arm64: bpf: optimize modulo operation

Optimize modulo operation instruction generation by
using single MSUB instruction vs MUL followed by SUB
instruction scheme.
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>

arch/arm64/net/bpf_jit.h

@@ -171,6 +171,9 @@
 /* Rd = Ra + Rn * Rm */
 #define A64_MADD(sf, Rd, Ra, Rn, Rm) aarch64_insn_gen_data3(Rd, Ra, Rn, Rm, \
 	A64_VARIANT(sf), AARCH64_INSN_DATA3_MADD)
+/* Rd = Ra - Rn * Rm */
+#define A64_MSUB(sf, Rd, Ra, Rn, Rm) aarch64_insn_gen_data3(Rd, Ra, Rn, Rm, \
+	A64_VARIANT(sf), AARCH64_INSN_DATA3_MSUB)
 /* Rd = Rn * Rm */
 #define A64_MUL(sf, Rd, Rn, Rm) A64_MADD(sf, Rd, A64_ZR, Rn, Rm)
 

arch/arm64/net/bpf_jit_comp.c

@@ -409,8 +409,7 @@ static int build_insn(const struct bpf_insn *insn, struct jit_ctx *ctx,
 			break;
 		case BPF_MOD:
 			emit(A64_UDIV(is64, tmp, dst, src), ctx);
-			emit(A64_MUL(is64, tmp, tmp, src), ctx);
-			emit(A64_SUB(is64, dst, dst, tmp), ctx);
+			emit(A64_MSUB(is64, dst, dst, tmp, src), ctx);
 			break;
 		}
 		break;
@@ -516,8 +515,7 @@ static int build_insn(const struct bpf_insn *insn, struct jit_ctx *ctx,
 	case BPF_ALU64 | BPF_MOD | BPF_K:
 		emit_a64_mov_i(is64, tmp2, imm, ctx);
 		emit(A64_UDIV(is64, tmp, dst, tmp2), ctx);
-		emit(A64_MUL(is64, tmp, tmp, tmp2), ctx);
-		emit(A64_SUB(is64, dst, dst, tmp), ctx);
+		emit(A64_MSUB(is64, dst, dst, tmp, tmp2), ctx);
 		break;
 	case BPF_ALU | BPF_LSH | BPF_K:
 	case BPF_ALU64 | BPF_LSH | BPF_K: