On Wed, Mar 8, 2023 at 8:35 PM Dan Carpenter <error27@xxxxxxxxx> wrote: > > Hi Andrii, > > url: https://github.com/intel-lab-lkp/linux/commits/Andrii-Nakryiko/bpf-factor-out-fetching-basic-kfunc-metadata/20230308-115539 > base: https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git master > patch link: https://lore.kernel.org/r/20230308035416.2591326-4-andrii%40kernel.org > patch subject: [PATCH v4 bpf-next 3/8] bpf: add support for open-coded iterator loops > config: loongarch-randconfig-m041-20230305 (https://download.01.org/0day-ci/archive/20230309/202303090153.YeswNcW4-lkp@xxxxxxxxx/config) > compiler: loongarch64-linux-gcc (GCC) 12.1.0 > > If you fix the issue, kindly add following tag where applicable > | Reported-by: kernel test robot <lkp@xxxxxxxxx> > | Reported-by: Dan Carpenter <error27@xxxxxxxxx> > | Link: https://lore.kernel.org/r/202303090153.YeswNcW4-lkp@xxxxxxxxx/ > > smatch warnings: > kernel/bpf/verifier.c:1244 is_iter_reg_valid_uninit() error: uninitialized symbol 'j'. > > vim +/j +1244 kernel/bpf/verifier.c > > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1225 static bool is_iter_reg_valid_uninit(struct bpf_verifier_env *env, > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1226 struct bpf_reg_state *reg, int nr_slots) > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1227 { > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1228 struct bpf_func_state *state = func(env, reg); > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1229 int spi, i, j; > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1230 > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1231 /* For -ERANGE (i.e. spi not falling into allocated stack slots), we > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1232 * will do check_mem_access to check and update stack bounds later, so > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1233 * return true for that case. > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1234 */ > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1235 spi = iter_get_spi(env, reg, nr_slots); > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1236 if (spi == -ERANGE) > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1237 return true; > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1238 if (spi < 0) > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1239 return spi; > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1240 > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1241 for (i = 0; i < nr_slots; i++) { > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1242 struct bpf_stack_state *slot = &state->stack[spi - i]; > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1243 > 8f263e1296a91f Andrii Nakryiko 2023-03-07 @1244 if (slot->slot_type[j] == STACK_ITER) > ^ > s/j/i/? > nope, I accidentally removed the inner for loop. I fixed all that in v5, which was applied today. But thanks for the notification! > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1245 return false; > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1246 } > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1247 > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1248 return true; > 8f263e1296a91f Andrii Nakryiko 2023-03-07 1249 } > > -- > 0-DAY CI Kernel Test Service > https://github.com/intel/lkp-tests >
