x86/insn,uprobes,alternative: Unify insn_is_nop()

Both uprobes and alternatives have insn_is_nop() variants, unify them
and make sure insn_is_nop() works for both x86_64 and i386.

Specifically, uprobe must not compare userspace instructions to kernel
nops as that does not work right in the compat case.

For the uprobe case we therefore must recognise common 32bit and 64bit
nops. Because uprobe will consume the instruction as a nop, it must
not mistakenly claim a non-nop instruction to be a nop. Eg. 'REX.b3
NOP' is 'xchg %r8,%rax' - not a nop.

For the kernel case similar constraints apply, is it used to optimize
NOPs by replacing strings of short(er) nops with longer nops. Must not
claim an instruction is a nop if it really isn't. Not recognising a
nop is non-fatal.

Signed-off-by: Peter Zijlstra (Intel) <[e-mail]>

arch/x86/include/asm/insn-eval.h

···
 
 enum insn_mmio_type insn_decode_mmio(struct insn *insn, int *bytes);
 
+bool insn_is_nop(struct insn *insn);
+
 #endif /* _ASM_X86_INSN_EVAL_H */

arch/x86/kernel/alternative.c

···
 
 #include <asm/text-patching.h>
 #include <asm/insn.h>
+#include <asm/insn-eval.h>
 #include <asm/ibt.h>
 #include <asm/set_memory.h>
 #include <asm/nmi.h>
···
 
 	for (;buf < target; buf++)
 		*buf = INT3_INSN_OPCODE;
-}
-
-/*
- * Matches NOP and NOPL, not any of the other possible NOPs.
- */
-static bool insn_is_nop(struct insn *insn)
-{
-	/* Anything NOP, but no REP NOP */
-	if (insn->opcode.bytes[0] == 0x90 &&
-	    (!insn->prefixes.nbytes || insn->prefixes.bytes[0] != 0xF3))
-		return true;
-
-	/* NOPL */
-	if (insn->opcode.bytes[0] == 0x0F && insn->opcode.bytes[1] == 0x1F)
-		return true;
-
-	/* TODO: more nops */
-
-	return false;
 }
 
 /*

arch/x86/kernel/uprobes.c

···
 #include <linux/kdebug.h>
 #include <asm/processor.h>
 #include <asm/insn.h>
+#include <asm/insn-eval.h>
 #include <asm/mmu_context.h>
 #include <asm/nops.h>
 
···
 	mmap_write_unlock(mm);
 }
 
-static bool insn_is_nop(struct insn *insn)
-{
-	return insn->opcode.nbytes == 1 && insn->opcode.bytes[0] == 0x90;
-}
-
-static bool insn_is_nopl(struct insn *insn)
-{
-	if (insn->opcode.nbytes != 2)
-		return false;
-
-	if (insn->opcode.bytes[0] != 0x0f || insn->opcode.bytes[1] != 0x1f)
-		return false;
-
-	if (!insn->modrm.nbytes)
-		return false;
-
-	if (X86_MODRM_REG(insn->modrm.bytes[0]) != 0)
-		return false;
-
-	/* 0f 1f /0 - NOPL */
-	return true;
-}
-
 static bool can_optimize(struct insn *insn, unsigned long vaddr)
 {
 	if (!insn->x86_64 || insn->length != 5)
 		return false;
 
-	if (!insn_is_nop(insn) && !insn_is_nopl(insn))
+	if (!insn_is_nop(insn))
 		return false;
 
 	/* We can't do cross page atomic writes yet. */
···
 	insn_byte_t p;
 	int i;
 
-	/* x86_nops[insn->length]; same as jmp with .offs = 0 */
-	if (insn->length <= ASM_NOP_MAX &&
-	    !memcmp(insn->kaddr, x86_nops[insn->length], insn->length))
+	if (insn_is_nop(insn))
 		goto setup;
 
 	switch (opc1) {
 	case 0xeb:	/* jmp 8 */
 	case 0xe9:	/* jmp 32 */
 		break;
-	case 0x90:	/* prefix* + nop; same as jmp with .offs = 0 */
-		goto setup;
 
 	case 0xe8:	/* call relative */
 		branch_clear_offset(auprobe, insn);

arch/x86/lib/insn-eval.c

···
 
 	return type;
 }
+
+/*
+ * Recognise typical NOP patterns for both 32bit and 64bit.
+ *
+ * Notably:
+ *  - NOP, but not: REP NOP aka PAUSE
+ *  - NOPL
+ *  - MOV %reg, %reg
+ *  - LEA 0(%reg),%reg
+ *  - JMP +0
+ *
+ * Must not have false-positives; instructions identified as a NOP might be
+ * emulated as a NOP (uprobe) or Run Length Encoded in a larger NOP
+ * (alternatives).
+ *
+ * False-negatives are fine; need not be exhaustive.
+ */
+bool insn_is_nop(struct insn *insn)
+{
+	u8 b3 = 0, x3 = 0, r3 = 0;
+	u8 b4 = 0, x4 = 0, r4 = 0, m = 0;
+	u8 modrm, modrm_mod, modrm_reg, modrm_rm;
+	u8 sib = 0, sib_scale, sib_index, sib_base;
+	u8 nrex, rex;
+	u8 p, rep = 0;
+	int i;
+
+	if ((nrex = insn->rex_prefix.nbytes)) {
+		rex = insn->rex_prefix.bytes[nrex-1];
+
+		r3 = !!X86_REX_R(rex);
+		x3 = !!X86_REX_X(rex);
+		b3 = !!X86_REX_B(rex);
+		if (nrex > 1) {
+			r4 = !!X86_REX2_R(rex);
+			x4 = !!X86_REX2_X(rex);
+			b4 = !!X86_REX2_B(rex);
+			m  = !!X86_REX2_M(rex);
+		}
+
+	} else if (insn->vex_prefix.nbytes) {
+		/*
+		 * Ignore VEX encoded NOPs
+		 */
+		return false;
+	}
+
+	if (insn->modrm.nbytes) {
+		modrm = insn->modrm.bytes[0];
+		modrm_mod = X86_MODRM_MOD(modrm);
+		modrm_reg = X86_MODRM_REG(modrm) + 8*r3 + 16*r4;
+		modrm_rm  = X86_MODRM_RM(modrm)  + 8*b3 + 16*b4;
+		modrm = 1;
+	}
+
+	if (insn->sib.nbytes) {
+		sib = insn->sib.bytes[0];
+		sib_scale = X86_SIB_SCALE(sib);
+		sib_index = X86_SIB_INDEX(sib) + 8*x3 + 16*x4;
+		sib_base  = X86_SIB_BASE(sib)  + 8*b3 + 16*b4;
+		sib = 1;
+
+		modrm_rm = sib_base;
+	}
+
+	for_each_insn_prefix(insn, i, p) {
+		if (p == 0xf3) /* REPE */
+			rep = 1;
+	}
+
+	/*
+	 * Opcode map munging:
+	 *
+	 * REX2: 0 - single byte opcode
+	 *       1 - 0f second byte opcode
+	 */
+	switch (m) {
+	case 0: break;
+	case 1: insn->opcode.value <<= 8;
+		insn->opcode.value |= 0x0f;
+		break;
+	default:
+		return false;
+	}
+
+	switch (insn->opcode.bytes[0]) {
+	case 0x0f: /* 2nd byte */
+		break;
+
+	case 0x89: /* MOV */
+		if (modrm_mod != 3) /* register-direct */
+			return false;
+
+		/* native size */
+		if (insn->opnd_bytes != 4 * (1 + insn->x86_64))
+			return false;
+
+		return modrm_reg == modrm_rm; /* MOV %reg, %reg */
+
+	case 0x8d: /* LEA */
+		if (modrm_mod == 0 || modrm_mod == 3) /* register-indirect with disp */
+			return false;
+
+		/* native size */
+		if (insn->opnd_bytes != 4 * (1 + insn->x86_64))
+			return false;
+
+		if (insn->displacement.value != 0)
+			return false;
+
+		if (sib && (sib_scale != 0 || sib_index != 4)) /* (%reg, %eiz, 1) */
+			return false;
+
+		for_each_insn_prefix(insn, i, p) {
+			if (p != 0x3e) /* DS */
+				return false;
+		}
+
+		return modrm_reg == modrm_rm; /* LEA 0(%reg), %reg */
+
+	case 0x90: /* NOP */
+		if (b3 || b4) /* XCHG %r{8,16,24},%rax */
+			return false;
+
+		if (rep) /* REP NOP := PAUSE */
+			return false;
+
+		return true;
+
+	case 0xe9: /* JMP.d32 */
+	case 0xeb: /* JMP.d8 */
+		return insn->immediate.value == 0; /* JMP +0 */
+
+	default:
+		return false;
+	}
+
+	switch (insn->opcode.bytes[1]) {
+	case 0x1f:
+		return modrm_reg == 0; /* 0f 1f /0 -- NOPL */
+
+	default:
+		return false;
+	}
+}