From 1d509f2a6ebca1aea3089c769f6375f01a832e9b Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 13 Nov 2020 00:03:23 +0100 Subject: x86/insn: Support big endian cross-compiles The x86 instruction decoder code is shared across the kernel source and the tools. Currently objtool seems to be the only tool from build tools needed which breaks x86 cross-compilation on big endian systems. Make the x86 instruction decoder build host endianness agnostic to support x86 cross-compilation and enable objtool to implement endianness awareness for big endian architectures support. Signed-off-by: Martin Schwidefsky Co-developed-by: Vasily Gorbik Signed-off-by: Vasily Gorbik Acked-by: Peter Zijlstra (Intel) Acked-by: Masami Hiramatsu Signed-off-by: Josh Poimboeuf --- tools/arch/x86/include/asm/insn.h | 33 +++++++++++++ tools/arch/x86/lib/insn.c | 101 ++++++++++++++++++-------------------- 2 files changed, 80 insertions(+), 54 deletions(-) (limited to 'tools/arch') diff --git a/tools/arch/x86/include/asm/insn.h b/tools/arch/x86/include/asm/insn.h index 52c6262e6bfd..c1fab7a570be 100644 --- a/tools/arch/x86/include/asm/insn.h +++ b/tools/arch/x86/include/asm/insn.h @@ -7,9 +7,12 @@ * Copyright (C) IBM Corporation, 2009 */ +#include /* insn_attr_t is defined in inat.h */ #include "inat.h" +#if defined(__BYTE_ORDER) ? __BYTE_ORDER == __LITTLE_ENDIAN : defined(__LITTLE_ENDIAN) + struct insn_field { union { insn_value_t value; @@ -20,6 +23,36 @@ struct insn_field { unsigned char nbytes; }; +static inline void insn_field_set(struct insn_field *p, insn_value_t v, + unsigned char n) +{ + p->value = v; + p->nbytes = n; +} + +#else + +struct insn_field { + insn_value_t value; + union { + insn_value_t little; + insn_byte_t bytes[4]; + }; + /* !0 if we've run insn_get_xxx() for this field */ + unsigned char got; + unsigned char nbytes; +}; + +static inline void insn_field_set(struct insn_field *p, insn_value_t v, + unsigned char n) +{ + p->value = v; + p->little = __cpu_to_le32(v); + p->nbytes = n; +} + +#endif + struct insn { struct insn_field prefixes; /* * Prefixes diff --git a/tools/arch/x86/lib/insn.c b/tools/arch/x86/lib/insn.c index 0151dfc6da61..77e92aa52cdc 100644 --- a/tools/arch/x86/lib/insn.c +++ b/tools/arch/x86/lib/insn.c @@ -5,6 +5,7 @@ * Copyright (C) IBM Corporation, 2002, 2004, 2009 */ +#include #ifdef __KERNEL__ #include #else @@ -15,15 +16,28 @@ #include "../include/asm/emulate_prefix.h" +#define leXX_to_cpu(t, r) \ +({ \ + __typeof__(t) v; \ + switch (sizeof(t)) { \ + case 4: v = le32_to_cpu(r); break; \ + case 2: v = le16_to_cpu(r); break; \ + case 1: v = r; break; \ + default: \ + BUILD_BUG(); break; \ + } \ + v; \ +}) + /* Verify next sizeof(t) bytes can be on the same instruction */ #define validate_next(t, insn, n) \ ((insn)->next_byte + sizeof(t) + n <= (insn)->end_kaddr) #define __get_next(t, insn) \ - ({ t r = *(t*)insn->next_byte; insn->next_byte += sizeof(t); r; }) + ({ t r = *(t*)insn->next_byte; insn->next_byte += sizeof(t); leXX_to_cpu(t, r); }) #define __peek_nbyte_next(t, insn, n) \ - ({ t r = *(t*)((insn)->next_byte + n); r; }) + ({ t r = *(t*)((insn)->next_byte + n); leXX_to_cpu(t, r); }) #define get_next(t, insn) \ ({ if (unlikely(!validate_next(t, insn, 0))) goto err_out; __get_next(t, insn); }) @@ -157,8 +171,7 @@ found: b = peek_next(insn_byte_t, insn); attr = inat_get_opcode_attribute(b); if (inat_is_rex_prefix(attr)) { - insn->rex_prefix.value = b; - insn->rex_prefix.nbytes = 1; + insn_field_set(&insn->rex_prefix, b, 1); insn->next_byte++; if (X86_REX_W(b)) /* REX.W overrides opnd_size */ @@ -295,8 +308,7 @@ void insn_get_modrm(struct insn *insn) if (inat_has_modrm(insn->attr)) { mod = get_next(insn_byte_t, insn); - modrm->value = mod; - modrm->nbytes = 1; + insn_field_set(modrm, mod, 1); if (inat_is_group(insn->attr)) { pfx_id = insn_last_prefix_id(insn); insn->attr = inat_get_group_attribute(mod, pfx_id, @@ -334,7 +346,7 @@ int insn_rip_relative(struct insn *insn) * For rip-relative instructions, the mod field (top 2 bits) * is zero and the r/m field (bottom 3 bits) is 0x5. */ - return (modrm->nbytes && (modrm->value & 0xc7) == 0x5); + return (modrm->nbytes && (modrm->bytes[0] & 0xc7) == 0x5); } /** @@ -353,11 +365,11 @@ void insn_get_sib(struct insn *insn) if (!insn->modrm.got) insn_get_modrm(insn); if (insn->modrm.nbytes) { - modrm = (insn_byte_t)insn->modrm.value; + modrm = insn->modrm.bytes[0]; if (insn->addr_bytes != 2 && X86_MODRM_MOD(modrm) != 3 && X86_MODRM_RM(modrm) == 4) { - insn->sib.value = get_next(insn_byte_t, insn); - insn->sib.nbytes = 1; + insn_field_set(&insn->sib, + get_next(insn_byte_t, insn), 1); } } insn->sib.got = 1; @@ -407,19 +419,18 @@ void insn_get_displacement(struct insn *insn) if (mod == 3) goto out; if (mod == 1) { - insn->displacement.value = get_next(signed char, insn); - insn->displacement.nbytes = 1; + insn_field_set(&insn->displacement, + get_next(signed char, insn), 1); } else if (insn->addr_bytes == 2) { if ((mod == 0 && rm == 6) || mod == 2) { - insn->displacement.value = - get_next(short, insn); - insn->displacement.nbytes = 2; + insn_field_set(&insn->displacement, + get_next(short, insn), 2); } } else { if ((mod == 0 && rm == 5) || mod == 2 || (mod == 0 && base == 5)) { - insn->displacement.value = get_next(int, insn); - insn->displacement.nbytes = 4; + insn_field_set(&insn->displacement, + get_next(int, insn), 4); } } } @@ -435,18 +446,14 @@ static int __get_moffset(struct insn *insn) { switch (insn->addr_bytes) { case 2: - insn->moffset1.value = get_next(short, insn); - insn->moffset1.nbytes = 2; + insn_field_set(&insn->moffset1, get_next(short, insn), 2); break; case 4: - insn->moffset1.value = get_next(int, insn); - insn->moffset1.nbytes = 4; + insn_field_set(&insn->moffset1, get_next(int, insn), 4); break; case 8: - insn->moffset1.value = get_next(int, insn); - insn->moffset1.nbytes = 4; - insn->moffset2.value = get_next(int, insn); - insn->moffset2.nbytes = 4; + insn_field_set(&insn->moffset1, get_next(int, insn), 4); + insn_field_set(&insn->moffset2, get_next(int, insn), 4); break; default: /* opnd_bytes must be modified manually */ goto err_out; @@ -464,13 +471,11 @@ static int __get_immv32(struct insn *insn) { switch (insn->opnd_bytes) { case 2: - insn->immediate.value = get_next(short, insn); - insn->immediate.nbytes = 2; + insn_field_set(&insn->immediate, get_next(short, insn), 2); break; case 4: case 8: - insn->immediate.value = get_next(int, insn); - insn->immediate.nbytes = 4; + insn_field_set(&insn->immediate, get_next(int, insn), 4); break; default: /* opnd_bytes must be modified manually */ goto err_out; @@ -487,18 +492,15 @@ static int __get_immv(struct insn *insn) { switch (insn->opnd_bytes) { case 2: - insn->immediate1.value = get_next(short, insn); - insn->immediate1.nbytes = 2; + insn_field_set(&insn->immediate1, get_next(short, insn), 2); break; case 4: - insn->immediate1.value = get_next(int, insn); + insn_field_set(&insn->immediate1, get_next(int, insn), 4); insn->immediate1.nbytes = 4; break; case 8: - insn->immediate1.value = get_next(int, insn); - insn->immediate1.nbytes = 4; - insn->immediate2.value = get_next(int, insn); - insn->immediate2.nbytes = 4; + insn_field_set(&insn->immediate1, get_next(int, insn), 4); + insn_field_set(&insn->immediate2, get_next(int, insn), 4); break; default: /* opnd_bytes must be modified manually */ goto err_out; @@ -515,12 +517,10 @@ static int __get_immptr(struct insn *insn) { switch (insn->opnd_bytes) { case 2: - insn->immediate1.value = get_next(short, insn); - insn->immediate1.nbytes = 2; + insn_field_set(&insn->immediate1, get_next(short, insn), 2); break; case 4: - insn->immediate1.value = get_next(int, insn); - insn->immediate1.nbytes = 4; + insn_field_set(&insn->immediate1, get_next(int, insn), 4); break; case 8: /* ptr16:64 is not exist (no segment) */ @@ -528,8 +528,7 @@ static int __get_immptr(struct insn *insn) default: /* opnd_bytes must be modified manually */ goto err_out; } - insn->immediate2.value = get_next(unsigned short, insn); - insn->immediate2.nbytes = 2; + insn_field_set(&insn->immediate2, get_next(unsigned short, insn), 2); insn->immediate1.got = insn->immediate2.got = 1; return 1; @@ -565,22 +564,17 @@ void insn_get_immediate(struct insn *insn) switch (inat_immediate_size(insn->attr)) { case INAT_IMM_BYTE: - insn->immediate.value = get_next(signed char, insn); - insn->immediate.nbytes = 1; + insn_field_set(&insn->immediate, get_next(signed char, insn), 1); break; case INAT_IMM_WORD: - insn->immediate.value = get_next(short, insn); - insn->immediate.nbytes = 2; + insn_field_set(&insn->immediate, get_next(short, insn), 2); break; case INAT_IMM_DWORD: - insn->immediate.value = get_next(int, insn); - insn->immediate.nbytes = 4; + insn_field_set(&insn->immediate, get_next(int, insn), 4); break; case INAT_IMM_QWORD: - insn->immediate1.value = get_next(int, insn); - insn->immediate1.nbytes = 4; - insn->immediate2.value = get_next(int, insn); - insn->immediate2.nbytes = 4; + insn_field_set(&insn->immediate1, get_next(int, insn), 4); + insn_field_set(&insn->immediate2, get_next(int, insn), 4); break; case INAT_IMM_PTR: if (!__get_immptr(insn)) @@ -599,8 +593,7 @@ void insn_get_immediate(struct insn *insn) goto err_out; } if (inat_has_second_immediate(insn->attr)) { - insn->immediate2.value = get_next(signed char, insn); - insn->immediate2.nbytes = 1; + insn_field_set(&insn->immediate2, get_next(signed char, insn), 1); } done: insn->immediate.got = 1; -- cgit From 8bfe273238d77d3cee18e4c03b2f26ae360b5661 Mon Sep 17 00:00:00 2001 From: Vasily Gorbik Date: Fri, 13 Nov 2020 00:03:29 +0100 Subject: objtool: Fix x86 orc generation on big endian cross-compiles Correct objtool orc generation endianness problems to enable fully functional x86 cross-compiles on big endian hardware. Introduce bswap_if_needed() macro, which does a byte swap if target endianness doesn't match the host, i.e. cross-compilation for little endian on big endian and vice versa. The macro is used for conversion of multi-byte values which are read from / about to be written to a target native endianness ELF file. Signed-off-by: Vasily Gorbik Acked-by: Peter Zijlstra (Intel) Acked-by: Masami Hiramatsu Signed-off-by: Josh Poimboeuf --- tools/arch/x86/include/asm/orc_types.h | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'tools/arch') diff --git a/tools/arch/x86/include/asm/orc_types.h b/tools/arch/x86/include/asm/orc_types.h index fdbffec4cfde..5a2baf28a1dc 100644 --- a/tools/arch/x86/include/asm/orc_types.h +++ b/tools/arch/x86/include/asm/orc_types.h @@ -40,6 +40,8 @@ #define ORC_REG_MAX 15 #ifndef __ASSEMBLY__ +#include + /* * This struct is more or less a vastly simplified version of the DWARF Call * Frame Information standard. It contains only the necessary parts of DWARF @@ -51,10 +53,18 @@ struct orc_entry { s16 sp_offset; s16 bp_offset; +#if defined(__LITTLE_ENDIAN_BITFIELD) unsigned sp_reg:4; unsigned bp_reg:4; unsigned type:2; unsigned end:1; +#elif defined(__BIG_ENDIAN_BITFIELD) + unsigned bp_reg:4; + unsigned sp_reg:4; + unsigned unused:5; + unsigned end:1; + unsigned type:2; +#endif } __packed; #endif /* __ASSEMBLY__ */ -- cgit From 5ed934e57e712b676ca62e1904ad672a9fa1505a Mon Sep 17 00:00:00 2001 From: Vasily Gorbik Date: Fri, 13 Nov 2020 17:09:54 +0100 Subject: x86/insn: Fix vector instruction decoding on big endian cross-compiles Running instruction decoder posttest on an s390 host with an x86 target with allyesconfig shows errors. Instructions used in a couple of kernel objects could not be correctly decoded on big endian system. insn_decoder_test: warning: objdump says 6 bytes, but insn_get_length() says 5 insn_decoder_test: warning: Found an x86 instruction decoder bug, please report this. insn_decoder_test: warning: ffffffff831eb4e1: 62 d1 fd 48 7f 04 24 vmovdqa64 %zmm0,(%r12) insn_decoder_test: warning: objdump says 7 bytes, but insn_get_length() says 6 insn_decoder_test: warning: Found an x86 instruction decoder bug, please report this. insn_decoder_test: warning: ffffffff831eb4e8: 62 51 fd 48 7f 44 24 01 vmovdqa64 %zmm8,0x40(%r12) insn_decoder_test: warning: objdump says 8 bytes, but insn_get_length() says 6 This is because in a few places instruction field bytes are set directly with further usage of "value". To address that introduce and use a insn_set_byte() helper, which correctly updates "value" on big endian systems. Signed-off-by: Vasily Gorbik Acked-by: Masami Hiramatsu Signed-off-by: Josh Poimboeuf --- tools/arch/x86/include/asm/insn.h | 12 ++++++++++++ tools/arch/x86/lib/insn.c | 18 +++++++++--------- 2 files changed, 21 insertions(+), 9 deletions(-) (limited to 'tools/arch') diff --git a/tools/arch/x86/include/asm/insn.h b/tools/arch/x86/include/asm/insn.h index c1fab7a570be..cc777c185212 100644 --- a/tools/arch/x86/include/asm/insn.h +++ b/tools/arch/x86/include/asm/insn.h @@ -30,6 +30,12 @@ static inline void insn_field_set(struct insn_field *p, insn_value_t v, p->nbytes = n; } +static inline void insn_set_byte(struct insn_field *p, unsigned char n, + insn_byte_t v) +{ + p->bytes[n] = v; +} + #else struct insn_field { @@ -51,6 +57,12 @@ static inline void insn_field_set(struct insn_field *p, insn_value_t v, p->nbytes = n; } +static inline void insn_set_byte(struct insn_field *p, unsigned char n, + insn_byte_t v) +{ + p->bytes[n] = v; + p->value = __le32_to_cpu(p->little); +} #endif struct insn { diff --git a/tools/arch/x86/lib/insn.c b/tools/arch/x86/lib/insn.c index 77e92aa52cdc..3d9355ed1246 100644 --- a/tools/arch/x86/lib/insn.c +++ b/tools/arch/x86/lib/insn.c @@ -161,9 +161,9 @@ found: b = insn->prefixes.bytes[3]; for (i = 0; i < nb; i++) if (prefixes->bytes[i] == lb) - prefixes->bytes[i] = b; + insn_set_byte(prefixes, i, b); } - insn->prefixes.bytes[3] = lb; + insn_set_byte(&insn->prefixes, 3, lb); } /* Decode REX prefix */ @@ -194,13 +194,13 @@ found: if (X86_MODRM_MOD(b2) != 3) goto vex_end; } - insn->vex_prefix.bytes[0] = b; - insn->vex_prefix.bytes[1] = b2; + insn_set_byte(&insn->vex_prefix, 0, b); + insn_set_byte(&insn->vex_prefix, 1, b2); if (inat_is_evex_prefix(attr)) { b2 = peek_nbyte_next(insn_byte_t, insn, 2); - insn->vex_prefix.bytes[2] = b2; + insn_set_byte(&insn->vex_prefix, 2, b2); b2 = peek_nbyte_next(insn_byte_t, insn, 3); - insn->vex_prefix.bytes[3] = b2; + insn_set_byte(&insn->vex_prefix, 3, b2); insn->vex_prefix.nbytes = 4; insn->next_byte += 4; if (insn->x86_64 && X86_VEX_W(b2)) @@ -208,7 +208,7 @@ found: insn->opnd_bytes = 8; } else if (inat_is_vex3_prefix(attr)) { b2 = peek_nbyte_next(insn_byte_t, insn, 2); - insn->vex_prefix.bytes[2] = b2; + insn_set_byte(&insn->vex_prefix, 2, b2); insn->vex_prefix.nbytes = 3; insn->next_byte += 3; if (insn->x86_64 && X86_VEX_W(b2)) @@ -220,7 +220,7 @@ found: * Makes it easier to decode vex.W, vex.vvvv, * vex.L and vex.pp. Masking with 0x7f sets vex.W == 0. */ - insn->vex_prefix.bytes[2] = b2 & 0x7f; + insn_set_byte(&insn->vex_prefix, 2, b2 & 0x7f); insn->vex_prefix.nbytes = 2; insn->next_byte += 2; } @@ -256,7 +256,7 @@ void insn_get_opcode(struct insn *insn) /* Get first opcode */ op = get_next(insn_byte_t, insn); - opcode->bytes[0] = op; + insn_set_byte(opcode, 0, op); opcode->nbytes = 1; /* Check if there is VEX prefix or not */ -- cgit