Skip to contentSkip to conversation

x86: coding style fixes to arch/x86/math-emu/reg_constant

PATCH Applied Paolo Ciarrocchi 5 participants 7 messages Jun 14, 2008 – Jun 19, 2008 x86

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 x86: coding style fixes to arch/x86/math-emu/reg_constant

Before:
total: 6 errors, 1 warnings, 117 lines checked

After:
total: 0 errors, 1 warnings, 117 lines checked

paolo@paolo-desktop:~/linux.trees.git$ md5sum /tmp/reg_constant.o.*
780388a3056d58fb759efaf190d5d3d1  /tmp/reg_constant.o.after
780388a3056d58fb759efaf190d5d3d1  /tmp/reg_constant.o.before
paolo@paolo-desktop:~/linux.trees.git$ size /tmp/reg_constant.o.*
   text    data     bss     dec     hex filename
    457       0       0     457     1c9 /tmp/reg_constant.o.after
    457       0       0     457     1c9 /tmp/reg_constant.o.before
---
 arch/x86/math-emu/reg_constant.c |    8 ++++----
 1 files changed, 4 insertions(+), 4 deletions(-)
arch/x86/math-emu/reg_constant.c +4 -4
index 04869e6..0054835 100644
@@ -16,8 +16,8 @@
 #include "reg_constant.h"
 #include "control_w.h"
 
-#define MAKE_REG(s,e,l,h) { l, h, \
-                            ((EXTENDED_Ebias+(e)) | ((SIGN_##s != 0)*0x8000)) }
+#define MAKE_REG(s, e, l, h) { l, h, \
+		((EXTENDED_Ebias+(e)) | ((SIGN_##s != 0)*0x8000)) }
 
 FPU_REG const CONST_1 = MAKE_REG(POS, 0, 0x00000000, 0x80000000);
 #if 0
@@ -40,7 +40,7 @@ FPU_REG const CONST_PI2extra = MAKE_REG(NEG, -66,
 FPU_REG const CONST_Z = MAKE_REG(POS, EXP_UNDER, 0x0, 0x0);
 
 /* Only the sign and significand (and tag) are used in internal NaNs */
-/* The 80486 never generates one of these 
+/* The 80486 never generates one of these
 FPU_REG const CONST_SNAN = MAKE_REG(POS, EXP_OVER, 0x00000001, 0x80000000);
  */
 /* This is the real indefinite QNaN */
@@ -49,7 +49,7 @@ FPU_REG const CONST_QNaN = MAKE_REG(NEG, EXP_OVER, 0x00000000, 0xC0000000);
 /* Only the sign (and tag) is used in internal infinities */
 FPU_REG const CONST_INF = MAKE_REG(POS, EXP_OVER, 0x00000000, 0x80000000);
 
-static void fld_const(FPU_REG const *c, int adj, u_char tag)
+static void fld_const(FPU_REG const * c, int adj, u_char tag)
 {
 	FPU_REG *st_new_ptr;
 
Signature · 1 line
1.5.6.rc1.21.g03300
Applied
#1 lore raw
Quoted 5 lines · Paolo Ciarrocchi

Before:
total: 6 errors, 1 warnings, 117 lines checked

After:
total: 0 errors, 1 warnings, 117 lines checked

applied to tip/x86/cleanups, thanks.

Ingo

#2 lore raw
In reply to Paolo Ciarrocchi

arch/x86/math-emu/reg_constant.c FPU_REG const CONST_QNaN = MAKE_REG(NEG, EXP_OVER, 0x00000000, 0xC0000000); · line 52

Quoted 3 lines · Paolo Ciarrocchi
arch/x86/math-emu/reg_constant.c +1 -1
-static void fld_const(FPU_REG const *c, int adj, u_char tag)
+static void fld_const(FPU_REG const * c, int adj, u_char tag)

________________________________________^

The above does not seem to me like a big improvement in code quality. A false positive from checkpatch.pl?

Dmitri

Quoted 3 lines

{
FPU_REG *st_new_ptr;

#3 lore raw
Quoted 8 lines · Dmitri Vorobiev

arch/x86/math-emu/reg_constant.c FPU_REG const CONST_QNaN = MAKE_REG(NEG, EXP_OVER, 0x00000000, 0xC0000000); · line 52

Quoted 3 lines · Paolo Ciarrocchi
arch/x86/math-emu/reg_constant.c +1 -1
-static void fld_const(FPU_REG const *c, int adj, u_char tag)
+static void fld_const(FPU_REG const * c, int adj, u_char tag)

________________________________________^

The above does not seem to me like a big improvement in code quality. A false positive from checkpatch.pl?

Yes, I think you are right.
Andy CC'ed so that he can have a look at that.

Thanks.

regards,

Signature · 2 lines
Paolo
http://paolo.ciarrocchi.googlepages.com/
#4 lore raw
Quoted 10 lines
Quoted 8 lines · Dmitri Vorobiev

arch/x86/math-emu/reg_constant.c FPU_REG const CONST_QNaN = MAKE_REG(NEG, EXP_OVER, 0x00000000, 0xC0000000); · line 52

Quoted 3 lines · Paolo Ciarrocchi
arch/x86/math-emu/reg_constant.c +1 -1
-static void fld_const(FPU_REG const *c, int adj, u_char tag)
+static void fld_const(FPU_REG const * c, int adj, u_char tag)

________________________________________^

The above does not seem to me like a big improvement in code quality. A false positive from checkpatch.pl?

It should obviously be

static void fld_const(const FPU_REG *c, int adj, u_char tag)

FPU_REG is not an appropriate type name though. Is "typedef struct fpu__reg FPU_REG;" necessary at all? (Removing it would create some code churn all over x86/math-emu/ though. Maybe not desirable.)

Signature · 3 lines
Stefan Richter
-=====-==--- -==- =--=-
http://arcgraph.de/sr/
#5 lore raw
In reply to Paolo Ciarrocchi
Quoted 8 lines · Paolo Ciarrocchi
Quoted 6 lines · Dmitri Vorobiev

arch/x86/math-emu/reg_constant.c FPU_REG const CONST_QNaN = MAKE_REG(NEG, EXP_OVER, 0x00000000, 0xC0000000); · line 52

Quoted 3 lines · Paolo Ciarrocchi
arch/x86/math-emu/reg_constant.c +1 -1
-static void fld_const(FPU_REG const *c, int adj, u_char tag)
+static void fld_const(FPU_REG const * c, int adj, u_char tag)

________________________________________^

Quoted 5 lines
Quoted 2 lines

The above does not seem to me like a big improvement in code quality. A false positive from checkpatch.pl?

Yes, I think you are right.
Andy CC'ed so that he can have a look at that.

That would be a false positive if we are recommending that. I have pushed this fragment in isolation through all versions back to 0.15 and none of them recommend it.

Ahh, but if I push the whole patch through 0.19 spits it out. Ok this is a subtle bug in modifier procesing and is fixed in the head of my tree. Will be fixed in the next batch which goes up to -mm.

-apw

#6 lore raw
Quoted 13 lines · Andy Whitcroft
Quoted 11 lines · Paolo Ciarrocchi
Quoted 7 lines · Dmitri Vorobiev

arch/x86/math-emu/reg_constant.c FPU_REG const CONST_QNaN = MAKE_REG(NEG, EXP_OVER, 0x00000000, 0xC0000000); · line 52

Quoted 2 lines · Paolo Ciarrocchi
arch/x86/math-emu/reg_constant.c +1 -1
-static void fld_const(FPU_REG const *c, int adj, u_char tag)
+static void fld_const(FPU_REG const * c, int adj, u_char tag)

________________________________________^

The above does not seem to me like a big improvement in code quality. A false positive from checkpatch.pl?

Yes, I think you are right.
Andy CC'ed so that he can have a look at that.

Quoted 7 lines

That would be a false positive if we are recommending that. I have pushed this fragment in isolation through all versions back to 0.15 and none of them recommend it.

Ahh, but if I push the whole patch through 0.19 spits it out. Ok this is a subtle bug in modifier procesing and is fixed in the head of my tree. Will be fixed in the next batch which goes up to -mm.

Maybe it makes sense for Ingo to revert the original Paolo's patch now?

Dmitri