Skip to contentSkip to conversation

i2c via686a.c: save at least 0.5k of space by long v[256] -> u16 v[256]

PATCH In review Denis Vlasenko 3 participants 9 messages Sep 1, 2005 – Sep 10, 2005 Hardware monitoring

lore
Share & tools
lore
#0 lore raw

Not tested, but it's rather obvious.
--
vda

drivers/i2c/chips/via686a.c renamed from drivers/i2c/chips/via686a.c.orig +1 -1
@@ -205,7 +205,7 @@ static inline u8 FAN_TO_REG(long rpm, in
  but the function is very linear in the useful range (0-80 deg C), so 
  we'll just use linear interpolation for 10-bit readings.)  So, tempLUT 
  is the temp at via register values 0-255: */
-static const long tempLUT[] =
+static const int16_t tempLUT[] =
     { -709, -688, -667, -646, -627, -607, -589, -570, -553, -536, -519,
 	    -503, -487, -471, -456, -442, -428, -414, -400, -387, -375,
 	    -362, -350, -339, -327, -316, -305, -295, -285, -275, -265,
#1 lore raw
Quoted 1 line · Denis Vlasenko

Not tested, but it's rather obvious.

Except you forgot a "Signed-off-by:" line...

drivers/i2c/chips/via686a.c static inline u8 FAN_TO_REG(long rpm, in · line 208

Quoted 8 lines
drivers/i2c/chips/via686a.c renamed from drivers/i2c/chips/via686a.c.orig +1 -1
@@ -205,7 +205,7 @@ static inline u8 FAN_TO_REG(long rpm, in
  but the function is very linear in the useful range (0-80 deg C), so 
  we'll just use linear interpolation for 10-bit readings.)  So, tempLUT 
  is the temp at via register values 0-255: */
-static const long tempLUT[] =
+static const int16_t tempLUT[] =

int16_t is not a proper kernel type. Do you really mean s16 instead? Care to redo this?

thanks,

greg k-h

#2 lore raw
Quoted 5 lines · Greg KH
Quoted 1 line · Denis Vlasenko

Not tested, but it's rather obvious.

Except you forgot a "Signed-off-by:" line...

Quoted 10 lines

drivers/i2c/chips/via686a.c static inline u8 FAN_TO_REG(long rpm, in · line 208

Quoted 8 lines
drivers/i2c/chips/via686a.c renamed from drivers/i2c/chips/via686a.c.orig +1 -1
@@ -205,7 +205,7 @@ static inline u8 FAN_TO_REG(long rpm, in
  but the function is very linear in the useful range (0-80 deg C), so 
  we'll just use linear interpolation for 10-bit readings.)  So, tempLUT 
  is the temp at via register values 0-255: */
-static const long tempLUT[] =
+static const int16_t tempLUT[] =

int16_t is not a proper kernel type. Do you really mean s16 instead?

Ok. Please be informed that there are lots of intNN_t's in i2c dir tho...

Quoted 1 line

Care to redo this?

--
vda

drivers/i2c/chips/via686a.c renamed from drivers/i2c/chips/via686a.c.orig +1 -1
@@ -205,7 +205,7 @@ static inline u8 FAN_TO_REG(long rpm, in
  but the function is very linear in the useful range (0-80 deg C), so 
  we'll just use linear interpolation for 10-bit readings.)  So, tempLUT 
  is the temp at via register values 0-255: */
-static const long tempLUT[] =
+static const s16 tempLUT[] =
     { -709, -688, -667, -646, -627, -607, -589, -570, -553, -536, -519,
 	    -503, -487, -471, -456, -442, -428, -414, -400, -387, -375,
 	    -362, -350, -339, -327, -316, -305, -295, -285, -275, -265,
#3 lore raw

Hi Denis,

drivers/i2c/chips/via686a.c static inline u8 FAN_TO_REG(long rpm, in · line 211

Quoted 11 lines
drivers/i2c/chips/via686a.c renamed from drivers/i2c/chips/via686a.c.orig +1 -1
@@ -205,7 +205,7 @@ static inline u8 FAN_TO_REG(long rpm, in
  but the function is very linear in the useful range (0-80 deg C), so 
  we'll just use linear interpolation for 10-bit readings.)  So, tempLUT 
  is the temp at via register values 0-255: */
-static const long tempLUT[] =
+static const s16 tempLUT[] =
     { -709, -688, -667, -646, -627, -607, -589, -570, -553, -536, -519,
 	    -503, -487, -471, -456, -442, -428, -414, -400, -387, -375,
 	    -362, -350, -339, -327, -316, -305, -295, -285, -275, -265,

This patch doesn't apply on top of my stack, first because the hardware monitoring drivers have been moved to drivers/hwmon, second because the via686a driver had indentation cleanups since 2.6.12.

Could you please provide this patch against 2.6.13-mm1?

Thanks,

Signature · 1 line
Jean Delvare
#4 lore raw

PATCH 1/1 hwmon: via686a: save 0.5k by long v[256] -> s16 v[256]

Hi Greg, all,

Quoted 5 lines

This patch doesn't apply on top of my stack, first because the hardware monitoring drivers have been moved to drivers/hwmon, second because the via686a driver had indentation cleanups since 2.6.12.

Could you please provide this patch against 2.6.13-mm1?

On Denis' request, I have done that myself.

------

We can save 0.5kB of data in the via686a driver.

From: Denis Vlasenko <[email protected]>

 drivers/hwmon/via686a.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/hwmon/via686a.c +1 -1
@@ -198,7 +198,7 @@
  but the function is very linear in the useful range (0-80 deg C), so
  we'll just use linear interpolation for 10-bit readings.)  So, tempLUT
  is the temp at via register values 0-255: */
-static const long tempLUT[] =
+static const s16 tempLUT[] =
 { -709, -688, -667, -646, -627, -607, -589, -570, -553, -536, -519,
 	-503, -487, -471, -456, -442, -428, -414, -400, -387, -375,
 	-362, -350, -339, -327, -316, -305, -295, -285, -275, -265,
Signature · 1 line
Jean Delvare
#5 lore raw

Re: [PATCH 2.6] hwmon: via686a: save 0.5k by long v[256] -> s16 v[256]

Quoted 9 lines · Jean Delvare

Hi Greg, all,

Quoted 5 lines

This patch doesn't apply on top of my stack, first because the hardware monitoring drivers have been moved to drivers/hwmon, second because the via686a driver had indentation cleanups since 2.6.12.

Could you please provide this patch against 2.6.13-mm1?

On Denis' request, I have done that myself.

Unfortunatly, no one noticed that this patch adds a build warning :(

So I'm not going to apply it, sorry.

thanks,

greg k-h

#6 lore raw

Re: [PATCH 2.6] hwmon: via686a: save 0.5k by long v[256] -> s16 v[256]

Hi Greg,

Quoted 1 line

Unfortunatly, no one noticed that this patch adds a build warning :(

I'm sorry about that, I though I had checked but now it seems not. A new patch addressing the issue follows.

Thanks.

----------------------

We can save 0.5kB of data in the via686a driver.

From: Denis Vlasenko <[email protected]>

 drivers/hwmon/via686a.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/hwmon/via686a.c +2 -2
@@ -198,7 +198,7 @@
  but the function is very linear in the useful range (0-80 deg C), so
  we'll just use linear interpolation for 10-bit readings.)  So, tempLUT
  is the temp at via register values 0-255: */
-static const long tempLUT[] =
+static const s16 tempLUT[] =
 { -709, -688, -667, -646, -627, -607, -589, -570, -553, -536, -519,
 	-503, -487, -471, -456, -442, -428, -414, -400, -387, -375,
 	-362, -350, -339, -327, -316, -305, -295, -285, -275, -265,
@@ -270,7 +270,7 @@
 }
 
 /* for 8-bit temperature hyst and over registers */
-#define TEMP_FROM_REG(val) (tempLUT[(val)] * 100)
+#define TEMP_FROM_REG(val)	((long)tempLUT[val] * 100)
 
 /* for 10-bit temperature readings */
 static inline long TEMP_FROM_REG10(u16 val)
Signature · 1 line
Jean Delvare
#7 lore raw
In reply to Denis Vlasenko

Hi Denis,

BTW...

Quoted 2 lines

Please be informed that there are lots of intNN_t's in i2c dir tho...

I couldn't find any. What were you refering to exactly?

Thanks,

Signature · 1 line
Jean Delvare
#8 lore raw
Quoted 8 lines · Jean Delvare

Hi Denis,

BTW...

Quoted 2 lines

Please be informed that there are lots of intNN_t's in i2c dir tho...

I couldn't find any. What were you refering to exactly?

Sorry I was wrong. While kernel has ~15000 [u]intNN_t's they are all _not_ in i2c.
--
vda