Skip to contentSkip to conversation

Checkpatch query

PATCH In review Lee Jones 2 participants 8 messages Jul 21, 2014 – Jul 22, 2014 1 message not in the archive Checkpatch

lore
Share & tools
lore

Lee Jones

author
#0 lore raw

Hi Andy, Joe,

When running checkpatch on drivers/mfd/tps80031.c I recieve the following warning:

Quoted 6 lines

WARNING: Possible switch case/default not preceeded by break or
fallthrough comment
#337: FILE: drivers/mfd/tps80031.c:337:
+ case TPS80031_BACKUP_REG:

total: 0 errors, 1 warnings, 573 lines checked

... but we use switch statement fall through all the time when identifying different types of registers used with Regmap. Placing a fall-through comment between them all sounds very messy to me.

Also, the warning only fires on 'case's which do not use '...' notation, which seems a little odd.

Anyway, I'm not sure placing a 'fall through' comment makes the code any cleaner or reduces possible failure rate in any way. Is there any chance that this check can be removed?

Quoted 44 lines

static bool rd_wr_reg_id0(struct device *dev, unsigned int reg)

{
	switch (reg) {
	case TPS80031_SMPS1_CFG_FORCE ... TPS80031_SMPS2_CFG_VOLTAGE:
		return true;
	default:
		return false;
	}
}

static bool rd_wr_reg_id1(struct device *dev, unsigned int reg)

{
	switch (reg) {
	case TPS80031_SECONDS_REG ... TPS80031_RTC_RESET_STATUS_REG:
	case TPS80031_VALIDITY0 ... TPS80031_VALIDITY7:
	case TPS80031_PHOENIX_START_CONDITION ... TPS80031_KEY_PRESS_DUR_CFG:
	case TPS80031_SMPS4_CFG_TRANS ... TPS80031_SMPS3_CFG_VOLTAGE:
	case TPS80031_BROADCAST_ADDR_ALL ... TPS80031_BROADCAST_ADDR_CLK_RST:
	case TPS80031_VANA_CFG_TRANS ... TPS80031_LDO7_CFG_VOLTAGE:
	case TPS80031_REGEN1_CFG_TRANS ... TPS80031_TMP_CFG_STATE:
	case TPS80031_PREQ1_RES_ASS_A ... TPS80031_PREQ3_RES_ASS_C:
	case TPS80031_SMPS_OFFSET ... TPS80031_BATDEBOUNCING:
	case TPS80031_CFG_INPUT_PUPD1 ... TPS80031_CFG_SMPS_PD:
	case TPS80031_BACKUP_REG:
		return true;
	default:
		return false;
	}
}

static bool is_volatile_reg_id1(struct device *dev, unsigned int reg)

{
	switch (reg) {
	case TPS80031_SMPS4_CFG_TRANS ... TPS80031_SMPS3_CFG_VOLTAGE:
	case TPS80031_VANA_CFG_TRANS ... TPS80031_LDO7_CFG_VOLTAGE:
	case TPS80031_REGEN1_CFG_TRANS ... TPS80031_TMP_CFG_STATE:
	case TPS80031_PREQ1_RES_ASS_A ... TPS80031_PREQ3_RES_ASS_C:
	case TPS80031_SMPS_OFFSET ... TPS80031_BATDEBOUNCING:
	case TPS80031_CFG_INPUT_PUPD1 ... TPS80031_CFG_SMPS_PD:
		return true;
	default:
		return false;
	}
}
Signature · 4 lines
Lee Jones
Linaro STMicroelectronics Landing Team Lead
Linaro.org │ Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog
maintainer
#1 lore raw
Quoted 1 line · Lee Jones

Hi Andy, Joe,

Hi Lee.

Quoted 8 lines

When running checkpatch on drivers/mfd/tps80031.c I recieve the following warning:

Quoted 5 lines

WARNING: Possible switch case/default not preceeded by break or
fallthrough comment
#337: FILE: drivers/mfd/tps80031.c:337:
+ case TPS80031_BACKUP_REG:

Quoted 5 lines
Quoted 1 line

total: 0 errors, 1 warnings, 573 lines checked

... but we use switch statement fall through all the time when identifying different types of registers used with Regmap. Placing a fall-through comment between them all sounds very messy to me.

Yes, it can be messy, but it's found some
actual bugs/defects.

Quoted 2 lines

Also, the warning only fires on 'case's which do not use '...' notation, which seems a little odd.

That's a checkpatch defect. Thanks for the report.

Quoted 3 lines

Anyway, I'm not sure placing a 'fall through' comment makes the code any cleaner or reduces possible failure rate in any way. Is there any chance that this check can be removed?

You could use the --ignore= cmd-line option or add it to a .checkpatch.conf file.

$ ./scripts/checkpatch.pl -f drivers/mfd/tps80031.c  --ignore=missing_break
WARNING: Possible unnecessary 'out of memory' message
#435: FILE: drivers/mfd/tps80031.c:435:
+	if (!tps80031) {
+		dev_err(&client->dev, "Malloc failed for tps80031\n");

total: 0 errors, 1 warnings, 573 lines checked

NOTE: Ignored message types: MISSING_BREAK

drivers/mfd/tps80031.c has style problems, please review.

If any of these errors are false positives, please report them to the maintainer, see CHECKPATCH in MAINTAINERS.

maintainer
#2 lore raw

PATCH 1/1 checkpatch: Allow case labels with ranges

In reply to Lee Jones

The MISSING_BREAK / fallthrough test does not support gcc's extension for case labels with ranges.

Add it.

---
 scripts/checkpatch.pl | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
scripts/checkpatch.pl +2 -2
index dc72a9b..2258497 100755
@@ -4767,7 +4767,7 @@ sub process {
 		}
 
 # check for case / default statements not preceeded by break/fallthrough/switch
-		if ($line =~ /^.\s*(?:case\s+(?:$Ident|$Constant)\s*|default):/) {
+		if ($line =~ /^.\s*(?:case\s+(?:$Ident|$Constant)\s*(?:\.\.\.\s*(?:$Ident|$Constant)\s*)?|default):/) {
 			my $has_break = 0;
 			my $has_statement = 0;
 			my $count = 0;
@@ -4778,7 +4778,7 @@ sub process {
 				my $fline = $lines[$prevline - 1];
 				last if ($fline =~ /^\@\@/);
 				next if ($fline =~ /^\-/);
-				next if ($fline =~ /^.(?:\s*(?:case\s+(?:$Ident|$Constant)[\s$;]*|default):[\s$;]*)*$/);
+				next if ($fline =~ /^.(?:\s*(?:case\s+(?:$Ident|$Constant(?:\s*\.\.\.\s*(?:$Ident|$Constant))?)[\s$;]*|default):[\s$;]*)*$/);
 				$has_break = 1 if ($rline =~ /fall[\s_-]*(through|thru)/i);
 				next if ($fline =~ /^.[\s$;]*$/);
 				$has_statement = 1;
author Acked-by on v1 1/1
#3 lore raw

Re: [PATCH] checkpatch: Allow case labels with ranges

Quoted 10 lines · Joe Perches

The MISSING_BREAK / fallthrough test does not support gcc's extension for case labels with ranges.

Add it.

---
 scripts/checkpatch.pl | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

[...]

Yep, that's much worse [1] - I wish I never mentioned it. ;)

[1]

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #330: FILE: drivers/mfd/tps80031.c:330:
+ case TPS80031_SMPS4_CFG_TRANS ... TPS80031_SMPS3_CFG_VOLTAGE:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #331: FILE: drivers/mfd/tps80031.c:331:
+ case TPS80031_BROADCAST_ADDR_ALL ... TPS80031_BROADCAST_ADDR_CLK_RST:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #332: FILE: drivers/mfd/tps80031.c:332:
+ case TPS80031_VANA_CFG_TRANS ... TPS80031_LDO7_CFG_VOLTAGE:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #333: FILE: drivers/mfd/tps80031.c:333:
+ case TPS80031_REGEN1_CFG_TRANS ... TPS80031_TMP_CFG_STATE:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #334: FILE: drivers/mfd/tps80031.c:334:
+ case TPS80031_PREQ1_RES_ASS_A ... TPS80031_PREQ3_RES_ASS_C:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #335: FILE: drivers/mfd/tps80031.c:335:
+ case TPS80031_SMPS_OFFSET ... TPS80031_BATDEBOUNCING:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #336: FILE: drivers/mfd/tps80031.c:336:
+ case TPS80031_CFG_INPUT_PUPD1 ... TPS80031_CFG_SMPS_PD:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #351: FILE: drivers/mfd/tps80031.c:351:
+ case TPS80031_PREQ1_RES_ASS_A ... TPS80031_PREQ3_RES_ASS_C:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #352: FILE: drivers/mfd/tps80031.c:352:
+ case TPS80031_SMPS_OFFSET ... TPS80031_BATDEBOUNCING:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #353: FILE: drivers/mfd/tps80031.c:353:
+ case TPS80031_CFG_INPUT_PUPD1 ... TPS80031_CFG_SMPS_PD:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #366: FILE: drivers/mfd/tps80031.c:366:
+ case TPS80031_TOGGLE1 ... TPS80031_VIBMODE:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #367: FILE: drivers/mfd/tps80031.c:367:
+ case TPS80031_PWM1ON ... TPS80031_PWM2OFF:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #368: FILE: drivers/mfd/tps80031.c:368:
+ case TPS80031_FG_REG_00 ... TPS80031_FG_REG_11:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #369: FILE: drivers/mfd/tps80031.c:369:
+ case TPS80031_INT_STS_A ... TPS80031_INT_MSK_STS_C:

WARNING: Possible switch case/default not preceeded by break or fallthrough comment #370: FILE: drivers/mfd/tps80031.c:370:
+ case TPS80031_CONTROLLER_CTRL2 ... TPS80031_LED_PWM_CTRL2:

total: 0 errors, 15 warnings, 574 lines checked

Signature · 4 lines
Lee Jones
Linaro STMicroelectronics Landing Team Lead
Linaro.org │ Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog
maintainer
#4 lore raw

Re: [PATCH] checkpatch: Allow case labels with ranges

Quoted 3 lines · Lee Jones
Quoted 2 lines · Joe Perches

The MISSING_BREAK / fallthrough test does not support gcc's extension for case labels with ranges.

[]

Quoted 1 line

Yep, that's much worse [1] - I wish I never mentioned it. ;)

I'm glad you did actually.

The code itself:

bool foo(int bar)

{
	case (bar) {
	case 1:
	case 2:
	case 4:
		return true;
	default:
		return false;
}

should not emit any error.

I'll fix it soon.

author
#5 lore raw

Re: [PATCH] checkpatch: Allow case labels with ranges

Quoted 22 lines · Joe Perches
Quoted 3 lines · Lee Jones
Quoted 2 lines · Joe Perches

The MISSING_BREAK / fallthrough test does not support gcc's extension for case labels with ranges.

[]

Quoted 1 line

Yep, that's much worse [1] - I wish I never mentioned it. ;)

I'm glad you did actually.

The code itself:

bool foo(int bar)

{
	case (bar) {
	case 1:
	case 2:
	case 4:
		return true;
	default:
		return false;
}
Quoted 1 line

should not emit any error.

That's why I submitted the report.

Quoted 1 line

I'll fix it soon.

Brilliant, thanks Joe!

Signature · 4 lines
Lee Jones
Linaro STMicroelectronics Landing Team Lead
Linaro.org │ Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog

A message that is not in the archive (sent off-list or not yet received): [email protected]

maintainer
#7 lore raw

PATCH 1/1 checkpatch: Fix false positive MISSING_BREAK warnings with --file

Using --file mode can give false positives with MISSING_BREAK fall-through warnings on simple but long multiple consecutive case statements.

Look for all lines before a case statement for a switch or a statement when using --file mode.

Fix a misspelling of preceded while there.

---

Quoted 5 lines · Andrew Morton
Quoted 1 line · On Mon, 21 Jul 2014 07:28:49 -0700 Joe Perches <[email protected]> wrote:

I'll fix it soon.

wanna do s/preceeded/preceded/g while you're in there?

Sure. It might be a few days though as
a generic fix seems non-trivial.

There's a trivial --file fix, but I'm not
sure the patch one is appropriate.

In the mean time, the case range label patch
should be OK to commit.

 scripts/checkpatch.pl | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
scripts/checkpatch.pl +2 -2
index 2258497..74afc7c 100755
@@ -4766,13 +4766,13 @@ sub process {
 			}
 		}
 
-# check for case / default statements not preceeded by break/fallthrough/switch
+# check for case / default statements not preceded by break/fallthrough/switch
 		if ($line =~ /^.\s*(?:case\s+(?:$Ident|$Constant)\s*(?:\.\.\.\s*(?:$Ident|$Constant)\s*)?|default):/) {
 			my $has_break = 0;
 			my $has_statement = 0;
 			my $count = 0;
 			my $prevline = $linenr;
-			while ($prevline > 1 && $count < 3 && !$has_break) {
+			while ($prevline > 1 && ($file || $count < 3) && !$has_break) {
 				$prevline--;
 				my $rline = $rawlines[$prevline - 1];
 				my $fline = $lines[$prevline - 1];
author Acked-by on v1 1/1
#8 lore raw

Re: [PATCH] checkpatch: Fix false positive MISSING_BREAK warnings with --file

Quoted 11 lines · Joe Perches

Using --file mode can give false positives with MISSING_BREAK fall-through warnings on simple but long multiple consecutive case statements.

Look for all lines before a case statement for a switch or a statement when using --file mode.

Fix a misspelling of preceded while there.

Solution works for me.

Quoted 4 lines

total: 0 errors, 0 warnings, 574 lines checked

drivers/mfd/tps80031.c has no obvious style problems and is ready for submission.

Signature · 4 lines
Lee Jones
Linaro STMicroelectronics Landing Team Lead
Linaro.org │ Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog