From: kernel test robot <lkp@intel.com>
To: oe-kbuild@lists.linux.dev
Cc: lkp@intel.com, Dan Carpenter <error27@gmail.com>
Subject: [bpf-next:master 35/51] kernel/bpf/precision.c:1418 bpf_undo_precision() error: buffer overflow 'partition_by' 8 <= 8
Date: Sat, 21 Nov 2026 17:42:10 +0100 [thread overview]
Message-ID: <20261121173744.q01c3spZ-lkp@intel.com> (raw)
Unfiltered warning: this smatch report is sent as-is and has not been
reviewed for false positives, so please treat it with due skepticism.
BCC: lkp@intel.com
CC: oe-kbuild-all@lists.linux.dev
TO: Alexei Starovoitov <ast@kernel.org>
tree: https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git master
head: 9867c61abbd44f7053a871566b3d982f62c6d8dc
commit: ecb43124db3947b1d632e2fe651d994a399186ef [35/51] bpf: Partition the states after an error
:::::: branch date: 12 hours ago
:::::: commit date: 12 hours ago
config: parisc-randconfig-2200-20261121 (https://download.01.org/0day-ci/archive/20261121/20261121173744.q01c3spZ-lkp@intel.com/config)
compiler: hppa-linux-gcc (GCC) 16.2.0
smatch: v0.5.0-9187-g5189e3fb
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Reported-by: Dan Carpenter <error27@gmail.com>
| Closes: https://lore.kernel.org/r/20261121173744.q01c3spZ-lkp@intel.com/
smatch warnings:
kernel/bpf/precision.c:1418 bpf_undo_precision() error: buffer overflow 'partition_by' 8 <= 8
vim +/partition_by +1418 kernel/bpf/precision.c
ecb43124db3947 Alexei Starovoitov 2026-11-20 1380
ecb43124db3947 Alexei Starovoitov 2026-11-20 1381 /* The pass is over its budget. Returns true if another partition replaces the last one. */
ecb43124db3947 Alexei Starovoitov 2026-11-20 1382 bool bpf_undo_precision(struct bpf_verifier_env *env)
ecb43124db3947 Alexei Starovoitov 2026-11-20 1383 {
ecb43124db3947 Alexei Starovoitov 2026-11-20 1384 struct bpf_precision *prec = env->precision;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1385 bool over_budget = false, again = false;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1386 struct bpf_loop *loop;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1387 u32 i;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1388
ecb43124db3947 Alexei Starovoitov 2026-11-20 1389 prec->this_pass.over_budget = env->over_limit;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1390 prec->this_pass.calls_joined = env->calls_joined;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1391 if (!env->over_limit || env->cost > env->max_cost)
ecb43124db3947 Alexei Starovoitov 2026-11-20 1392 return false;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1393 if (prec->partitioned_by <= BIT(PARTITION_ADD_DEPS) ||
ecb43124db3947 Alexei Starovoitov 2026-11-20 1394 (env->err && (prec->last_pass.over_budget || partition_helped(env)))) {
ecb43124db3947 Alexei Starovoitov 2026-11-20 1395 env->over_limit = !env->err;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1396 return false;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1397 }
ecb43124db3947 Alexei Starovoitov 2026-11-20 1398 env->over_limit = false;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1399 memcpy(prec->data_deps, prec->undo, 5 * prec->words * sizeof(long));
ecb43124db3947 Alexei Starovoitov 2026-11-20 1400 for (i = 0; i < prec->loop_cnt; i++)
ecb43124db3947 Alexei Starovoitov 2026-11-20 1401 over_budget |= prec->loops[i]->partitioned &&
ecb43124db3947 Alexei Starovoitov 2026-11-20 1402 !bpf_loop_has_budget(prec->loops[i]);
ecb43124db3947 Alexei Starovoitov 2026-11-20 1403 for (i = 0; i < prec->loop_cnt; i++) {
ecb43124db3947 Alexei Starovoitov 2026-11-20 1404 loop = prec->loops[i];
ecb43124db3947 Alexei Starovoitov 2026-11-20 1405 if (loop->partitioned)
ecb43124db3947 Alexei Starovoitov 2026-11-20 1406 undo_loop(prec, loop);
ecb43124db3947 Alexei Starovoitov 2026-11-20 1407 /* if the old budget was enough, the pass possibly did not get to the loop */
ecb43124db3947 Alexei Starovoitov 2026-11-20 1408 if (loop->partitioned && over_budget && bpf_loop_has_budget(loop))
ecb43124db3947 Alexei Starovoitov 2026-11-20 1409 loop->tried &= ~prec->partitioned_by;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1410 /* peeled_apart did not help. The partition can fit without it. */
ecb43124db3947 Alexei Starovoitov 2026-11-20 1411 if (loop->peeled_apart && !loop->kept.peeled_apart) {
ecb43124db3947 Alexei Starovoitov 2026-11-20 1412 loop->peeled_apart = false;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1413 loop->tried |= BIT(PARTITION_PEELED_APART) |
ecb43124db3947 Alexei Starovoitov 2026-11-20 1414 BIT(PARTITION_TOO_COSTLY);
ecb43124db3947 Alexei Starovoitov 2026-11-20 1415 loop->changed = again = true;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1416 }
ecb43124db3947 Alexei Starovoitov 2026-11-20 1417 }
ecb43124db3947 Alexei Starovoitov 2026-11-20 @1418 if (partition_by[__ffs(prec->partitioned_by)] != partition_suspect_loops && !again)
ecb43124db3947 Alexei Starovoitov 2026-11-20 1419 prec->too_costly |= prec->partitioned_by;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1420 return partition_next(env) > 0;
ecb43124db3947 Alexei Starovoitov 2026-11-20 1421 }
ecb43124db3947 Alexei Starovoitov 2026-11-20 1422
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
reply other threads:[~2026-11-21 16:42 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20261121173744.q01c3spZ-lkp@intel.com \
--to=lkp@intel.com \
--cc=error27@gmail.com \
--cc=oe-kbuild@lists.linux.dev \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.