vmalloc() and vzalloc() functions have now 2-factor multiplication
argument forms vmalloc_array() and vcalloc(), correspondingly.
Add alloc-with-multiplies checks for these new functions.
Simplify the original codes repeated else to use a hash.
Link: https://github.com/KSPP/linux/issues/342
Original-patch-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Link: https://lore.kernel.org/lkml/ZQCaO+tYycDxVLy7@work/
Signed-off-by: Joe Perches <joe@perches.com>
---
scripts/checkpatch.pl | 21 ++++++++++++++-------
1 file changed, 14 insertions(+), 7 deletions(-)
diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 617f9e53bacdf..4cb248985eefc 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -834,6 +834,16 @@ our %deprecated_apis = (
#Create a search pattern for all these strings to speed up a loop below
our $deprecated_apis_search = '(?:' . join('|', keys %deprecated_apis) . ')';
+our %alloc_with_multiply_apis = (
+ "kmalloc" => "kmalloc_array",
+ "kvmalloc" => "kvmalloc_array",
+ "vmalloc" => "vmalloc_array",
+ "kvzalloc" => "kvcalloc",
+ "kzalloc" => "kcalloc",
+ "vzalloc" => "vcalloc",
+);
+our $alloc_with_multiply_search = '(?:' . join('|', keys %alloc_with_multiply_apis) . ')';
+
our $mode_perms_world_writable = qr{
S_IWUGO |
S_IWOTH |
@@ -7187,17 +7197,14 @@ sub process {
"Prefer $3(sizeof(*$1)...) over $3($4...)\n" . $herecurr);
}
-# check for (kv|k)[mz]alloc with multiplies that could be kmalloc_array/kvmalloc_array/kvcalloc/kcalloc
+# check for various allocs with multiplies that should use safer functions
if ($perl_version_ok &&
defined $stat &&
- $stat =~ /^\+\s*($Lval)\s*\=\s*(?:$balanced_parens)?\s*((?:kv|k)[mz]alloc)\s*\(\s*($FuncArg)\s*\*\s*($FuncArg)\s*,/) {
+ $stat =~ /^\+\s*($Lval)\s*\=\s*(?:$balanced_parens)?\s*($alloc_with_multiply_search)\s*\(\s*($FuncArg)\s*\*\s*($FuncArg)\s*,/) {
my $oldfunc = $3;
+ my $newfunc = $alloc_with_multiply_apis{$oldfunc};
my $a1 = $4;
my $a2 = $10;
- my $newfunc = "kmalloc_array";
- $newfunc = "kvmalloc_array" if ($oldfunc eq "kvmalloc");
- $newfunc = "kvcalloc" if ($oldfunc eq "kvzalloc");
- $newfunc = "kcalloc" if ($oldfunc eq "kzalloc");
my $r1 = $a1;
my $r2 = $a2;
if ($a1 =~ /^sizeof\s*\S/) {
@@ -7213,7 +7220,7 @@ sub process {
"Prefer $newfunc over $oldfunc with multiply\n" . $herectx) &&
$cnt == 1 &&
$fix) {
- $fixed[$fixlinenr] =~ s/\b($Lval)\s*\=\s*(?:$balanced_parens)?\s*((?:kv|k)[mz]alloc)\s*\(\s*($FuncArg)\s*\*\s*($FuncArg)/$1 . ' = ' . "$newfunc(" . trim($r1) . ', ' . trim($r2)/e;
+ $fixed[$fixlinenr] =~ s/\b($Lval)\s*\=\s*(?:$balanced_parens)?\s*($oldfunc)\s*\(\s*($FuncArg)\s*\*\s*($FuncArg)/$1 . ' = ' . "$newfunc(" . trim($r1) . ', ' . trim($r2)/e;
}
}
}
--
2.41.0On 9/13/23 14:37, Joe Perches wrote: > vmalloc() and vzalloc() functions have now 2-factor multiplication > argument forms vmalloc_array() and vcalloc(), correspondingly. > > Add alloc-with-multiplies checks for these new functions. > > Simplify the original codes repeated else to use a hash. > > Link: https://github.com/KSPP/linux/issues/342 > > Original-patch-by: Gustavo A. R. Silva <gustavoars@kernel.org> Why don't you wait for a response or a v2 from the original submitter? -- Gustavo
On Wed, 2023-09-13 at 16:14 -0600, Gustavo A. R. Silva wrote: > > On 9/13/23 14:37, Joe Perches wrote: > > vmalloc() and vzalloc() functions have now 2-factor multiplication > > argument forms vmalloc_array() and vcalloc(), correspondingly. > > > > Add alloc-with-multiplies checks for these new functions. > > > > Simplify the original codes repeated else to use a hash. > > > > Link: https://github.com/KSPP/linux/issues/342 > > > > Original-patch-by: Gustavo A. R. Silva <gustavoars@kernel.org> > > Why don't you wait for a response or a v2 from the original > submitter? Because there really is no need to wait.
On 9/13/23 16:25, Joe Perches wrote: > On Wed, 2023-09-13 at 16:14 -0600, Gustavo A. R. Silva wrote: >> >> On 9/13/23 14:37, Joe Perches wrote: >>> vmalloc() and vzalloc() functions have now 2-factor multiplication >>> argument forms vmalloc_array() and vcalloc(), correspondingly. >>> >>> Add alloc-with-multiplies checks for these new functions. >>> >>> Simplify the original codes repeated else to use a hash. >>> >>> Link: https://github.com/KSPP/linux/issues/342 >>> >>> Original-patch-by: Gustavo A. R. Silva <gustavoars@kernel.org> >> >> Why don't you wait for a response or a v2 from the original >> submitter? > > Because there really is no need to wait. By the way, did you test it? -- Gustavo
On Wed, 2023-09-13 at 16:32 -0600, Gustavo A. R. Silva wrote: > > On 9/13/23 16:25, Joe Perches wrote: > > On Wed, 2023-09-13 at 16:14 -0600, Gustavo A. R. Silva wrote: > > > > > > On 9/13/23 14:37, Joe Perches wrote: > > > > vmalloc() and vzalloc() functions have now 2-factor multiplication > > > > argument forms vmalloc_array() and vcalloc(), correspondingly. > > > > > > > > Add alloc-with-multiplies checks for these new functions. > > > > > > > > Simplify the original codes repeated else to use a hash. > > > > > > > > Link: https://github.com/KSPP/linux/issues/342 > > > > > > > > Original-patch-by: Gustavo A. R. Silva <gustavoars@kernel.org> > > > > > > Why don't you wait for a response or a v2 from the original > > > submitter? > > > > Because there really is no need to wait. > > By the way, did you test it? Yes, against arch/s390/include/asm/idals.h Did you? If so, you could also add your own tested-by...
On 9/13/23 16:34, Joe Perches wrote: > On Wed, 2023-09-13 at 16:32 -0600, Gustavo A. R. Silva wrote: >> >> On 9/13/23 16:25, Joe Perches wrote: >>> On Wed, 2023-09-13 at 16:14 -0600, Gustavo A. R. Silva wrote: >>>> >>>> On 9/13/23 14:37, Joe Perches wrote: >>>>> vmalloc() and vzalloc() functions have now 2-factor multiplication >>>>> argument forms vmalloc_array() and vcalloc(), correspondingly. >>>>> >>>>> Add alloc-with-multiplies checks for these new functions. >>>>> >>>>> Simplify the original codes repeated else to use a hash. >>>>> >>>>> Link: https://github.com/KSPP/linux/issues/342 >>>>> >>>>> Original-patch-by: Gustavo A. R. Silva <gustavoars@kernel.org> >>>> >>>> Why don't you wait for a response or a v2 from the original >>>> submitter? >>> >>> Because there really is no need to wait. >> >> By the way, did you test it? > > Yes, against arch/s390/include/asm/idals.h I don't see any instances of vmalloc() or vzalloc() in that file. -- Gustavo
On Wed, 2023-09-13 at 16:41 -0600, Gustavo A. R. Silva wrote: > > On 9/13/23 16:34, Joe Perches wrote: > > On Wed, 2023-09-13 at 16:32 -0600, Gustavo A. R. Silva wrote: > > > > > > On 9/13/23 16:25, Joe Perches wrote: > > > > On Wed, 2023-09-13 at 16:14 -0600, Gustavo A. R. Silva wrote: > > > > > > > > > > On 9/13/23 14:37, Joe Perches wrote: > > > > > > vmalloc() and vzalloc() functions have now 2-factor multiplication > > > > > > argument forms vmalloc_array() and vcalloc(), correspondingly. > > > > > > > > > > > > Add alloc-with-multiplies checks for these new functions. > > > > > > > > > > > > Simplify the original codes repeated else to use a hash. > > > > > > > > > > > > Link: https://github.com/KSPP/linux/issues/342 > > > > > > > > > > > > Original-patch-by: Gustavo A. R. Silva <gustavoars@kernel.org> > > > > > > > > > > Why don't you wait for a response or a v2 from the original > > > > > submitter? > > > > > > > > Because there really is no need to wait. > > > > > > By the way, did you test it? > > > > Yes, against arch/s390/include/asm/idals.h > > I don't see any instances of vmalloc() or vzalloc() in that file. <snort> Cute. Missing a ? in the original search '\s*,?' or just its removal... Likely the removal would be better. thanks, cheers, Joe
© 2016 - 2026 Red Hat, Inc.