Skip to contentSkip to conversation

checkpatch: Add support for floating point constants

PATCH In review Joe Perches 3 participants 8 messages Nov 1, 2012 – Nov 7, 2012 Checkpatch

lore
Share & tools
lore

Joe Perches

author maintainer
#0 lore raw

v2 cover checkpatch: Add support for floating point constants

The kernel doesn't support them though.

v2: Make it clearer that checkpatch understands floating point constants
Support floating point hex constants too

Joe Perches (3):
checkpatch: Find hex constants as a single IDENT
checkpatch: Add support for floating point constants
checkpatch: Emit an warning when floating point values are used

 scripts/checkpatch.pl |   13 ++++++++++++-
 1 files changed, 12 insertions(+), 1 deletions(-)
Signature · 1 line
1.7.8.112.g3fd21
author maintainer
#1 lore raw

PATCH v2 1/3 checkpatch: Find hex constants as a single IDENT

Hexadecimal values are current found in 2 parts.
A hex constant like 0x123456abcdef is found as
0 and then x123456abcdef and later coalesced.

Instead, reverse the order of the 2 searches in
$Constant to find 0x first, then 0 so that the
entire hex constant is found all at once.

Signed-off-by Joe Perches <[email protected]>
---
 scripts/checkpatch.pl |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
scripts/checkpatch.pl +1 -1
index f18750e..099a0ad 100755
@@ -227,7 +227,7 @@ our $Inline	= qr{inline|__always_inline|noinline};
 our $Member	= qr{->$Ident|\.$Ident|\[[^]]*\]};
 our $Lval	= qr{$Ident(?:$Member)*};
 
-our $Constant	= qr{(?i:(?:[0-9]+|0x[0-9a-f]+)[ul]*)};
+our $Constant	= qr{(?i:(?:0x[0-9a-f]+|[0-9]+)[ul]*)};
 our $Assignment	= qr{(?:\*\=|/=|%=|\+=|-=|<<=|>>=|&=|\^=|\|=|=)};
 our $Compare    = qr{<=|>=|==|!=|<|>};
 our $Operators	= qr{
Signature · 1 line
1.7.8.112.g3fd21
author maintainer
#2 lore raw

PATCH v2 2/3 checkpatch: Add support for floating point constants

In reply to Joe Perches

Even though the kernel doesn't support using floating point constants, add a regex for them.

Support forms like: 0x123p1, 123e-1, 1.23, 1.5e23f

Signed-off-by Joe Perches <[email protected]>
---
 scripts/checkpatch.pl |    6 +++++-
 1 files changed, 5 insertions(+), 1 deletions(-)
scripts/checkpatch.pl +5 -1
index 099a0ad..c3a2162 100755
@@ -227,7 +227,11 @@ our $Inline	= qr{inline|__always_inline|noinline};
 our $Member	= qr{->$Ident|\.$Ident|\[[^]]*\]};
 our $Lval	= qr{$Ident(?:$Member)*};
 
-our $Constant	= qr{(?i:(?:0x[0-9a-f]+|[0-9]+)[ul]*)};
+our $Float_hex	= qr{(?i:0x[0-9a-f]+p-?[0-9]+[fl]?)};
+our $Float_dec	= qr{(?i:((?:[0-9]+\.[0-9]*|[0-9]*\.[0-9]+)(?:e-?[0-9]+)?[fl]?))};
+our $Float_int	= qr{(?i:[0-9]+e-?[0-9]+[fl]?)};
+our $Float	= qr{$Float_hex|$Float_dec|$Float_int};
+our $Constant	= qr{(?:$Float|(?i:(?:0x[0-9a-f]+|[0-9]+)[ul]*))};
 our $Assignment	= qr{(?:\*\=|/=|%=|\+=|-=|<<=|>>=|&=|\^=|\|=|=)};
 our $Compare    = qr{<=|>=|==|!=|<|>};
 our $Operators	= qr{
Signature · 1 line
1.7.8.112.g3fd21
author maintainer
#3 lore raw

PATCH v2 3/3 checkpatch: Emit an warning when floating point values are used

In reply to Joe Perches

Linux kernel doesn't like floating point, say so.

Signed-off-by Joe Perches <[email protected]>
---
 scripts/checkpatch.pl |    7 +++++++
 1 files changed, 7 insertions(+), 0 deletions(-)
scripts/checkpatch.pl +7 -0
index c3a2162..96685c6 100755
@@ -2332,6 +2332,13 @@ sub process {
 			     "do not add new typedefs\n" . $herecurr);
 		}
 
+# check for floating point constants
+
+		if ($line =~ /\b$Float\b/) {
+			WARN("KERNEL_FLOAT",
+			     "Floating point is not supported in linux kernel source\n" . $herecurr);
+		}
+
 # * goes on variable not on type
 		# (char*[ const])
 		while ($line =~ m{(\($NonptrType(\s*(?:$Modifier\b\s*|\*\s*)+)\))}g) {
Signature · 1 line
1.7.8.112.g3fd21
#4 lore raw

Re: [PATCH V2 3/3] checkpatch: Emit an warning when floating point values are used

Quoted 10 lines · On Thu, 1 Nov 2012 00:12:18 -0700 Joe Perches <[email protected]> wrote:

Linux kernel doesn't like floating point, say so.

...

scripts/checkpatch.pl +0 -0
@@ -2332,6 +2332,13 @@ sub process {
 			     "do not add new typedefs\n" . $herecurr);
 		}

scripts/checkpatch.pl sub process { · line 2344

Quoted 10 lines
scripts/checkpatch.pl +7 -0
+# check for floating point constants
+
+		if ($line =~ /\b$Float\b/) {
+			WARN("KERNEL_FLOAT",
+			     "Floating point is not supported in linux kernel source\n" . $herecurr);
+		}
+
 # * goes on variable not on type
 		# (char*[ const])
 		while ($line =~ m{(\($NonptrType(\s*(?:$Modifier\b\s*|\*\s*)+)\))}g) {

The earlier review comments were not addressed.

In particular, I don't see a problem with people doing

	int foo = 1.1 * 2.2;
maintainer
#5 lore raw

Re: [PATCH V2 3/3] checkpatch: Emit an warning when floating point values are used

Quoted 26 lines · Andrew Morton
Quoted 10 lines · On Thu, 1 Nov 2012 00:12:18 -0700 Joe Perches <[email protected]> wrote:

Linux kernel doesn't like floating point, say so.

...

scripts/checkpatch.pl +0 -0
@@ -2332,6 +2332,13 @@ sub process {
 			     "do not add new typedefs\n" . $herecurr);
 		}

scripts/checkpatch.pl sub process { · line 2344

Quoted 10 lines
scripts/checkpatch.pl +7 -0
+# check for floating point constants
+
+		if ($line =~ /\b$Float\b/) {
+			WARN("KERNEL_FLOAT",
+			     "Floating point is not supported in linux kernel source\n" . $herecurr);
+		}
+
 # * goes on variable not on type
 		# (char*[ const])
 		while ($line =~ m{(\($NonptrType(\s*(?:$Modifier\b\s*|\*\s*)+)\))}g) {

The earlier review comments were not addressed.

Quoted 3 lines

In particular, I don't see a problem with people doing

	int foo = 1.1 * 2.2;

I thought there was also now a way to enable floats in kernel space and some use thereof.

-apw

author maintainer
#6 lore raw

Re: [PATCH V2 3/3] checkpatch: Emit an warning when floating point values are used

In reply to Andrew Morton
Quoted 3 lines · Andrew Morton
Quoted 1 line · On Thu, 1 Nov 2012 00:12:18 -0700 Joe Perches <[email protected]> wrote:

Linux kernel doesn't like floating point, say so.

[]

Quoted 6 lines
Quoted 6 lines
scripts/checkpatch.pl +0 -0
@@ -2332,6 +2332,13 @@ sub process {
 			     "do not add new typedefs\n" . $herecurr);
 		}
Quoted 12 lines

scripts/checkpatch.pl sub process { · line 2344

Quoted 10 lines
scripts/checkpatch.pl +7 -0
+# check for floating point constants
+
+		if ($line =~ /\b$Float\b/) {
+			WARN("KERNEL_FLOAT",
+			     "Floating point is not supported in linux kernel source\n" . $herecurr);
+		}
+
 # * goes on variable not on type
 		# (char*[ const])
 		while ($line =~ m{(\($NonptrType(\s*(?:$Modifier\b\s*|\*\s*)+)\))}g) {

The earlier review comments were not addressed.

You didn't read my response in the thread then. I "addressed" it by saying that the problem exists without solution.

Quoted 3 lines

In particular, I don't see a problem with people doing

	int foo = 1.1 * 2.2;

Neither do I but I know of no way to determine that any calc is a constant.

#7 lore raw

Re: [PATCH V2 3/3] checkpatch: Emit an warning when floating point values are used

Quoted 24 lines · On Wed, 07 Nov 2012 03:09:35 -0800 Joe Perches <[email protected]> wrote:
Quoted 3 lines · Andrew Morton
Quoted 1 line · On Thu, 1 Nov 2012 00:12:18 -0700 Joe Perches <[email protected]> wrote:

Linux kernel doesn't like floating point, say so.

[]

Quoted 6 lines
Quoted 6 lines
scripts/checkpatch.pl +0 -0
@@ -2332,6 +2332,13 @@ sub process {
 			     "do not add new typedefs\n" . $herecurr);
 		}
Quoted 12 lines

scripts/checkpatch.pl sub process { · line 2344

Quoted 10 lines
scripts/checkpatch.pl +7 -0
+# check for floating point constants
+
+		if ($line =~ /\b$Float\b/) {
+			WARN("KERNEL_FLOAT",
+			     "Floating point is not supported in linux kernel source\n" . $herecurr);
+		}
+
 # * goes on variable not on type
 		# (char*[ const])
 		while ($line =~ m{(\($NonptrType(\s*(?:$Modifier\b\s*|\*\s*)+)\))}g) {

The earlier review comments were not addressed.

Quoted 1 line

You didn't read my response in the thread then.

There was no response afaict. Not here, not on lkml.org?