bpf: Clear zext_dst of dead insns
PATCH 1/1 bpf: Clear zext_dst of dead insns
"access skb fields ok" verifier test fails on s390 with the "verifier bug. zext_dst is set, but no reg is defined" message. The first insns of the test prog are:
0: 61 01 00 00 00 00 00 00 ldxw %r0,[%r1+0]
8: 35 00 00 01 00 00 00 00 jge %r0,0,1
10: 61 01 00 08 00 00 00 00 ldxw %r0,[%r1+8]
and the 3rd one is dead (this does not look intentional to me, but this is a separate topic).
sanitize_dead_code() converts dead insns into "ja -1", but keeps zext_dst. When opt_subreg_zext_lo32_rnd_hi32() tries to parse such an insn, it sees this discrepancy and bails. This problem can be seen only with JITs whose bpf_jit_needs_zext() returns true.
Fix by clearning dead insns' zext_dst.
--- kernel/bpf/verifier.c | 1 + 1 file changed, 1 insertion(+)
kernel/bpf/verifier.c +1 -0
@@ -11955,6 +11955,7 @@ static void sanitize_dead_code(struct bpf_verifier_env *env)if (aux_data[i].seen)continue ;memcpy (insn + i, &trap,sizeof (trap)); + aux_data[i].zext_dst = false; } }
Signature · 1 line
2.31.1
Daniel Borkmann
maintainerHey Ilya,
Quoted 11 lines · Ilya Leoshkevich
"access skb fields ok" verifier test fails on s390 with the "verifier bug.
zext_dstis set, but no reg is defined" message. The first insns of the test prog are:0: 61 01 00 00 00 00 00 00 ldxw %r0,[%r1+0] 8: 35 00 00 01 00 00 00 00 jge %r0,0,110: 61 01 00 08 00 00 00 00 ldxw %r0,[%r1+8]
and the 3rd one is dead (this does not look intentional to me, but this is a separate topic).
Quoted 8 lines
sanitize_dead_code()converts dead insns into "ja -1", but keepszext_dst. Whenopt_subreg_zext_lo32_rnd_hi32()tries to parse such an insn, it sees this discrepancy and bails. This problem can be seen only with JITs whosebpf_jit_needs_zext()returns true.Fix by clearning dead insns'
zext_dst.Signed-off-by Ilya Leoshkevich <[email protected]>
I presume this would rather be bpf tree material, no? Do you also have a Fixes tag I could add?
And one last small request: if this is not already covered by test_verifier selftest, could you add one along with the fix?
Thanks a lot,
Daniel