Skip to contentSkip to conversation

update checkpatch to v0.29

PATCH In review Andy Whitcroft 4 participants 11 messages Aug 21, 2009 – Aug 25, 2009

Subsystems (6)
lore
Share & tools
lore

Andy Whitcroft

author maintainer
#0 lore raw

cover update checkpatch to v0.29

Here is a few update for checkpatch that seem to have gotten lost somewhere along the line, I though you had everything I had queued. These have passed testing here, I have some more stuff in testing and I will get those out soon.

This update brings an a number of minor fixes. Thanks for the contributions.

Complete changelog below.

-apw

Andy Whitcroft (5):
checkpatch: possible types -- else cannot start a type
checkpatch: indent checks -- stop when we run out of continuation
lines
checkpatch: format strings should not have brackets in macros
checkpatch: limit sN/uN matches to actual bit sizes
checkpatch: version 0.29

Daniel Walker (1):
checkpatch: handle C99 comments correctly (performance issue)

Hannes Eder (1):
checkpatch: make -f alias --file, add --help, more verbose help
message

 scripts/checkpatch.pl |   84 ++++++++++++++++++++++++++++++++++++++-----------
 1 files changed, 65 insertions(+), 19 deletions(-)
author maintainer
#1 lore raw

PATCH 1/7 checkpatch: possible types -- else cannot start a type

An else cannot start a type, it would have to be within a block after the else. This can trigger false modifier matching.

---
 scripts/checkpatch.pl |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)
scripts/checkpatch.pl +2 -0
index 2d5ece7..fd64816 100755
@@ -1372,6 +1372,8 @@ sub process {
 			# Ignore functions being called
 			} elsif ($s =~ /^.\s*$Ident\s*\(/s) {
 
+			} elsif ($s =~ /^.\s*else\b/s) {
+
 			# declarations always start with types
 			} elsif ($prev_values eq 'E' && $s =~ /^.\s*(?:$Storage\s+)?(?:$Inline\s+)?(?:const\s+)?((?:\s*$Ident)+?)\b(?:\s+$Sparse)?\s*\**\s*(?:$Ident|\(\*[^\)]*\))(?:\s*$Modifier)?\s*(?:;|=|,|\()/s) {
 				my $type = $1;
Signature · 1 line
1.6.3.3
author maintainer
#2 lore raw

PATCH 2/7 checkpatch: handle C99 comments correctly (performance issue)

In reply to Andy Whitcroft

From: Daniel Walker <[email protected]>

This fixes the sanitation process in checkpatch.pl so that it blocks out the text after a C99 style comment the same way it does with block style comments. This prevents the text from getting processed as regular code.

---
 scripts/checkpatch.pl |   13 +++++++++++++
 1 files changed, 13 insertions(+), 0 deletions(-)
scripts/checkpatch.pl +13 -0
index fd64816..aa009a3 100755
@@ -356,6 +356,13 @@ sub sanitise_line {
 			$off++;
 			next;
 		}
+		if ($sanitise_quote eq '' && substr($line, $off, 2) eq '//') {
+			$sanitise_quote = '//';
+
+			substr($res, $off, 2, $sanitise_quote);
+			$off++;
+			next;
+		}
 
 		# A \ in a string means ignore the next character.
 		if (($sanitise_quote eq "'" || $sanitise_quote eq '"') &&
@@ -379,6 +386,8 @@ sub sanitise_line {
 		#print "c<$c> SQ<$sanitise_quote>\n";
 		if ($off != 0 && $sanitise_quote eq '*/' && $c ne "\t") {
 			substr($res, $off, 1, $;);
+		} elsif ($off != 0 && $sanitise_quote eq '//' && $c ne "\t") {
+			substr($res, $off, 1, $;);
 		} elsif ($off != 0 && $sanitise_quote && $c ne "\t") {
 			substr($res, $off, 1, 'X');
 		} else {
@@ -386,6 +395,10 @@ sub sanitise_line {
 		}
 	}
 
+	if ($sanitise_quote eq '//') {
+		$sanitise_quote = '';
+	}
+
 	# The pathname on a #include may be surrounded by '<' and '>'.
 	if ($res =~ /^.\s*\#\s*include\s+\<(.*)\>/) {
 		my $clean = 'X' x length($1);
Signature · 1 line
1.6.3.3
author maintainer
#3 lore raw

PATCH 3/7 checkpatch: indent checks -- stop when we run out of continuation lines

In reply to Andy Whitcroft

Ensure we terminate when there are no futher continuation lines when trying to determine relative indent of conditionals and their blocks.

---
 scripts/checkpatch.pl |    5 +++--
 1 files changed, 3 insertions(+), 2 deletions(-)
scripts/checkpatch.pl +3 -2
index aa009a3..b6f267b 100755
@@ -1547,8 +1547,9 @@ sub process {
 				    $s =~ /^\s*#\s*?/ ||
 				    $s =~ /^\s*$Ident\s*:/) {
 					$continuation = ($s =~ /^.*?\\\n/) ? 1 : 0;
-					$s =~ s/^.*?\n//;
-					$cond_lines++;
+					if ($s =~ s/^.*?\n//) {
+						$cond_lines++;
+					}
 				}
 			}
 
Signature · 1 line
1.6.3.3
author maintainer
#4 lore raw

PATCH 4/7 checkpatch: make -f alias --file, add --help, more verbose help message

In reply to Andy Whitcroft

From: Hannes Eder <[email protected]>

Impact:
- More verbose help/usage message.
- Make the option -f an alias for --file.
- On -h, --help, and --version display help message and exit(0).
- With no FILE(s) given, exit(1) with "no input files".
- On invalid options display help/usage and exit(1).

Based on a patch by Pavel Machek.

---
 scripts/checkpatch.pl |   55 +++++++++++++++++++++++++++++++++++++-----------
 1 files changed, 42 insertions(+), 13 deletions(-)
scripts/checkpatch.pl +42 -13
index b6f267b..e3c6b49 100755
@@ -28,6 +28,41 @@ my $mailback = 0;
 my $summary_file = 0;
 my $root;
 my %debug;
+my $help = 0;
+
+sub help {
+	my ($exitcode) = @_;
+
+	print << "EOM";
+Usage: $P [OPTION]... [FILE]...
+Version: $V
+
+Options:
+  -q, --quiet                quiet
+  --no-tree                  run without a kernel tree
+  --no-signoff               do not check for 'Signed-off-by' line
+  --patch                    treat FILE as patchfile (default)
+  --emacs                    emacs compile window format
+  --terse                    one line per report
+  -f, --file                 treat FILE as regular source file
+  --subjective, --strict     enable more subjective tests
+  --root=PATH                PATH to the kernel tree root
+  --no-summary               suppress the per-file summary
+  --mailback                 only produce a report in case of warnings/errors
+  --summary-file             include the filename in summary
+  --debug KEY=[0|1]          turn on/off debugging of KEY, where KEY is one of
+                             'values', 'possible', 'type', and 'attr' (default
+                             is all off)
+  --test-only=WORD           report only warnings/errors containing WORD
+                             literally
+  -h, --help, --version      display this help and exit
+
+When FILE is - read standard input.
+EOM
+
+	exit($exitcode);
+}
+
 GetOptions(
 	'q|quiet+'	=> \$quiet,
 	'tree!'		=> \$tree,
@@ -35,7 +70,7 @@ GetOptions(
 	'patch!'	=> \$chk_patch,
 	'emacs!'	=> \$emacs,
 	'terse!'	=> \$terse,
-	'file!'		=> \$file,
+	'f|file!'	=> \$file,
 	'subjective!'	=> \$check,
 	'strict!'	=> \$check,
 	'root=s'	=> \$root,
@@ -45,22 +80,16 @@ GetOptions(
 
 	'debug=s'	=> \%debug,
 	'test-only=s'	=> \$tst_only,
-) or exit;
+	'h|help'	=> \$help,
+	'version'	=> \$help
+) or help(1);
+
+help(0) if ($help);
 
 my $exit = 0;
 
 if ($#ARGV < 0) {
-	print "usage: $P [options] patchfile\n";
-	print "version: $V\n";
-	print "options: -q               => quiet\n";
-	print "         --no-tree        => run without a kernel tree\n";
-	print "         --terse          => one line per report\n";
-	print "         --emacs          => emacs compile window format\n";
-	print "         --file           => check a source file\n";
-	print "         --strict         => enable more subjective tests\n";
-	print "         --root           => path to the kernel tree root\n";
-	print "         --no-summary     => suppress the per-file summary\n";
-	print "         --summary-file   => include the filename in summary\n";
+	print "$P: no input files\n";
 	exit(1);
 }
 
Signature · 1 line
1.6.3.3
Acked-by on v1 4/7
#5 lore raw

Re: [PATCH 4/7] checkpatch: make -f alias --file, add --help, more verbose help message

Quoted 9 lines · Andy Whitcroft

From: Hannes Eder <[email protected]>

Impact:
- More verbose help/usage message.
- Make the option -f an alias for --file.
- On -h, --help, and --version display help message and exit(0).
- With no FILE(s) given, exit(1) with "no input files".
- On invalid options display help/usage and exit(1).

Quoted 3 lines

Based on a patch by Pavel Machek.

Quoted 1 line

Thanks!

Quoted 88 lines
---
 scripts/checkpatch.pl |   55 +++++++++++++++++++++++++++++++++++++-----------
 1 files changed, 42 insertions(+), 13 deletions(-)
scripts/checkpatch.pl +42 -13
index b6f267b..e3c6b49 100755
@@ -28,6 +28,41 @@ my $mailback = 0;
 my $summary_file = 0;
 my $root;
 my %debug;
+my $help = 0;
+
+sub help {
+	my ($exitcode) = @_;
+
+	print << "EOM";
+Usage: $P [OPTION]... [FILE]...
+Version: $V
+
+Options:
+  -q, --quiet                quiet
+  --no-tree                  run without a kernel tree
+  --no-signoff               do not check for 'Signed-off-by' line
+  --patch                    treat FILE as patchfile (default)
+  --emacs                    emacs compile window format
+  --terse                    one line per report
+  -f, --file                 treat FILE as regular source file
+  --subjective, --strict     enable more subjective tests
+  --root=PATH                PATH to the kernel tree root
+  --no-summary               suppress the per-file summary
+  --mailback                 only produce a report in case of warnings/errors
+  --summary-file             include the filename in summary
+  --debug KEY=[0|1]          turn on/off debugging of KEY, where KEY is one of
+                             'values', 'possible', 'type', and 'attr' (default
+                             is all off)
+  --test-only=WORD           report only warnings/errors containing WORD
+                             literally
+  -h, --help, --version      display this help and exit
+
+When FILE is - read standard input.
+EOM
+
+	exit($exitcode);
+}
+
 GetOptions(
 	'q|quiet+'	=> \$quiet,
 	'tree!'		=> \$tree,
@@ -35,7 +70,7 @@ GetOptions(
 	'patch!'	=> \$chk_patch,
 	'emacs!'	=> \$emacs,
 	'terse!'	=> \$terse,
-	'file!'		=> \$file,
+	'f|file!'	=> \$file,
 	'subjective!'	=> \$check,
 	'strict!'	=> \$check,
 	'root=s'	=> \$root,
@@ -45,22 +80,16 @@ GetOptions(
 
 	'debug=s'	=> \%debug,
 	'test-only=s'	=> \$tst_only,
-) or exit;
+	'h|help'	=> \$help,
+	'version'	=> \$help
+) or help(1);
+
+help(0) if ($help);
 
 my $exit = 0;
 
 if ($#ARGV < 0) {
-	print "usage: $P [options] patchfile\n";
-	print "version: $V\n";
-	print "options: -q               => quiet\n";
-	print "         --no-tree        => run without a kernel tree\n";
-	print "         --terse          => one line per report\n";
-	print "         --emacs          => emacs compile window format\n";
-	print "         --file           => check a source file\n";
-	print "         --strict         => enable more subjective tests\n";
-	print "         --root           => path to the kernel tree root\n";
-	print "         --no-summary     => suppress the per-file summary\n";
-	print "         --summary-file   => include the filename in summary\n";
+	print "$P: no input files\n";
 	exit(1);
 }
 
Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
author maintainer
#6 lore raw

PATCH 5/7 checkpatch: format strings should not have brackets in macros

In reply to Andy Whitcroft

We should not recommend braces for the following:

    #define pr_fmt(fmt)    "%s: " fmt, __func__

allow things with double quotes round them to avoid this check.

---
 scripts/checkpatch.pl |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)
scripts/checkpatch.pl +2 -1
index e3c6b49..57df906 100755
@@ -2288,7 +2288,8 @@ sub process {
 				DECLARE_PER_CPU|
 				DEFINE_PER_CPU|
 				__typeof__\(|
-				\.$Ident\s*=\s*
+				\.$Ident\s*=\s*|
+				^\"|\"$
 			}x;
 			#print "REST<$rest> dstat<$dstat>\n";
 			if ($rest ne '') {
Signature · 1 line
1.6.3.3
author maintainer
#7 lore raw

PATCH 6/7 checkpatch: limit sN/uN matches to actual bit sizes

In reply to Andy Whitcroft

Limit our type matcher to the s/u/le/be etc sizes that actually exist to prevent miss categorising s2 as a type. Fix up the spelling of the error also.

---
 scripts/checkpatch.pl |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)
scripts/checkpatch.pl +2 -2
index 57df906..7fee823 100755
@@ -182,7 +182,7 @@ our $UTF8	= qr {
 }x;
 
 our $typeTypedefs = qr{(?x:
-	(?:__)?(?:u|s|be|le)(?:\d|\d\d)|
+	(?:__)?(?:u|s|be|le)(?:8|16|32|64)|
 	atomic_t
 )};
 
@@ -1936,7 +1936,7 @@ sub process {
 						# A unary '*' may be const
 
 					} elsif ($ctx =~ /.xW/) {
-						ERROR("Aspace prohibited after that '$op' $at\n" . $hereptr);
+						ERROR("space prohibited after that '$op' $at\n" . $hereptr);
 					}
 
 				# unary ++ and unary -- are allowed no space on one side.
Signature · 1 line
1.6.3.3
author maintainer
#8 lore raw

PATCH 7/7 checkpatch: version 0.29

In reply to Andy Whitcroft
---
 scripts/checkpatch.pl |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
scripts/checkpatch.pl +1 -1
index 7fee823..8b1dfd5 100755
@@ -10,7 +10,7 @@ use strict;
 my $P = $0;
 $P =~ s@.*/@@g;
 
-my $V = '0.28';
+my $V = '0.29';
 
 use Getopt::Long qw(:config no_auto_abbrev);
 
Signature · 1 line
1.6.3.3
maintainer Reviewed-by KOSAKI Motohiro Acked-by Balbir Singh Reviewed-by Minchan Kim on the v1 series
#9 lore raw
In reply to Andy Whitcroft
Quoted 7 lines · On Fri, 21 Aug 2009 22:46:03 +0100 Andy Whitcroft <[email protected]> wrote:

Here is a few update for checkpatch that seem to have gotten lost somewhere along the line, I though you had everything I had queued. These have passed testing here, I have some more stuff in testing and I will get those out soon.

This update brings an a number of minor fixes. Thanks for the contributions.

Running checkpatch 0.29 against the below diff produces a warning spew:

Use of uninitialized value in length at scripts/checkpatch.pl line 1480. Use of uninitialized value in substr at scripts/checkpatch.pl line 1480. Use of uninitialized value in substr at scripts/checkpatch.pl line 1480. Use of uninitialized value in pattern match (m//) at scripts/checkpatch.pl line 1488. Use of uninitialized value in pattern match (m//) at scripts/checkpatch.pl line 1545. Use of uninitialized value in length at scripts/checkpatch.pl line 1480. Use of uninitialized value in substr at scripts/checkpatch.pl line 1480. Use of uninitialized value in substr at scripts/checkpatch.pl line 1480. Use of uninitialized value in pattern match (m//) at scripts/checkpatch.pl line 1488. Use of uninitialized value in pattern match (m//) at scripts/checkpatch.pl line 1545.
total: 0 errors, 0 warnings, 72 lines checked

From: Wu Fengguang <[email protected]>

For mem_cgroup, shrink_zone() may call shrink_list() with nr_to_scan=1, in which case shrink_list() _still_ calls isolate_pages() with the much larger SWAP_CLUSTER_MAX. It effectively scales up the inactive list scan rate by up to 32 times.

For example, with 16k inactive pages and DEF_PRIORITY=12, (16k >> 12)=4. So when shrink_zone() expects to scan 4 pages in the active/inactive list, the active list will be scanned 4 pages, while the inactive list will be (over) scanned SWAP_CLUSTER_MAX=32 pages in effect. And that could break the balance between the two lists.

It can further impact the scan of anon active list, due to the anon active/inactive ratio rebalance logic in balance_pgdat()/shrink_zone():

inactive anon list over scanned => inactive_anon_is_low() == TRUE
=> shrink_active_list() => active anon list over scanned

So the end result may be

- anon inactive  => over scanned
- anon active    => over scanned (maybe not as much)
- file inactive  => over scanned
- file active    => under scanned (relatively)

The accesses to nr_saved_scan are not lock protected and so not 100% accurate, however we can tolerate small errors and the resulted small imbalanced scan rates between zones.

---

 include/linux/mmzone.h |    6 +++++-
 mm/page_alloc.c        |    2 +-
 mm/vmscan.c            |   20 +++++++++++---------
 3 files changed, 17 insertions(+), 11 deletions(-)
include/linux/mmzone.h renamed from include/linux/mmzone.h~mm-do-batched-scans-for-mem_cgroup +5 -1
diff -puN include/linux/mmzone.h~mm-do-batched-scans-for-mem_cgroup include/linux/mmzone.h
@@ -273,6 +273,11 @@ struct zone_reclaim_stat {
 	 */
 	unsigned long		recent_rotated[2];
 	unsigned long		recent_scanned[2];
+
+	/*
+	 * accumulated for batching
+	 */
+	unsigned long		nr_saved_scan[NR_LRU_LISTS];
 };
 
 struct zone {
@@ -327,7 +332,6 @@ struct zone {
 	spinlock_t		lru_lock;	
 	struct zone_lru {
 		struct list_head list;
-		unsigned long nr_saved_scan;	/* accumulated for batching */
 	} lru[NR_LRU_LISTS];
 
 	struct zone_reclaim_stat reclaim_stat;
mm/page_alloc.c renamed from mm/page_alloc.c~mm-do-batched-scans-for-mem_cgroup +1 -1
diff -puN mm/page_alloc.c~mm-do-batched-scans-for-mem_cgroup mm/page_alloc.c
@@ -3830,7 +3830,7 @@ static void __paginginit free_area_init_
 		zone_pcp_init(zone);
 		for_each_lru(l) {
 			INIT_LIST_HEAD(&zone->lru[l].list);
-			zone->lru[l].nr_saved_scan = 0;
+			zone->reclaim_stat.nr_saved_scan[l] = 0;
 		}
 		zone->reclaim_stat.recent_rotated[0] = 0;
 		zone->reclaim_stat.recent_rotated[1] = 0;
mm/vmscan.c renamed from mm/vmscan.c~mm-do-batched-scans-for-mem_cgroup +11 -9
diff -puN mm/vmscan.c~mm-do-batched-scans-for-mem_cgroup mm/vmscan.c
@@ -1586,6 +1586,7 @@ static void shrink_zone(int priority, st
 	enum lru_list l;
 	unsigned long nr_reclaimed = sc->nr_reclaimed;
 	unsigned long swap_cluster_max = sc->swap_cluster_max;
+	struct zone_reclaim_stat *reclaim_stat = get_reclaim_stat(zone, sc);
 	int noswap = 0;
 
 	/* If we have no swap space, do not bother scanning anon pages. */
@@ -1605,12 +1606,9 @@ static void shrink_zone(int priority, st
 			scan >>= priority;
 			scan = (scan * percent[file]) / 100;
 		}
-		if (scanning_global_lru(sc))
-			nr[l] = nr_scan_try_batch(scan,
-						  &zone->lru[l].nr_saved_scan,
-						  swap_cluster_max);
-		else
-			nr[l] = scan;
+		nr[l] = nr_scan_try_batch(scan,
+					  &reclaim_stat->nr_saved_scan[l],
+					  swap_cluster_max);
 	}
 
 	while (nr[LRU_INACTIVE_ANON] || nr[LRU_ACTIVE_FILE] ||
@@ -2220,6 +2218,7 @@ static void shrink_all_zones(unsigned lo
 {
 	struct zone *zone;
 	unsigned long nr_reclaimed = 0;
+	struct zone_reclaim_stat *reclaim_stat;
 
 	for_each_populated_zone(zone) {
 		enum lru_list l;
@@ -2236,11 +2235,14 @@ static void shrink_all_zones(unsigned lo
 						l == LRU_ACTIVE_FILE))
 				continue;
 
-			zone->lru[l].nr_saved_scan += (lru_pages >> prio) + 1;
-			if (zone->lru[l].nr_saved_scan >= nr_pages || pass > 3) {
+			reclaim_stat = get_reclaim_stat(zone, sc);
+			reclaim_stat->nr_saved_scan[l] +=
+						(lru_pages >> prio) + 1;
+			if (reclaim_stat->nr_saved_scan[l]
+						>= nr_pages || pass > 3) {
 				unsigned long nr_to_scan;
 
-				zone->lru[l].nr_saved_scan = 0;
+				reclaim_stat->nr_saved_scan[l] = 0;
 				nr_to_scan = min(nr_pages, lru_pages);
 				nr_reclaimed += shrink_list(l, nr_to_scan, zone,
 								sc, prio);

_

#10 lore raw

checkpatch infinite loop ?

In reply to Andy Whitcroft

Hi Andy

Following checkpatch.pl invocation has a problem in current linux-2.6 tree :

scripts/checkpatch.pl --file include/linux/inetdevice.h

linux-2.6.27.10/scripts/checkpatch.pl is OK, but starting from 2.6.28, there is an infinite loop...

Thanks