Re: [PATCH bpf-next v3 3/6] bpf: Add a bpf_snprintf helper

From: kernel test robot
Date: Mon Apr 12 2021 - 15:11:47 EST


Hi Florent,

I love your patch! Perhaps something to improve:

[auto build test WARNING on bpf-next/master]

url: https://github.com/0day-ci/linux/commits/Florent-Revest/Add-a-snprintf-eBPF-helper/20210412-233921
base: https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git master
config: i386-randconfig-s002-20210412 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-22) 9.3.0
reproduce:
# apt-get install sparse
# sparse version: v0.6.3-280-g2cd6d34e-dirty
# https://github.com/0day-ci/linux/commit/168301dda58989eca274d5f5c926675540010e13
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Florent-Revest/Add-a-snprintf-eBPF-helper/20210412-233921
git checkout 168301dda58989eca274d5f5c926675540010e13
# save the attached .config to linux build tree
make W=1 C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' ARCH=i386

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@xxxxxxxxx>


sparse warnings: (new ones prefixed by >>)
kernel/bpf/verifier.c:1674:41: sparse: sparse: subtraction of functions? Share your drugs
kernel/bpf/verifier.c:12051:76: sparse: sparse: subtraction of functions? Share your drugs
kernel/bpf/verifier.c:12489:81: sparse: sparse: subtraction of functions? Share your drugs
kernel/bpf/verifier.c:12493:81: sparse: sparse: subtraction of functions? Share your drugs
kernel/bpf/verifier.c:12497:81: sparse: sparse: subtraction of functions? Share your drugs
kernel/bpf/verifier.c:12501:79: sparse: sparse: subtraction of functions? Share your drugs
kernel/bpf/verifier.c:12505:78: sparse: sparse: subtraction of functions? Share your drugs
kernel/bpf/verifier.c:12509:79: sparse: sparse: subtraction of functions? Share your drugs
kernel/bpf/verifier.c:12513:78: sparse: sparse: subtraction of functions? Share your drugs
kernel/bpf/verifier.c:12557:38: sparse: sparse: subtraction of functions? Share your drugs
>> kernel/bpf/verifier.c:5944:23: sparse: sparse: non size-preserving integer to pointer cast

vim +5944 kernel/bpf/verifier.c

5920
5921 static int check_bpf_snprintf_call(struct bpf_verifier_env *env,
5922 struct bpf_reg_state *regs)
5923 {
5924 struct bpf_reg_state *fmt_reg = &regs[BPF_REG_3];
5925 struct bpf_reg_state *data_len_reg = &regs[BPF_REG_5];
5926 struct bpf_map *fmt_map = fmt_reg->map_ptr;
5927 int err, fmt_map_off, num_args;
5928 u64 fmt_addr;
5929 char *fmt;
5930
5931 /* data must be an array of u64 */
5932 if (data_len_reg->var_off.value % 8)
5933 return -EINVAL;
5934 num_args = data_len_reg->var_off.value / 8;
5935
5936 /* fmt being ARG_PTR_TO_CONST_STR guarantees that var_off is const
5937 * and map_direct_value_addr is set.
5938 */
5939 fmt_map_off = fmt_reg->off + fmt_reg->var_off.value;
5940 err = fmt_map->ops->map_direct_value_addr(fmt_map, &fmt_addr,
5941 fmt_map_off);
5942 if (err)
5943 return err;
> 5944 fmt = (char *)fmt_addr + fmt_map_off;
5945
5946 /* We are also guaranteed that fmt+fmt_map_off is NULL terminated, we
5947 * can focus on validating the format specifiers.
5948 */
5949 err = bpf_printf_prepare(fmt, UINT_MAX, NULL, NULL, NULL, num_args);
5950 if (err < 0)
5951 verbose(env, "Invalid format string\n");
5952
5953 return err;
5954 }
5955

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@xxxxxxxxxxxx

Attachment: .config.gz
Description: application/gzip