fs/udf/directory.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
Use unsigned type in call to macro mint_t(). This avoids confusing the
compiler about possible negative values that would cause the value in
_len_ to wrap around.
Fixes the following -Wstringop-warnings seen when building ARM
architecture with allyesconfig (GCC 13):
fs/udf/directory.c: In function 'udf_copy_fi':
include/linux/fortify-string.h:57:33: warning: '__builtin_memcpy' specified bound between 2147483648 and 4294967295 exceeds maximum object size 2147483647 [-Wstringop-overflow=]
57 | #define __underlying_memcpy __builtin_memcpy
| ^
include/linux/fortify-string.h:648:9: note: in expansion of macro '__underlying_memcpy'
648 | __underlying_##op(p, q, __fortify_size); \
| ^~~~~~~~~~~~~
include/linux/fortify-string.h:693:26: note: in expansion of macro '__fortify_memcpy_chk'
693 | #define memcpy(p, q, s) __fortify_memcpy_chk(p, q, s, \
| ^~~~~~~~~~~~~~~~~~~~
fs/udf/directory.c:99:9: note: in expansion of macro 'memcpy'
99 | memcpy(&iter->fi, iter->bh[0]->b_data + off, len);
| ^~~~~~
include/linux/fortify-string.h:57:33: warning: '__builtin_memcpy' specified bound between 2147483648 and 4294967295 exceeds maximum object size 2147483647 [-Wstringop-overflow=]
57 | #define __underlying_memcpy __builtin_memcpy
| ^
include/linux/fortify-string.h:648:9: note: in expansion of macro '__underlying_memcpy'
648 | __underlying_##op(p, q, __fortify_size); \
| ^~~~~~~~~~~~~
include/linux/fortify-string.h:693:26: note: in expansion of macro '__fortify_memcpy_chk'
693 | #define memcpy(p, q, s) __fortify_memcpy_chk(p, q, s, \
| ^~~~~~~~~~~~~~~~~~~~
fs/udf/directory.c:99:9: note: in expansion of macro 'memcpy'
99 | memcpy(&iter->fi, iter->bh[0]->b_data + off, len);
| ^~~~~~
AR fs/udf/built-in.a
This helps with the ongoing efforts to globally enable
-Wstringop-overflow.
Link: https://github.com/KSPP/linux/issues/329
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
---
fs/udf/directory.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/udf/directory.c b/fs/udf/directory.c
index 1c775e072b2f..93153665eb37 100644
--- a/fs/udf/directory.c
+++ b/fs/udf/directory.c
@@ -95,7 +95,7 @@ static int udf_copy_fi(struct udf_fileident_iter *iter)
}
off = iter->pos & (blksize - 1);
- len = min_t(int, sizeof(struct fileIdentDesc), blksize - off);
+ len = min_t(u32, sizeof(struct fileIdentDesc), blksize - off);
memcpy(&iter->fi, iter->bh[0]->b_data + off, len);
if (len < sizeof(struct fileIdentDesc))
memcpy((char *)(&iter->fi) + len, iter->bh[1]->b_data,
--
2.34.1
On Wed, Jul 12, 2023 at 12:25:45PM -0600, Gustavo A. R. Silva wrote: > Use unsigned type in call to macro mint_t(). This avoids confusing the > compiler about possible negative values that would cause the value in > _len_ to wrap around. > > Fixes the following -Wstringop-warnings seen when building ARM > architecture with allyesconfig (GCC 13): > fs/udf/directory.c: In function 'udf_copy_fi': > include/linux/fortify-string.h:57:33: warning: '__builtin_memcpy' specified bound between 2147483648 and 4294967295 exceeds maximum object size 2147483647 [-Wstringop-overflow=] > 57 | #define __underlying_memcpy __builtin_memcpy > | ^ > include/linux/fortify-string.h:648:9: note: in expansion of macro '__underlying_memcpy' > 648 | __underlying_##op(p, q, __fortify_size); \ > | ^~~~~~~~~~~~~ > include/linux/fortify-string.h:693:26: note: in expansion of macro '__fortify_memcpy_chk' > 693 | #define memcpy(p, q, s) __fortify_memcpy_chk(p, q, s, \ > | ^~~~~~~~~~~~~~~~~~~~ > fs/udf/directory.c:99:9: note: in expansion of macro 'memcpy' > 99 | memcpy(&iter->fi, iter->bh[0]->b_data + off, len); > | ^~~~~~ > include/linux/fortify-string.h:57:33: warning: '__builtin_memcpy' specified bound between 2147483648 and 4294967295 exceeds maximum object size 2147483647 [-Wstringop-overflow=] > 57 | #define __underlying_memcpy __builtin_memcpy > | ^ > include/linux/fortify-string.h:648:9: note: in expansion of macro '__underlying_memcpy' > 648 | __underlying_##op(p, q, __fortify_size); \ > | ^~~~~~~~~~~~~ > include/linux/fortify-string.h:693:26: note: in expansion of macro '__fortify_memcpy_chk' > 693 | #define memcpy(p, q, s) __fortify_memcpy_chk(p, q, s, \ > | ^~~~~~~~~~~~~~~~~~~~ > fs/udf/directory.c:99:9: note: in expansion of macro 'memcpy' > 99 | memcpy(&iter->fi, iter->bh[0]->b_data + off, len); > | ^~~~~~ > AR fs/udf/built-in.a > > This helps with the ongoing efforts to globally enable > -Wstringop-overflow. > > Link: https://github.com/KSPP/linux/issues/329 > Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org> > --- > fs/udf/directory.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/fs/udf/directory.c b/fs/udf/directory.c > index 1c775e072b2f..93153665eb37 100644 > --- a/fs/udf/directory.c > +++ b/fs/udf/directory.c > @@ -95,7 +95,7 @@ static int udf_copy_fi(struct udf_fileident_iter *iter) > } > > off = iter->pos & (blksize - 1); > - len = min_t(int, sizeof(struct fileIdentDesc), blksize - off); > + len = min_t(u32, sizeof(struct fileIdentDesc), blksize - off); > memcpy(&iter->fi, iter->bh[0]->b_data + off, len); > if (len < sizeof(struct fileIdentDesc)) > memcpy((char *)(&iter->fi) + len, iter->bh[1]->b_data, len is u32, "off" can't be less than blksize, so this all looks correct to me. Thanks! Reviewed-by: Kees Cook <keescook@chromium.org> -- Kees Cook
On Wed 12-07-23 12:05:34, Kees Cook wrote: > On Wed, Jul 12, 2023 at 12:25:45PM -0600, Gustavo A. R. Silva wrote: > > Use unsigned type in call to macro mint_t(). This avoids confusing the > > compiler about possible negative values that would cause the value in > > _len_ to wrap around. > > > > Fixes the following -Wstringop-warnings seen when building ARM > > architecture with allyesconfig (GCC 13): > > fs/udf/directory.c: In function 'udf_copy_fi': > > include/linux/fortify-string.h:57:33: warning: '__builtin_memcpy' specified bound between 2147483648 and 4294967295 exceeds maximum object size 2147483647 [-Wstringop-overflow=] > > 57 | #define __underlying_memcpy __builtin_memcpy > > | ^ > > include/linux/fortify-string.h:648:9: note: in expansion of macro '__underlying_memcpy' > > 648 | __underlying_##op(p, q, __fortify_size); \ > > | ^~~~~~~~~~~~~ > > include/linux/fortify-string.h:693:26: note: in expansion of macro '__fortify_memcpy_chk' > > 693 | #define memcpy(p, q, s) __fortify_memcpy_chk(p, q, s, \ > > | ^~~~~~~~~~~~~~~~~~~~ > > fs/udf/directory.c:99:9: note: in expansion of macro 'memcpy' > > 99 | memcpy(&iter->fi, iter->bh[0]->b_data + off, len); > > | ^~~~~~ > > include/linux/fortify-string.h:57:33: warning: '__builtin_memcpy' specified bound between 2147483648 and 4294967295 exceeds maximum object size 2147483647 [-Wstringop-overflow=] > > 57 | #define __underlying_memcpy __builtin_memcpy > > | ^ > > include/linux/fortify-string.h:648:9: note: in expansion of macro '__underlying_memcpy' > > 648 | __underlying_##op(p, q, __fortify_size); \ > > | ^~~~~~~~~~~~~ > > include/linux/fortify-string.h:693:26: note: in expansion of macro '__fortify_memcpy_chk' > > 693 | #define memcpy(p, q, s) __fortify_memcpy_chk(p, q, s, \ > > | ^~~~~~~~~~~~~~~~~~~~ > > fs/udf/directory.c:99:9: note: in expansion of macro 'memcpy' > > 99 | memcpy(&iter->fi, iter->bh[0]->b_data + off, len); > > | ^~~~~~ > > AR fs/udf/built-in.a > > > > This helps with the ongoing efforts to globally enable > > -Wstringop-overflow. > > > > Link: https://github.com/KSPP/linux/issues/329 > > Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org> > > --- > > fs/udf/directory.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > diff --git a/fs/udf/directory.c b/fs/udf/directory.c > > index 1c775e072b2f..93153665eb37 100644 > > --- a/fs/udf/directory.c > > +++ b/fs/udf/directory.c > > @@ -95,7 +95,7 @@ static int udf_copy_fi(struct udf_fileident_iter *iter) > > } > > > > off = iter->pos & (blksize - 1); > > - len = min_t(int, sizeof(struct fileIdentDesc), blksize - off); > > + len = min_t(u32, sizeof(struct fileIdentDesc), blksize - off); > > memcpy(&iter->fi, iter->bh[0]->b_data + off, len); > > if (len < sizeof(struct fileIdentDesc)) > > memcpy((char *)(&iter->fi) + len, iter->bh[1]->b_data, > > len is u32, "off" can't be less than blksize, so this all looks correct > to me. Thanks! > > Reviewed-by: Kees Cook <keescook@chromium.org> Thanks! I've merged the patch to my tree. Honza -- Jan Kara <jack@suse.com> SUSE Labs, CR
© 2016 - 2025 Red Hat, Inc.