[<prev] [next>] [day] [month] [year] [list]
Message-ID: <202112081218.SHPzkHBb-lkp@intel.com>
Date: Wed, 8 Dec 2021 13:08:59 +0800
From: kernel test robot <lkp@...el.com>
To: Kees Cook <keescook@...omium.org>
Cc: kbuild-all@...ts.01.org, linux-kernel@...r.kernel.org
Subject: [kees:memcpy/step2/next-20211206 7/19]
include/linux/fortify-string.h:267:25: warning: call to
'__write_overflow_field' declared with attribute warning: detected write
beyond size of field (1st parameter); maybe use struct_group()?
tree: https://git.kernel.org/pub/scm/linux/kernel/git/kees/linux.git memcpy/step2/next-20211206
head: 082faead4a3c2e5d9f541f97d8d4d5fa0f88dce0
commit: a91965903a44bf236856efc7e20c6334c4e07388 [7/19] fortify: Detect struct member overflows in memcpy() at compile-time
config: powerpc-randconfig-c023-20211207 (https://download.01.org/0day-ci/archive/20211208/202112081218.SHPzkHBb-lkp@intel.com/config)
compiler: powerpc-linux-gcc (GCC) 11.2.0
reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# https://git.kernel.org/pub/scm/linux/kernel/git/kees/linux.git/commit/?id=a91965903a44bf236856efc7e20c6334c4e07388
git remote add kees https://git.kernel.org/pub/scm/linux/kernel/git/kees/linux.git
git fetch --no-tags kees memcpy/step2/next-20211206
git checkout a91965903a44bf236856efc7e20c6334c4e07388
# save the config file to linux build tree
mkdir build_dir
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-11.2.0 make.cross O=build_dir ARCH=powerpc SHELL=/bin/bash drivers/net/ethernet/intel/iavf/ fs/dlm/
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@...el.com>
All warnings (new ones prefixed by >>):
In file included from include/linux/string.h:253,
from include/linux/bitmap.h:11,
from include/linux/cpumask.h:12,
from include/linux/smp.h:13,
from include/linux/tracepoint.h:15,
from include/trace/events/dlm.h:10,
from fs/dlm/lock.c:56:
In function 'fortify_memcpy_chk',
inlined from 'send_repeat_remove' at fs/dlm/lock.c:4075:2:
>> include/linux/fortify-string.h:267:25: warning: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning]
267 | __write_overflow_field(p_size_field, size);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In function 'fortify_memcpy_chk',
inlined from 'shrink_bucket' at fs/dlm/lock.c:1792:3:
>> include/linux/fortify-string.h:267:25: warning: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning]
267 | __write_overflow_field(p_size_field, size);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
vim +/__write_overflow_field +267 include/linux/fortify-string.h
213
214 /*
215 * To make sure the compiler can enforce protection against buffer overflows,
216 * memcpy(), memmove(), and memset() must not be used beyond individual
217 * struct members. If you need to copy across multiple members, please use
218 * struct_group() to create a named mirror of an anonymous struct union.
219 * (e.g. see struct sk_buff.)
220 *
221 * Mitigation coverage
222 * Bounds checking at:
223 * +-------+-------+-------+-------+
224 * | Compile time | Run time |
225 * memcpy() argument sizes: | write | read | write | read |
226 * +-------+-------+-------+-------+
227 * memcpy(known, known, constant) | y | y | n/a | n/a |
228 * memcpy(unknown, known, constant) | n | y | V | n/a |
229 * memcpy(known, unknown, constant) | y | n | n/a | V |
230 * memcpy(unknown, unknown, constant) | n | n | V | V |
231 * memcpy(known, known, dynamic) | n | n | b | B |
232 * memcpy(unknown, known, dynamic) | n | n | V | B |
233 * memcpy(known, unknown, dynamic) | n | n | b | V |
234 * memcpy(unknown, unknown, dynamic) | n | n | V | V |
235 * +-------+-------+-------+-------+
236 *
237 * y = deterministic compile-time bounds checking
238 * n = cannot do deterministic compile-time bounds checking
239 * n/a = no run-time bounds checking needed since compile-time deterministic
240 * b = perform run-time bounds checking
241 * B = can perform run-time bounds checking, but current unenforced
242 * V = vulnerable to run-time overflow
243 *
244 */
245 __FORTIFY_INLINE void fortify_memcpy_chk(__kernel_size_t size,
246 const size_t p_size,
247 const size_t q_size,
248 const size_t p_size_field,
249 const size_t q_size_field,
250 const char *func)
251 {
252 if (__builtin_constant_p(size)) {
253 /*
254 * Length argument is a constant expression, so we
255 * can perform compile-time bounds checking where
256 * buffer sizes are known.
257 */
258
259 /* Error when size is larger than enclosing struct. */
260 if (p_size > p_size_field && p_size < size)
261 __write_overflow();
262 if (q_size > q_size_field && q_size < size)
263 __read_overflow2();
264
265 /* Warn when write size argument larger than dest field. */
266 if (p_size_field < size)
> 267 __write_overflow_field(p_size_field, size);
268 /*
269 * Warn for source field over-read when building with W=1
270 * or when an over-write happened, so both can be fixed at
271 * the same time.
272 */
273 if ((IS_ENABLED(KBUILD_EXTRA_WARN1) || p_size_field < size) &&
274 q_size_field < size)
275 __read_overflow2_field(q_size_field, size);
276 }
277 /*
278 * At this point, length argument may not be a constant expression,
279 * so run-time bounds checking can be done where buffer sizes are
280 * known. (This is not an "else" because the above checks may only
281 * be compile-time warnings, and we want to still warn for run-time
282 * overflows.)
283 */
284
285 /*
286 * Always stop accesses beyond the struct that contains the
287 * field, when the buffer's remaining size is known.
288 * (The -1 test is to optimize away checks where the buffer
289 * lengths are unknown.)
290 */
291 if ((p_size != (size_t)(-1) && p_size < size) ||
292 (q_size != (size_t)(-1) && q_size < size))
293 fortify_panic(func);
294 }
295
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
Powered by blists - more mailing lists