linux.git/scripts/gcc-plugins/randomize_layout_plugin.c, branch v6.18.21

randstruct: gcc-plugin: Fix attribute addition

2025-06-01T15:41:11+00:00

Based on changes in the 2021 public version of the randstruct
out-of-tree GCC plugin[1], more carefully update the attributes on
resulting decls, to avoid tripping checks in GCC 15's
comptypes_check_enum_int() when it has been configured with
"--enable-checking=misc":

arch/arm64/kernel/kexec_image.c:132:14: internal compiler error: in comptypes_check_enum_int, at c/c-typeck.cc:1519
  132 | const struct kexec_file_ops kexec_image_ops = {
      |              ^~~~~~~~~~~~~~
 internal_error(char const*, ...), at gcc/gcc/diagnostic-global-context.cc:517
 fancy_abort(char const*, int, char const*), at gcc/gcc/diagnostic.cc:1803
 comptypes_check_enum_int(tree_node*, tree_node*, bool*), at gcc/gcc/c/c-typeck.cc:1519
 ...

Link: https://archive.org/download/grsecurity/grsecurity-3.1-5.10.41-202105280954.patch.gz [1]
Reported-by: Thiago Jung Bauermann 
Closes: https://github.com/KSPP/linux/issues/367
Closes: https://lore.kernel.org/lkml/20250530000646.104457-1-thiago.bauermann@linaro.org/
Reported-by: Ingo Saitz 
Closes: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1104745
Fixes: 313dd1b62921 ("gcc-plugins: Add the randstruct plugin")
Tested-by: Thiago Jung Bauermann 
Link: https://lore.kernel.org/r/20250530221824.work.623-kees@kernel.org
Signed-off-by: Kees Cook

randstruct: gcc-plugin: Remove bogus void member

2025-05-08T16:42:40+00:00

When building the randomized replacement tree of struct members, the
randstruct GCC plugin would insert, as the first member, a 0-sized void
member. This appears as though it was done to catch non-designated
("unnamed") static initializers, which wouldn't be stable since they
depend on the original struct layout order.

This was accomplished by having the side-effect of the "void member"
tripping an assert in GCC internals (count_type_elements) if the member
list ever needed to be counted (e.g. for figuring out the order of members
during a non-designated initialization), which would catch impossible type
(void) in the struct:

security/landlock/fs.c: In function ‘hook_file_ioctl_common’:
security/landlock/fs.c:1745:61: internal compiler error: in count_type_elements, at expr.cc:7075
 1745 |                         .u.op = &(struct lsm_ioctlop_audit) {
      |                                                             ^

static HOST_WIDE_INT
count_type_elements (const_tree type, bool for_ctor_p)
{
  switch (TREE_CODE (type))
...
    case VOID_TYPE:
    default:
      gcc_unreachable ();
    }
}

However this is a redundant safety measure since randstruct uses the
__designated_initializer attribute both internally and within the
__randomized_layout attribute macro so that this would be enforced
by the compiler directly even when randstruct was not enabled (via
-Wdesignated-init).

A recent change in Landlock ended up tripping the same member counting
routine when using a full-struct copy initializer as part of an anonymous
initializer. This, however, is a false positive as the initializer is
copying between identical structs (and hence identical layouts). The
"path" member is "struct path", a randomized struct, and is being copied
to from another "struct path", the "f_path" member:

        landlock_log_denial(landlock_cred(file->f_cred), &(struct landlock_request) {
                .type = LANDLOCK_REQUEST_FS_ACCESS,
                .audit = {
                        .type = LSM_AUDIT_DATA_IOCTL_OP,
                        .u.op = &(struct lsm_ioctlop_audit) {
                                .path = file->f_path,
                                .cmd = cmd,
                        },
                },
	...

As can be seen with the coming randstruct KUnit test, there appears to
be no behavioral problems with this kind of initialization when the void
member is removed from the randstruct GCC plugin, so remove it.

Reported-by: "Dr. David Alan Gilbert" 
Closes: https://lore.kernel.org/lkml/Z_PRaKx7q70MKgCA@gallifrey/
Reported-by: Mark Brown 
Closes: https://lore.kernel.org/lkml/20250407-kbuild-disable-gcc-plugins-v1-1-5d46ae583f5e@kernel.org/
Reported-by: WangYuli 
Closes: https://lore.kernel.org/lkml/337D5D4887277B27+3c677db3-a8b9-47f0-93a4-7809355f1381@uniontech.com/
Fixes: 313dd1b62921 ("gcc-plugins: Add the randstruct plugin")
Signed-off-by: Kees Cook

gcc-plugins: randstruct: Remove GCC 4.7 or newer requirement

2024-08-05T21:34:23+00:00

Since the kernel currently requires GCC 5.1 as a minimum, remove the
unnecessary GCC version >= 4.7 check.

Signed-off-by: Thorsten Blum 
Link: https://lore.kernel.org/r/20240723165332.1947-1-thorsten.blum@toblux.com
Signed-off-by: Kees Cook

gcc-plugins: randstruct: Update code comment in relayout_struct()

2023-11-28T00:30:05+00:00

Update code comment to clarify that the only element whose layout is
not randomized is a proper C99 flexible-array member. This update is
complementary to commit 1ee60356c2dc ("gcc-plugins: randstruct: Only
warn about true flexible arrays")

Signed-off-by: "Gustavo A. R. Silva" 
Link: https://lore.kernel.org/r/ZWJr2MWDjXLHE8ap@work
Fixes: 1ee60356c2dc ("gcc-plugins: randstruct: Only warn about true flexible arrays")
Signed-off-by: Kees Cook

gcc-plugins: randstruct: Only warn about true flexible arrays

2023-11-08T22:18:57+00:00

The randstruct GCC plugin tried to discover "fake" flexible arrays
to issue warnings about them in randomized structs. In the future
LSM overhead reduction series, it would be legal to have a randomized
struct with a 1-element array, and this should _not_ be treated as a
flexible array, especially since commit df8fc4e934c1 ("kbuild: Enable
-fstrict-flex-arrays=3"). Disable the 0-sized and 1-element array
discovery logic in the plugin, but keep the "true" flexible array check.

Cc: KP Singh 
Cc: linux-hardening@vger.kernel.org
Reported-by: kernel test robot 
Closes: https://lore.kernel.org/oe-kbuild-all/202311021532.iBwuZUZ0-lkp@intel.com/
Fixes: df8fc4e934c1 ("kbuild: Enable -fstrict-flex-arrays=3")
Reviewed-by: Bill Wendling 
Acked-by: "Gustavo A. R. Silva" 
Link: https://lore.kernel.org/r/20231104204334.work.160-kees@kernel.org
Signed-off-by: Kees Cook

randstruct: Fix gcc-plugin performance mode to stay in group

2023-10-09T05:46:05+00:00

The performance mode of the gcc-plugin randstruct was shuffling struct
members outside of the cache-line groups. Limit the range to the
specified group indexes.

Cc: linux-hardening@vger.kernel.org
Cc: stable@vger.kernel.org
Reported-by: Lukas Loidolt 
Closes: https://lore.kernel.org/all/f3ca77f0-e414-4065-83a5-ae4c4d25545d@student.tuwien.ac.at
Fixes: 313dd1b62921 ("gcc-plugins: Add the randstruct plugin")
Signed-off-by: Kees Cook

gcc-plugins: use KERNELVERSION for plugin version

2022-05-24T15:25:16+00:00

Commit 61f60bac8c05 ("gcc-plugins: Change all version strings match
kernel") broke parallel builds.

Instead of adding the dependency between GCC plugins and utsrelease.h,
let's use KERNELVERSION, which does not require any build artifact.

Another reason why I want to avoid utsrelease.h is because it depends
on CONFIG_LOCALVERSION(_AUTO) and localversion* files.

(include/generated/utsrelease.h depends on include/config/kernel.release,
which is generated by scripts/setlocalversion)

I want to keep host tools independent of the kernel configuration.

There is no good reason to rebuild GCC plugins just because of
CONFIG_LOCALVERSION being changed.

We just want to associate the plugin versions with the kernel source
version. KERNELVERSION should be enough for our purpose.

Fixes: 61f60bac8c05 ("gcc-plugins: Change all version strings match kernel")
Reported-by: kernel test robot 
Link: https://lore.kernel.org/linux-mm/202205230239.EZxeZ3Fv-lkp@intel.com
Reported-by: Guenter Roeck 
Signed-off-by: Masahiro Yamada 
Signed-off-by: Kees Cook 
Link: https://lore.kernel.org/r/20220524135541.1453693-1-masahiroy@kernel.org

gcc-plugins: randstruct: Remove cast exception handling

2022-05-16T23:02:21+00:00

With all randstruct exceptions removed, remove all the exception
handling code. Any future warnings are likely to be shared between
this plugin and Clang randstruct, and will need to be addressed in a
more wholistic fashion.

Cc: Christoph Hellwig 
Cc: linux-hardening@vger.kernel.org
Signed-off-by: Kees Cook

af_unix: Silence randstruct GCC plugin warning

2022-05-16T23:02:21+00:00

While preparing for Clang randstruct support (which duplicated many of
the warnings the randstruct GCC plugin warned about), one strange one
remained only for the randstruct GCC plugin. Eliminating this rids
the plugin of the last exception.

It seems the plugin is happy to dereference individual members of
a cross-struct cast, but it is upset about casting to a whole object
pointer. This only manifests in one place in the kernel, so just replace
the variable with individual member accesses. There is no change in
executable instruction output.

Drop the last exception from the randstruct GCC plugin.

Cc: "David S. Miller" 
Cc: Christoph Hellwig 
Cc: Paolo Abeni 
Cc: Alexei Starovoitov 
Cc: Cong Wang 
Cc: Al Viro 
Cc: netdev@vger.kernel.org
Cc: linux-hardening@vger.kernel.org
Acked-by: Kuniyuki Iwashima 
Link: https://lore.kernel.org/lkml/20220511022217.58586-1-kuniyu@amazon.co.jp
Acked-by: Jakub Kicinski 
Link: https://lore.kernel.org/lkml/20220511151542.4cb3ff17@kernel.org
Signed-off-by: Kees Cook

niu: Silence randstruct warnings

2022-05-16T23:02:21+00:00

Clang randstruct gets upset when it sees struct addresspace (which is
randomized) being assigned to a struct page (which is not randomized):

drivers/net/ethernet/sun/niu.c:3385:12: error: casting from randomized structure pointer type 'struct address_space *' to 'struct page *'
                        *link = (struct page *) page->mapping;
                                ^

It looks like niu.c is looking for an in-line place to chain its allocated
pages together and is overloading the "mapping" member, as it is unused.
This is very non-standard, and is expected to be cleaned up in the
future[1], but there is no "correct" way to handle it today.

No meaningful machine code changes result after this change, and source
readability is improved.

Drop the randstruct exception now that there is no "confusing" cross-type
assignment.

[1] https://lore.kernel.org/lkml/YnqgjVoMDu5v9PNG@casper.infradead.org/

Cc: "Matthew Wilcox (Oracle)" 
Cc: Christoph Hellwig 
Cc: "David S. Miller" 
Cc: Jakub Kicinski 
Cc: Paolo Abeni 
Cc: Du Cheng 
Cc: Christophe JAILLET 
Cc: Vlastimil Babka 
Cc: William Kucharski 
Cc: Arnd Bergmann 
Cc: Nathan Chancellor 
Cc: netdev@vger.kernel.org
Cc: linux-mm@kvack.org
Cc: linux-hardening@vger.kernel.org
Acked-by: Jakub Kicinski 
Link: https://lore.kernel.org/lkml/20220511151647.7290adbe@kernel.org
Signed-off-by: Kees Cook