bpf: fix 64-bit divide

ALU64_DIV instruction should be dividing 64-bit by 64-bit,
whereas do_div() does 64-bit by 32-bit divide.
x64 and arm64 JITs correctly implement 64 by 64 unsigned divide.
llvm BPF backend emits code assuming that ALU64_DIV does 64 by 64.

Fixes: 89aa0758 ("net: sock: allow eBPF programs to be attached to sockets")
Reported-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NAlexei Starovoitov <ast@plumgrid.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

kernel/bpf/core.c

@@ -357,8 +357,8 @@ static unsigned int __bpf_prog_run(void *ctx, const struct bpf_insn *insn)
 	ALU64_MOD_X:
 		if (unlikely(SRC == 0))
 			return 0;
-		tmp = DST;
-		DST = do_div(tmp, SRC);
+		div64_u64_rem(DST, SRC, &tmp);
+		DST = tmp;
 		CONT;
 	ALU_MOD_X:
 		if (unlikely(SRC == 0))
@@ -367,8 +367,8 @@ static unsigned int __bpf_prog_run(void *ctx, const struct bpf_insn *insn)
 		DST = do_div(tmp, (u32) SRC);
 		CONT;
 	ALU64_MOD_K:
-		tmp = DST;
-		DST = do_div(tmp, IMM);
+		div64_u64_rem(DST, IMM, &tmp);
+		DST = tmp;
 		CONT;
 	ALU_MOD_K:
 		tmp = (u32) DST;
@@ -377,7 +377,7 @@ static unsigned int __bpf_prog_run(void *ctx, const struct bpf_insn *insn)
 	ALU64_DIV_X:
 		if (unlikely(SRC == 0))
 			return 0;
-		do_div(DST, SRC);
+		DST = div64_u64(DST, SRC);
 		CONT;
 	ALU_DIV_X:
 		if (unlikely(SRC == 0))
@@ -387,7 +387,7 @@ static unsigned int __bpf_prog_run(void *ctx, const struct bpf_insn *insn)
 		DST = (u32) tmp;
 		CONT;
 	ALU64_DIV_K:
-		do_div(DST, IMM);
+		DST = div64_u64(DST, IMM);
 		CONT;
 	ALU_DIV_K:
 		tmp = (u32) DST;