[PATCH] [v2] iommu, debugobjects: avoid gcc-16.1 section mismatch warnings

[PATCH] [v2] iommu, debugobjects: avoid gcc-16.1 section mismatch warnings

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

gcc-16 has gained some more advanced inter-procedual optimization
techniques that enable it to inline the dummy_tlb_add_page() and
dummy_tlb_flush() function pointers into a specialized version of
__arm_v7s_unmap:

WARNING: modpost: vmlinux: section mismatch in reference: __arm_v7s_unmap+0x2cc (section: .text) -> dummy_tlb_add_page (section: .init.text)
ERROR: modpost: Section mismatches detected.

From what I can tell, the transformation is correct, as this is only
called when __arm_v7s_unmap() is called from arm_v7s_do_selftests(),
which is also __init. Since __arm_v7s_unmap() however is not __init,
gcc cannot inline the inner function calls directly.

In debug_objects_selftest(), the same thing happens. Both the
caller and the leaf function are __init, but the IPA pulls
it into a non-init one:

WARNING: modpost: vmlinux: section mismatch in reference: lookup_object_or_alloc+0x7c (section: .text.lookup_object_or_alloc) -> is_static_object (section: .init.text)

Marking the affected functions as not "__init" would reliably avoid this
issue but is not a good solution because it removes an otherwise correct
annotation. I tried marking the functions as 'noinline', but that ended
up not covering all the affected configurations.

With some more experimenting, I found that marking these functions as
__attribute__((noipa)) is both logical and reliable.

In order to keep the syntax readable, add a custom macro for this in
include/linux/compiler_attributes.h next to other related macros and
use it to annotate both files.

Link: https://lore.kernel.org/all/abRB6g-48ZX6Yl2r@willie-the-truck/
Cc: Will Deacon <will@kernel.org>
Cc: Thomas Gleixner <tglx@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Miguel Ojeda <ojeda@kernel.org>
Cc: linux-kbuild@vger.kernel.org
Cc: stable@vger.kernel.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
v2: I merged both patches into one, to simplify the dependency
on the new compiler_attributes.h macro.
---
 drivers/iommu/io-pgtable-arm-v7s.c  | 18 ++++++++++++------
 include/linux/compiler_attributes.h | 11 +++++++++++
 lib/debugobjects.c                  |  2 +-
 3 files changed, 24 insertions(+), 7 deletions(-)

diff --git a/drivers/iommu/io-pgtable-arm-v7s.c b/drivers/iommu/io-pgtable-arm-v7s.c
index 40e33257d3c2..1dbef8c55007 100644
--- a/drivers/iommu/io-pgtable-arm-v7s.c
+++ b/drivers/iommu/io-pgtable-arm-v7s.c
@@ -777,21 +777,27 @@ struct io_pgtable_init_fns io_pgtable_arm_v7s_init_fns = {
 
 static struct io_pgtable_cfg *cfg_cookie __initdata;
 
-static void __init dummy_tlb_flush_all(void *cookie)
+/*
+ * __noipa prevents gcc from turning indirect iommu_flush_ops calls
+ * into direct calls from a specialized __arm_v7s_unmap() that triggers
+ * a build time section mismatch assertion.
+ */
+static __noipa void __init dummy_tlb_flush_all(void *cookie)
 {
 	WARN_ON(cookie != cfg_cookie);
 }
 
-static void __init dummy_tlb_flush(unsigned long iova, size_t size,
-				   size_t granule, void *cookie)
+static __noipa void __init dummy_tlb_flush(unsigned long iova, size_t size,
+					   size_t granule, void *cookie)
 {
 	WARN_ON(cookie != cfg_cookie);
 	WARN_ON(!(size & cfg_cookie->pgsize_bitmap));
 }
 
-static void __init dummy_tlb_add_page(struct iommu_iotlb_gather *gather,
-				      unsigned long iova, size_t granule,
-				      void *cookie)
+static __noipa void __init dummy_tlb_add_page(struct iommu_iotlb_gather *gather,
+					      unsigned long iova,
+					      size_t granule,
+					      void *cookie)
 {
 	dummy_tlb_flush(iova, granule, granule, cookie);
 }
diff --git a/include/linux/compiler_attributes.h b/include/linux/compiler_attributes.h
index c16d4199bf92..836a50f5917a 100644
--- a/include/linux/compiler_attributes.h
+++ b/include/linux/compiler_attributes.h
@@ -396,6 +396,17 @@
 # define __disable_sanitizer_instrumentation
 #endif
 
+/*
+ * Optional: not supported by clang
+ *
+ *   gcc: https://gcc.gnu.org/onlinedocs/gcc/Common-Attributes.html#index-noipa
+ */
+#if __has_attribute(noipa)
+# define __noipa __attribute__((noipa))
+#else
+# define __noipa
+#endif
+
 /*
  *   gcc: https://gcc.gnu.org/onlinedocs/gcc/Common-Function-Attributes.html#index-weak-function-attribute
  *   gcc: https://gcc.gnu.org/onlinedocs/gcc/Common-Variable-Attributes.html#index-weak-variable-attribute
diff --git a/lib/debugobjects.c b/lib/debugobjects.c
index 12e2e42e6a31..c93b7ca3e1ab 100644
--- a/lib/debugobjects.c
+++ b/lib/debugobjects.c
@@ -1212,7 +1212,7 @@ struct self_test {
 
 static __initconst const struct debug_obj_descr descr_type_test;
 
-static bool __init is_static_object(void *addr)
+static __noipa bool __init is_static_object(void *addr)
 {
 	struct self_test *obj = addr;
 
-- 
2.39.5

Re: [PATCH] [v2] iommu, debugobjects: avoid gcc-16.1 section mismatch warnings

[Miguel Ojeda]

On Wed, May 13, 2026 at 4:54 PM Arnd Bergmann <arnd@kernel.org> wrote:
>
> With some more experimenting, I found that marking these functions as
> __attribute__((noipa)) is both logical and reliable.
>
> In order to keep the syntax readable, add a custom macro for this in
> include/linux/compiler_attributes.h next to other related macros and
> use it to annotate both files.

Acked-by: Miguel Ojeda <ojeda@kernel.org>

I double-checked that GCC 8.1 implemented the attribute in commit
036ea39917b0 ("attribs.c (decl_attributes): Imply noinline, noclone
and no_icf attributes for noipa attribute."), which also happens to be
our minimum in Linux; and that Clang indeed does not seem to support
it (not even a mention of it in their repository).

If you don't mind, please add underscores (i.e. `((__noipa__))`) and
place it after `__noinline__` to keep it sorted (the file is meant to
be sorted by actual attribute name, though some entries were added
that break that, but I will clean that and a couple other things up at
some point).

Thanks!

Cheers,
Miguel