[PATCH v2] checkpatch: Warn about 0-length and 1-element arrays

[PATCH v2] checkpatch: Warn about 0-length and 1-element arrays

[Kees Cook]

Fake flexible arrays have been deprecated since last millennium. Proper
C99 flexible arrays must be used throughout the kernel so
CONFIG_FORTIFY_SOURCE and CONFIG_UBSAN_BOUNDS can provide proper array
bounds checking.

Cc: Andy Whitcroft <apw@canonical.com>
Cc: Dwaipayan Ray <dwaipayanray1@gmail.com>
Cc: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Cc: Gustavo A. R. Silva <gustavoars@kernel.org>
Fixed-by: Joe Perches <joe@perches.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20230517204530.never.151-kees@kernel.org
---
v2: find using "end of struct" marking instead of $context_function (joe)
v1: https://lore.kernel.org/all/20230517204530.never.151-kees@kernel.org/
---
 scripts/checkpatch.pl | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 30b0b4fdb3bf..cfa19bfc857c 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -7430,6 +7430,21 @@ sub process {
 			}
 		}
 
+# check for array definition/declarations that should use flexible arrays instead
+		if ($sline =~ /^[\+ ]\s*}\s*;\s*$/ &&
+		    $prevline =~ /^\+\s*(?:(?:struct|union|enum)\s+$Ident|\}|$Type)\s*$Ident\s*\[\s*(0|1)\s*\]\s*;\s*$/) {
+			if ($1 == '0') {
+				if (WARN("ZERO_LENGTH_ARRAY",
+					 "Use C99 flexible arrays instead of zero-length arrays - see https://github.com/KSPP/linux/issues/78\n" . $hereprev) &&
+				    $fix) {
+					$fixed[$fixlinenr - 1] =~ s/\[0\]/[]/g;
+				}
+			} else {
+				WARN("ONE_ELEMENT_ARRAY",
+				     "Use C99 flexible arrays instead of one-element arrays - see https://github.com/KSPP/linux/issues/79\n" . $hereprev);
+			}
+		}
+
 # nested likely/unlikely calls
 		if ($line =~ /\b(?:(?:un)?likely)\s*\(\s*!?\s*(IS_ERR(?:_OR_NULL|_VALUE)?|WARN)/) {
 			WARN("LIKELY_MISUSE",
-- 
2.34.1

Re: [PATCH v2] checkpatch: Warn about 0-length and 1-element arrays

[Joe Perches]

On Fri, 2023-05-26 at 10:39 -0700, Kees Cook wrote:
> Fake flexible arrays have been deprecated since last millennium. Proper
> C99 flexible arrays must be used throughout the kernel so
> CONFIG_FORTIFY_SOURCE and CONFIG_UBSAN_BOUNDS can provide proper array
> bounds checking.
[]
> diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
> []
> @@ -7430,6 +7430,21 @@ sub process {
>  			}
>  		}
>  
> +# check for array definition/declarations that should use flexible arrays instead
> +		if ($sline =~ /^[\+ ]\s*}\s*;\s*$/ &&
> +		    $prevline =~ /^\+\s*(?:(?:struct|union|enum)\s+$Ident|\}|$Type)\s*$Ident\s*\[\s*(0|1)\s*\]\s*;\s*$/) {

I think this is overly complicated and not necessary

		    $prevline =~ /^\+\s*$Type\s*$Ident\s*\[\s*(0|1)\s*\]\s*;\s*$/) {

should work no?

($Type already includes this from @typeList):
	qr{struct\s+$Ident},
	qr{union\s+$Ident},
	qr{enum\s+$Ident},

> +			if ($1 == '0') {
> +				if (WARN("ZERO_LENGTH_ARRAY",
> +					 "Use C99 flexible arrays instead of zero-length arrays - see https://github.com/KSPP/linux/issues/78\n" . $hereprev) &&
> +				    $fix) {
> +					$fixed[$fixlinenr - 1] =~ s/\[0\]/[]/g;

And this $fix doesn't work if the line is struct foo bar[ 0 ];
and the use of /g is odd.

Because the message is a WARN and not an ERR, please use
"Prefer/over" and not "Use/instead of"

					$fixed[$fixlinenr - 1] =~ s/\[\s*0\s*\]/[]/;
> +				}
> +			} else {
> +				WARN("ONE_ELEMENT_ARRAY",
> +				     "Use C99 flexible arrays instead of one-element arrays - see https://github.com/KSPP/linux/issues/79\n" . $hereprev);
> +			}

And this could have a $fix change too

				if (WARN("ONE_ELEMENT_ARRAY",
					 ...) &&
				    $fix) {
					$fixed[$fixlinenr - 1] =~ s/\[\s*1\s*\]/[]/;
> +		}
> +
>  # nested likely/unlikely calls
>  		if ($line =~ /\b(?:(?:un)?likely)\s*\(\s*!?\s*(IS_ERR(?:_OR_NULL|_VALUE)?|WARN)/) {
>  			WARN("LIKELY_MISUSE",

Re: [PATCH v2] checkpatch: Warn about 0-length and 1-element arrays

[Kees Cook]

On Fri, May 26, 2023 at 06:38:27PM -0700, Joe Perches wrote:
> On Fri, 2023-05-26 at 10:39 -0700, Kees Cook wrote:
> > Fake flexible arrays have been deprecated since last millennium. Proper
> > C99 flexible arrays must be used throughout the kernel so
> > CONFIG_FORTIFY_SOURCE and CONFIG_UBSAN_BOUNDS can provide proper array
> > bounds checking.
> []
> > diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
> > []
> > @@ -7430,6 +7430,21 @@ sub process {
> >  			}
> >  		}
> >  
> > +# check for array definition/declarations that should use flexible arrays instead
> > +		if ($sline =~ /^[\+ ]\s*}\s*;\s*$/ &&
> > +		    $prevline =~ /^\+\s*(?:(?:struct|union|enum)\s+$Ident|\}|$Type)\s*$Ident\s*\[\s*(0|1)\s*\]\s*;\s*$/) {
> 
> I think this is overly complicated and not necessary
> 
> 		    $prevline =~ /^\+\s*$Type\s*$Ident\s*\[\s*(0|1)\s*\]\s*;\s*$/) {
> 
> should work no?
> 
> ($Type already includes this from @typeList):
> 	qr{struct\s+$Ident},
> 	qr{union\s+$Ident},
> 	qr{enum\s+$Ident},

Hm, I didn't when I originally tried it. I will double-check.

> 
> > +			if ($1 == '0') {
> > +				if (WARN("ZERO_LENGTH_ARRAY",
> > +					 "Use C99 flexible arrays instead of zero-length arrays - see https://github.com/KSPP/linux/issues/78\n" . $hereprev) &&
> > +				    $fix) {
> > +					$fixed[$fixlinenr - 1] =~ s/\[0\]/[]/g;
> 
> And this $fix doesn't work if the line is struct foo bar[ 0 ];
> and the use of /g is odd.

Thanks!

> Because the message is a WARN and not an ERR, please use
> "Prefer/over" and not "Use/instead of"

I will change to ERR, then. We must not use [0]-sized arrays ever.

> 
> 					$fixed[$fixlinenr - 1] =~ s/\[\s*0\s*\]/[]/;
> > +				}
> > +			} else {
> > +				WARN("ONE_ELEMENT_ARRAY",
> > +				     "Use C99 flexible arrays instead of one-element arrays - see https://github.com/KSPP/linux/issues/79\n" . $hereprev);
> > +			}
> 
> And this could have a $fix change too
> 
> 				if (WARN("ONE_ELEMENT_ARRAY",
> 					 ...) &&
> 				    $fix) {
> 					$fixed[$fixlinenr - 1] =~ s/\[\s*1\s*\]/[]/;

Almost never is it possible to only fix the struct when it's a 1-element
array (due to the impact of the size changes that be present in other
code), so I left it out.

I'll send a v3.

-Kees

-- 
Kees Cook