[PATCH v3] checkpatch: Check WQ_PERCPU or WQ_UNBOUND presence in alloc_workqueue() users

[PATCH v3] checkpatch: Check WQ_PERCPU or WQ_UNBOUND presence in alloc_workqueue() users

[Marco Crivellari]

The workqueue API introduced a new flag, WQ_PERCPU, that has to be used when
WQ_UNBOUND is not present. One of these flags must be present, but not
both of them.

To limit usage mistakes, emit an ERROR if one of the below condition is met:
- alloc_workqueue() is called without WQ_PERCPU nor WQ_UNBOUND
- alloc_workqueue() is called with both WQ_PERCPU and WQ_UNBOUND

Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
---
Changes in v3:
- code aligned to open parens

- ERROR changed with WARN

Link to v2: https://lore.kernel.org/all/20260604154447.381477-1-marco.crivellari@suse.com/

Changes in v2:
- removed $line from the test (Joe Perches)

- devm_alloc_workqueue() regex take into account also (Joe Perches)

- defined() on variables and other improvements (Joe Perches)

Link to v1: https://lore.kernel.org/all/20260603140941.320063-1-marco.crivellari@suse.com/

 scripts/checkpatch.pl | 25 ++++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 0492d6afc9a1..f1b03897b6e7 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -7804,8 +7804,31 @@ sub process {
 			ERROR("UNINITIALIZED_PTR_WITH_FREE",
 			      "pointer '$1' with __free attribute should be initialized\n" . $herecurr);
 		}
-	}
 
+# check alloc_workqueue() parameters for WQ_PERCPU and WQ_UNBOUND uses
+		if (defined($stat) &&
+			$stat =~ /^[ \+]\s*(?:$Lval\s*=\s*)?((?:devm_)?alloc_workqueue)\s*($balanced_parens)/) {
+
+			my $func = $1;
+			my $args = $2;
+			my $has_percpu = $args =~ /\bWQ_PERCPU\b/;
+			my $has_unbound = $args =~ /\bWQ_UNBOUND\b/;
+			my $error_msg;
+
+			if ($has_percpu && $has_unbound) {
+				$error_msg = "$func() should not contain both WQ_PERCPU and WQ_UNBOUND\n";
+			} elsif (!$has_percpu && !$has_unbound) {
+				$error_msg = "$func() must specify either WQ_PERCPU or WQ_UNBOUND\n";
+			}
+
+			if (defined($error_msg)) {
+				my $stmt_cnt = statement_rawlines($stat);
+				my $herectx = get_stat_here($linenr, $stmt_cnt, $here);
+				WARN("ALLOC_WORKQUEUE_FLAGS",
+					 $error_msg . $herectx);
+			}
+		}
+	}
 	# If we have no input at all, then there is nothing to report on
 	# so just keep quiet.
 	if ($#rawlines == -1) {
-- 
2.54.0

Re: [PATCH v3] checkpatch: Check WQ_PERCPU or WQ_UNBOUND presence in alloc_workqueue() users

[Joe Perches]

On Fri, 2026-06-05 at 09:43 +0200, Marco Crivellari wrote:
> The workqueue API introduced a new flag, WQ_PERCPU, that has to be used when
> WQ_UNBOUND is not present. One of these flags must be present, but not
> both of them.
[]
> Signed-off-by: Marco Crivellari <[marco.crivellari@suse.com](mailto:marco.crivellari@suse.com)>
> ---
> Changes in v3:
> - code aligned to open parens

Not so.

> diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
[]
> @@ -7804,8 +7804,31 @@ sub process {
>  			ERROR("UNINITIALIZED_PTR_WITH_FREE",
>  			      "pointer '$1' with __free attribute should be initialized\n" . $herecurr);
>  		}
> -	}
>  
> +# check alloc_workqueue() parameters for WQ_PERCPU and WQ_UNBOUND uses
> +		if (defined($stat) &&
> +			$stat =~ /^[ \+]\s*(?:$Lval\s*=\s*)?((?:devm_)?alloc_workqueue)\s*($balanced_parens)/) {

Properly aligned would be 2 tabs then 4 spaces

		    $stat =~ /^[ \+]\s*(?:$Lval\s*=\s*)?((?:devm_)?alloc_workqueue)\s*($balanced_parens)/) {
[]

> +			if (defined($error_msg)) {
> +				my $stmt_cnt = statement_rawlines($stat);
> +				my $herectx = get_stat_here($linenr, $stmt_cnt, $here);
> +				WARN("ALLOC_WORKQUEUE_FLAGS",
> +					 $error_msg . $herectx);

4 tabs then 5 spaces

Re: [PATCH v3] checkpatch: Check WQ_PERCPU or WQ_UNBOUND presence in alloc_workqueue() users

[Marco Crivellari]

On Sat, Jun 6, 2026 at 3:22 AM Joe Perches <joe@perches.com> wrote:
>
> On Fri, 2026-06-05 at 09:43 +0200, Marco Crivellari wrote:
> > The workqueue API introduced a new flag, WQ_PERCPU, that has to be used when
> > WQ_UNBOUND is not present. One of these flags must be present, but not
> > both of them.
> []
> > Signed-off-by: Marco Crivellari <[marco.crivellari@suse.com](mailto:marco.crivellari@suse.com)>
> > ---
> > Changes in v3:
> > - code aligned to open parens
>
> Not so.

Aha, thanks, I will send a new version.

-- 

Marco Crivellari

SUSE Labs