and then use that as a temporary variable for all 
> three calls.
> 
> However, now I see that the first call, unlike the other two doesn't cause
> 
> a return from the function.
> 
>                 if (IS_ERR(drvdata->mi2s_osr_clk[dai_id])) {
>                         dev_warn(&pdev->dev,
>                                 "%s() error getting mi2s-osr-clk: %ld\n",
> 				__func__,
> 		                PTR_ERR(drvdata->mi2s_osr_clk[dai_id]));
>                 }
> 
> Is that intentional?

Yes, that was intentional as the presense of the OSR clock in the DT
node is optional.

-- 
Kenneth Westfield
Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, 
a Linux Foundation Collaborative Project

Re: [PATCH 1/2] ASoC: qcom: change PTR_ERR argument

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 461 bytes --]

On Sun, Aug 30, 2015 at 08:05:10PM +0200, Julia Lawall wrote:
> Apply PTR_ERR to the value that was recently assigned.
> 
> The semantic match that finds this problem is as follows:
> (http://coccinelle.lip6.fr/)

I have no idea what's going on with this stuff without spending more
time than it looks like should need, there's a moderately big thread and
some patches posted in the middle of it.  Can you repost whatever the
current state is please?

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]

Re: [PATCH 1/2] ASoC: qcom: change PTR_ERR argument

[Julia Lawall]

On Mon, 14 Sep 2015, Mark Brown wrote:

> On Sun, Aug 30, 2015 at 08:05:10PM +0200, Julia Lawall wrote:
> > Apply PTR_ERR to the value that was recently assigned.
> >
> > The semantic match that finds this problem is as follows:
> > (http://coccinelle.lip6.fr/)
>
> I have no idea what's going on with this stuff without spending more
> time than it looks like should need, there's a moderately big thread and
> some patches posted in the middle of it.  Can you repost whatever the
> current state is please?

The discussion was about introducing a temporary variable to simplify the
code.  But that makes a lot of changes, so I think it would be better to
just apply the original bug fixing patch as is, and then the cleanup could
be applied on top of that.  I will submit the original patch again.

julia

Re: [PATCH 1/2] ASoC: qcom: change PTR_ERR argument

[Julia Lawall]

Apply PTR_ERR to the value that was recently assigned.

The semantic match that finds this problem is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@@
expression x,y;
@@

if (IS_ERR(x) || ...) {
  ... when any
      when != IS_ERR(...)
(
   PTR_ERR(x)
|
*  PTR_ERR(y)
)
  ... when any
}
// </smpl>

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>

---
 sound/soc/qcom/lpass-cpu.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/sound/soc/qcom/lpass-cpu.c b/sound/soc/qcom/lpass-cpu.c
index 23f3d59..94beb99 100644
--- a/sound/soc/qcom/lpass-cpu.c
+++ b/sound/soc/qcom/lpass-cpu.c
@@ -438,7 +438,8 @@ int asoc_qcom_lpass_cpu_platform_probe(struct platform_device *pdev)
 		if (IS_ERR(drvdata->mi2s_bit_clk[dai_id])) {
 			dev_err(&pdev->dev,
 				"%s() error getting mi2s-bit-clk: %ld\n",
-				__func__, PTR_ERR(drvdata->mi2s_bit_clk[i]));
+				__func__,
+				PTR_ERR(drvdata->mi2s_bit_clk[dai_id]));
 			return PTR_ERR(drvdata->mi2s_bit_clk[dai_id]);
 		}
 	}

Re: [PATCH 1/2] ASoC: qcom: change PTR_ERR argument

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 388 bytes --]

On Thu, Sep 17, 2015 at 10:46:16AM +0200, Julia Lawall wrote:

> The discussion was about introducing a temporary variable to simplify the
> code.  But that makes a lot of changes, so I think it would be better to
> just apply the original bug fixing patch as is, and then the cleanup could
> be applied on top of that.  I will submit the original patch again.

OK, makes sense - thanks.

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]