bpf, riscv64: Skip redundant zext instruction after load-acquire

Currently, the verifier inserts a zext instruction right after every 8-,
16- or 32-bit load-acquire, which is already zero-extending.  Skip such
redundant zext instructions.

While we are here, update that already-obsolete comment about "skip the
next instruction" in build_body().  Also change emit_atomic_rmw()'s
parameters to keep it consistent with emit_atomic_ld_st().

Note that checking 'insn[1]' relies on 'insn' not being the last
instruction, which should have been guaranteed by the verifier; we
already use 'insn[1]' elsewhere in the file for similar purposes.
Additionally, we don't check if 'insn[1]' is actually a zext for our
load-acquire's dst_reg, or some other registers - in other words, here
we are relying on the verifier to always insert a redundant zext right
after a 8/16/32-bit load-acquire, for its dst_reg.

Acked-by: Björn Töpel <bjorn@kernel.org>
Reviewed-by: Pu Lehui <pulehui@huawei.com>
Tested-by: Björn Töpel <bjorn@rivosinc.com> # QEMU/RVA23
Signed-off-by: Peilin Ye <yepeilin@google.com>
Link: https://lore.kernel.org/r/10e90e0eab042f924d35ad0d1c1f7ca29f673152.1746588351.git.yepeilin@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

2 files changed, 19 insertions, 7 deletions

diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c
index 8767f032f2de..10e01ff06312 100644
--- a/arch/riscv/net/bpf_jit_comp64.c
+++ b/arch/riscv/net/bpf_jit_comp64.c
@@ -607,8 +607,13 @@ static void emit_store_64(u8 rd, s32 off, u8 rs, struct rv_jit_context *ctx)
 	emit_sd(RV_REG_T1, 0, rs, ctx);
 }
 
-static int emit_atomic_ld_st(u8 rd, u8 rs, s16 off, s32 imm, u8 code, struct rv_jit_context *ctx)
+static int emit_atomic_ld_st(u8 rd, u8 rs, const struct bpf_insn *insn,
+			     struct rv_jit_context *ctx)
 {
+	u8 code = insn->code;
+	s32 imm = insn->imm;
+	s16 off = insn->off;
+
 	switch (imm) {
 	/* dst_reg = load_acquire(src_reg + off16) */
 	case BPF_LOAD_ACQ:
@@ -627,6 +632,12 @@ static int emit_atomic_ld_st(u8 rd, u8 rs, s16 off, s32 imm, u8 code, struct rv_
 			break;
 		}
 		emit_fence_r_rw(ctx);
+
+		/* If our next insn is a redundant zext, return 1 to tell
+		 * build_body() to skip it.
+		 */
+		if (BPF_SIZE(code) != BPF_DW && insn_is_zext(&insn[1]))
+			return 1;
 		break;
 	/* store_release(dst_reg + off16, src_reg) */
 	case BPF_STORE_REL:
@@ -654,10 +665,12 @@ static int emit_atomic_ld_st(u8 rd, u8 rs, s16 off, s32 imm, u8 code, struct rv_
 	return 0;
 }
 
-static int emit_atomic_rmw(u8 rd, u8 rs, s16 off, s32 imm, u8 code,
+static int emit_atomic_rmw(u8 rd, u8 rs, const struct bpf_insn *insn,
 			   struct rv_jit_context *ctx)
 {
-	u8 r0;
+	u8 r0, code = insn->code;
+	s16 off = insn->off;
+	s32 imm = insn->imm;
 	int jmp_offset;
 	bool is64;
 
@@ -2026,9 +2039,9 @@ int bpf_jit_emit_insn(const struct bpf_insn *insn, struct rv_jit_context *ctx,
 	case BPF_STX | BPF_ATOMIC | BPF_W:
 	case BPF_STX | BPF_ATOMIC | BPF_DW:
 		if (bpf_atomic_is_load_store(insn))
-			ret = emit_atomic_ld_st(rd, rs, off, imm, code, ctx);
+			ret = emit_atomic_ld_st(rd, rs, insn, ctx);
 		else
-			ret = emit_atomic_rmw(rd, rs, off, imm, code, ctx);
+			ret = emit_atomic_rmw(rd, rs, insn, ctx);
 		if (ret)
 			return ret;
 		break;
diff --git a/arch/riscv/net/bpf_jit_core.c b/arch/riscv/net/bpf_jit_core.c
index f8cd2f70a7fb..f6ca5cfa6b2f 100644
--- a/arch/riscv/net/bpf_jit_core.c
+++ b/arch/riscv/net/bpf_jit_core.c
@@ -26,9 +26,8 @@ static int build_body(struct rv_jit_context *ctx, bool extra_pass, int *offset)
 		int ret;
 
 		ret = bpf_jit_emit_insn(insn, ctx, extra_pass);
-		/* BPF_LD | BPF_IMM | BPF_DW: skip the next instruction. */
 		if (ret > 0)
-			i++;
+			i++; /* skip the next instruction */
 		if (offset)
 			offset[i] = ctx->ninsns;
 		if (ret < 0)