Merge branch 'bpf,x64: implement jump padding in jit'

Gary Lin says:

====================
This patch series implements jump padding to x64 jit to cover some
corner cases that used to consume more than 20 jit passes and caused
failure.

v4:
  - Add the detailed comments about the possible padding bytes
  - Add the second test case which triggers jmp_cond padding and imm32 nop
    jmp padding.
  - Add the new test case as another subprog

v3:
  - Copy the instructions of prologue separately or the size calculation
    of the first BPF instruction would include the prologue.
  - Replace WARN_ONCE() with pr_err() and EFAULT
  - Use MAX_PASSES in the for loop condition check
  - Remove the "padded" flag from x64_jit_data. For the extra pass of
    subprogs, padding is always enabled since it won't hurt the images
    that converge without padding.
v2:
  - Simplify the sample code in the commit description and provide the
    jit code
  - Check the expected padding bytes with WARN_ONCE
  - Move the 'padded' flag to 'struct x64_jit_data'
  - Remove the EXPECTED_FAIL flag from bpf_fill_maxinsns11() in test_bpf
  - Add 2 verifier tests
====================

Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Author: Alexei Starovoitov

arch/x86/net/bpf_jit_comp.c

@@ -869,8 +869,31 @@ static void detect_reg_usage(struct bpf_insn *insn, int insn_cnt,
 	}
 }
 
+static int emit_nops(u8 **pprog, int len)
+{
+	u8 *prog = *pprog;
+	int i, noplen, cnt = 0;
+
+	while (len > 0) {
+		noplen = len;
+
+		if (noplen > ASM_NOP_MAX)
+			noplen = ASM_NOP_MAX;
+
+		for (i = 0; i < noplen; i++)
+			EMIT1(ideal_nops[noplen][i]);
+		len -= noplen;
+	}
+
+	*pprog = prog;
+
+	return cnt;
+}
+
+#define INSN_SZ_DIFF (((addrs[i] - addrs[i - 1]) - (prog - temp)))
+
 static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image,
-		  int oldproglen, struct jit_context *ctx)
+		  int oldproglen, struct jit_context *ctx, bool jmp_padding)
 {
 	bool tail_call_reachable = bpf_prog->aux->tail_call_reachable;
 	struct bpf_insn *insn = bpf_prog->insnsi;
@@ -880,7 +903,7 @@ static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image,
 	bool seen_exit = false;
 	u8 temp[BPF_MAX_INSN_SIZE + BPF_INSN_SAFETY];
 	int i, cnt = 0, excnt = 0;
-	int proglen = 0;
+	int ilen, proglen = 0;
 	u8 *prog = temp;
 	int err;
 
@@ -894,7 +917,13 @@ static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image,
 		      bpf_prog_was_classic(bpf_prog), tail_call_reachable,
 		      bpf_prog->aux->func_idx != 0);
 	push_callee_regs(&prog, callee_regs_used);
-	addrs[0] = prog - temp;
+
+	ilen = prog - temp;
+	if (image)
+		memcpy(image + proglen, temp, ilen);
+	proglen += ilen;
+	addrs[0] = proglen;
+	prog = temp;
 
 	for (i = 1; i <= insn_cnt; i++, insn++) {
 		const s32 imm32 = insn->imm;
@@ -903,8 +932,8 @@ static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image,
 		u8 b2 = 0, b3 = 0;
 		s64 jmp_offset;
 		u8 jmp_cond;
-		int ilen;
 		u8 *func;
+		int nops;
 
 		switch (insn->code) {
 			/* ALU */
@@ -1502,6 +1531,30 @@ st:			if (is_imm8(insn->off))
 			}
 			jmp_offset = addrs[i + insn->off] - addrs[i];
 			if (is_imm8(jmp_offset)) {
+				if (jmp_padding) {
+					/* To keep the jmp_offset valid, the extra bytes are
+					 * padded before the jump insn, so we substract the
+					 * 2 bytes of jmp_cond insn from INSN_SZ_DIFF.
+					 *
+					 * If the previous pass already emits an imm8
+					 * jmp_cond, then this BPF insn won't shrink, so
+					 * "nops" is 0.
+					 *
+					 * On the other hand, if the previous pass emits an
+					 * imm32 jmp_cond, the extra 4 bytes(*) is padded to
+					 * keep the image from shrinking further.
+					 *
+					 * (*) imm32 jmp_cond is 6 bytes, and imm8 jmp_cond
+					 *     is 2 bytes, so the size difference is 4 bytes.
+					 */
+					nops = INSN_SZ_DIFF - 2;
+					if (nops != 0 && nops != 4) {
+						pr_err("unexpected jmp_cond padding: %d bytes\n",
+						       nops);
+						return -EFAULT;
+					}
+					cnt += emit_nops(&prog, nops);
+				}
 				EMIT2(jmp_cond, jmp_offset);
 			} else if (is_simm32(jmp_offset)) {
 				EMIT2_off32(0x0F, jmp_cond + 0x10, jmp_offset);
@@ -1524,11 +1577,55 @@ st:			if (is_imm8(insn->off))
 			else
 				jmp_offset = addrs[i + insn->off] - addrs[i];
 
-			if (!jmp_offset)
-				/* Optimize out nop jumps */
+			if (!jmp_offset) {
+				/*
+				 * If jmp_padding is enabled, the extra nops will
+				 * be inserted. Otherwise, optimize out nop jumps.
+				 */
+				if (jmp_padding) {
+					/* There are 3 possible conditions.
+					 * (1) This BPF_JA is already optimized out in
+					 *     the previous run, so there is no need
+					 *     to pad any extra byte (0 byte).
+					 * (2) The previous pass emits an imm8 jmp,
+					 *     so we pad 2 bytes to match the previous
+					 *     insn size.
+					 * (3) Similarly, the previous pass emits an
+					 *     imm32 jmp, and 5 bytes is padded.
+					 */
+					nops = INSN_SZ_DIFF;
+					if (nops != 0 && nops != 2 && nops != 5) {
+						pr_err("unexpected nop jump padding: %d bytes\n",
+						       nops);
+						return -EFAULT;
+					}
+					cnt += emit_nops(&prog, nops);
+				}
 				break;
+			}
 emit_jmp:
 			if (is_imm8(jmp_offset)) {
+				if (jmp_padding) {
+					/* To avoid breaking jmp_offset, the extra bytes
+					 * are padded before the actual jmp insn, so
+					 * 2 bytes is substracted from INSN_SZ_DIFF.
+					 *
+					 * If the previous pass already emits an imm8
+					 * jmp, there is nothing to pad (0 byte).
+					 *
+					 * If it emits an imm32 jmp (5 bytes) previously
+					 * and now an imm8 jmp (2 bytes), then we pad
+					 * (5 - 2 = 3) bytes to stop the image from
+					 * shrinking further.
+					 */
+					nops = INSN_SZ_DIFF - 2;
+					if (nops != 0 && nops != 3) {
+						pr_err("unexpected jump padding: %d bytes\n",
+						       nops);
+						return -EFAULT;
+					}
+					cnt += emit_nops(&prog, INSN_SZ_DIFF - 2);
+				}
 				EMIT2(0xEB, jmp_offset);
 			} else if (is_simm32(jmp_offset)) {
 				EMIT1_off32(0xE9, jmp_offset);
@@ -1671,26 +1768,6 @@ static int invoke_bpf_prog(const struct btf_func_model *m, u8 **pprog,
 	return 0;
 }
 
-static void emit_nops(u8 **pprog, unsigned int len)
-{
-	unsigned int i, noplen;
-	u8 *prog = *pprog;
-	int cnt = 0;
-
-	while (len > 0) {
-		noplen = len;
-
-		if (noplen > ASM_NOP_MAX)
-			noplen = ASM_NOP_MAX;
-
-		for (i = 0; i < noplen; i++)
-			EMIT1(ideal_nops[noplen][i]);
-		len -= noplen;
-	}
-
-	*pprog = prog;
-}
-
 static void emit_align(u8 **pprog, u32 align)
 {
 	u8 *target, *prog = *pprog;
@@ -2065,6 +2142,9 @@ struct x64_jit_data {
 	struct jit_context ctx;
 };
 
+#define MAX_PASSES 20
+#define PADDING_PASSES (MAX_PASSES - 5)
+
 struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 {
 	struct bpf_binary_header *header = NULL;
@@ -2074,6 +2154,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 	struct jit_context ctx = {};
 	bool tmp_blinded = false;
 	bool extra_pass = false;
+	bool padding = false;
 	u8 *image = NULL;
 	int *addrs;
 	int pass;
@@ -2110,6 +2191,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 		image = jit_data->image;
 		header = jit_data->header;
 		extra_pass = true;
+		padding = true;
 		goto skip_init_addrs;
 	}
 	addrs = kmalloc_array(prog->len + 1, sizeof(*addrs), GFP_KERNEL);
@@ -2135,8 +2217,10 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 	 * may converge on the last pass. In such case do one more
 	 * pass to emit the final image.
 	 */
-	for (pass = 0; pass < 20 || image; pass++) {
-		proglen = do_jit(prog, addrs, image, oldproglen, &ctx);
+	for (pass = 0; pass < MAX_PASSES || image; pass++) {
+		if (!padding && pass >= PADDING_PASSES)
+			padding = true;
+		proglen = do_jit(prog, addrs, image, oldproglen, &ctx, padding);
 		if (proglen <= 0) {
 out_image:
 			image = NULL;

lib/test_bpf.c

@@ -345,7 +345,7 @@ static int __bpf_fill_ja(struct bpf_test *self, unsigned int len,
 
 static int bpf_fill_maxinsns11(struct bpf_test *self)
 {
-	/* Hits 70 passes on x86_64, so cannot get JITed there. */
+	/* Hits 70 passes on x86_64 and triggers NOPs padding. */
 	return __bpf_fill_ja(self, BPF_MAXINSNS, 68);
 }
 
@@ -5318,15 +5318,10 @@ static struct bpf_test tests[] = {
 	{
 		"BPF_MAXINSNS: Jump, gap, jump, ...",
 		{ },
-#if defined(CONFIG_BPF_JIT_ALWAYS_ON) && defined(CONFIG_X86)
-		CLASSIC | FLAG_NO_DATA | FLAG_EXPECTED_FAIL,
-#else
 		CLASSIC | FLAG_NO_DATA,
-#endif
 		{ },
 		{ { 0, 0xababcbac } },
 		.fill_helper = bpf_fill_maxinsns11,
-		.expected_errcode = -ENOTSUPP,
 	},
 	{
 		"BPF_MAXINSNS: jump over MSH",

tools/testing/selftests/bpf/test_verifier.c

@@ -296,6 +296,78 @@ static void bpf_fill_scale(struct bpf_test *self)
 	}
 }
 
+static int bpf_fill_torturous_jumps_insn_1(struct bpf_insn *insn)
+{
+	unsigned int len = 259, hlen = 128;
+	int i;
+
+	insn[0] = BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32);
+	for (i = 1; i <= hlen; i++) {
+		insn[i]        = BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, i, hlen);
+		insn[i + hlen] = BPF_JMP_A(hlen - i);
+	}
+	insn[len - 2] = BPF_MOV64_IMM(BPF_REG_0, 1);
+	insn[len - 1] = BPF_EXIT_INSN();
+
+	return len;
+}
+
+static int bpf_fill_torturous_jumps_insn_2(struct bpf_insn *insn)
+{
+	unsigned int len = 4100, jmp_off = 2048;
+	int i, j;
+
+	insn[0] = BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32);
+	for (i = 1; i <= jmp_off; i++) {
+		insn[i] = BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, i, jmp_off);
+	}
+	insn[i++] = BPF_JMP_A(jmp_off);
+	for (; i <= jmp_off * 2 + 1; i+=16) {
+		for (j = 0; j < 16; j++) {
+			insn[i + j] = BPF_JMP_A(16 - j - 1);
+		}
+	}
+
+	insn[len - 2] = BPF_MOV64_IMM(BPF_REG_0, 2);
+	insn[len - 1] = BPF_EXIT_INSN();
+
+	return len;
+}
+
+static void bpf_fill_torturous_jumps(struct bpf_test *self)
+{
+	struct bpf_insn *insn = self->fill_insns;
+	int i = 0;
+
+	switch (self->retval) {
+	case 1:
+		self->prog_len = bpf_fill_torturous_jumps_insn_1(insn);
+		return;
+	case 2:
+		self->prog_len = bpf_fill_torturous_jumps_insn_2(insn);
+		return;
+	case 3:
+		/* main */
+		insn[i++] = BPF_RAW_INSN(BPF_JMP|BPF_CALL, 0, 1, 0, 4);
+		insn[i++] = BPF_RAW_INSN(BPF_JMP|BPF_CALL, 0, 1, 0, 262);
+		insn[i++] = BPF_ST_MEM(BPF_B, BPF_REG_10, -32, 0);
+		insn[i++] = BPF_MOV64_IMM(BPF_REG_0, 3);
+		insn[i++] = BPF_EXIT_INSN();
+
+		/* subprog 1 */
+		i += bpf_fill_torturous_jumps_insn_1(insn + i);
+
+		/* subprog 2 */
+		i += bpf_fill_torturous_jumps_insn_2(insn + i);
+
+		self->prog_len = i;
+		return;
+	default:
+		self->prog_len = 0;
+		break;
+	}
+}
+
 /* BPF_SK_LOOKUP contains 13 instructions, if you need to fix up maps */
 #define BPF_SK_LOOKUP(func)						\
 	/* struct bpf_sock_tuple tuple = {} */				\

tools/testing/selftests/bpf/verifier/jit.c

@@ -105,3 +105,27 @@
 	.result = ACCEPT,
 	.retval = 2,
 },
+{
+	"jit: torturous jumps, imm8 nop jmp and pure jump padding",
+	.insns = { },
+	.fill_helper = bpf_fill_torturous_jumps,
+	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
+	.result = ACCEPT,
+	.retval = 1,
+},
+{
+	"jit: torturous jumps, imm32 nop jmp and jmp_cond padding",
+	.insns = { },
+	.fill_helper = bpf_fill_torturous_jumps,
+	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
+	.result = ACCEPT,
+	.retval = 2,
+},
+{
+	"jit: torturous jumps in subprog",
+	.insns = { },
+	.fill_helper = bpf_fill_torturous_jumps,
+	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
+	.result = ACCEPT,
+	.retval = 3,
+},