Re: [cocci] [PATCH v6 5/5] coccinelle: Add kmalloc_objs conversion script

On Wed, Dec 03, 2025 at 03:30:35PM -0800, Kees Cook wrote:
> Finds and converts sized kmalloc-family of allocations into the
> typed kmalloc_obj-family of allocations.
> 
> Signed-off-by: Kees Cook <[email protected]>
> ---
> Cc: Julia Lawall <[email protected]>
> Cc: Nicolas Palix <[email protected]>
> Cc: [email protected]
> ---
>  scripts/coccinelle/api/kmalloc_objs.cocci | 109 ++++++++++++++++++++++
>  1 file changed, 109 insertions(+)
>  create mode 100644 scripts/coccinelle/api/kmalloc_objs.cocci

Is there a way to reproduce commit 69050f8d6d075dc using this semantic
patch as claimed?  I spend a while installing the latest version of
spatch (which was quite hard to do due to all the unusual dependencies),
but it complains that no rules apply:

$ /usr/local/bin/spatch --sp-file scripts/coccinelle/api/kmalloc_objs.cocci --dir . --in-place
init_defs_builtins: /usr/local/lib/coccinelle/standard.h
SPECIAL NAMES: adding u8 as a type
SPECIAL NAMES: adding u16 as a type
SPECIAL NAMES: adding u32 as a type
SPECIAL NAMES: adding u64 as a type
SPECIAL NAMES: adding __u8 as a type
SPECIAL NAMES: adding __u16 as a type
SPECIAL NAMES: adding __u32 as a type
SPECIAL NAMES: adding __u64 as a type
SPECIAL NAMES: adding uint8_t as a type
SPECIAL NAMES: adding uint16_t as a type
SPECIAL NAMES: adding uint32_t as a type
SPECIAL NAMES: adding uint64_t as a type
SPECIAL NAMES: adding uchar as a type
SPECIAL NAMES: adding ushort as a type
SPECIAL NAMES: adding uint as a type
SPECIAL NAMES: adding ulong as a type
SPECIAL NAMES: adding __le16 as a type
SPECIAL NAMES: adding __le32 as a type
SPECIAL NAMES: adding __le64 as a type
SPECIAL NAMES: adding __be16 as a type
SPECIAL NAMES: adding __be32 as a type
SPECIAL NAMES: adding __be64 as a type
SPECIAL NAMES: adding wchar_t as a type
No rules apply.  Perhaps your semantic patch doesn't contain any +/-/* code, or you have a failed dependency.  If the problem is not clear, try --debug-parse-cocci or check whether any virtual rules (e.g., patch) should be defined.