Merge tag 'hardening-v6.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux

Pull hardening updates from Kees Cook:
 "As usual, it's scattered changes all over. Patches touching things
  outside of our traditional areas in the tree have been Acked by
  maintainers or were trivial changes:

   - loadpin: remove unsupported MODULE_COMPRESS_NONE (Arulpandiyan
     Vadivel)

   - samples/check-exec: Fix script name (Mickaël Salaün)

   - yama: remove needless locking in yama_task_prctl() (Oleg Nesterov)

   - lib/string_choices: Sort by function name (R Sundar)

   - hardening: Allow default HARDENED_USERCOPY to be set at compile
     time (Mel Gorman)

   - uaccess: Split out compile-time checks into ucopysize.h

   - kbuild: clang: Support building UM with SUBARCH=i386

   - x86: Enable i386 FORTIFY_SOURCE on Clang 16+

   - ubsan/overflow: Rework integer overflow sanitizer option

   - Add missing __nonstring annotations for callers of
     memtostr*()/strtomem*()

   - Add __must_be_noncstr() and have memtostr*()/strtomem*() check for
     it

   - Introduce __nonstring_array for silencing future GCC 15 warnings"

* tag 'hardening-v6.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux: (26 commits)
  compiler_types: Introduce __nonstring_array
  hardening: Enable i386 FORTIFY_SOURCE on Clang 16+
  x86/build: Remove -ffreestanding on i386 with GCC
  ubsan/overflow: Enable ignorelist parsing and add type filter
  ubsan/overflow: Enable pattern exclusions
  ubsan/overflow: Rework integer overflow sanitizer option to turn on everything
  samples/check-exec: Fix script name
  yama: don't abuse rcu_read_lock/get_task_struct in yama_task_prctl()
  kbuild: clang: Support building UM with SUBARCH=i386
  loadpin: remove MODULE_COMPRESS_NONE as it is no longer supported
  lib/string_choices: Rearrange functions in sorted order
  string.h: Validate memtostr*()/strtomem*() arguments more carefully
  compiler.h: Introduce __must_be_noncstr()
  nilfs2: Mark on-disk strings as nonstring
  uapi: stddef.h: Introduce __kernel_nonstring
  x86/tdx: Mark message.bytes as nonstring
  string: kunit: Mark nonstring test strings as __nonstring
  scsi: qla2xxx: Mark device strings as nonstring
  scsi: mpt3sas: Mark device strings as nonstring
  scsi: mpi3mr: Mark device strings as nonstring
  ...

1 files changed, 17 insertions, 1 deletions

diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index 155385754824..6db889c9efcc 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -206,9 +206,25 @@ void ftrace_likely_update(struct ftrace_likely_data *f, int val,
 #define __must_be_byte_array(a)	__BUILD_BUG_ON_ZERO_MSG(!__is_byte_array(a), \
 							"must be byte array")
 
+/*
+ * If the "nonstring" attribute isn't available, we have to return true
+ * so the __must_*() checks pass when "nonstring" isn't supported.
+ */
+#if __has_attribute(__nonstring__) && defined(__annotated)
+#define __is_cstr(a)		(!__annotated(a, nonstring))
+#define __is_noncstr(a)		(__annotated(a, nonstring))
+#else
+#define __is_cstr(a)		(true)
+#define __is_noncstr(a)		(true)
+#endif
+
 /* Require C Strings (i.e. NUL-terminated) lack the "nonstring" attribute. */
 #define __must_be_cstr(p) \
-	__BUILD_BUG_ON_ZERO_MSG(__annotated(p, nonstring), "must be cstr (NUL-terminated)")
+	__BUILD_BUG_ON_ZERO_MSG(!__is_cstr(p), \
+				"must be C-string (NUL-terminated)")
+#define __must_be_noncstr(p) \
+	__BUILD_BUG_ON_ZERO_MSG(!__is_noncstr(p), \
+				"must be non-C-string (not NUL-terminated)")
 
 #endif /* __KERNEL__ */