Skip to contentSkip to conversation

cpufreq: speedstep: remove unnecessary static in speedstep_detect_chipset()

PATCH Applied Gustavo A. R. Silva 4 participants 4 messages Jul 5, 2017 – Jul 21, 2017 CPU frequency scaling framework

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 cpufreq: speedstep: remove unnecessary static in speedstep_detect_chipset()

Remove unnecessary static on local variable hostbridge. Such variable is initialized before being used, on every execution path throughout the function. The static has no benefit and, removing it reduces the code size.

This issue was detected using Coccinelle and the following semantic patch:

@bad exists@
position p;
identifier x;
type T;
@@

static T x@p;
...
x = <+...x...+>

@@
identifier x;
expression e;
type T;
position p != bad.p;
@@

-static
T x@p;
... when != x
when strict
?x = e;

In the following log you can see the difference in the code size. Also, there is a significant difference in the bss segment. This log is the output of the size command, before and after the code change:

before:

   text    data     bss     dec     hex filename
   5084    3392     256    8732    221c drivers/cpufreq/speedstep-ich.o

after:

   text    data     bss     dec     hex filename
   5062    3304     192    8558    216e drivers/cpufreq/speedstep-ich.o
---
 drivers/cpufreq/speedstep-ich.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/cpufreq/speedstep-ich.c +1 -1
index b86953a..0412a24 100644
@@ -207,7 +207,7 @@ static unsigned int speedstep_detect_chipset(void)
 		 * 8100 which use a pretty old revision of the 82815
 		 * host bridge. Abort on these systems.
 		 */
-		static struct pci_dev *hostbridge;
+		struct pci_dev *hostbridge;
 
 		hostbridge  = pci_get_subsys(PCI_VENDOR_ID_INTEL,
 			      PCI_DEVICE_ID_INTEL_82815_MC,
Signature · 1 line
2.5.0
maintainer Acked-by on v1 1/1
#1 lore raw

+ Dominic,

Quoted 53 lines · Gustavo A. R. Silva

Remove unnecessary static on local variable hostbridge. Such variable is initialized before being used, on every execution path throughout the function. The static has no benefit and, removing it reduces the code size.

This issue was detected using Coccinelle and the following semantic patch:

@bad exists@
position p;
identifier x;
type T;
@@

static T x@p;
...
x = <+...x...+>

@@
identifier x;
expression e;
type T;
position p != bad.p;
@@

-static
T x@p;
... when != x
when strict
?x = e;

In the following log you can see the difference in the code size. Also, there is a significant difference in the bss segment. This log is the output of the size command, before and after the code change:

before:

   text    data     bss     dec     hex filename
   5084    3392     256    8732    221c drivers/cpufreq/speedstep-ich.o

after:

   text    data     bss     dec     hex filename
   5062    3304     192    8558    216e drivers/cpufreq/speedstep-ich.o
---
 drivers/cpufreq/speedstep-ich.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/cpufreq/speedstep-ich.c +0 -0
index b86953a..0412a24 100644

drivers/cpufreq/speedstep-ich.c static unsigned int speedstep_detect_chipset(void) · line 210

Quoted 5 lines
drivers/cpufreq/speedstep-ich.c +1 -1
@@ -207,7 +207,7 @@ static unsigned int speedstep_detect_chipset(void)
 		 * 8100 which use a pretty old revision of the 82815
 		 * host bridge. Abort on these systems.
 		 */
-		static struct pci_dev *hostbridge;
+		struct pci_dev *hostbridge;

This was added way back Sep'02 by Dominic (@Dominic: I hope you are the same Dominic with a different email id) and I don't think we need it to be static here. Anyway, Dominic can respond on that.

commit 1673f3b404bd ("[PATCH] CPUfreq i386 drivers update")

(Note: You can find it in history kernel git repo, not the current one).

Signature · 1 line
viresh
#2 lore raw

Viresh, Gustavo,

Quoted 66 lines · Viresh Kumar

+ Dominic,

Quoted 53 lines · Gustavo A. R. Silva

Remove unnecessary static on local variable hostbridge. Such variable is initialized before being used, on every execution path throughout the function. The static has no benefit and, removing it reduces the code size.

This issue was detected using Coccinelle and the following semantic patch:

@bad exists@
position p;
identifier x;
type T;
@@

static T x@p;
...
x = <+...x...+>

@@
identifier x;
expression e;
type T;
position p != bad.p;
@@

-static
T x@p;
... when != x
when strict
?x = e;

In the following log you can see the difference in the code size. Also, there is a significant difference in the bss segment. This log is the output of the size command, before and after the code change:

before:

   text    data     bss     dec     hex filename
   5084    3392     256    8732    221c drivers/cpufreq/speedstep-ich.o

after:

   text    data     bss     dec     hex filename
   5062    3304     192    8558    216e drivers/cpufreq/speedstep-ich.o
---
 drivers/cpufreq/speedstep-ich.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/cpufreq/speedstep-ich.c +0 -0
index b86953a..0412a24 100644

drivers/cpufreq/speedstep-ich.c static unsigned int speedstep_detect_chipset(void) · line 210

Quoted 5 lines
drivers/cpufreq/speedstep-ich.c +1 -1
@@ -207,7 +207,7 @@ static unsigned int speedstep_detect_chipset(void)
 		 * 8100 which use a pretty old revision of the 82815
 		 * host bridge. Abort on these systems.
 		 */
-		static struct pci_dev *hostbridge;
+		struct pci_dev *hostbridge;

This was added way back Sep'02 by Dominic (@Dominic: I hope you are the same Dominic with a different email id) and I don't think we need it to be static here. Anyway, Dominic can respond on that.

Quoted 5 lines

commit 1673f3b404bd ("[PATCH] CPUfreq i386 drivers update")

(Note: You can find it in history kernel git repo, not the current one).

indeed, that's me, and the patch seems to be valid and therefore is a useful improvement:

Thanks,
Dominik

#3 lore raw
In reply to Gustavo A. R. Silva
Quoted 54 lines · Gustavo A. R. Silva

Remove unnecessary static on local variable hostbridge. Such variable is initialized before being used, on every execution path throughout the function. The static has no benefit and, removing it reduces the code size.

This issue was detected using Coccinelle and the following semantic patch:

@bad exists@
position p;
identifier x;
type T;
@@

static T x@p;
...
x = <+...x...+>

@@
identifier x;
expression e;
type T;
position p != bad.p;
@@

-static
T x@p;
... when != x
when strict
?x = e;

In the following log you can see the difference in the code size. Also, there is a significant difference in the bss segment. This log is the output of the size command, before and after the code change:

before:

   text    data     bss     dec     hex filename
   5084    3392     256    8732    221c drivers/cpufreq/speedstep-ich.o

after:

   text    data     bss     dec     hex filename
   5062    3304     192    8558    216e drivers/cpufreq/speedstep-ich.o
---
 drivers/cpufreq/speedstep-ich.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/cpufreq/speedstep-ich.c +0 -0
index b86953a..0412a24 100644

drivers/cpufreq/speedstep-ich.c static unsigned int speedstep_detect_chipset(void) · line 213

Quoted 8 lines
drivers/cpufreq/speedstep-ich.c +1 -1
@@ -207,7 +207,7 @@ static unsigned int speedstep_detect_chipset(void)
 		 * 8100 which use a pretty old revision of the 82815
 		 * host bridge. Abort on these systems.
 		 */
-		static struct pci_dev *hostbridge;
+		struct pci_dev *hostbridge;
 
 		hostbridge  = pci_get_subsys(PCI_VENDOR_ID_INTEL,
 			      PCI_DEVICE_ID_INTEL_82815_MC,

Applied, thanks!