Add x86 instruction decoder to arch-specific libraries. This decoder
can decode all x86 instructions into prefix, opcode, modrm, sib,
displacement and immediates. This can also show the length of
instructions.
Signed-off-by: Jim Keniston <[email protected]>
Signed-off-by: Masami Hiramatsu <[email protected]>
Cc: Ananth N Mavinakayanahalli <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: [email protected]
---
arch/x86/include/asm/insn.h | 130 +++++++++
arch/x86/lib/Makefile | 1
arch/x86/lib/insn.c | 627 +++++++++++++++++++++++++++++++++++++++++++
3 files changed, 758 insertions(+), 0 deletions(-)
create mode 100644 arch/x86/include/asm/insn.h
create mode 100644 arch/x86/lib/insn.c
diff --git a/arch/x86/include/asm/insn.h b/arch/x86/include/asm/insn.h
new file mode 100644
index 0000000..488001f
--- /dev/null
+++ b/arch/x86/include/asm/insn.h
@@ -0,0 +1,130 @@
+#ifndef _ASM_X86_INSN_H
+#define _ASM_X86_INSN_H
+/*
+ * x86 instruction analysis
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2009
+ */
+
+#include <linux/types.h>
+
+/* legacy instruction prefixes */
+#define X86_PFX_OPNDSZ 0x1 /* 0x66 */
+#define X86_PFX_ADDRSZ 0x2 /* 0x67 */
+#define X86_PFX_CS 0x4 /* 0x2E */
+#define X86_PFX_DS 0x8 /* 0x3E */
+#define X86_PFX_ES 0x10 /* 0x26 */
+#define X86_PFX_FS 0x20 /* 0x64 */
+#define X86_PFX_GS 0x40 /* 0x65 */
+#define X86_PFX_SS 0x80 /* 0x36 */
+#define X86_PFX_LOCK 0x100 /* 0xF0 */
+#define X86_PFX_REPE 0x200 /* 0xF3 */
+#define X86_PFX_REPNE 0x400 /* 0xF2 */
+/* REX prefix */
+#define X86_PFX_REX 0x800 /* 0x4X */
+/* REX prefix dissected */
+#define X86_PFX_REX_BASE 0x1000
+#define X86_PFX_REXB 0x1000 /* 0x41 bit */
+#define X86_PFX_REXX 0x2000 /* 0x42 bit */
+#define X86_PFX_REXR 0x4000 /* 0x44 bit */
+#define X86_PFX_REXW 0x8000 /* 0x48 bit */
+
+struct insn_field {
+ union {
+ s32 value;
+ u8 bytes[4];
+ };
+ bool got; /* true if we've run insn_get_xxx() for this field */
+ u8 nbytes;
+};
+
+struct insn {
+ struct insn_field prefixes; /* prefixes.value is a bitmap */
+ struct insn_field opcode; /*
+ * opcode.bytes[0]: opcode1
+ * opcode.bytes[1]: opcode2
+ * opcode.bytes[2]: opcode3
+ */
+ struct insn_field modrm;
+ struct insn_field sib;
+ struct insn_field displacement;
+ union {
+ struct insn_field immediate;
+ struct insn_field moffset1; /* for 64bit MOV */
+ struct insn_field immediate1; /* for 64bit imm or off16/32 */
+ };
+ union {
+ struct insn_field moffset2; /* for 64bit MOV */
+ struct insn_field immediate2; /* for 64bit imm or seg16 */
+ };
+
+ u8 opnd_bytes;
+ u8 addr_bytes;
+ u8 length;
+ bool x86_64;
+
+ const u8 *kaddr; /* kernel address of insn (copy) to analyze */
+ const u8 *next_byte;
+};
+
+#define OPCODE1(insn) ((insn)->opcode.bytes[0])
+#define OPCODE2(insn) ((insn)->opcode.bytes[1])
+#define OPCODE3(insn) ((insn)->opcode.bytes[2])
+
+#define MODRM_MOD(insn) (((insn)->modrm.value & 0xc0) >> 6)
+#define MODRM_REG(insn) (((insn)->modrm.value & 0x38) >> 3)
+#define MODRM_RM(insn) ((insn)->modrm.value & 0x07)
+
+#define SIB_SCALE(insn) (((insn)->sib.value & 0xc0) >> 6)
+#define SIB_INDEX(insn) (((insn)->sib.value & 0x38) >> 3)
+#define SIB_BASE(insn) ((insn)->sib.value & 0x07)
+
+#define MOFFSET64(insn) (((u64)((insn)->moffset2.value) << 32) | \
+ (u32)((insn)->moffset1.value))
+
+#define IMMEDIATE64(insn) (((u64)((insn)->immediate2.value) << 32) | \
+ (u32)((insn)->immediate1.value))
+
+extern void insn_init(struct insn *insn, const u8 *kaddr, bool x86_64);
+extern void insn_get_prefixes(struct insn *insn);
+extern void insn_get_opcode(struct insn *insn);
+extern void insn_get_modrm(struct insn *insn);
+extern void insn_get_sib(struct insn *insn);
+extern void insn_get_displacement(struct insn *insn);
+extern void insn_get_immediate(struct insn *insn);
+extern void insn_get_length(struct insn *insn);
+
+#ifdef CONFIG_X86_64
+/* Init insn for kernel text */
+#define insn_init_kernel(insn, kaddr) insn_init(insn, kaddr, 1)
+extern bool insn_rip_relative(struct insn *insn);
+
+#else /* CONFIG_X86_32 */
+
+#define insn_init_kernel(insn, kaddr) insn_init(insn, kaddr, 0)
+static inline bool insn_rip_relative(struct insn *insn)
+{
+ return false;
+}
+#endif
+
+static inline bool insn_field_exists(const struct insn_field *field)
+{
+ return (field->nbytes > 0);
+}
+
+#endif /* _ASM_X86_INSN_H */
diff --git a/arch/x86/lib/Makefile b/arch/x86/lib/Makefile
index 55e11aa..0f81979 100644
--- a/arch/x86/lib/Makefile
+++ b/arch/x86/lib/Makefile
@@ -8,6 +8,7 @@ lib-y := delay.o
lib-y += thunk_$(BITS).o
lib-y += usercopy_$(BITS).o getuser.o putuser.o
lib-y += memcpy_$(BITS).o
+lib-y += insn.o
ifeq ($(CONFIG_X86_32),y)
lib-y += checksum_32.o
diff --git a/arch/x86/lib/insn.c b/arch/x86/lib/insn.c
new file mode 100644
index 0000000..f5b13b0
--- /dev/null
+++ b/arch/x86/lib/insn.c
@@ -0,0 +1,627 @@
+/*
+ * x86 instruction analysis
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2002, 2004, 2009
+ */
+
+#include <linux/string.h>
+#include <linux/module.h>
+#include <asm/insn.h>
+
+#undef W
+#define W(row, b0, b1, b2, b3, b4, b5, b6, b7, b8, b9, ba, bb, bc, bd, be, bf)\
+ (((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) | \
+ (b4##UL << 0x4)|(b5##UL << 0x5)|(b6##UL << 0x6)|(b7##UL << 0x7) | \
+ (b8##UL << 0x8)|(b9##UL << 0x9)|(ba##UL << 0xa)|(bb##UL << 0xb) | \
+ (bc##UL << 0xc)|(bd##UL << 0xd)|(be##UL << 0xe)|(bf##UL << 0xf)) \
+ << (row % 32))
+
+/**
+ * insn_init() - initialize struct insn
+ * @insn: &struct insn to be initialized
+ * @kaddr: address (in kernel memory) of instruction (or copy thereof)
+ * @x86_64: true for 64-bit kernel or 64-bit app
+ */
+void insn_init(struct insn *insn, const u8 *kaddr, bool x86_64)
+{
+ memset(insn, 0, sizeof(*insn));
+ insn->kaddr = kaddr;
+ insn->next_byte = kaddr;
+ insn->x86_64 = x86_64;
+ insn->opnd_bytes = 4;
+ if (x86_64)
+ insn->addr_bytes = 8;
+ else
+ insn->addr_bytes = 4;
+}
+EXPORT_SYMBOL_GPL(insn_init);
+
+/**
+ * insn_get_prefixes - scan x86 instruction prefix bytes
+ * @insn: &struct insn containing instruction
+ *
+ * Populates the @insn->prefixes bitmap, and updates @insn->next_byte
+ * to point to the (first) opcode. No effect if @insn->prefixes.got
+ * is already true.
+ */
+void insn_get_prefixes(struct insn *insn)
+{
+ u32 pfx;
+ struct insn_field *prefixes = &insn->prefixes;
+ if (prefixes->got)
+ return;
+ for (;; insn->next_byte++, prefixes->nbytes++) {
+ u8 b = *(insn->next_byte);
+#ifdef CONFIG_X86_64
+ if ((b & 0xf0) == 0x40 && insn->x86_64) {
+ prefixes->value |= X86_PFX_REX;
+ prefixes->value |= (b & 0x0f) * X86_PFX_REX_BASE;
+ /* REX prefix is always last. */
+ insn->next_byte++;
+ prefixes->nbytes++;
+ break;
+ }
+#endif
+ switch (b) {
+ case 0x26:
+ pfx = X86_PFX_ES;
+ break;
+ case 0x2E:
+ pfx = X86_PFX_CS;
+ break;
+ case 0x36:
+ pfx = X86_PFX_SS;
+ break;
+ case 0x3E:
+ pfx = X86_PFX_DS;
+ break;
+ case 0x64:
+ pfx = X86_PFX_FS;
+ break;
+ case 0x65:
+ pfx = X86_PFX_GS;
+ break;
+ case 0x66:
+ pfx = X86_PFX_OPNDSZ;
+ break;
+ case 0x67:
+ pfx = X86_PFX_ADDRSZ;
+ break;
+ case 0xF0:
+ pfx = X86_PFX_LOCK;
+ break;
+ case 0xF2:
+ pfx = X86_PFX_REPNE;
+ break;
+ case 0xF3:
+ pfx = X86_PFX_REPE;
+ break;
+ default:
+ pfx = 0x0;
+ break;
+ }
+ if (!pfx)
+ break;
+ prefixes->value |= pfx;
+ }
+ if (prefixes->value & X86_PFX_OPNDSZ) {
+ /* oprand size switches 2/4 */
+ insn->opnd_bytes ^= 6;
+ }
+ if (prefixes->value & X86_PFX_ADDRSZ) {
+ /* address size switches 2/4 or 4/8 */
+#ifdef CONFIG_X86_64
+ if (insn->x86_64)
+ insn->addr_bytes ^= 12;
+ else
+#endif
+ insn->addr_bytes ^= 6;
+ }
+#ifdef CONFIG_X86_64
+ if (prefixes->value & X86_PFX_REXW)
+ insn->opnd_bytes = 8;
+#endif
+ prefixes->got = true;
+}
+EXPORT_SYMBOL_GPL(insn_get_prefixes);
+
+/**
+ * insn_get_opcode - collect opcode(s)
+ * @insn: &struct insn containing instruction
+ *
+ * Populates @insn->opcode1 (and @insn->opcode2, if it's a 2-byte opcode)
+ * and updates @insn->next_byte to point past the opcode byte(s).
+ * If necessary, first collects any preceding (prefix) bytes.
+ * Sets @insn->opcode.value = opcode1. No effect if @insn->opcode.got
+ * is already true.
+ */
+void insn_get_opcode(struct insn *insn)
+{
+ struct insn_field *opcode = &insn->opcode;
+ if (opcode->got)
+ return;
+ if (!insn->prefixes.got)
+ insn_get_prefixes(insn);
+ OPCODE1(insn) = *insn->next_byte++;
+ if (OPCODE1(insn) == 0x0f) {
+ OPCODE2(insn) = *insn->next_byte++;
+ if (OPCODE2(insn) == 0x38 || OPCODE2(insn) == 0x3a) {
+ OPCODE3(insn) = *insn->next_byte++;
+ opcode->nbytes = 3;
+ } else
+ opcode->nbytes = 2;
+ } else
+ opcode->nbytes = 1;
+ opcode->got = true;
+}
+EXPORT_SYMBOL_GPL(insn_get_opcode);
+
+const u32 onebyte_has_modrm[256 / 32] = {
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+ /* ----------------------------------------------- */
+ W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 0f */
+ W(0x10, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 1f */
+ W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 2f */
+ W(0x30, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 3f */
+ W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 4f */
+ W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 5f */
+ W(0x60, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0) | /* 6f */
+ W(0x70, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 7f */
+ W(0x80, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 8f */
+ W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 9f */
+ W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* af */
+ W(0xb0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* bf */
+ W(0xc0, 1, 1, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* cf */
+ W(0xd0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) , /* df */
+ W(0xe0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* ef */
+ W(0xf0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1) /* ff */
+ /* ----------------------------------------------- */
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+};
+
+const u32 twobyte_has_modrm[256 / 32] = {
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+ /* ----------------------------------------------- */
+ W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1) | /* 0f */
+ W(0x10, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 1f */
+ W(0x20, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* 2f */
+ W(0x30, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 3f */
+ W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 4f */
+ W(0x50, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 5f */
+ W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 6f */
+ W(0x70, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1) , /* 7f */
+ W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 8f */
+ W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 9f */
+ W(0xa0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) | /* af */
+ W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1) , /* bf */
+ W(0xc0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* cf */
+ W(0xd0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* df */
+ W(0xe0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* ef */
+ W(0xf0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0) /* ff */
+ /* ----------------------------------------------- */
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+};
+
+#ifdef CONFIG_X86_64
+const u32 onebyte_force_64[256 / 32] = {
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+ /* ----------------------------------------------- */
+ W(0x00, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 0f */
+ W(0x10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 1f */
+ W(0x20, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 2f */
+ W(0x30, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 3f */
+ W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 4f */
+ W(0x50, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 5f */
+ W(0x60, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0) | /* 6f */
+ W(0x70, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 7f */
+ W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1) | /* 8f */
+ W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 9f */
+ W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* af */
+ W(0xb0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* bf */
+ W(0xc0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0) | /* cf */
+ W(0xd0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* df */
+ W(0xe0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 0, 1, 0, 0, 0, 0) | /* ef */
+ W(0xf0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) /* ff */
+ /* ----------------------------------------------- */
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+};
+
+/* force 64 or default 64 bits operand opcodes */
+static bool __operand_64(struct insn *insn)
+{
+ u8 reg = MODRM_REG(insn);
+ if (insn->opcode.nbytes == 1) {
+ if (test_bit(OPCODE1(insn),
+ (const unsigned long *) onebyte_force_64) ||
+ (OPCODE1(insn) == 0xff &&
+ (reg == 2 || reg == 4 || reg == 6)))
+ return true;
+ }
+ return false;
+}
+#endif
+
+/**
+ * insn_get_modrm - collect ModRM byte, if any
+ * @insn: &struct insn containing instruction
+ *
+ * Populates @insn->modrm and updates @insn->next_byte to point past the
+ * ModRM byte, if any. If necessary, first collects the preceding bytes
+ * (prefixes and opcode(s)). No effect if @insn->modrm.got is already true.
+ */
+void insn_get_modrm(struct insn *insn)
+{
+ struct insn_field *modrm = &insn->modrm;
+ if (modrm->got)
+ return;
+ if (!insn->opcode.got)
+ insn_get_opcode(insn);
+ switch (insn->opcode.nbytes) {
+ case 1:
+ modrm->nbytes = test_bit(OPCODE1(insn),
+ (const unsigned long *) onebyte_has_modrm);
+ break;
+ case 2:
+ modrm->nbytes = test_bit(OPCODE2(insn),
+ (const unsigned long *) twobyte_has_modrm);
+ break;
+ case 3:
+ /* Three bytes opcodes always have modrm */
+ modrm->nbytes = 1;
+ break;
+ }
+ if (modrm->nbytes)
+ modrm->value = *(insn->next_byte++);
+
+#ifdef CONFIG_X86_64
+ if (insn->x86_64 && __operand_64(insn))
+ insn->opnd_bytes = 8;
+#endif
+ modrm->got = true;
+}
+EXPORT_SYMBOL_GPL(insn_get_modrm);
+
+#ifdef CONFIG_X86_64
+/**
+ * insn_rip_relative() - Does instruction use RIP-relative addressing mode?
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * ModRM byte. No effect if @insn->x86_64 is false.
+ */
+bool insn_rip_relative(struct insn *insn)
+{
+ struct insn_field *modrm = &insn->modrm;
+
+ if (!insn->x86_64)
+ return false;
+ if (!modrm->got)
+ insn_get_modrm(insn);
+ /*
+ * For rip-relative instructions, the mod field (top 2 bits)
+ * is zero and the r/m field (bottom 3 bits) is 0x5.
+ */
+ return (insn_field_exists(modrm) && (modrm->value & 0xc7) == 0x5);
+}
+EXPORT_SYMBOL_GPL(insn_rip_relative);
+#endif
+
+/**
+ *
+ * insn_get_sib() - Get the SIB byte of instruction
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * ModRM byte.
+ */
+void insn_get_sib(struct insn *insn)
+{
+ if (insn->sib.got)
+ return;
+ if (!insn->modrm.got)
+ insn_get_modrm(insn);
+ if (insn->modrm.nbytes)
+ if (insn->addr_bytes != 2 &&
+ MODRM_MOD(insn) != 3 && MODRM_RM(insn) == 4) {
+ insn->sib.value = *(insn->next_byte++);
+ insn->sib.nbytes = 1;
+ }
+ insn->sib.got = true;
+}
+EXPORT_SYMBOL_GPL(insn_get_sib);
+
+#define get_next(t, insn) \
+ ({t r; r = *(t*)insn->next_byte; insn->next_byte += sizeof(t); r; })
+
+/**
+ *
+ * insn_get_displacement() - Get the displacement of instruction
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * SIB byte.
+ * Displacement value is sign-expanded.
+ */
+void insn_get_displacement(struct insn *insn)
+{
+ u8 mod;
+ if (insn->displacement.got)
+ return;
+ if (!insn->sib.got)
+ insn_get_sib(insn);
+ if (insn->modrm.nbytes) {
+ /*
+ * Interpreting the modrm byte:
+ * mod = 00 - no displacement fields (exceptions below)
+ * mod = 01 - 1-byte displacement field
+ * mod = 10 - displacement field is 4 bytes, or 2 bytes if
+ * address size = 2 (0x67 prefix in 32-bit mode)
+ * mod = 11 - no memory operand
+ *
+ * If address size = 2...
+ * mod = 00, r/m = 110 - displacement field is 2 bytes
+ *
+ * If address size != 2...
+ * mod != 11, r/m = 100 - SIB byte exists
+ * mod = 00, SIB base = 101 - displacement field is 4 bytes
+ * mod = 00, r/m = 101 - rip-relative addressing, displacement
+ * field is 4 bytes
+ */
+ mod = MODRM_MOD(insn);
+ if (mod == 3)
+ goto out;
+ if (mod == 1) {
+ insn->displacement.value = *((s8 *)insn->next_byte++);
+ insn->displacement.nbytes = 1;
+ } else if (insn->addr_bytes == 2) {
+ if ((mod == 0 && MODRM_RM(insn) == 6) || mod == 2) {
+ insn->displacement.value = get_next(s16, insn);
+ insn->displacement.nbytes = 2;
+ }
+ } else {
+ if ((mod == 0 && MODRM_RM(insn) == 5) || mod == 2 ||
+ (mod == 0 && SIB_BASE(insn) == 5)) {
+ insn->displacement.value = get_next(s32, insn);
+ insn->displacement.nbytes = 4;
+ }
+ }
+ }
+out:
+ insn->displacement.got = true;
+}
+EXPORT_SYMBOL_GPL(insn_get_displacement);
+
+const u32 onebyte_has_immb[256 / 32] = {
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+ /* ----------------------------------------------- */
+ W(0x00, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0) | /* 0f */
+ W(0x10, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0) , /* 1f */
+ W(0x20, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0) | /* 2f */
+ W(0x30, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0) , /* 3f */
+ W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 4f */
+ W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 5f */
+ W(0x60, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0) | /* 6f */
+ W(0x70, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 7f */
+ W(0x80, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 8f */
+ W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 9f */
+ W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0) | /* af */
+ W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) , /* bf */
+ W(0xc0, 1, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0) | /* cf */
+ W(0xd0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* df */
+ W(0xe0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 0) | /* ef */
+ W(0xf0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) /* ff */
+ /* ----------------------------------------------- */
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+};
+
+const u32 onebyte_has_imm[256 / 32] = {
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+ /* ----------------------------------------------- */
+ W(0x00, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0) | /* 0f */
+ W(0x10, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0) , /* 1f */
+ W(0x20, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0) | /* 2f */
+ W(0x30, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0) , /* 3f */
+ W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 4f */
+ W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 5f */
+ W(0x60, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0) | /* 6f */
+ W(0x70, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 7f */
+ W(0x80, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 8f */
+ W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 9f */
+ W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0) | /* af */
+ W(0xb0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* bf */
+ W(0xc0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* cf */
+ W(0xd0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* df */
+ W(0xe0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0) | /* ef */
+ W(0xf0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) /* ff */
+ /* ----------------------------------------------- */
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+};
+
+/* Decode moffset16/32/64 */
+static void __get_moffset(struct insn *insn)
+{
+ switch (insn->addr_bytes) {
+ case 2:
+ insn->moffset1.value = get_next(s16, insn);
+ insn->moffset1.nbytes = 2;
+ break;
+ case 4:
+ insn->moffset1.value = get_next(s32, insn);
+ insn->moffset1.nbytes = 4;
+ break;
+ case 8:
+ insn->moffset1.value = get_next(s32, insn);
+ insn->moffset1.nbytes = 4;
+ insn->moffset2.value = get_next(s32, insn);
+ insn->moffset2.nbytes = 4;
+ break;
+ }
+ insn->moffset1.got = insn->moffset2.got = true;
+}
+
+/* Decode imm(Iz) */
+static void __get_imm(struct insn *insn)
+{
+ switch (insn->opnd_bytes) {
+ case 2:
+ insn->immediate.value = get_next(s16, insn);
+ insn->immediate.nbytes = 2;
+ break;
+ case 4:
+ case 8:
+ insn->immediate.value = get_next(s32, insn);
+ insn->immediate.nbytes = 4;
+ break;
+ }
+}
+
+/* Decode imm64(Iv) */
+static void __get_imm64(struct insn *insn)
+{
+ switch (insn->opnd_bytes) {
+ case 2:
+ insn->immediate1.value = get_next(s16, insn);
+ insn->immediate1.nbytes = 2;
+ break;
+ case 4:
+ insn->immediate1.value = get_next(s32, insn);
+ insn->immediate1.nbytes = 4;
+ break;
+ case 8:
+ insn->immediate1.value = get_next(s32, insn);
+ insn->immediate1.nbytes = 4;
+ insn->immediate2.value = get_next(s32, insn);
+ insn->immediate2.nbytes = 4;
+ break;
+ }
+ insn->immediate1.got = insn->immediate2.got = true;
+}
+
+/* Decode ptr16:16/32(AP) */
+static void __get_immptr(struct insn *insn)
+{
+ switch (insn->opnd_bytes) {
+ case 2:
+ insn->immediate1.value = get_next(s16, insn);
+ insn->immediate1.nbytes = 2;
+ break;
+ case 4:
+ insn->immediate1.value = get_next(s32, insn);
+ insn->immediate1.nbytes = 4;
+ break;
+ case 8:
+ /* ptr16:64 is not supported (no segment) */
+ WARN_ON(1);
+ return;
+ }
+ insn->immediate2.value = get_next(u16, insn);
+ insn->immediate2.nbytes = 2;
+ insn->immediate1.got = insn->immediate2.got = true;
+}
+
+/**
+ *
+ * insn_get_immediate() - Get the immediates of instruction
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * displacement bytes.
+ * Basically, most of immediates are sign-expanded. Unsigned-value can be
+ * get by bit masking with ((1 << (nbytes * 8)) - 1)
+ */
+void insn_get_immediate(struct insn *insn)
+{
+ u8 opcode;
+ if (insn->immediate.got)
+ return;
+ if (!insn->displacement.got)
+ insn_get_displacement(insn);
+ if (insn->opcode.nbytes == 1) {
+ opcode = OPCODE1(insn);
+ if (opcode >= 0xa0 && opcode <= 0xa3) { /* direct moffset mov */
+ __get_moffset(insn);
+ } else if (test_bit(opcode,
+ (const unsigned long *)onebyte_has_immb) ||
+ (opcode == 0xf6 && MODRM_REG(insn) == 0)) {
+ insn->immediate.value = get_next(s8, insn);
+ insn->immediate.nbytes = 1;
+ } else if (test_bit(opcode,
+ (const unsigned long *)onebyte_has_imm) ||
+ (opcode == 0xf7 && MODRM_REG(insn) == 0)) {
+ __get_imm(insn);
+ } else if (0xb8 <= opcode && opcode <= 0xbf /* mov immv */) {
+ __get_imm64(insn);
+ } else if (opcode == 0xea /* jmp far seg:offs */) {
+ __get_immptr(insn);
+ } else if (opcode == 0xc2 /* retn immw */ ||
+ opcode == 0xca /* retf immw */) {
+ insn->immediate.value = get_next(u16, insn);
+ insn->immediate.nbytes = 2;
+ } else if (opcode == 0xc8 /* enter immw, immb */) {
+ insn->immediate1.value = get_next(u16, insn);
+ insn->immediate1.nbytes = 2;
+ insn->immediate2.value = get_next(u8, insn);
+ insn->immediate2.nbytes = 1;
+ }
+ } else if (insn->opcode.nbytes == 2) {
+ opcode = OPCODE2(insn);
+ if ((opcode & 0xf0) == 0x80 /* Jcc imm32 */) {
+ __get_imm(insn);
+ } else
+ switch (opcode) {
+ case 0x70: /* pshuf* %1, %2, immb */
+ case 0x71: /* Group12 %1, immb */
+ case 0x72: /* Group13 %1, immb */
+ case 0x73: /* Group14 %1, immb */
+ case 0xa4: /* shld %1, %2, immb */
+ case 0xac: /* shrd %1, %2, immb */
+ case 0xba: /* Group8 %1, immb */
+ case 0xc2: /* cmpps %1, %2, immb */
+ case 0xc4: /* pinsw %1, %2, immb */
+ case 0xc5: /* pextrw %1, %2, immb */
+ case 0xc6: /* shufps/d %1, %2, immb */
+ insn->immediate.value = get_next(u8, insn);
+ insn->immediate.nbytes = 1;
+ default:
+ break;
+ }
+ } else if (OPCODE3(insn) == 0x0f /* pailgnr %1, %2, immb */) {
+ insn->immediate.value = get_next(u8, insn);
+ insn->immediate.nbytes = 1;
+ }
+ insn->immediate.got = true;
+}
+EXPORT_SYMBOL_GPL(insn_get_immediate);
+
+/**
+ *
+ * insn_get_length() - Get the length of instruction
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * immediates bytes.
+ */
+void insn_get_length(struct insn *insn)
+{
+ if (insn->length)
+ return;
+ if (!insn->immediate.got)
+ insn_get_immediate(insn);
+ insn->length = (u8)((unsigned long)insn->next_byte
+ - (unsigned long)insn->kaddr);
+}
+EXPORT_SYMBOL_GPL(insn_get_length);
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division
e-mail: [email protected]
Add x86 instruction decoder to arch-specific libraries. This decoder
can decode all x86 instructions into prefix, opcode, modrm, sib,
displacement and immediates. This can also show the length of
instructions.
changes from v4:
- make bitmap tables static.
Signed-off-by: Jim Keniston <[email protected]>
Signed-off-by: Masami Hiramatsu <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Ananth N Mavinakayanahalli <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
Cc: Andi Kleen <[email protected]>
---
arch/x86/include/asm/insn.h | 130 +++++++++
arch/x86/lib/Makefile | 1
arch/x86/lib/insn.c | 627 +++++++++++++++++++++++++++++++++++++++++++
3 files changed, 758 insertions(+), 0 deletions(-)
create mode 100644 arch/x86/include/asm/insn.h
create mode 100644 arch/x86/lib/insn.c
diff --git a/arch/x86/include/asm/insn.h b/arch/x86/include/asm/insn.h
new file mode 100644
index 0000000..488001f
--- /dev/null
+++ b/arch/x86/include/asm/insn.h
@@ -0,0 +1,130 @@
+#ifndef _ASM_X86_INSN_H
+#define _ASM_X86_INSN_H
+/*
+ * x86 instruction analysis
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2009
+ */
+
+#include <linux/types.h>
+
+/* legacy instruction prefixes */
+#define X86_PFX_OPNDSZ 0x1 /* 0x66 */
+#define X86_PFX_ADDRSZ 0x2 /* 0x67 */
+#define X86_PFX_CS 0x4 /* 0x2E */
+#define X86_PFX_DS 0x8 /* 0x3E */
+#define X86_PFX_ES 0x10 /* 0x26 */
+#define X86_PFX_FS 0x20 /* 0x64 */
+#define X86_PFX_GS 0x40 /* 0x65 */
+#define X86_PFX_SS 0x80 /* 0x36 */
+#define X86_PFX_LOCK 0x100 /* 0xF0 */
+#define X86_PFX_REPE 0x200 /* 0xF3 */
+#define X86_PFX_REPNE 0x400 /* 0xF2 */
+/* REX prefix */
+#define X86_PFX_REX 0x800 /* 0x4X */
+/* REX prefix dissected */
+#define X86_PFX_REX_BASE 0x1000
+#define X86_PFX_REXB 0x1000 /* 0x41 bit */
+#define X86_PFX_REXX 0x2000 /* 0x42 bit */
+#define X86_PFX_REXR 0x4000 /* 0x44 bit */
+#define X86_PFX_REXW 0x8000 /* 0x48 bit */
+
+struct insn_field {
+ union {
+ s32 value;
+ u8 bytes[4];
+ };
+ bool got; /* true if we've run insn_get_xxx() for this field */
+ u8 nbytes;
+};
+
+struct insn {
+ struct insn_field prefixes; /* prefixes.value is a bitmap */
+ struct insn_field opcode; /*
+ * opcode.bytes[0]: opcode1
+ * opcode.bytes[1]: opcode2
+ * opcode.bytes[2]: opcode3
+ */
+ struct insn_field modrm;
+ struct insn_field sib;
+ struct insn_field displacement;
+ union {
+ struct insn_field immediate;
+ struct insn_field moffset1; /* for 64bit MOV */
+ struct insn_field immediate1; /* for 64bit imm or off16/32 */
+ };
+ union {
+ struct insn_field moffset2; /* for 64bit MOV */
+ struct insn_field immediate2; /* for 64bit imm or seg16 */
+ };
+
+ u8 opnd_bytes;
+ u8 addr_bytes;
+ u8 length;
+ bool x86_64;
+
+ const u8 *kaddr; /* kernel address of insn (copy) to analyze */
+ const u8 *next_byte;
+};
+
+#define OPCODE1(insn) ((insn)->opcode.bytes[0])
+#define OPCODE2(insn) ((insn)->opcode.bytes[1])
+#define OPCODE3(insn) ((insn)->opcode.bytes[2])
+
+#define MODRM_MOD(insn) (((insn)->modrm.value & 0xc0) >> 6)
+#define MODRM_REG(insn) (((insn)->modrm.value & 0x38) >> 3)
+#define MODRM_RM(insn) ((insn)->modrm.value & 0x07)
+
+#define SIB_SCALE(insn) (((insn)->sib.value & 0xc0) >> 6)
+#define SIB_INDEX(insn) (((insn)->sib.value & 0x38) >> 3)
+#define SIB_BASE(insn) ((insn)->sib.value & 0x07)
+
+#define MOFFSET64(insn) (((u64)((insn)->moffset2.value) << 32) | \
+ (u32)((insn)->moffset1.value))
+
+#define IMMEDIATE64(insn) (((u64)((insn)->immediate2.value) << 32) | \
+ (u32)((insn)->immediate1.value))
+
+extern void insn_init(struct insn *insn, const u8 *kaddr, bool x86_64);
+extern void insn_get_prefixes(struct insn *insn);
+extern void insn_get_opcode(struct insn *insn);
+extern void insn_get_modrm(struct insn *insn);
+extern void insn_get_sib(struct insn *insn);
+extern void insn_get_displacement(struct insn *insn);
+extern void insn_get_immediate(struct insn *insn);
+extern void insn_get_length(struct insn *insn);
+
+#ifdef CONFIG_X86_64
+/* Init insn for kernel text */
+#define insn_init_kernel(insn, kaddr) insn_init(insn, kaddr, 1)
+extern bool insn_rip_relative(struct insn *insn);
+
+#else /* CONFIG_X86_32 */
+
+#define insn_init_kernel(insn, kaddr) insn_init(insn, kaddr, 0)
+static inline bool insn_rip_relative(struct insn *insn)
+{
+ return false;
+}
+#endif
+
+static inline bool insn_field_exists(const struct insn_field *field)
+{
+ return (field->nbytes > 0);
+}
+
+#endif /* _ASM_X86_INSN_H */
diff --git a/arch/x86/lib/Makefile b/arch/x86/lib/Makefile
index 55e11aa..0f81979 100644
--- a/arch/x86/lib/Makefile
+++ b/arch/x86/lib/Makefile
@@ -8,6 +8,7 @@ lib-y := delay.o
lib-y += thunk_$(BITS).o
lib-y += usercopy_$(BITS).o getuser.o putuser.o
lib-y += memcpy_$(BITS).o
+lib-y += insn.o
ifeq ($(CONFIG_X86_32),y)
lib-y += checksum_32.o
diff --git a/arch/x86/lib/insn.c b/arch/x86/lib/insn.c
new file mode 100644
index 0000000..28a57ce
--- /dev/null
+++ b/arch/x86/lib/insn.c
@@ -0,0 +1,627 @@
+/*
+ * x86 instruction analysis
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2002, 2004, 2009
+ */
+
+#include <linux/string.h>
+#include <linux/module.h>
+#include <asm/insn.h>
+
+#undef W
+#define W(row, b0, b1, b2, b3, b4, b5, b6, b7, b8, b9, ba, bb, bc, bd, be, bf)\
+ (((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) | \
+ (b4##UL << 0x4)|(b5##UL << 0x5)|(b6##UL << 0x6)|(b7##UL << 0x7) | \
+ (b8##UL << 0x8)|(b9##UL << 0x9)|(ba##UL << 0xa)|(bb##UL << 0xb) | \
+ (bc##UL << 0xc)|(bd##UL << 0xd)|(be##UL << 0xe)|(bf##UL << 0xf)) \
+ << (row % 32))
+
+/**
+ * insn_init() - initialize struct insn
+ * @insn: &struct insn to be initialized
+ * @kaddr: address (in kernel memory) of instruction (or copy thereof)
+ * @x86_64: true for 64-bit kernel or 64-bit app
+ */
+void insn_init(struct insn *insn, const u8 *kaddr, bool x86_64)
+{
+ memset(insn, 0, sizeof(*insn));
+ insn->kaddr = kaddr;
+ insn->next_byte = kaddr;
+ insn->x86_64 = x86_64;
+ insn->opnd_bytes = 4;
+ if (x86_64)
+ insn->addr_bytes = 8;
+ else
+ insn->addr_bytes = 4;
+}
+EXPORT_SYMBOL_GPL(insn_init);
+
+/**
+ * insn_get_prefixes - scan x86 instruction prefix bytes
+ * @insn: &struct insn containing instruction
+ *
+ * Populates the @insn->prefixes bitmap, and updates @insn->next_byte
+ * to point to the (first) opcode. No effect if @insn->prefixes.got
+ * is already true.
+ */
+void insn_get_prefixes(struct insn *insn)
+{
+ u32 pfx;
+ struct insn_field *prefixes = &insn->prefixes;
+ if (prefixes->got)
+ return;
+ for (;; insn->next_byte++, prefixes->nbytes++) {
+ u8 b = *(insn->next_byte);
+#ifdef CONFIG_X86_64
+ if ((b & 0xf0) == 0x40 && insn->x86_64) {
+ prefixes->value |= X86_PFX_REX;
+ prefixes->value |= (b & 0x0f) * X86_PFX_REX_BASE;
+ /* REX prefix is always last. */
+ insn->next_byte++;
+ prefixes->nbytes++;
+ break;
+ }
+#endif
+ switch (b) {
+ case 0x26:
+ pfx = X86_PFX_ES;
+ break;
+ case 0x2E:
+ pfx = X86_PFX_CS;
+ break;
+ case 0x36:
+ pfx = X86_PFX_SS;
+ break;
+ case 0x3E:
+ pfx = X86_PFX_DS;
+ break;
+ case 0x64:
+ pfx = X86_PFX_FS;
+ break;
+ case 0x65:
+ pfx = X86_PFX_GS;
+ break;
+ case 0x66:
+ pfx = X86_PFX_OPNDSZ;
+ break;
+ case 0x67:
+ pfx = X86_PFX_ADDRSZ;
+ break;
+ case 0xF0:
+ pfx = X86_PFX_LOCK;
+ break;
+ case 0xF2:
+ pfx = X86_PFX_REPNE;
+ break;
+ case 0xF3:
+ pfx = X86_PFX_REPE;
+ break;
+ default:
+ pfx = 0x0;
+ break;
+ }
+ if (!pfx)
+ break;
+ prefixes->value |= pfx;
+ }
+ if (prefixes->value & X86_PFX_OPNDSZ) {
+ /* oprand size switches 2/4 */
+ insn->opnd_bytes ^= 6;
+ }
+ if (prefixes->value & X86_PFX_ADDRSZ) {
+ /* address size switches 2/4 or 4/8 */
+#ifdef CONFIG_X86_64
+ if (insn->x86_64)
+ insn->addr_bytes ^= 12;
+ else
+#endif
+ insn->addr_bytes ^= 6;
+ }
+#ifdef CONFIG_X86_64
+ if (prefixes->value & X86_PFX_REXW)
+ insn->opnd_bytes = 8;
+#endif
+ prefixes->got = true;
+}
+EXPORT_SYMBOL_GPL(insn_get_prefixes);
+
+/**
+ * insn_get_opcode - collect opcode(s)
+ * @insn: &struct insn containing instruction
+ *
+ * Populates @insn->opcode1 (and @insn->opcode2, if it's a 2-byte opcode)
+ * and updates @insn->next_byte to point past the opcode byte(s).
+ * If necessary, first collects any preceding (prefix) bytes.
+ * Sets @insn->opcode.value = opcode1. No effect if @insn->opcode.got
+ * is already true.
+ */
+void insn_get_opcode(struct insn *insn)
+{
+ struct insn_field *opcode = &insn->opcode;
+ if (opcode->got)
+ return;
+ if (!insn->prefixes.got)
+ insn_get_prefixes(insn);
+ OPCODE1(insn) = *insn->next_byte++;
+ if (OPCODE1(insn) == 0x0f) {
+ OPCODE2(insn) = *insn->next_byte++;
+ if (OPCODE2(insn) == 0x38 || OPCODE2(insn) == 0x3a) {
+ OPCODE3(insn) = *insn->next_byte++;
+ opcode->nbytes = 3;
+ } else
+ opcode->nbytes = 2;
+ } else
+ opcode->nbytes = 1;
+ opcode->got = true;
+}
+EXPORT_SYMBOL_GPL(insn_get_opcode);
+
+static const u32 onebyte_has_modrm[256 / 32] = {
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+ /* ----------------------------------------------- */
+ W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 0f */
+ W(0x10, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 1f */
+ W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 2f */
+ W(0x30, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 3f */
+ W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 4f */
+ W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 5f */
+ W(0x60, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0) | /* 6f */
+ W(0x70, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 7f */
+ W(0x80, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 8f */
+ W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 9f */
+ W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* af */
+ W(0xb0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* bf */
+ W(0xc0, 1, 1, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* cf */
+ W(0xd0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) , /* df */
+ W(0xe0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* ef */
+ W(0xf0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1) /* ff */
+ /* ----------------------------------------------- */
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+};
+
+static const u32 twobyte_has_modrm[256 / 32] = {
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+ /* ----------------------------------------------- */
+ W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1) | /* 0f */
+ W(0x10, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 1f */
+ W(0x20, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* 2f */
+ W(0x30, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 3f */
+ W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 4f */
+ W(0x50, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 5f */
+ W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 6f */
+ W(0x70, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1) , /* 7f */
+ W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 8f */
+ W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 9f */
+ W(0xa0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) | /* af */
+ W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1) , /* bf */
+ W(0xc0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* cf */
+ W(0xd0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* df */
+ W(0xe0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* ef */
+ W(0xf0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0) /* ff */
+ /* ----------------------------------------------- */
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+};
+
+#ifdef CONFIG_X86_64
+static const u32 onebyte_force_64[256 / 32] = {
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+ /* ----------------------------------------------- */
+ W(0x00, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 0f */
+ W(0x10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 1f */
+ W(0x20, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 2f */
+ W(0x30, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 3f */
+ W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 4f */
+ W(0x50, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 5f */
+ W(0x60, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0) | /* 6f */
+ W(0x70, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 7f */
+ W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1) | /* 8f */
+ W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 9f */
+ W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* af */
+ W(0xb0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* bf */
+ W(0xc0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0) | /* cf */
+ W(0xd0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* df */
+ W(0xe0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 0, 1, 0, 0, 0, 0) | /* ef */
+ W(0xf0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) /* ff */
+ /* ----------------------------------------------- */
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+};
+
+/* force 64 or default 64 bits operand opcodes */
+static bool __operand_64(struct insn *insn)
+{
+ u8 reg = MODRM_REG(insn);
+ if (insn->opcode.nbytes == 1) {
+ if (test_bit(OPCODE1(insn),
+ (const unsigned long *) onebyte_force_64) ||
+ (OPCODE1(insn) == 0xff &&
+ (reg == 2 || reg == 4 || reg == 6)))
+ return true;
+ }
+ return false;
+}
+#endif
+
+/**
+ * insn_get_modrm - collect ModRM byte, if any
+ * @insn: &struct insn containing instruction
+ *
+ * Populates @insn->modrm and updates @insn->next_byte to point past the
+ * ModRM byte, if any. If necessary, first collects the preceding bytes
+ * (prefixes and opcode(s)). No effect if @insn->modrm.got is already true.
+ */
+void insn_get_modrm(struct insn *insn)
+{
+ struct insn_field *modrm = &insn->modrm;
+ if (modrm->got)
+ return;
+ if (!insn->opcode.got)
+ insn_get_opcode(insn);
+ switch (insn->opcode.nbytes) {
+ case 1:
+ modrm->nbytes = test_bit(OPCODE1(insn),
+ (const unsigned long *) onebyte_has_modrm);
+ break;
+ case 2:
+ modrm->nbytes = test_bit(OPCODE2(insn),
+ (const unsigned long *) twobyte_has_modrm);
+ break;
+ case 3:
+ /* Three bytes opcodes always have modrm */
+ modrm->nbytes = 1;
+ break;
+ }
+ if (modrm->nbytes)
+ modrm->value = *(insn->next_byte++);
+
+#ifdef CONFIG_X86_64
+ if (insn->x86_64 && __operand_64(insn))
+ insn->opnd_bytes = 8;
+#endif
+ modrm->got = true;
+}
+EXPORT_SYMBOL_GPL(insn_get_modrm);
+
+#ifdef CONFIG_X86_64
+/**
+ * insn_rip_relative() - Does instruction use RIP-relative addressing mode?
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * ModRM byte. No effect if @insn->x86_64 is false.
+ */
+bool insn_rip_relative(struct insn *insn)
+{
+ struct insn_field *modrm = &insn->modrm;
+
+ if (!insn->x86_64)
+ return false;
+ if (!modrm->got)
+ insn_get_modrm(insn);
+ /*
+ * For rip-relative instructions, the mod field (top 2 bits)
+ * is zero and the r/m field (bottom 3 bits) is 0x5.
+ */
+ return (insn_field_exists(modrm) && (modrm->value & 0xc7) == 0x5);
+}
+EXPORT_SYMBOL_GPL(insn_rip_relative);
+#endif
+
+/**
+ *
+ * insn_get_sib() - Get the SIB byte of instruction
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * ModRM byte.
+ */
+void insn_get_sib(struct insn *insn)
+{
+ if (insn->sib.got)
+ return;
+ if (!insn->modrm.got)
+ insn_get_modrm(insn);
+ if (insn->modrm.nbytes)
+ if (insn->addr_bytes != 2 &&
+ MODRM_MOD(insn) != 3 && MODRM_RM(insn) == 4) {
+ insn->sib.value = *(insn->next_byte++);
+ insn->sib.nbytes = 1;
+ }
+ insn->sib.got = true;
+}
+EXPORT_SYMBOL_GPL(insn_get_sib);
+
+#define get_next(t, insn) \
+ ({t r; r = *(t*)insn->next_byte; insn->next_byte += sizeof(t); r; })
+
+/**
+ *
+ * insn_get_displacement() - Get the displacement of instruction
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * SIB byte.
+ * Displacement value is sign-expanded.
+ */
+void insn_get_displacement(struct insn *insn)
+{
+ u8 mod;
+ if (insn->displacement.got)
+ return;
+ if (!insn->sib.got)
+ insn_get_sib(insn);
+ if (insn->modrm.nbytes) {
+ /*
+ * Interpreting the modrm byte:
+ * mod = 00 - no displacement fields (exceptions below)
+ * mod = 01 - 1-byte displacement field
+ * mod = 10 - displacement field is 4 bytes, or 2 bytes if
+ * address size = 2 (0x67 prefix in 32-bit mode)
+ * mod = 11 - no memory operand
+ *
+ * If address size = 2...
+ * mod = 00, r/m = 110 - displacement field is 2 bytes
+ *
+ * If address size != 2...
+ * mod != 11, r/m = 100 - SIB byte exists
+ * mod = 00, SIB base = 101 - displacement field is 4 bytes
+ * mod = 00, r/m = 101 - rip-relative addressing, displacement
+ * field is 4 bytes
+ */
+ mod = MODRM_MOD(insn);
+ if (mod == 3)
+ goto out;
+ if (mod == 1) {
+ insn->displacement.value = *((s8 *)insn->next_byte++);
+ insn->displacement.nbytes = 1;
+ } else if (insn->addr_bytes == 2) {
+ if ((mod == 0 && MODRM_RM(insn) == 6) || mod == 2) {
+ insn->displacement.value = get_next(s16, insn);
+ insn->displacement.nbytes = 2;
+ }
+ } else {
+ if ((mod == 0 && MODRM_RM(insn) == 5) || mod == 2 ||
+ (mod == 0 && SIB_BASE(insn) == 5)) {
+ insn->displacement.value = get_next(s32, insn);
+ insn->displacement.nbytes = 4;
+ }
+ }
+ }
+out:
+ insn->displacement.got = true;
+}
+EXPORT_SYMBOL_GPL(insn_get_displacement);
+
+static const u32 onebyte_has_immb[256 / 32] = {
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+ /* ----------------------------------------------- */
+ W(0x00, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0) | /* 0f */
+ W(0x10, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0) , /* 1f */
+ W(0x20, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0) | /* 2f */
+ W(0x30, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0) , /* 3f */
+ W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 4f */
+ W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 5f */
+ W(0x60, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0) | /* 6f */
+ W(0x70, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 7f */
+ W(0x80, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 8f */
+ W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 9f */
+ W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0) | /* af */
+ W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) , /* bf */
+ W(0xc0, 1, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0) | /* cf */
+ W(0xd0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* df */
+ W(0xe0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 0) | /* ef */
+ W(0xf0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) /* ff */
+ /* ----------------------------------------------- */
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+};
+
+static const u32 onebyte_has_imm[256 / 32] = {
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+ /* ----------------------------------------------- */
+ W(0x00, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0) | /* 0f */
+ W(0x10, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0) , /* 1f */
+ W(0x20, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0) | /* 2f */
+ W(0x30, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0) , /* 3f */
+ W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 4f */
+ W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 5f */
+ W(0x60, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0) | /* 6f */
+ W(0x70, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 7f */
+ W(0x80, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 8f */
+ W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 9f */
+ W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0) | /* af */
+ W(0xb0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* bf */
+ W(0xc0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* cf */
+ W(0xd0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* df */
+ W(0xe0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0) | /* ef */
+ W(0xf0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) /* ff */
+ /* ----------------------------------------------- */
+ /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
+};
+
+/* Decode moffset16/32/64 */
+static void __get_moffset(struct insn *insn)
+{
+ switch (insn->addr_bytes) {
+ case 2:
+ insn->moffset1.value = get_next(s16, insn);
+ insn->moffset1.nbytes = 2;
+ break;
+ case 4:
+ insn->moffset1.value = get_next(s32, insn);
+ insn->moffset1.nbytes = 4;
+ break;
+ case 8:
+ insn->moffset1.value = get_next(s32, insn);
+ insn->moffset1.nbytes = 4;
+ insn->moffset2.value = get_next(s32, insn);
+ insn->moffset2.nbytes = 4;
+ break;
+ }
+ insn->moffset1.got = insn->moffset2.got = true;
+}
+
+/* Decode imm(Iz) */
+static void __get_imm(struct insn *insn)
+{
+ switch (insn->opnd_bytes) {
+ case 2:
+ insn->immediate.value = get_next(s16, insn);
+ insn->immediate.nbytes = 2;
+ break;
+ case 4:
+ case 8:
+ insn->immediate.value = get_next(s32, insn);
+ insn->immediate.nbytes = 4;
+ break;
+ }
+}
+
+/* Decode imm64(Iv) */
+static void __get_imm64(struct insn *insn)
+{
+ switch (insn->opnd_bytes) {
+ case 2:
+ insn->immediate1.value = get_next(s16, insn);
+ insn->immediate1.nbytes = 2;
+ break;
+ case 4:
+ insn->immediate1.value = get_next(s32, insn);
+ insn->immediate1.nbytes = 4;
+ break;
+ case 8:
+ insn->immediate1.value = get_next(s32, insn);
+ insn->immediate1.nbytes = 4;
+ insn->immediate2.value = get_next(s32, insn);
+ insn->immediate2.nbytes = 4;
+ break;
+ }
+ insn->immediate1.got = insn->immediate2.got = true;
+}
+
+/* Decode ptr16:16/32(AP) */
+static void __get_immptr(struct insn *insn)
+{
+ switch (insn->opnd_bytes) {
+ case 2:
+ insn->immediate1.value = get_next(s16, insn);
+ insn->immediate1.nbytes = 2;
+ break;
+ case 4:
+ insn->immediate1.value = get_next(s32, insn);
+ insn->immediate1.nbytes = 4;
+ break;
+ case 8:
+ /* ptr16:64 is not supported (no segment) */
+ WARN_ON(1);
+ return;
+ }
+ insn->immediate2.value = get_next(u16, insn);
+ insn->immediate2.nbytes = 2;
+ insn->immediate1.got = insn->immediate2.got = true;
+}
+
+/**
+ *
+ * insn_get_immediate() - Get the immediates of instruction
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * displacement bytes.
+ * Basically, most of immediates are sign-expanded. Unsigned-value can be
+ * get by bit masking with ((1 << (nbytes * 8)) - 1)
+ */
+void insn_get_immediate(struct insn *insn)
+{
+ u8 opcode;
+ if (insn->immediate.got)
+ return;
+ if (!insn->displacement.got)
+ insn_get_displacement(insn);
+ if (insn->opcode.nbytes == 1) {
+ opcode = OPCODE1(insn);
+ if (opcode >= 0xa0 && opcode <= 0xa3) { /* direct moffset mov */
+ __get_moffset(insn);
+ } else if (test_bit(opcode,
+ (const unsigned long *)onebyte_has_immb) ||
+ (opcode == 0xf6 && MODRM_REG(insn) == 0)) {
+ insn->immediate.value = get_next(s8, insn);
+ insn->immediate.nbytes = 1;
+ } else if (test_bit(opcode,
+ (const unsigned long *)onebyte_has_imm) ||
+ (opcode == 0xf7 && MODRM_REG(insn) == 0)) {
+ __get_imm(insn);
+ } else if (0xb8 <= opcode && opcode <= 0xbf /* mov immv */) {
+ __get_imm64(insn);
+ } else if (opcode == 0xea /* jmp far seg:offs */) {
+ __get_immptr(insn);
+ } else if (opcode == 0xc2 /* retn immw */ ||
+ opcode == 0xca /* retf immw */) {
+ insn->immediate.value = get_next(u16, insn);
+ insn->immediate.nbytes = 2;
+ } else if (opcode == 0xc8 /* enter immw, immb */) {
+ insn->immediate1.value = get_next(u16, insn);
+ insn->immediate1.nbytes = 2;
+ insn->immediate2.value = get_next(u8, insn);
+ insn->immediate2.nbytes = 1;
+ }
+ } else if (insn->opcode.nbytes == 2) {
+ opcode = OPCODE2(insn);
+ if ((opcode & 0xf0) == 0x80 /* Jcc imm32 */) {
+ __get_imm(insn);
+ } else
+ switch (opcode) {
+ case 0x70: /* pshuf* %1, %2, immb */
+ case 0x71: /* Group12 %1, immb */
+ case 0x72: /* Group13 %1, immb */
+ case 0x73: /* Group14 %1, immb */
+ case 0xa4: /* shld %1, %2, immb */
+ case 0xac: /* shrd %1, %2, immb */
+ case 0xba: /* Group8 %1, immb */
+ case 0xc2: /* cmpps %1, %2, immb */
+ case 0xc4: /* pinsw %1, %2, immb */
+ case 0xc5: /* pextrw %1, %2, immb */
+ case 0xc6: /* shufps/d %1, %2, immb */
+ insn->immediate.value = get_next(u8, insn);
+ insn->immediate.nbytes = 1;
+ default:
+ break;
+ }
+ } else if (OPCODE3(insn) == 0x0f /* pailgnr %1, %2, immb */) {
+ insn->immediate.value = get_next(u8, insn);
+ insn->immediate.nbytes = 1;
+ }
+ insn->immediate.got = true;
+}
+EXPORT_SYMBOL_GPL(insn_get_immediate);
+
+/**
+ *
+ * insn_get_length() - Get the length of instruction
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * immediates bytes.
+ */
+void insn_get_length(struct insn *insn)
+{
+ if (insn->length)
+ return;
+ if (!insn->immediate.got)
+ insn_get_immediate(insn);
+ insn->length = (u8)((unsigned long)insn->next_byte
+ - (unsigned long)insn->kaddr);
+}
+EXPORT_SYMBOL_GPL(insn_get_length);
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division
e-mail: [email protected]
Masami Hiramatsu wrote:
> Add x86 instruction decoder to arch-specific libraries. This decoder
> can decode all x86 instructions into prefix, opcode, modrm, sib,
> displacement and immediates. This can also show the length of
> instructions.
>
> changes from v4:
> - make bitmap tables static.
Hi Masami,
On the surface the overall structure looks fine, but I have a couple of
concerns:
1. is this meant to be able to decode userspace code or just kernel
code? If it is supposed to be able to decode userspace code, is there a
reason you're not dealing with 16-bit or V86 mode code at all? If not,
why are you including the 32-bit decoder in a 64-bit kernel (as well as
instructions which we're pretty much guaranteed to never use in the
kernel, such as ENTER.)
2. you're already not dealing with all existing three-byte opcode
spaces, nor with DREX or VEX encodings for upcoming processors. This
doesn't matter so much for the kernel, but it does matter if this is
supposed to be used for user-space code.
3. is there any need to deal with instruction set differences among
processors? (Again, this depends on the usage model.)
4. you have a bunch of magic opcode constants all over the place. This
means that as new instructions come in -- and they're going to be coming
in -- this is going to be hard to update. It would be cleaner if we
could have an intermediate format that preprocesses down to all the
relevant tables and perhaps even some of the code rather than
open-coding everything in C.
This matters... for example you have:
+ } else if (opcode == 0xea /* jmp far seg:offs */) {
+ __get_immptr(insn);
... but nothing similar for opcode 0x9a. This is extremely hard to spot
with this kind of structure.
The more data-driven we can make it (without bloating the code too much)
the better off we are, I believe.
-hpa
Hi Peter,
H. Peter Anvin wrote:
> Masami Hiramatsu wrote:
>> Add x86 instruction decoder to arch-specific libraries. This decoder
>> can decode all x86 instructions into prefix, opcode, modrm, sib,
>> displacement and immediates. This can also show the length of
>> instructions.
>>
>> changes from v4:
>> - make bitmap tables static.
>
> Hi Masami,
>
> On the surface the overall structure looks fine, but I have a couple of
> concerns:
>
> 1. is this meant to be able to decode userspace code or just kernel
> code? If it is supposed to be able to decode userspace code, is there a
> reason you're not dealing with 16-bit or V86 mode code at all? If not,
> why are you including the 32-bit decoder in a 64-bit kernel (as well as
> instructions which we're pretty much guaranteed to never use in the
> kernel, such as ENTER.)
Actually, this aims to decode both of user space and kernel code.
At this point, it just needs to cover kernel code, because kprobes
just want to decode kernel binary.
However, this is just a starting point, uprobe developers want to
use it to decode user-space code. In that case, it needs to be
enhanced.
> 2. you're already not dealing with all existing three-byte opcode
> spaces, nor with DREX or VEX encodings for upcoming processors. This
> doesn't matter so much for the kernel, but it does matter if this is
> supposed to be used for user-space code.
>
> 3. is there any need to deal with instruction set differences among
> processors? (Again, this depends on the usage model.)
Agreed. When it support decoding user-space code, it should
support all kind of instructions.
>
> 4. you have a bunch of magic opcode constants all over the place. This
> means that as new instructions come in -- and they're going to be coming
> in -- this is going to be hard to update. It would be cleaner if we
> could have an intermediate format that preprocesses down to all the
> relevant tables and perhaps even some of the code rather than
> open-coding everything in C.
>
> This matters... for example you have:
>
> + } else if (opcode == 0xea /* jmp far seg:offs */) {
> + __get_immptr(insn);
>
> ... but nothing similar for opcode 0x9a. This is extremely hard to spot
> with this kind of structure.
Oops, that should be a bug. Hmm, I think we'd better bit-flags tables
for classifying opcodes.
Jim, can your INAT idea help this situation?
http://sources.redhat.com/ml/systemtap/2009-q2/msg00109.html
>
> The more data-driven we can make it (without bloating the code too much)
> the better off we are, I believe.
>
> -hpa
Thank you for good advice!
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division
e-mail: [email protected]
On Fri, 2009-04-03 at 20:37 -0400, Masami Hiramatsu wrote:
> Hi Peter,
>
> H. Peter Anvin wrote:
> > Masami Hiramatsu wrote:
> >> Add x86 instruction decoder to arch-specific libraries. This decoder
> >> can decode all x86 instructions into prefix, opcode, modrm, sib,
> >> displacement and immediates. This can also show the length of
> >> instructions.
> >>
...
> >
> > Hi Masami,
> >
> > On the surface the overall structure looks fine, but I have a couple of
> > concerns:
> >
> > 1. is this meant to be able to decode userspace code or just kernel
> > code? If it is supposed to be able to decode userspace code, is there a
> > reason you're not dealing with 16-bit or V86 mode code at all? If not,
> > why are you including the 32-bit decoder in a 64-bit kernel (as well as
> > instructions which we're pretty much guaranteed to never use in the
> > kernel, such as ENTER.)
>
> Actually, this aims to decode both of user space and kernel code.
> At this point, it just needs to cover kernel code, because kprobes
> just want to decode kernel binary.
> However, this is just a starting point, uprobe developers want to
> use it to decode user-space code. In that case, it needs to be
> enhanced.
For user-space probing, we've been concentrating on native-built
executables. Am I correct in thinking that we'll see 16-bit or V86 mode
only on legacy apps built elsewhere? In any case, it only makes sense
to build on the kvm folks' work in this regard.
...
>
> >
> > 4. you have a bunch of magic opcode constants all over the place. This
> > means that as new instructions come in -- and they're going to be coming
> > in -- this is going to be hard to update. It would be cleaner if we
> > could have an intermediate format that preprocesses down to all the
> > relevant tables and perhaps even some of the code rather than
> > open-coding everything in C.
> >
> > This matters... for example you have:
> >
> > + } else if (opcode == 0xea /* jmp far seg:offs */) {
> > + __get_immptr(insn);
> >
> > ... but nothing similar for opcode 0x9a. This is extremely hard to spot
> > with this kind of structure.
>
> Oops, that should be a bug. Hmm, I think we'd better bit-flags tables
> for classifying opcodes.
> Jim, can your INAT idea help this situation?
>
> http://sources.redhat.com/ml/systemtap/2009-q2/msg00109.html
>
As noted, the INAT tables follow the kvm model of one fat bitmap of
attributes per opcode, rather than the kprobes/uprobes model of one or
two 256-bit tables per attribute. (This latter approach was due to the
gradual accumulation of tables over the years.)
I like the bitmap-per-opcode approach because it's relatively easy to
see in one place everything you're saying about a particular opcode.
But with all the potential clients for this service, it's not clear that
we'll get by with a single bitmap for every opcode. (x86 kvm uses 32
bits per opcode, I think, and the INAT tables use 10. Seems like we
could overrun 64 bits pretty quickly.) So I guess that means we'll have
to get a little creative as to how we expose these attribute sets to the
client.
...
>
> Thank you for good advice!
>
Ditto.
Jim Keniston
Jim Keniston wrote:
>
> For user-space probing, we've been concentrating on native-built
> executables. Am I correct in thinking that we'll see 16-bit or V86 mode
> only on legacy apps built elsewhere? In any case, it only makes sense
> to build on the kvm folks' work in this regard.
>
That's a fair assumption; you will of course need to test it and take
appropriate action if it doesn't pan out.
>
> As noted, the INAT tables follow the kvm model of one fat bitmap of
> attributes per opcode, rather than the kprobes/uprobes model of one or
> two 256-bit tables per attribute. (This latter approach was due to the
> gradual accumulation of tables over the years.)
>
> I like the bitmap-per-opcode approach because it's relatively easy to
> see in one place everything you're saying about a particular opcode.
> But with all the potential clients for this service, it's not clear that
> we'll get by with a single bitmap for every opcode. (x86 kvm uses 32
> bits per opcode, I think, and the INAT tables use 10. Seems like we
> could overrun 64 bits pretty quickly.) So I guess that means we'll have
> to get a little creative as to how we expose these attribute sets to the
> client.
>
This is another very good reason to use an instruction table which is
preprocessed into a usable format: it means that if the internal data
structures change -- and they almost certainly will have to at some
point -- the raw data isn't lost.
-hpa
--
H. Peter Anvin, Intel Open Source Technology Center
I work for Intel. I don't speak on their behalf.
H. Peter Anvin wrote:
> Jim Keniston wrote:
>> For user-space probing, we've been concentrating on native-built
>> executables. Am I correct in thinking that we'll see 16-bit or V86 mode
>> only on legacy apps built elsewhere? In any case, it only makes sense
>> to build on the kvm folks' work in this regard.
>>
>
> That's a fair assumption; you will of course need to test it and take
> appropriate action if it doesn't pan out.
>
>> As noted, the INAT tables follow the kvm model of one fat bitmap of
>> attributes per opcode, rather than the kprobes/uprobes model of one or
>> two 256-bit tables per attribute. (This latter approach was due to the
>> gradual accumulation of tables over the years.)
>>
>> I like the bitmap-per-opcode approach because it's relatively easy to
>> see in one place everything you're saying about a particular opcode.
>> But with all the potential clients for this service, it's not clear that
>> we'll get by with a single bitmap for every opcode. (x86 kvm uses 32
>> bits per opcode, I think, and the INAT tables use 10. Seems like we
>> could overrun 64 bits pretty quickly.) So I guess that means we'll have
>> to get a little creative as to how we expose these attribute sets to the
>> client.
>>
>
> This is another very good reason to use an instruction table which is
> preprocessed into a usable format: it means that if the internal data
> structures change -- and they almost certainly will have to at some
> point -- the raw data isn't lost.
Hmm, I have an idea about instruction table. Usually, instruction tables
are encoded with code defined by each decoder/emulator. This method
will show their internal code directly, and is hard to maintain when
the opcode map is updated. Instead of that, I'd like to suggest using
the expressions in the opcode maps in a vender's genuine document (in
this case, Intel/AMD's manual) or http://www.sandpile.org for instruction
tables.
e.g.
const insn_attr_t onebyte_attr_table[ATTR_TABLE_SIZE] = {
/* 0x00-0x0f */
AT2(Eb,Gb), AT2(Ev,Gv), AT2(Gb,Eb), AT2(Gv,Ev),
AT2(AL,Ib), AT2(rAX,Iz), AT2(ES,i64), AT2(ES,i64),
AT2(Eb,Gb), AT2(Ev,Gv), AT2(Gb,Eb), AT2(Gv,Ev),
AT2(AL,Ib), AT2(rAX,Iz), AT2(CS,i64), AT(ESC),
...
Here, AT and AT2 macros are defined as follows:
#define AT(a) (INAT_OMEXP_##a)
#define AT2(a1, a2) (INAT_OMEXP_##a1 | INAT_OMEXP_##a2)
(OMEXP means Opcode Map Expression)
And each INAT_OMEXP_* is translated into internal format.
#define INAT_OMEXP_Eb INAT_ENCODE_RM(TYPE_MEMREG, SIZE_BYTE)
#define INAT_OMEXP_Gb INAT_ENCODE_REG(TYPE_MEMREG, SIZE_BYTE)
...
This idea will allow everyone to easily maintain instruction tables
by comparing instruction tables with vender's opcode map.
Designing internal instruction tables is harder. Currently, I'm
working on below layout.
Comments are welcome!
Instruction Attribute Encoding
==============================
Bitmap layout:
[ESC]
0 0 [(padding)][OPFLG][IMM][REG][RGT][RGS][RM][RMT][RMS]
0 1 [(padding)][PFXGRP][PFXEXT][Prefix code]
1 0 [(padding)][EID]
1 1 [(padding)][GID]
ESC(2): Switching normal/escape/group/prefix.
(0:normal opcode, 1:(Legacy)Prefix, 2:Escape, 3:Group)
- Normal opcodes
OPFLG(7): Flag bits: [REX][LPFX][I64][F64][NOPR][EREG][AIMM]
REX(1): Opcode is a REX prefix.
LPFX(1): Opcode can be modified by Last Prefix(SSE2-4)
I64(1): Opcode is invalid in 64bit mode.
F64(1): Oprand is 64bits width in 64bit mode.
NOPR(1): Opcode has no operand.
EREG(1) : Opcode byte encodes Registers
AIMM(1) : Opcode has another 1 byte Immediate(2nd Immediate).
IMM(3): Immediate size bits
(0:none, 1:byte, 2:word, 3:dword, 4:qword, 5:pointer,
6:word/dword, 7:word/dword/qword)
REG(1): Opcode has ModRM 'reg'
RGT(3): ModRM 'reg' type or special operand bits
(0:none,
REG=0: 1:DS/SI
REG=1: 1:GPR, 2:MMX, 3:XMM, 4:DBG, 5:CTR, 6:FP, 7:SR)
RGS(3): ModRM 'reg' or special operand size bits
(GPR: 1:byte, 2:word, 3:dword, 4:qword, 5:N/A, 6:word/dword,
7:word/dword/qword)
(MMX: 3:dword, 4:qword)
(XMM: 2:Scalar-SingleFP, 3:Scaler-DoubleFP, 4:qword, 5:d-qword,
6:Packed-SingleFP, 7:Packed-DoubleFP)
(FP: ?)
(Others: same as GPR)
RM(1) : Opcode has ModRM 'rm'
RMT(3) : ModRM 'rm' type or special operand bits
(0:none,
RM=0: 1:ES/DI
RM=1: 1:GPR, 2:MMX, 3:XMM, 4:Memory, 5:GPR/Mem, 6:MMX/MEM, 7:XMM/Mem)
RMS(3): ModRM 'rm' or special operand size bits. see RGS.
- Legacy prefixes
PFXGRP(4): Prefix group bits: [PGRP1][PGRP2][PGRP3][PGRP4]
PGRP1(1): opcode is prefix group1
PGRP2(1): opcode is prefix group2
PGRP3(1): opcode is prefix group3
PGRP4(1): opcode is prefix group4
PFXEXT(2): Mandatory prefix extent
(0:none, 1:66H, 2:F2H, 3:F3H)
Prefix code(11): Prefix code bits
- Escape opcode
EID(2): Escape code id.
(0:2byte escape, 1:FPU escape, 2:3byte escape1, 3:3byte escape2)
- Group opcode
GID(5): Group Number
(0:Group1, 1:Group1A, 2:Group2, ... 16:Group16)
Thanks,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division
e-mail: [email protected]
Masami Hiramatsu wrote:
>
> Hmm, I have an idea about instruction table. Usually, instruction tables
> are encoded with code defined by each decoder/emulator. This method
> will show their internal code directly, and is hard to maintain when
> the opcode map is updated. Instead of that, I'd like to suggest using
> the expressions in the opcode maps in a vender's genuine document (in
> this case, Intel/AMD's manual) or http://www.sandpile.org for instruction
> tables.
>
Yes, we discussed this at the Collab Summit. I think it's the only sane
thing.
> e.g.
>
> const insn_attr_t onebyte_attr_table[ATTR_TABLE_SIZE] = {
> /* 0x00-0x0f */
> AT2(Eb,Gb), AT2(Ev,Gv), AT2(Gb,Eb), AT2(Gv,Ev),
> AT2(AL,Ib), AT2(rAX,Iz), AT2(ES,i64), AT2(ES,i64),
> AT2(Eb,Gb), AT2(Ev,Gv), AT2(Gb,Eb), AT2(Gv,Ev),
> AT2(AL,Ib), AT2(rAX,Iz), AT2(CS,i64), AT(ESC),
> ...
>
> Here, AT and AT2 macros are defined as follows:
>
I would suggest using an actual parser, rather than relying on cpp for
this. The parser will be much more powerful, and will make it much
easier to change data structure radically as we discussed.
-hpa
On Thu, 2009-04-16 at 19:31 -0400, Masami Hiramatsu wrote:
...
>
> Hmm, I have an idea about instruction table. Usually, instruction tables
> are encoded with code defined by each decoder/emulator. This method
> will show their internal code directly, and is hard to maintain when
> the opcode map is updated. Instead of that, I'd like to suggest using
> the expressions in the opcode maps in a vender's genuine document (in
> this case, Intel/AMD's manual) or http://www.sandpile.org for instruction
> tables.
>
> e.g.
>
> const insn_attr_t onebyte_attr_table[ATTR_TABLE_SIZE] = {
> /* 0x00-0x0f */
> AT2(Eb,Gb), AT2(Ev,Gv), AT2(Gb,Eb), AT2(Gv,Ev),
> AT2(AL,Ib), AT2(rAX,Iz), AT2(ES,i64), AT2(ES,i64),
> AT2(Eb,Gb), AT2(Ev,Gv), AT2(Gb,Eb), AT2(Gv,Ev),
> AT2(AL,Ib), AT2(rAX,Iz), AT2(CS,i64), AT(ESC),
> ...
>
> Here, AT and AT2 macros are defined as follows:
>
> #define AT(a) (INAT_OMEXP_##a)
> #define AT2(a1, a2) (INAT_OMEXP_##a1 | INAT_OMEXP_##a2)
>
...
It looks like AT2(Ev,Gv) would yield the same bits as AT2(Gv,Ev). It'd
be nice not to lose the operand-order information. And we'd have to
make clear whether which notation we're using -- src,dest as in the gnu
assembler, or dest,src as in the AMD (and Intel?) manuals.
Jim
Jim Keniston wrote:
>
> It looks like AT2(Ev,Gv) would yield the same bits as AT2(Gv,Ev). It'd
> be nice not to lose the operand-order information. And we'd have to
> make clear whether which notation we're using -- src,dest as in the gnu
> assembler, or dest,src as in the AMD (and Intel?) manuals.
>
Since the information would come from the manuals, I would recommend
following them (dst first.)
-hpa
H. Peter Anvin wrote:
> Masami Hiramatsu wrote:
>>
>> Hmm, I have an idea about instruction table. Usually, instruction tables
>> are encoded with code defined by each decoder/emulator. This method
>> will show their internal code directly, and is hard to maintain when
>> the opcode map is updated. Instead of that, I'd like to suggest using
>> the expressions in the opcode maps in a vender's genuine document (in
>> this case, Intel/AMD's manual) or http://www.sandpile.org for instruction
>> tables.
>>
>
> Yes, we discussed this at the Collab Summit. I think it's the only sane
> thing.
>
>> e.g.
>>
>> const insn_attr_t onebyte_attr_table[ATTR_TABLE_SIZE] = {
>> /* 0x00-0x0f */
>> AT2(Eb,Gb), AT2(Ev,Gv), AT2(Gb,Eb), AT2(Gv,Ev),
>> AT2(AL,Ib), AT2(rAX,Iz), AT2(ES,i64), AT2(ES,i64),
>> AT2(Eb,Gb), AT2(Ev,Gv), AT2(Gb,Eb), AT2(Gv,Ev),
>> AT2(AL,Ib), AT2(rAX,Iz), AT2(CS,i64), AT(ESC),
>> ...
>>
>> Here, AT and AT2 macros are defined as follows:
>>
>
> I would suggest using an actual parser, rather than relying on cpp for
> this. The parser will be much more powerful, and will make it much
> easier to change data structure radically as we discussed.
Aah, I see. So we'd better make a parser which generates internal
data structure from genuine opcode map in compilation time.
And I changed my mind about internal data structure too.
In this version, I'll use a smallest bits which are needed
for the decoder.
Thank you,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division
e-mail: [email protected]
Masami Hiramatsu wrote:
>
> Aah, I see. So we'd better make a parser which generates internal
> data structure from genuine opcode map in compilation time.
>
> And I changed my mind about internal data structure too.
> In this version, I'll use a smallest bits which are needed
> for the decoder.
>
Yes, and with a proper compile-time parser, that kind of things can be
done, and be changed later if it is no longer appropriate.
-hpa
H. Peter Anvin wrote:
> Jim Keniston wrote:
>> It looks like AT2(Ev,Gv) would yield the same bits as AT2(Gv,Ev). It'd
>> be nice not to lose the operand-order information. And we'd have to
>> make clear whether which notation we're using -- src,dest as in the gnu
>> assembler, or dest,src as in the AMD (and Intel?) manuals.
>>
>
> Since the information would come from the manuals, I would recommend
> following them (dst first.)
>
Hi Peter and Jim,
Now what I'm doing is making opcode tables like this.
Table: 1-byte opcode
Alias: none
00: ADD Eb,Gb
01: ADD Ev,Gv
02: ADD Gb,Eb
03: ADD Gv,Ev
04: ADD AL,Ib
05: ADD rAX,Iz
06: PUSH ES (i64)
07: POP ES (i64)
08: OR Eb,Gb
09: OR Ev,Gv
0a: OR Gb,Eb
0b: OR Gv,Ev
0c: OR AL,Ib
0d: OR rAX,Iz
0e: PUSH CS
0f: 2-byte escape
...
and a parser script which parses them into,
const insn_attr_t primary_table[INAT_TABLE_SIZE] = {
[0x04] = INAT_IMM(IMM_SIZE_BYTE)
[0x05] = INAT_IMM(IMM_SIZE_VWORD32)
[0x0c] = INAT_IMM(IMM_SIZE_BYTE)
[0x0d] = INAT_IMM(IMM_SIZE_VWORD32)
[0x0f] = INAT_ESC(IMM_ESC_2BYTE)
...
(note, instructions which has no attributes for decoder, are just ignored)
By the way, I'm worried about legal things of Intel's instruction
encoding expressions. Would you think there is any problem if we
have those tables in the kernel tree?
Thanks,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division
e-mail: [email protected]
On Tue, 2009-04-21 at 20:17 -0400, Masami Hiramatsu wrote:
...
> Hi Peter and Jim,
>
> Now what I'm doing is making opcode tables like this.
>
> Table: 1-byte opcode
> Alias: none
> 00: ADD Eb,Gb
> 01: ADD Ev,Gv
> 02: ADD Gb,Eb
> 03: ADD Gv,Ev
> 04: ADD AL,Ib
> 05: ADD rAX,Iz
> 06: PUSH ES (i64)
> 07: POP ES (i64)
> 08: OR Eb,Gb
> 09: OR Ev,Gv
> 0a: OR Gb,Eb
> 0b: OR Gv,Ev
> 0c: OR AL,Ib
> 0d: OR rAX,Iz
> 0e: PUSH CS
> 0f: 2-byte escape
> ...
We want to keep this info easy to parse. (Who knows how it might be
used, and by whom?) Your format seems to be
opcode: mnemonic [comma,separated,operands] [(extra_info)]
which is fine if you stick to it... but your entry for 0f doesn't match
that.
Also, something like
+ extra_info
would be easier to parse (using, say, awk) than
(extra_info)
>
> and a parser script which parses them into,
>
> const insn_attr_t primary_table[INAT_TABLE_SIZE] = {
> [0x04] = INAT_IMM(IMM_SIZE_BYTE)
> [0x05] = INAT_IMM(IMM_SIZE_VWORD32)
> [0x0c] = INAT_IMM(IMM_SIZE_BYTE)
> [0x0d] = INAT_IMM(IMM_SIZE_VWORD32)
> [0x0f] = INAT_ESC(IMM_ESC_2BYTE)
> ...
>
> (note, instructions which has no attributes for decoder, are just ignored)
>
>
> By the way, I'm worried about legal things of Intel's instruction
> encoding expressions. Would you think there is any problem if we
> have those tables in the kernel tree?
Good question. Sorry, I'm not a lawyer. Intel and AMD and sandpile.org
all seem to be using the same notation, so the notation's inventor must
not be feeling too proprietary. Interestingly, sandpile.org asserts a
copyright.
>
> Thanks,
>
Jim
Jim Keniston wrote:
> On Tue, 2009-04-21 at 20:17 -0400, Masami Hiramatsu wrote:
> ...
>> Hi Peter and Jim,
>>
>> Now what I'm doing is making opcode tables like this.
>>
>> Table: 1-byte opcode
>> Alias: none
>> 00: ADD Eb,Gb
>> 01: ADD Ev,Gv
>> 02: ADD Gb,Eb
>> 03: ADD Gv,Ev
>> 04: ADD AL,Ib
>> 05: ADD rAX,Iz
>> 06: PUSH ES (i64)
>> 07: POP ES (i64)
>> 08: OR Eb,Gb
>> 09: OR Ev,Gv
>> 0a: OR Gb,Eb
>> 0b: OR Gv,Ev
>> 0c: OR AL,Ib
>> 0d: OR rAX,Iz
>> 0e: PUSH CS
>> 0f: 2-byte escape
>> ...
>
> We want to keep this info easy to parse. (Who knows how it might be
> used, and by whom?) Your format seems to be
> opcode: mnemonic [comma,separated,operands] [(extra_info)]
> which is fine if you stick to it... but your entry for 0f doesn't match
> that.
Sure, it was just based on the original opcode map.
Maybe, we can have something special expressions for that.
e.g.
0f: ESC # 2-byte escape
> Also, something like
> + extra_info
> would be easier to parse (using, say, awk) than
> (extra_info)
Hmm, maybe, parser can handle "(extra_info)" as a solid keyword.
so let's define actual format.
<opcode maps>
Table: table-name
Referrer: escamed-name
opcode: mnemonic|Grp [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
opcode: ESC # escaped-name
<group maps>
reg: mnemonic ...
Thank you,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division
e-mail: [email protected]
On Thu, 2009-04-23 at 13:29 -0400, Masami Hiramatsu wrote:
...
>
> Hmm, maybe, parser can handle "(extra_info)" as a solid keyword.
> so let's define actual format.
>
> <opcode maps>
> Table: table-name
> Referrer: escamed-name
> opcode: mnemonic|Grp [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
> opcode: ESC # escaped-name
>
> <group maps>
> reg: mnemonic ...
For some instruction groups -- e.g., Groups 12, 13, 14 -- the
instruction prefix (66, f2, f3) and the reg field both affect the
instruction type. And for some x87 instructions, the value of the modrm
byte's rm field also affects the instruction type. (For others, rm just
selects among the st(0)..st(7) registers, as one might expect.)
Of course, that's all about floating-point instructions, which are of
more interest to uprobes than kprobes.
Jim
Jim Keniston wrote:
> On Thu, 2009-04-23 at 13:29 -0400, Masami Hiramatsu wrote:
> ...
>> Hmm, maybe, parser can handle "(extra_info)" as a solid keyword.
>> so let's define actual format.
>>
>> <opcode maps>
>> Table: table-name
>> Referrer: escamed-name
>> opcode: mnemonic|Grp [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
This should be:
opcode: mnemonic|GrpXXX ...
>> opcode: ESC # escaped-name
This should be
opcode: escape # escaped-name
because distinguishing from x87 ESC ops.
>>
>> <group maps>
>> reg: mnemonic ...
>
> For some instruction groups -- e.g., Groups 12, 13, 14 -- the
> instruction prefix (66, f2, f3) and the reg field both affect the
> instruction type. And for some x87 instructions, the value of the modrm
> byte's rm field also affects the instruction type. (For others, rm just
> selects among the st(0)..st(7) registers, as one might expect.)
Sure, I updated the format. There are some special cases,
(1) instructions which are switched by 64bit mode
40: INC eAX (i64) | REX (o64)
(2) instructions which are switched by the last prefix
13: movlps Mq,Vq | movlpd Mq,Vq (66)
(3) group instructions which are switched by modr/m
0: SGDT Ms | VMCALL (11B),(001) | VMLAUNCH (11B),(010) | VMRESUME (11B),(011) | VMXOFF (11B),(100)
>
> Of course, that's all about floating-point instructions, which are of
> more interest to uprobes than kprobes.
Hmm, x87 instructions may need to have some special format...
Thank you,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division
e-mail: [email protected]