| Message ID | 20210818214021.2476230-2-keescook@chromium.org (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | Add __alloc_size() for better bounds checking | expand |
On 8/18/2021 2:40 PM, Kees Cook wrote: > GCC and Clang can use the "alloc_size" attribute to better inform the > results of __builtin_object_size() (for compile-time constant values). > Clang can additionally use alloc_size to inform the results of > __builtin_dynamic_object_size() (for run-time values). > > Because GCC sees the frequent use of struct_size() as an allocator size > argument, and notices it can return SIZE_MAX (the overflow indication), > it complains about these call sites may overflow (since SIZE_MAX is > greater than the default -Walloc-size-larger-than=PTRDIFF_MAX). This > isn't helpful since we already know a SIZE_MAX will be caught at run-time > (this was an intentional design). Instead, just disable this check as > it is both a false positive and redundant. (Clang does not have this > warning option.) > > Cc: Miguel Ojeda <ojeda@kernel.org> > Cc: Nathan Chancellor <nathan@kernel.org> > Cc: Nick Desaulniers <ndesaulniers@google.com> > Cc: clang-built-linux@googlegroups.com > Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Nathan Chancellor <nathan@kernel.org> > --- > Makefile | 6 +++++- > include/linux/compiler_attributes.h | 6 ++++++ > 2 files changed, 11 insertions(+), 1 deletion(-) > > diff --git a/Makefile b/Makefile > index 72f9e2b0202c..34cffcdfd5dc 100644 > --- a/Makefile > +++ b/Makefile > @@ -1078,9 +1078,13 @@ KBUILD_CFLAGS += $(call cc-disable-warning, stringop-overflow) > # Another good warning that we'll want to enable eventually > KBUILD_CFLAGS += $(call cc-disable-warning, restrict) > > -# Enabled with W=2, disabled by default as noisy > ifdef CONFIG_CC_IS_GCC > +# Enabled with W=2, disabled by default as noisy > KBUILD_CFLAGS += -Wno-maybe-uninitialized > + > +# The allocators already balk at large sizes, so silence the compiler > +# warnings for bounds checks involving those possible values. > +KBUILD_CFLAGS += -Wno-alloc-size-larger-than > endif > > # disable invalid "can't wrap" optimizations for signed / pointers > diff --git a/include/linux/compiler_attributes.h b/include/linux/compiler_attributes.h > index 67c5667f8042..203b0ac62d15 100644 > --- a/include/linux/compiler_attributes.h > +++ b/include/linux/compiler_attributes.h > @@ -54,6 +54,12 @@ > #define __aligned(x) __attribute__((__aligned__(x))) > #define __aligned_largest __attribute__((__aligned__)) > > +/* > + * gcc: https://gcc.gnu.org/onlinedocs/gcc/Common-Function-Attributes.html#index-alloc_005fsize-function-attribute > + * clang: https://clang.llvm.org/docs/AttributeReference.html#alloc-size > + */ > +#define __alloc_size(x, ...) __attribute__((__alloc_size__(x, ## __VA_ARGS__))) > + > /* > * Note: users of __always_inline currently do not write "inline" themselves, > * which seems to be required by gcc to apply the attribute according >
On Wed, 18 Aug 2021 14:40:15 -0700 Kees Cook <keescook@chromium.org> wrote: > GCC and Clang can use the "alloc_size" attribute to better inform the > results of __builtin_object_size() (for compile-time constant values). > Clang can additionally use alloc_size to inform the results of > __builtin_dynamic_object_size() (for run-time values). > > Because GCC sees the frequent use of struct_size() as an allocator size > argument, and notices it can return SIZE_MAX (the overflow indication), > it complains about these call sites may overflow (since SIZE_MAX is > greater than the default -Walloc-size-larger-than=PTRDIFF_MAX). This > isn't helpful since we already know a SIZE_MAX will be caught at run-time > (this was an intentional design). Instead, just disable this check as > it is both a false positive and redundant. (Clang does not have this > warning option.) > > ... > > --- a/Makefile > +++ b/Makefile > @@ -1078,9 +1078,13 @@ KBUILD_CFLAGS += $(call cc-disable-warning, stringop-overflow) > # Another good warning that we'll want to enable eventually > KBUILD_CFLAGS += $(call cc-disable-warning, restrict) > > -# Enabled with W=2, disabled by default as noisy > ifdef CONFIG_CC_IS_GCC > +# Enabled with W=2, disabled by default as noisy > KBUILD_CFLAGS += -Wno-maybe-uninitialized > + > +# The allocators already balk at large sizes, so silence the compiler > +# warnings for bounds checks involving those possible values. > +KBUILD_CFLAGS += -Wno-alloc-size-larger-than > endif > > # disable invalid "can't wrap" optimizations for signed / pointers Makefile has changed. I did this: --- a/Makefile~compiler-attributes-add-__alloc_size-for-better-bounds-checking +++ a/Makefile @@ -1003,6 +1003,12 @@ KBUILD_CFLAGS += $(call cc-disable-warni # Enabled with W=2, disabled by default as noisy KBUILD_CFLAGS += $(call cc-disable-warning, maybe-uninitialized) +ifdef CONFIG_CC_IS_GCC +# The allocators already balk at large sizes, so silence the compiler +# warnings for bounds checks involving those possible values. +KBUILD_CFLAGS += -Wno-alloc-size-larger-than +endif + # disable invalid "can't wrap" optimizations for signed / pointers KBUILD_CFLAGS += -fno-strict-overflow
On Wed, Aug 18, 2021 at 11:40 PM Kees Cook <keescook@chromium.org> wrote: > > GCC and Clang can use the "alloc_size" attribute to better inform the > results of __builtin_object_size() (for compile-time constant values). > Clang can additionally use alloc_size to inform the results of > __builtin_dynamic_object_size() (for run-time values). > > Because GCC sees the frequent use of struct_size() as an allocator size > argument, and notices it can return SIZE_MAX (the overflow indication), > it complains about these call sites may overflow (since SIZE_MAX is > greater than the default -Walloc-size-larger-than=PTRDIFF_MAX). This > isn't helpful since we already know a SIZE_MAX will be caught at run-time > (this was an intentional design). Instead, just disable this check as > it is both a false positive and redundant. (Clang does not have this > warning option.) Thanks! Reviewed-by: Miguel Ojeda <ojeda@kernel.org> Cheers, Miguel
On Wed, Aug 18, 2021 at 04:19:12PM -0700, Andrew Morton wrote: > On Wed, 18 Aug 2021 14:40:15 -0700 Kees Cook <keescook@chromium.org> wrote: > > > GCC and Clang can use the "alloc_size" attribute to better inform the > > results of __builtin_object_size() (for compile-time constant values). > > Clang can additionally use alloc_size to inform the results of > > __builtin_dynamic_object_size() (for run-time values). > > > > Because GCC sees the frequent use of struct_size() as an allocator size > > argument, and notices it can return SIZE_MAX (the overflow indication), > > it complains about these call sites may overflow (since SIZE_MAX is > > greater than the default -Walloc-size-larger-than=PTRDIFF_MAX). This > > isn't helpful since we already know a SIZE_MAX will be caught at run-time > > (this was an intentional design). Instead, just disable this check as > > it is both a false positive and redundant. (Clang does not have this > > warning option.) > > > > ... > > > > --- a/Makefile > > +++ b/Makefile > > @@ -1078,9 +1078,13 @@ KBUILD_CFLAGS += $(call cc-disable-warning, stringop-overflow) > > # Another good warning that we'll want to enable eventually > > KBUILD_CFLAGS += $(call cc-disable-warning, restrict) > > > > -# Enabled with W=2, disabled by default as noisy > > ifdef CONFIG_CC_IS_GCC > > +# Enabled with W=2, disabled by default as noisy > > KBUILD_CFLAGS += -Wno-maybe-uninitialized > > + > > +# The allocators already balk at large sizes, so silence the compiler > > +# warnings for bounds checks involving those possible values. > > +KBUILD_CFLAGS += -Wno-alloc-size-larger-than > > endif > > > > # disable invalid "can't wrap" optimizations for signed / pointers > > Makefile has changed. I did this: > > --- a/Makefile~compiler-attributes-add-__alloc_size-for-better-bounds-checking > +++ a/Makefile > @@ -1003,6 +1003,12 @@ KBUILD_CFLAGS += $(call cc-disable-warni > # Enabled with W=2, disabled by default as noisy > KBUILD_CFLAGS += $(call cc-disable-warning, maybe-uninitialized) > > +ifdef CONFIG_CC_IS_GCC > +# The allocators already balk at large sizes, so silence the compiler > +# warnings for bounds checks involving those possible values. > +KBUILD_CFLAGS += -Wno-alloc-size-larger-than > +endif > + > # disable invalid "can't wrap" optimizations for signed / pointers > KBUILD_CFLAGS += -fno-strict-overflow Oh, er, where did "Makefile: remove stale cc-option checks" go? Ah, I see now: https://lore.kernel.org/mm-commits/20210814215814.W_qqW%25akpm@linux-foundation.org/T/#u Looks like I just happened to pick the wrong linux-next. ;) Thanks for the fix-up!
diff --git a/Makefile b/Makefile index 72f9e2b0202c..34cffcdfd5dc 100644 --- a/Makefile +++ b/Makefile @@ -1078,9 +1078,13 @@ KBUILD_CFLAGS += $(call cc-disable-warning, stringop-overflow) # Another good warning that we'll want to enable eventually KBUILD_CFLAGS += $(call cc-disable-warning, restrict) -# Enabled with W=2, disabled by default as noisy ifdef CONFIG_CC_IS_GCC +# Enabled with W=2, disabled by default as noisy KBUILD_CFLAGS += -Wno-maybe-uninitialized + +# The allocators already balk at large sizes, so silence the compiler +# warnings for bounds checks involving those possible values. +KBUILD_CFLAGS += -Wno-alloc-size-larger-than endif # disable invalid "can't wrap" optimizations for signed / pointers diff --git a/include/linux/compiler_attributes.h b/include/linux/compiler_attributes.h index 67c5667f8042..203b0ac62d15 100644 --- a/include/linux/compiler_attributes.h +++ b/include/linux/compiler_attributes.h @@ -54,6 +54,12 @@ #define __aligned(x) __attribute__((__aligned__(x))) #define __aligned_largest __attribute__((__aligned__)) +/* + * gcc: https://gcc.gnu.org/onlinedocs/gcc/Common-Function-Attributes.html#index-alloc_005fsize-function-attribute + * clang: https://clang.llvm.org/docs/AttributeReference.html#alloc-size + */ +#define __alloc_size(x, ...) __attribute__((__alloc_size__(x, ## __VA_ARGS__))) + /* * Note: users of __always_inline currently do not write "inline" themselves, * which seems to be required by gcc to apply the attribute according
GCC and Clang can use the "alloc_size" attribute to better inform the results of __builtin_object_size() (for compile-time constant values). Clang can additionally use alloc_size to inform the results of __builtin_dynamic_object_size() (for run-time values). Because GCC sees the frequent use of struct_size() as an allocator size argument, and notices it can return SIZE_MAX (the overflow indication), it complains about these call sites may overflow (since SIZE_MAX is greater than the default -Walloc-size-larger-than=PTRDIFF_MAX). This isn't helpful since we already know a SIZE_MAX will be caught at run-time (this was an intentional design). Instead, just disable this check as it is both a false positive and redundant. (Clang does not have this warning option.) Cc: Miguel Ojeda <ojeda@kernel.org> Cc: Nathan Chancellor <nathan@kernel.org> Cc: Nick Desaulniers <ndesaulniers@google.com> Cc: clang-built-linux@googlegroups.com Signed-off-by: Kees Cook <keescook@chromium.org> --- Makefile | 6 +++++- include/linux/compiler_attributes.h | 6 ++++++ 2 files changed, 11 insertions(+), 1 deletion(-)