[PATCH 1/7] drivers/ide/ide-cs.c: adjust suspicious bit operation

[PATCH 1/7] drivers/ide/ide-cs.c: adjust suspicious bit operation

[Julia Lawall]

From: Julia Lawall <Julia.Lawall@lip6.fr>

IO_DATA_PATH_WIDTH_8 is 0, so a bit-and with it is always false.  The
value IO_DATA_PATH_WIDTH covers the bits of the IO_DATA_PATH constants, so
first pick those bits and then make the test using !=.

This problem was found using Coccinelle (http://coccinelle.lip6.fr/).

Signed-off-by: Julia Lawall <julia@diku.dk>

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

diff --git a/drivers/ide/ide-cs.c b/drivers/ide/ide-cs.c
index 28e344e..a5e57bf 100644
--- a/drivers/ide/ide-cs.c
+++ b/drivers/ide/ide-cs.c
@@ -167,7 +167,8 @@ static int pcmcia_check_one_config(struct pcmcia_device *pdev, void *priv_data)
 {
 	int *is_kme = priv_data;
 
-	if (!(pdev->resource[0]->flags & IO_DATA_PATH_WIDTH_8)) {
+	if ((p1dev->resource[0]->flags & IO_DATA_PATH_WIDTH)
+	    != IO_DATA_PATH_WIDTH_8) {
 		pdev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
 		pdev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 	}

[PATCH 6/7] drivers/ata/pata_pcmcia.c: adjust suspicious bit operation

[Julia Lawall]

From: Julia Lawall <Julia.Lawall@lip6.fr>

IO_DATA_PATH_WIDTH_8 is 0, so a bit-and with it is always false.  The
value IO_DATA_PATH_WIDTH covers the bits of the IO_DATA_PATH constants, so
first pick those bits and then make the test using !=.

This problem was found using Coccinelle (http://coccinelle.lip6.fr/).

Signed-off-by: Julia Lawall <julia@diku.dk>

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

diff --git a/drivers/ata/pata_pcmcia.c b/drivers/ata/pata_pcmcia.c
index a808ba0..776e749 100644
--- a/drivers/ata/pata_pcmcia.c
+++ b/drivers/ata/pata_pcmcia.c
@@ -170,7 +170,8 @@ static int pcmcia_check_one_config(struct pcmcia_device *pdev, void *priv_data)
 {
 	int *is_kme = priv_data;
 
-	if (!(pdev->resource[0]->flags & IO_DATA_PATH_WIDTH_8)) {
+	if ((pdev->resource[0]->flags & IO_DATA_PATH_WIDTH)
+	    != IO_DATA_PATH_WIDTH_8) {
 		pdev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
 		pdev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 	}

Re: [PATCH 1/7] drivers/ide/ide-cs.c: adjust suspicious bit operation

[David Miller]

From: Julia Lawall <Julia.Lawall@lip6.fr>
Date: Wed,  6 Jun 2012 23:41:35 +0200

> -	if (!(pdev->resource[0]->flags & IO_DATA_PATH_WIDTH_8)) {
> +	if ((p1dev->resource[0]->flags & IO_DATA_PATH_WIDTH)

I'm really surprised someone as thorough as yourself did not
compile test this.

This is just more proof that it's absolutely pointless to make any
changes at all to the old IDE layer.  Nobody really cares, and the
risk %99.999 of the time is purely to introduce regressions rather
than make forward progress.

Re: [PATCH 1/7] drivers/ide/ide-cs.c: adjust suspicious bit operation

[Joe Perches]

On Thu, 2012-06-07 at 14:46 -0700, David Miller wrote:
> the
> risk %99.999 of the time is purely to introduce regressions rather
> than make forward progress.

 MAINTAINERS |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index b54f50c..41d4586 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3404,7 +3404,7 @@ M:	"David S. Miller" <davem@davemloft.net>
 L:	linux-ide@vger.kernel.org
 Q:	http://patchwork.ozlabs.org/project/linux-ide/list/
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/davem/ide.git
-S:	Maintained
+S:	Watchdogged - Patch at your own peril
 F:	Documentation/ide/
 F:	drivers/ide/
 F:	include/linux/ide.h

Re: [PATCH 1/7] drivers/ide/ide-cs.c: adjust suspicious bit operation

[Julia Lawall]

On Thu, 7 Jun 2012, David Miller wrote:

> From: Julia Lawall <Julia.Lawall@lip6.fr>
> Date: Wed,  6 Jun 2012 23:41:35 +0200
>
>> -	if (!(pdev->resource[0]->flags & IO_DATA_PATH_WIDTH_8)) {
>> +	if ((p1dev->resource[0]->flags & IO_DATA_PATH_WIDTH)
>
> I'm really surprised someone as thorough as yourself did not
> compile test this.

Sorry, I changed the patch at the last minute, and it was indeed not a 
good idea not to have compiled again.  I will send another version, in 
case it is useful.

julia

Re: [PATCH 1/7] drivers/ide/ide-cs.c: adjust suspicious bit operation

[Julia Lawall]

>From nobody Wed Jun  6 21:48:37 CEST 2012
From: Julia Lawall <Julia.Lawall@lip6.fr>
To: "David S. Miller" <davem@davemloft.net>
Cc: linux-ide@vger.kernel.org,linux-kernel@vger.kernel.org,joe@perches.com
Subject: [PATCH 1/7] drivers/ide/ide-cs.c: adjust suspicious bit operation

From: Julia Lawall <Julia.Lawall@lip6.fr>

IO_DATA_PATH_WIDTH_8 is 0, so a bit-and with it is always false.  The
value IO_DATA_PATH_WIDTH covers the bits of the IO_DATA_PATH constants, so
first pick those bits and then make the test using !=.

This problem was found using Coccinelle (http://coccinelle.lip6.fr/).

Signed-off-by: Julia Lawall <julia@diku.dk>

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

diff --git a/drivers/ide/ide-cs.c b/drivers/ide/ide-cs.c
index 28e344e..f1e922e 100644
--- a/drivers/ide/ide-cs.c
+++ b/drivers/ide/ide-cs.c
@@ -167,7 +167,8 @@ static int pcmcia_check_one_config(struct pcmcia_device *pdev, void *priv_data)
  {
  	int *is_kme = priv_data;

-	if (!(pdev->resource[0]->flags & IO_DATA_PATH_WIDTH_8)) {
+	if ((pdev->resource[0]->flags & IO_DATA_PATH_WIDTH)
+	    != IO_DATA_PATH_WIDTH_8) {
  		pdev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
  		pdev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
  	}

Re: [PATCH 1/7] drivers/ide/ide-cs.c: adjust suspicious bit operation

[walter harms]

Am 08.06.2012 07:14, schrieb Julia Lawall:
>> From nobody Wed Jun  6 21:48:37 CEST 2012
> From: Julia Lawall <Julia.Lawall@lip6.fr>
> To: "David S. Miller" <davem@davemloft.net>
> Cc: linux-ide@vger.kernel.org,linux-kernel@vger.kernel.org,joe@perches.com
> Subject: [PATCH 1/7] drivers/ide/ide-cs.c: adjust suspicious bit operation
> 
> From: Julia Lawall <Julia.Lawall@lip6.fr>
> 
> IO_DATA_PATH_WIDTH_8 is 0, so a bit-and with it is always false.  The
> value IO_DATA_PATH_WIDTH covers the bits of the IO_DATA_PATH constants, so
> first pick those bits and then make the test using !=.
> 
> This problem was found using Coccinelle (http://coccinelle.lip6.fr/).
> 
> Signed-off-by: Julia Lawall <julia@diku.dk>
> 
> ---
>  drivers/ide/ide-cs.c |    3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/ide/ide-cs.c b/drivers/ide/ide-cs.c
> index 28e344e..f1e922e 100644
> --- a/drivers/ide/ide-cs.c
> +++ b/drivers/ide/ide-cs.c
> @@ -167,7 +167,8 @@ static int pcmcia_check_one_config(struct
> pcmcia_device *pdev, void *priv_data)
>  {
>      int *is_kme = priv_data;
> 
> -    if (!(pdev->resource[0]->flags & IO_DATA_PATH_WIDTH_8)) {
> +    if ((pdev->resource[0]->flags & IO_DATA_PATH_WIDTH)
> +        != IO_DATA_PATH_WIDTH_8) {
>          pdev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
>          pdev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
>      }
> -- 

please patch until it is removed from the kernel code.
this will make checks happy and as fact of life such bug tend to show up
on surprising unpleasant times.

re,
 wh

Re: [PATCH 1/7] drivers/ide/ide-cs.c: adjust suspicious bit operation

[David Miller]

From: Julia Lawall <julia.lawall@lip6.fr>
Date: Fri, 8 Jun 2012 07:14:03 +0200 (CEST)

> @@ -167,7 +167,8 @@ static int pcmcia_check_one_config(struct
> pcmcia_device *pdev, void *priv_data)

Patch mangled by your email client or similar.

Re: [PATCH 1/7] drivers/ide/ide-cs.c: adjust suspicious bit operation

[Julia Lawall]

From: Julia Lawall <Julia.Lawall@lip6.fr>

IO_DATA_PATH_WIDTH_8 is 0, so a bit-and with it is always false.  The
value IO_DATA_PATH_WIDTH covers the bits of the IO_DATA_PATH constants, so
first pick those bits and then make the test using !=.

This problem was found using Coccinelle (http://coccinelle.lip6.fr/).

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

---
I hope to have fixed the mail client problem.

diff --git a/drivers/ide/ide-cs.c b/drivers/ide/ide-cs.c
index 28e344e..f1e922e 100644
--- a/drivers/ide/ide-cs.c
+++ b/drivers/ide/ide-cs.c
@@ -167,7 +167,8 @@ static int pcmcia_check_one_config(struct pcmcia_device *pdev, void *priv_data)
 {
 	int *is_kme = priv_data;

-	if (!(pdev->resource[0]->flags & IO_DATA_PATH_WIDTH_8)) {
+	if ((pdev->resource[0]->flags & IO_DATA_PATH_WIDTH)
+	    != IO_DATA_PATH_WIDTH_8) {
 		pdev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
 		pdev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 	}

Re: [PATCH 1/7] drivers/ide/ide-cs.c: adjust suspicious bit operation

[David Miller]

From: Julia Lawall <Julia.Lawall@lip6.fr>
Date: Tue, 12 Jun 2012 10:17:25 -0400 (EDT)

> From: Julia Lawall <Julia.Lawall@lip6.fr>
> 
> IO_DATA_PATH_WIDTH_8 is 0, so a bit-and with it is always false.  The
> value IO_DATA_PATH_WIDTH covers the bits of the IO_DATA_PATH constants, so
> first pick those bits and then make the test using !=.
> 
> This problem was found using Coccinelle (http://coccinelle.lip6.fr/).
> 
> Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>

Applied, thanks.