Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1452677 > unrolled thread
| Started by | Robert Jarzmik <robert.jarzmik@free.fr> |
|---|---|
| First post | 2016-07-30 13:30 +0200 |
| Last post | 2016-07-31 11:20 +0200 |
| Articles | 3 — 2 participants |
Back to article view | Back to linux.kernel
[PATCH v2] ARM: pxa: fix GPIO double shifts Robert Jarzmik <robert.jarzmik@free.fr> - 2016-07-30 13:30 +0200
Re: [PATCH v2] ARM: pxa: fix GPIO double shifts Joe Perches <joe@perches.com> - 2016-07-30 18:40 +0200
Re: [PATCH v2] ARM: pxa: fix GPIO double shifts Robert Jarzmik <robert.jarzmik@free.fr> - 2016-07-31 11:20 +0200
| From | Robert Jarzmik <robert.jarzmik@free.fr> |
|---|---|
| Date | 2016-07-30 13:30 +0200 |
| Subject | [PATCH v2] ARM: pxa: fix GPIO double shifts |
| Message-ID | <s0ACR-1QI-5@gated-at.bofh.it> |
The commit 9bf448c66d4b ("ARM: pxa: use generic gpio operation instead of
gpio register") from Oct 17, 2011, leads to the following static checker
warning:
arch/arm/mach-pxa/spitz_pm.c:172 spitz_charger_wakeup()
warn: double left shift '!gpio_get_value(SPITZ_GPIO_KEY_INT)
<< (1 << ((SPITZ_GPIO_KEY_INT) & 31))'
As Dan reported, the value is shifted three times :
- once by gpio_get_value(), which returns either 0 or BIT(gpio)
- once by the shift operation '<<'
- a last time by GPIO_bit(gpio) which is BIT(gpio)
Therefore the calculation lead to a chained or operator of :
- (1 << gpio) << (1 << gpio) = (2^gpio)^gpio = 2 ^ (gpio * gpio)
It is be sheer luck the former statement works, only because each gpio
used is strictly smaller than 6, and therefore 2^(gpio^2) never
overflows a 32 bits value, and because it is used as a boolean value to
check a gpio activation.
Fixes: 9bf448c66d4b ("ARM: pxa: use generic gpio operation instead of
gpio register")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
---
Since v1: replaced binary ORs with logical ORs after assembly comparison
---
arch/arm/mach-pxa/corgi_pm.c | 8 +++-----
arch/arm/mach-pxa/spitz_pm.c | 5 ++---
2 files changed, 5 insertions(+), 8 deletions(-)
diff --git a/arch/arm/mach-pxa/corgi_pm.c b/arch/arm/mach-pxa/corgi_pm.c
index d9206811be9b..8dc39d602884 100644
--- a/arch/arm/mach-pxa/corgi_pm.c
+++ b/arch/arm/mach-pxa/corgi_pm.c
@@ -135,11 +135,9 @@ static unsigned long corgi_charger_wakeup(void)
{
unsigned long ret;
- ret = (!gpio_get_value(CORGI_GPIO_AC_IN) << GPIO_bit(CORGI_GPIO_AC_IN))
- | (!gpio_get_value(CORGI_GPIO_KEY_INT)
- << GPIO_bit(CORGI_GPIO_KEY_INT))
- | (!gpio_get_value(CORGI_GPIO_WAKEUP)
- << GPIO_bit(CORGI_GPIO_WAKEUP));
+ ret = !gpio_get_value(CORGI_GPIO_AC_IN)
+ || !gpio_get_value(CORGI_GPIO_KEY_INT)
+ || !gpio_get_value(CORGI_GPIO_WAKEUP);
return ret;
}
diff --git a/arch/arm/mach-pxa/spitz_pm.c b/arch/arm/mach-pxa/spitz_pm.c
index ea9f9034cb54..dd85869f6f99 100644
--- a/arch/arm/mach-pxa/spitz_pm.c
+++ b/arch/arm/mach-pxa/spitz_pm.c
@@ -168,9 +168,8 @@ static int spitz_should_wakeup(unsigned int resume_on_alarm)
static unsigned long spitz_charger_wakeup(void)
{
unsigned long ret;
- ret = ((!gpio_get_value(SPITZ_GPIO_KEY_INT)
- << GPIO_bit(SPITZ_GPIO_KEY_INT))
- | gpio_get_value(SPITZ_GPIO_SYNC));
+ ret = !gpio_get_value(SPITZ_GPIO_KEY_INT)
+ || gpio_get_value(SPITZ_GPIO_SYNC);
return ret;
}
--
2.1.4
[toc] | [next] | [standalone]
| From | Joe Perches <joe@perches.com> |
|---|---|
| Date | 2016-07-30 18:40 +0200 |
| Message-ID | <s0FsR-4SM-15@gated-at.bofh.it> |
| In reply to | #1452677 |
On Sat, 2016-07-30 at 13:22 +0200, Robert Jarzmik wrote:
> The commit 9bf448c66d4b ("ARM: pxa: use generic gpio operation instead of
> gpio register") from Oct 17, 2011, leads to the following static checker
> warning:
> arch/arm/mach-pxa/spitz_pm.c:172 spitz_charger_wakeup()
> warn: double left shift '!gpio_get_value(SPITZ_GPIO_KEY_INT)
> << (1 << ((SPITZ_GPIO_KEY_INT) & 31))'
>
> As Dan reported, the value is shifted three times :
> - once by gpio_get_value(), which returns either 0 or BIT(gpio)
> - once by the shift operation '<<'
> - a last time by GPIO_bit(gpio) which is BIT(gpio)
>
> Therefore the calculation lead to a chained or operator of :
> - (1 << gpio) << (1 << gpio) = (2^gpio)^gpio = 2 ^ (gpio * gpio)
>
> It is be sheer luck the former statement works, only because each gpio
> used is strictly smaller than 6, and therefore 2^(gpio^2) never
> overflows a 32 bits value, and because it is used as a boolean value to
> check a gpio activation.
It may be better to change the charger_wakeup callback return
value from unsigned long to bool and modify the other use in
spitz_pm.c
$ git grep -w charger_wakeup
arch/arm/mach-pxa/corgi_pm.c: .charger_wakeup = corgi_charger_wakeup,
arch/arm/mach-pxa/sharpsl_pm.c: if (sharpsl_pm.machinfo->charger_wakeup() != 0)
arch/arm/mach-pxa/sharpsl_pm.c: if (sharpsl_pm.machinfo->charger_wakeup())
arch/arm/mach-pxa/sharpsl_pm.h: unsigned long (*charger_wakeup)(void);
arch/arm/mach-pxa/spitz_pm.c: .charger_wakeup = spitz_charger_wakeup,
> Fixes: 9bf448c66d4b ("ARM: pxa: use generic gpio operation instead of
> gpio register")
> Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
> Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
> ---
> Since v1: replaced binary ORs with logical ORs after assembly comparison
> ---
> arch/arm/mach-pxa/corgi_pm.c | 8 +++-----
> arch/arm/mach-pxa/spitz_pm.c | 5 ++---
> 2 files changed, 5 insertions(+), 8 deletions(-)
>
> diff --git a/arch/arm/mach-pxa/corgi_pm.c b/arch/arm/mach-pxa/corgi_pm.c
> index d9206811be9b..8dc39d602884 100644
> --- a/arch/arm/mach-pxa/corgi_pm.c
> +++ b/arch/arm/mach-pxa/corgi_pm.c
> @@ -135,11 +135,9 @@ static unsigned long corgi_charger_wakeup(void)
> {
> unsigned long ret;
>
> - ret = (!gpio_get_value(CORGI_GPIO_AC_IN) << GPIO_bit(CORGI_GPIO_AC_IN))
> - | (!gpio_get_value(CORGI_GPIO_KEY_INT)
> - << GPIO_bit(CORGI_GPIO_KEY_INT))
> - | (!gpio_get_value(CORGI_GPIO_WAKEUP)
> - << GPIO_bit(CORGI_GPIO_WAKEUP));
> + ret = !gpio_get_value(CORGI_GPIO_AC_IN)
> + || !gpio_get_value(CORGI_GPIO_KEY_INT)
> + || !gpio_get_value(CORGI_GPIO_WAKEUP);
> return ret;
> }
>
> diff --git a/arch/arm/mach-pxa/spitz_pm.c b/arch/arm/mach-pxa/spitz_pm.c
> index ea9f9034cb54..dd85869f6f99 100644
> --- a/arch/arm/mach-pxa/spitz_pm.c
> +++ b/arch/arm/mach-pxa/spitz_pm.c
> @@ -168,9 +168,8 @@ static int spitz_should_wakeup(unsigned int resume_on_alarm)
> static unsigned long spitz_charger_wakeup(void)
> {
> unsigned long ret;
> - ret = ((!gpio_get_value(SPITZ_GPIO_KEY_INT)
> - << GPIO_bit(SPITZ_GPIO_KEY_INT))
> - | gpio_get_value(SPITZ_GPIO_SYNC));
> + ret = !gpio_get_value(SPITZ_GPIO_KEY_INT)
> + || gpio_get_value(SPITZ_GPIO_SYNC);
> return ret;
> }
>
[toc] | [prev] | [next] | [standalone]
| From | Robert Jarzmik <robert.jarzmik@free.fr> |
|---|---|
| Date | 2016-07-31 11:20 +0200 |
| Message-ID | <s0V4B-6Dq-3@gated-at.bofh.it> |
| In reply to | #1452724 |
Joe Perches <joe@perches.com> writes: > On Sat, 2016-07-30 at 13:22 +0200, Robert Jarzmik wrote: ...zip... > $ git grep -w charger_wakeup > arch/arm/mach-pxa/corgi_pm.c: .charger_wakeup = corgi_charger_wakeup, > arch/arm/mach-pxa/sharpsl_pm.c: if (sharpsl_pm.machinfo->charger_wakeup() != 0) > arch/arm/mach-pxa/sharpsl_pm.c: if (sharpsl_pm.machinfo->charger_wakeup()) > arch/arm/mach-pxa/sharpsl_pm.h: unsigned long (*charger_wakeup)(void); > arch/arm/mach-pxa/spitz_pm.c: .charger_wakeup = spitz_charger_wakeup, ...zip... > It may be better to change the charger_wakeup callback return > value from unsigned long to bool and modify the other use in > spitz_pm.c You're most probably right. I'll do that for v3. Cheers. -- Robert
[toc] | [prev] | [standalone]
Back to top | Article view | linux.kernel
csiph-web