Merge tag 'objtool-core-2023-03-02' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull objtool updates from Ingo Molnar:

 - Shrink 'struct instruction', to improve objtool performance & memory
   footprint

 - Other maximum memory usage reductions - this makes the build both
   faster, and fixes kernel build OOM failures on allyesconfig and
   similar configs when they try to build the final (large) vmlinux.o

 - Fix ORC unwinding when a kprobe (INT3) is set on a stack-modifying
   single-byte instruction (PUSH/POP or LEAVE). This requires the
   extension of the ORC metadata structure with a 'signal' field

 - Misc fixes & cleanups

* tag 'objtool-core-2023-03-02' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (22 commits)
  objtool: Fix ORC 'signal' propagation
  objtool: Remove instruction::list
  x86: Fix FILL_RETURN_BUFFER
  objtool: Fix overlapping alternatives
  objtool: Union instruction::{call_dest,jump_table}
  objtool: Remove instruction::reloc
  objtool: Shrink instruction::{type,visited}
  objtool: Make instruction::alts a single-linked list
  objtool: Make instruction::stack_ops a single-linked list
  objtool: Change arch_decode_instruction() signature
  x86/entry: Fix unwinding from kprobe on PUSH/POP instruction
  x86/unwind/orc: Add 'signal' field to ORC metadata
  objtool: Optimize layout of struct special_alt
  objtool: Optimize layout of struct symbol
  objtool: Allocate multiple structures with calloc()
  objtool: Make struct check_options static
  objtool: Make struct entries[] static and const
  objtool: Fix HOSTCC flag usage
  objtool: Properly support make V=1
  objtool: Install libsubcmd in build
  ...

arch/x86/entry/entry_64.S

@@ -385,7 +385,14 @@ SYM_CODE_END(xen_error_entry)
  */
 .macro idtentry vector asmsym cfunc has_error_code:req
 SYM_CODE_START(\asmsym)
-	UNWIND_HINT_IRET_REGS offset=\has_error_code*8
+
+	.if \vector == X86_TRAP_BP
+		/* #BP advances %rip to the next instruction */
+		UNWIND_HINT_IRET_REGS offset=\has_error_code*8 signal=0
+	.else
+		UNWIND_HINT_IRET_REGS offset=\has_error_code*8
+	.endif
+
 	ENDBR
 	ASM_CLAC
 	cld

arch/x86/include/asm/nospec-branch.h

@@ -261,7 +261,7 @@
 .macro FILL_RETURN_BUFFER reg:req nr:req ftr:req ftr2=ALT_NOT(X86_FEATURE_ALWAYS)
 	ALTERNATIVE_2 "jmp .Lskip_rsb_\@", \
 		__stringify(__FILL_RETURN_BUFFER(\reg,\nr)), \ftr, \
-		__stringify(__FILL_ONE_RETURN), \ftr2
+		__stringify(nop;nop;__FILL_ONE_RETURN), \ftr2
 
 .Lskip_rsb_\@:
 .endm

arch/x86/include/asm/orc_types.h

@@ -57,12 +57,14 @@ struct orc_entry {
 	unsigned	sp_reg:4;
 	unsigned	bp_reg:4;
 	unsigned	type:2;
+	unsigned	signal:1;
 	unsigned	end:1;
 #elif defined(__BIG_ENDIAN_BITFIELD)
 	unsigned	bp_reg:4;
 	unsigned	sp_reg:4;
-	unsigned	unused:5;
+	unsigned	unused:4;
 	unsigned	end:1;
+	unsigned	signal:1;
 	unsigned	type:2;
 #endif
 } __packed;

arch/x86/include/asm/unwind_hints.h

@@ -15,7 +15,7 @@
 	UNWIND_HINT type=UNWIND_HINT_TYPE_ENTRY end=1
 .endm
 
-.macro UNWIND_HINT_REGS base=%rsp offset=0 indirect=0 extra=1 partial=0
+.macro UNWIND_HINT_REGS base=%rsp offset=0 indirect=0 extra=1 partial=0 signal=1
 	.if \base == %rsp
 		.if \indirect
 			.set sp_reg, ORC_REG_SP_INDIRECT
@@ -45,11 +45,11 @@
 		.set type, UNWIND_HINT_TYPE_REGS
 	.endif
 
-	UNWIND_HINT sp_reg=sp_reg sp_offset=sp_offset type=type
+	UNWIND_HINT sp_reg=sp_reg sp_offset=sp_offset type=type signal=\signal
 .endm
 
-.macro UNWIND_HINT_IRET_REGS base=%rsp offset=0
-	UNWIND_HINT_REGS base=\base offset=\offset partial=1
+.macro UNWIND_HINT_IRET_REGS base=%rsp offset=0 signal=1
+	UNWIND_HINT_REGS base=\base offset=\offset partial=1 signal=\signal
 .endm
 
 .macro UNWIND_HINT_FUNC
@@ -67,7 +67,7 @@
 #else
 
 #define UNWIND_HINT_FUNC \
-	UNWIND_HINT(ORC_REG_SP, 8, UNWIND_HINT_TYPE_FUNC, 0)
+	UNWIND_HINT(ORC_REG_SP, 8, UNWIND_HINT_TYPE_FUNC, 0, 0)
 
 #endif /* __ASSEMBLY__ */
 

arch/x86/kernel/unwind_orc.c

@@ -484,6 +484,8 @@ bool unwind_next_frame(struct unwind_state *state)
 		goto the_end;
 	}
 
+	state->signal = orc->signal;
+
 	/* Find the previous frame's stack: */
 	switch (orc->sp_reg) {
 	case ORC_REG_SP:
@@ -563,7 +565,6 @@ bool unwind_next_frame(struct unwind_state *state)
 		state->sp = sp;
 		state->regs = NULL;
 		state->prev_regs = NULL;
-		state->signal = false;
 		break;
 
 	case UNWIND_HINT_TYPE_REGS:
@@ -587,7 +588,6 @@ bool unwind_next_frame(struct unwind_state *state)
 		state->regs = (struct pt_regs *)sp;
 		state->prev_regs = NULL;
 		state->full_regs = true;
-		state->signal = true;
 		break;
 
 	case UNWIND_HINT_TYPE_REGS_PARTIAL:
@@ -604,7 +604,6 @@ bool unwind_next_frame(struct unwind_state *state)
 			state->prev_regs = state->regs;
 		state->regs = (void *)sp - IRET_FRAME_OFFSET;
 		state->full_regs = false;
-		state->signal = true;
 		break;
 
 	default:

include/linux/objtool.h

@@ -15,6 +15,7 @@ struct unwind_hint {
 	s16		sp_offset;
 	u8		sp_reg;
 	u8		type;
+	u8		signal;
 	u8		end;
 };
 #endif
@@ -49,7 +50,7 @@ struct unwind_hint {
 
 #ifndef __ASSEMBLY__
 
-#define UNWIND_HINT(sp_reg, sp_offset, type, end)		\
+#define UNWIND_HINT(sp_reg, sp_offset, type, signal, end)	\
 	"987: \n\t"						\
 	".pushsection .discard.unwind_hints\n\t"		\
 	/* struct unwind_hint */				\
@@ -57,6 +58,7 @@ struct unwind_hint {
 	".short " __stringify(sp_offset) "\n\t"			\
 	".byte " __stringify(sp_reg) "\n\t"			\
 	".byte " __stringify(type) "\n\t"			\
+	".byte " __stringify(signal) "\n\t"			\
 	".byte " __stringify(end) "\n\t"			\
 	".balign 4 \n\t"					\
 	".popsection\n\t"
@@ -129,7 +131,7 @@ struct unwind_hint {
  * the debuginfo as necessary.  It will also warn if it sees any
  * inconsistencies.
  */
-.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 end=0
+.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 signal=0 end=0
 .Lunwind_hint_ip_\@:
 	.pushsection .discard.unwind_hints
 		/* struct unwind_hint */
@@ -137,6 +139,7 @@ struct unwind_hint {
 		.short \sp_offset
 		.byte \sp_reg
 		.byte \type
+		.byte \signal
 		.byte \end
 		.balign 4
 	.popsection
@@ -174,7 +177,7 @@ struct unwind_hint {
 
 #ifndef __ASSEMBLY__
 
-#define UNWIND_HINT(sp_reg, sp_offset, type, end)	\
+#define UNWIND_HINT(sp_reg, sp_offset, type, signal, end) \
 	"\n\t"
 #define STACK_FRAME_NON_STANDARD(func)
 #define STACK_FRAME_NON_STANDARD_FP(func)
@@ -182,7 +185,7 @@ struct unwind_hint {
 #define ASM_REACHABLE
 #else
 #define ANNOTATE_INTRA_FUNCTION_CALL
-.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 end=0
+.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 signal=0 end=0
 .endm
 .macro STACK_FRAME_NON_STANDARD func:req
 .endm

tools/arch/x86/include/asm/orc_types.h

@@ -57,12 +57,14 @@ struct orc_entry {
 	unsigned	sp_reg:4;
 	unsigned	bp_reg:4;
 	unsigned	type:2;
+	unsigned	signal:1;
 	unsigned	end:1;
 #elif defined(__BIG_ENDIAN_BITFIELD)
 	unsigned	bp_reg:4;
 	unsigned	sp_reg:4;
-	unsigned	unused:5;
+	unsigned	unused:4;
 	unsigned	end:1;
+	unsigned	signal:1;
 	unsigned	type:2;
 #endif
 } __packed;

tools/include/linux/objtool.h

@@ -15,6 +15,7 @@ struct unwind_hint {
 	s16		sp_offset;
 	u8		sp_reg;
 	u8		type;
+	u8		signal;
 	u8		end;
 };
 #endif
@@ -49,7 +50,7 @@ struct unwind_hint {
 
 #ifndef __ASSEMBLY__
 
-#define UNWIND_HINT(sp_reg, sp_offset, type, end)		\
+#define UNWIND_HINT(sp_reg, sp_offset, type, signal, end)	\
 	"987: \n\t"						\
 	".pushsection .discard.unwind_hints\n\t"		\
 	/* struct unwind_hint */				\
@@ -57,6 +58,7 @@ struct unwind_hint {
 	".short " __stringify(sp_offset) "\n\t"			\
 	".byte " __stringify(sp_reg) "\n\t"			\
 	".byte " __stringify(type) "\n\t"			\
+	".byte " __stringify(signal) "\n\t"			\
 	".byte " __stringify(end) "\n\t"			\
 	".balign 4 \n\t"					\
 	".popsection\n\t"
@@ -129,7 +131,7 @@ struct unwind_hint {
  * the debuginfo as necessary.  It will also warn if it sees any
  * inconsistencies.
  */
-.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 end=0
+.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 signal=0 end=0
 .Lunwind_hint_ip_\@:
 	.pushsection .discard.unwind_hints
 		/* struct unwind_hint */
@@ -137,6 +139,7 @@ struct unwind_hint {
 		.short \sp_offset
 		.byte \sp_reg
 		.byte \type
+		.byte \signal
 		.byte \end
 		.balign 4
 	.popsection
@@ -174,7 +177,7 @@ struct unwind_hint {
 
 #ifndef __ASSEMBLY__
 
-#define UNWIND_HINT(sp_reg, sp_offset, type, end)	\
+#define UNWIND_HINT(sp_reg, sp_offset, type, signal, end) \
 	"\n\t"
 #define STACK_FRAME_NON_STANDARD(func)
 #define STACK_FRAME_NON_STANDARD_FP(func)
@@ -182,7 +185,7 @@ struct unwind_hint {
 #define ASM_REACHABLE
 #else
 #define ANNOTATE_INTRA_FUNCTION_CALL
-.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 end=0
+.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 signal=0 end=0
 .endm
 .macro STACK_FRAME_NON_STANDARD func:req
 .endm

tools/objtool/.gitignore

@@ -2,3 +2,4 @@
 arch/x86/lib/inat-tables.c
 /objtool
 fixdep
+libsubcmd/

tools/objtool/Build

@@ -16,8 +16,6 @@ objtool-y += libctype.o
 objtool-y += str_error_r.o
 objtool-y += librbtree.o
 
-CFLAGS += -I$(srctree)/tools/lib
-
 $(OUTPUT)libstring.o: ../lib/string.c FORCE
 	$(call rule_mkdir)
 	$(call if_changed_dep,cc_o_c)

tools/objtool/Documentation/objtool.txt

@@ -410,6 +410,14 @@ the objtool maintainers.
    can remove this warning by putting the ANNOTATE_INTRA_FUNCTION_CALL
    directive right before the call.
 
+12. file.o: warning: func(): not an indirect call target
+
+   This means that objtool is running with --ibt and a function expected
+   to be an indirect call target is not. In particular, this happens for
+   init_module() or cleanup_module() if a module relies on these special
+   names and does not use module_init() / module_exit() macros to create
+   them.
+
 
 If the error doesn't seem to make sense, it could be a bug in objtool.
 Feel free to ask the objtool maintainer for help.

tools/objtool/Makefile

@@ -2,19 +2,18 @@
 include ../scripts/Makefile.include
 include ../scripts/Makefile.arch
 
-# always use the host compiler
-AR	 = $(HOSTAR)
-CC	 = $(HOSTCC)
-LD	 = $(HOSTLD)
-
 ifeq ($(srctree),)
 srctree := $(patsubst %/,%,$(dir $(CURDIR)))
 srctree := $(patsubst %/,%,$(dir $(srctree)))
 endif
 
-SUBCMD_SRCDIR		= $(srctree)/tools/lib/subcmd/
-LIBSUBCMD_OUTPUT	= $(or $(OUTPUT),$(CURDIR)/)
-LIBSUBCMD		= $(LIBSUBCMD_OUTPUT)libsubcmd.a
+LIBSUBCMD_DIR = $(srctree)/tools/lib/subcmd/
+ifneq ($(OUTPUT),)
+  LIBSUBCMD_OUTPUT = $(abspath $(OUTPUT))/libsubcmd
+else
+  LIBSUBCMD_OUTPUT = $(CURDIR)/libsubcmd
+endif
+LIBSUBCMD = $(LIBSUBCMD_OUTPUT)/libsubcmd.a
 
 OBJTOOL    := $(OUTPUT)objtool
 OBJTOOL_IN := $(OBJTOOL)-in.o
@@ -28,16 +27,29 @@ INCLUDES := -I$(srctree)/tools/include \
 	    -I$(srctree)/tools/arch/$(HOSTARCH)/include/uapi \
 	    -I$(srctree)/tools/arch/$(SRCARCH)/include	\
 	    -I$(srctree)/tools/objtool/include \
-	    -I$(srctree)/tools/objtool/arch/$(SRCARCH)/include
+	    -I$(srctree)/tools/objtool/arch/$(SRCARCH)/include \
+	    -I$(LIBSUBCMD_OUTPUT)/include
+# Note, EXTRA_WARNINGS here was determined for CC and not HOSTCC, it
+# is passed here to match a legacy behavior.
 WARNINGS := $(EXTRA_WARNINGS) -Wno-switch-default -Wno-switch-enum -Wno-packed -Wno-nested-externs
-CFLAGS   := -Werror $(WARNINGS) $(KBUILD_HOSTCFLAGS) -g $(INCLUDES) $(LIBELF_FLAGS)
-LDFLAGS  += $(LIBELF_LIBS) $(LIBSUBCMD) $(KBUILD_HOSTLDFLAGS)
+OBJTOOL_CFLAGS := -Werror $(WARNINGS) $(KBUILD_HOSTCFLAGS) -g $(INCLUDES) $(LIBELF_FLAGS)
+OBJTOOL_LDFLAGS := $(LIBELF_LIBS) $(LIBSUBCMD) $(KBUILD_HOSTLDFLAGS)
 
 # Allow old libelf to be used:
-elfshdr := $(shell echo '$(pound)include <libelf.h>' | $(CC) $(CFLAGS) -x c -E - | grep elf_getshdr)
-CFLAGS += $(if $(elfshdr),,-DLIBELF_USE_DEPRECATED)
+elfshdr := $(shell echo '$(pound)include <libelf.h>' | $(HOSTCC) $(OBJTOOL_CFLAGS) -x c -E - | grep elf_getshdr)
+OBJTOOL_CFLAGS += $(if $(elfshdr),,-DLIBELF_USE_DEPRECATED)
+
+# Always want host compilation.
+HOST_OVERRIDES := CC="$(HOSTCC)" LD="$(HOSTLD)" AR="$(HOSTAR)"
 
 AWK = awk
+MKDIR = mkdir
+
+ifeq ($(V),1)
+  Q =
+else
+  Q = @
+endif
 
 BUILD_ORC := n
 
@@ -49,21 +61,33 @@ export BUILD_ORC
 export srctree OUTPUT CFLAGS SRCARCH AWK
 include $(srctree)/tools/build/Makefile.include
 
-$(OBJTOOL_IN): fixdep FORCE
-	@$(CONFIG_SHELL) ./sync-check.sh
-	@$(MAKE) $(build)=objtool
+$(OBJTOOL_IN): fixdep $(LIBSUBCMD) FORCE
+	$(Q)$(CONFIG_SHELL) ./sync-check.sh
+	$(Q)$(MAKE) $(build)=objtool $(HOST_OVERRIDES) CFLAGS="$(OBJTOOL_CFLAGS)" \
+		LDFLAGS="$(OBJTOOL_LDFLAGS)"
+
 
 $(OBJTOOL): $(LIBSUBCMD) $(OBJTOOL_IN)
-	$(QUIET_LINK)$(CC) $(OBJTOOL_IN) $(LDFLAGS) -o $@
+	$(QUIET_LINK)$(HOSTCC) $(OBJTOOL_IN) $(OBJTOOL_LDFLAGS) -o $@
+
+
+$(LIBSUBCMD_OUTPUT):
+	$(Q)$(MKDIR) -p $@
 
+$(LIBSUBCMD): fixdep $(LIBSUBCMD_OUTPUT) FORCE
+	$(Q)$(MAKE) -C $(LIBSUBCMD_DIR) O=$(LIBSUBCMD_OUTPUT) \
+		DESTDIR=$(LIBSUBCMD_OUTPUT) prefix= subdir= \
+		$(HOST_OVERRIDES) EXTRA_CFLAGS="$(OBJTOOL_CFLAGS)" \
+		$@ install_headers
 
-$(LIBSUBCMD): fixdep FORCE
-	$(Q)$(MAKE) -C $(SUBCMD_SRCDIR) OUTPUT=$(LIBSUBCMD_OUTPUT)
+$(LIBSUBCMD)-clean:
+	$(call QUIET_CLEAN, libsubcmd)
+	$(Q)$(RM) -r -- $(LIBSUBCMD_OUTPUT)
 
-clean:
+clean: $(LIBSUBCMD)-clean
 	$(call QUIET_CLEAN, objtool) $(RM) $(OBJTOOL)
 	$(Q)find $(OUTPUT) -name '*.o' -delete -o -name '\.*.cmd' -delete -o -name '\.*.d' -delete
-	$(Q)$(RM) $(OUTPUT)arch/x86/lib/inat-tables.c $(OUTPUT)fixdep $(LIBSUBCMD)
+	$(Q)$(RM) $(OUTPUT)arch/x86/lib/inat-tables.c $(OUTPUT)fixdep
 
 FORCE:
 

tools/objtool/arch/powerpc/decode.c

@@ -41,38 +41,36 @@ const char *arch_ret_insn(int len)
 
 int arch_decode_instruction(struct objtool_file *file, const struct section *sec,
 			    unsigned long offset, unsigned int maxlen,
-			    unsigned int *len, enum insn_type *type,
-			    unsigned long *immediate,
-			    struct list_head *ops_list)
+			    struct instruction *insn)
 {
 	unsigned int opcode;
 	enum insn_type typ;
 	unsigned long imm;
-	u32 insn;
+	u32 ins;
 
-	insn = bswap_if_needed(file->elf, *(u32 *)(sec->data->d_buf + offset));
-	opcode = insn >> 26;
+	ins = bswap_if_needed(file->elf, *(u32 *)(sec->data->d_buf + offset));
+	opcode = ins >> 26;
 	typ = INSN_OTHER;
 	imm = 0;
 
 	switch (opcode) {
 	case 18: /* b[l][a] */
-		if ((insn & 3) == 1) /* bl */
+		if ((ins & 3) == 1) /* bl */
 			typ = INSN_CALL;
 
-		imm = insn & 0x3fffffc;
+		imm = ins & 0x3fffffc;
 		if (imm & 0x2000000)
 			imm -= 0x4000000;
 		break;
 	}
 
 	if (opcode == 1)
-		*len = 8;
+		insn->len = 8;
 	else
-		*len = 4;
+		insn->len = 4;
 
-	*type = typ;
-	*immediate = imm;
+	insn->type = typ;
+	insn->immediate = imm;
 
 	return 0;
 }

tools/objtool/arch/x86/decode.c

@@ -105,7 +105,7 @@ bool arch_pc_relative_reloc(struct reloc *reloc)
 #define ADD_OP(op) \
 	if (!(op = calloc(1, sizeof(*op)))) \
 		return -1; \
-	else for (list_add_tail(&op->list, ops_list); op; op = NULL)
+	else for (*ops_list = op, ops_list = &op->next; op; op = NULL)
 
 /*
  * Helpers to decode ModRM/SIB:
@@ -146,12 +146,11 @@ static bool has_notrack_prefix(struct insn *insn)
 
 int arch_decode_instruction(struct objtool_file *file, const struct section *sec,
 			    unsigned long offset, unsigned int maxlen,
-			    unsigned int *len, enum insn_type *type,
-			    unsigned long *immediate,
-			    struct list_head *ops_list)
+			    struct instruction *insn)
 {
+	struct stack_op **ops_list = &insn->stack_ops;
 	const struct elf *elf = file->elf;
-	struct insn insn;
+	struct insn ins;
 	int x86_64, ret;
 	unsigned char op1, op2, op3, prefix,
 		      rex = 0, rex_b = 0, rex_r = 0, rex_w = 0, rex_x = 0,
@@ -165,42 +164,42 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 	if (x86_64 == -1)
 		return -1;
 
-	ret = insn_decode(&insn, sec->data->d_buf + offset, maxlen,
+	ret = insn_decode(&ins, sec->data->d_buf + offset, maxlen,
 			  x86_64 ? INSN_MODE_64 : INSN_MODE_32);
 	if (ret < 0) {
 		WARN("can't decode instruction at %s:0x%lx", sec->name, offset);
 		return -1;
 	}
 
-	*len = insn.length;
-	*type = INSN_OTHER;
+	insn->len = ins.length;
+	insn->type = INSN_OTHER;
 
-	if (insn.vex_prefix.nbytes)
+	if (ins.vex_prefix.nbytes)
 		return 0;
 
-	prefix = insn.prefixes.bytes[0];
+	prefix = ins.prefixes.bytes[0];
 
-	op1 = insn.opcode.bytes[0];
-	op2 = insn.opcode.bytes[1];
-	op3 = insn.opcode.bytes[2];
+	op1 = ins.opcode.bytes[0];
+	op2 = ins.opcode.bytes[1];
+	op3 = ins.opcode.bytes[2];
 
-	if (insn.rex_prefix.nbytes) {
-		rex = insn.rex_prefix.bytes[0];
+	if (ins.rex_prefix.nbytes) {
+		rex = ins.rex_prefix.bytes[0];
 		rex_w = X86_REX_W(rex) >> 3;
 		rex_r = X86_REX_R(rex) >> 2;
 		rex_x = X86_REX_X(rex) >> 1;
 		rex_b = X86_REX_B(rex);
 	}
 
-	if (insn.modrm.nbytes) {
-		modrm = insn.modrm.bytes[0];
+	if (ins.modrm.nbytes) {
+		modrm = ins.modrm.bytes[0];
 		modrm_mod = X86_MODRM_MOD(modrm);
 		modrm_reg = X86_MODRM_REG(modrm) + 8*rex_r;
 		modrm_rm  = X86_MODRM_RM(modrm)  + 8*rex_b;
 	}
 
-	if (insn.sib.nbytes) {
-		sib = insn.sib.bytes[0];
+	if (ins.sib.nbytes) {
+		sib = ins.sib.bytes[0];
 		/* sib_scale = X86_SIB_SCALE(sib); */
 		sib_index = X86_SIB_INDEX(sib) + 8*rex_x;
 		sib_base  = X86_SIB_BASE(sib)  + 8*rex_b;
@@ -254,7 +253,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 		break;
 
 	case 0x70 ... 0x7f:
-		*type = INSN_JUMP_CONDITIONAL;
+		insn->type = INSN_JUMP_CONDITIONAL;
 		break;
 
 	case 0x80 ... 0x83:
@@ -278,7 +277,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 		if (!rm_is_reg(CFI_SP))
 			break;
 
-		imm = insn.immediate.value;
+		imm = ins.immediate.value;
 		if (op1 & 2) { /* sign extend */
 			if (op1 & 1) { /* imm32 */
 				imm <<= 32;
@@ -309,7 +308,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 			ADD_OP(op) {
 				op->src.type = OP_SRC_AND;
 				op->src.reg = CFI_SP;
-				op->src.offset = insn.immediate.value;
+				op->src.offset = ins.immediate.value;
 				op->dest.type = OP_DEST_REG;
 				op->dest.reg = CFI_SP;
 			}
@@ -356,7 +355,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 					op->src.reg = CFI_SP;
 					op->dest.type = OP_DEST_REG_INDIRECT;
 					op->dest.reg = modrm_rm;
-					op->dest.offset = insn.displacement.value;
+					op->dest.offset = ins.displacement.value;
 				}
 				break;
 			}
@@ -389,7 +388,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 				op->src.reg = modrm_reg;
 				op->dest.type = OP_DEST_REG_INDIRECT;
 				op->dest.reg = CFI_BP;
-				op->dest.offset = insn.displacement.value;
+				op->dest.offset = ins.displacement.value;
 			}
 			break;
 		}
@@ -402,7 +401,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 				op->src.reg = modrm_reg;
 				op->dest.type = OP_DEST_REG_INDIRECT;
 				op->dest.reg = CFI_SP;
-				op->dest.offset = insn.displacement.value;
+				op->dest.offset = ins.displacement.value;
 			}
 			break;
 		}
@@ -419,7 +418,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 			ADD_OP(op) {
 				op->src.type = OP_SRC_REG_INDIRECT;
 				op->src.reg = CFI_BP;
-				op->src.offset = insn.displacement.value;
+				op->src.offset = ins.displacement.value;
 				op->dest.type = OP_DEST_REG;
 				op->dest.reg = modrm_reg;
 			}
@@ -432,7 +431,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 			ADD_OP(op) {
 				op->src.type = OP_SRC_REG_INDIRECT;
 				op->src.reg = CFI_SP;
-				op->src.offset = insn.displacement.value;
+				op->src.offset = ins.displacement.value;
 				op->dest.type = OP_DEST_REG;
 				op->dest.reg = modrm_reg;
 			}
@@ -464,7 +463,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 
 		/* lea disp(%src), %dst */
 		ADD_OP(op) {
-			op->src.offset = insn.displacement.value;
+			op->src.offset = ins.displacement.value;
 			if (!op->src.offset) {
 				/* lea (%src), %dst */
 				op->src.type = OP_SRC_REG;
@@ -487,7 +486,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 		break;
 
 	case 0x90:
-		*type = INSN_NOP;
+		insn->type = INSN_NOP;
 		break;
 
 	case 0x9c:
@@ -511,39 +510,39 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 		if (op2 == 0x01) {
 
 			if (modrm == 0xca)
-				*type = INSN_CLAC;
+				insn->type = INSN_CLAC;
 			else if (modrm == 0xcb)
-				*type = INSN_STAC;
+				insn->type = INSN_STAC;
 
 		} else if (op2 >= 0x80 && op2 <= 0x8f) {
 
-			*type = INSN_JUMP_CONDITIONAL;
+			insn->type = INSN_JUMP_CONDITIONAL;
 
 		} else if (op2 == 0x05 || op2 == 0x07 || op2 == 0x34 ||
 			   op2 == 0x35) {
 
 			/* sysenter, sysret */
-			*type = INSN_CONTEXT_SWITCH;
+			insn->type = INSN_CONTEXT_SWITCH;
 
 		} else if (op2 == 0x0b || op2 == 0xb9) {
 
 			/* ud2 */
-			*type = INSN_BUG;
+			insn->type = INSN_BUG;
 
 		} else if (op2 == 0x0d || op2 == 0x1f) {
 
 			/* nopl/nopw */
-			*type = INSN_NOP;
+			insn->type = INSN_NOP;
 
 		} else if (op2 == 0x1e) {
 
 			if (prefix == 0xf3 && (modrm == 0xfa || modrm == 0xfb))
-				*type = INSN_ENDBR;
+				insn->type = INSN_ENDBR;
 
 
 		} else if (op2 == 0x38 && op3 == 0xf8) {
-			if (insn.prefixes.nbytes == 1 &&
-			    insn.prefixes.bytes[0] == 0xf2) {
+			if (ins.prefixes.nbytes == 1 &&
+			    ins.prefixes.bytes[0] == 0xf2) {
 				/* ENQCMD cannot be used in the kernel. */
 				WARN("ENQCMD instruction at %s:%lx", sec->name,
 				     offset);
@@ -591,29 +590,29 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 
 	case 0xcc:
 		/* int3 */
-		*type = INSN_TRAP;
+		insn->type = INSN_TRAP;
 		break;
 
 	case 0xe3:
 		/* jecxz/jrcxz */
-		*type = INSN_JUMP_CONDITIONAL;
+		insn->type = INSN_JUMP_CONDITIONAL;
 		break;
 
 	case 0xe9:
 	case 0xeb:
-		*type = INSN_JUMP_UNCONDITIONAL;
+		insn->type = INSN_JUMP_UNCONDITIONAL;
 		break;
 
 	case 0xc2:
 	case 0xc3:
-		*type = INSN_RETURN;
+		insn->type = INSN_RETURN;
 		break;
 
 	case 0xc7: /* mov imm, r/m */
 		if (!opts.noinstr)
 			break;
 
-		if (insn.length == 3+4+4 && !strncmp(sec->name, ".init.text", 10)) {
+		if (ins.length == 3+4+4 && !strncmp(sec->name, ".init.text", 10)) {
 			struct reloc *immr, *disp;
 			struct symbol *func;
 			int idx;
@@ -661,17 +660,17 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 
 	case 0xca: /* retf */
 	case 0xcb: /* retf */
-		*type = INSN_CONTEXT_SWITCH;
+		insn->type = INSN_CONTEXT_SWITCH;
 		break;
 
 	case 0xe0: /* loopne */
 	case 0xe1: /* loope */
 	case 0xe2: /* loop */
-		*type = INSN_JUMP_CONDITIONAL;
+		insn->type = INSN_JUMP_CONDITIONAL;
 		break;
 
 	case 0xe8:
-		*type = INSN_CALL;
+		insn->type = INSN_CALL;
 		/*
 		 * For the impact on the stack, a CALL behaves like
 		 * a PUSH of an immediate value (the return address).
@@ -683,30 +682,30 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 		break;
 
 	case 0xfc:
-		*type = INSN_CLD;
+		insn->type = INSN_CLD;
 		break;
 
 	case 0xfd:
-		*type = INSN_STD;
+		insn->type = INSN_STD;
 		break;
 
 	case 0xff:
 		if (modrm_reg == 2 || modrm_reg == 3) {
 
-			*type = INSN_CALL_DYNAMIC;
-			if (has_notrack_prefix(&insn))
+			insn->type = INSN_CALL_DYNAMIC;
+			if (has_notrack_prefix(&ins))
 				WARN("notrack prefix found at %s:0x%lx", sec->name, offset);
 
 		} else if (modrm_reg == 4) {
 
-			*type = INSN_JUMP_DYNAMIC;
-			if (has_notrack_prefix(&insn))
+			insn->type = INSN_JUMP_DYNAMIC;
+			if (has_notrack_prefix(&ins))
 				WARN("notrack prefix found at %s:0x%lx", sec->name, offset);
 
 		} else if (modrm_reg == 5) {
 
 			/* jmpf */
-			*type = INSN_CONTEXT_SWITCH;
+			insn->type = INSN_CONTEXT_SWITCH;
 
 		} else if (modrm_reg == 6) {
 
@@ -723,7 +722,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 		break;
 	}
 
-	*immediate = insn.immediate.nbytes ? insn.immediate.value : 0;
+	insn->immediate = ins.immediate.nbytes ? ins.immediate.value : 0;
 
 	return 0;
 }

tools/objtool/builtin-check.c

@@ -65,7 +65,7 @@ static int parse_hacks(const struct option *opt, const char *str, int unset)
 	return found ? 0 : -1;
 }
 
-const struct option check_options[] = {
+static const struct option check_options[] = {
 	OPT_GROUP("Actions:"),
 	OPT_CALLBACK_OPTARG('h', "hacks", NULL, NULL, "jump_label,noinstr,skylake", "patch toolchain bugs/limitations", parse_hacks),
 	OPT_BOOLEAN('i', "ibt", &opts.ibt, "validate and annotate IBT"),

tools/objtool/elf.c

@@ -284,13 +284,13 @@ static int read_sections(struct elf *elf)
 	    !elf_alloc_hash(section_name, sections_nr))
 		return -1;
 
+	elf->section_data = calloc(sections_nr, sizeof(*sec));
+	if (!elf->section_data) {
+		perror("calloc");
+		return -1;
+	}
 	for (i = 0; i < sections_nr; i++) {
-		sec = malloc(sizeof(*sec));
-		if (!sec) {
-			perror("malloc");
-			return -1;
-		}
-		memset(sec, 0, sizeof(*sec));
+		sec = &elf->section_data[i];
 
 		INIT_LIST_HEAD(&sec->symbol_list);
 		INIT_LIST_HEAD(&sec->reloc_list);
@@ -422,13 +422,13 @@ static int read_symbols(struct elf *elf)
 	    !elf_alloc_hash(symbol_name, symbols_nr))
 		return -1;
 
+	elf->symbol_data = calloc(symbols_nr, sizeof(*sym));
+	if (!elf->symbol_data) {
+		perror("calloc");
+		return -1;
+	}
 	for (i = 0; i < symbols_nr; i++) {
-		sym = malloc(sizeof(*sym));
-		if (!sym) {
-			perror("malloc");
-			return -1;
-		}
-		memset(sym, 0, sizeof(*sym));
+		sym = &elf->symbol_data[i];
 
 		sym->idx = i;
 
@@ -918,13 +918,13 @@ static int read_relocs(struct elf *elf)
 		sec->base->reloc = sec;
 
 		nr_reloc = 0;
+		sec->reloc_data = calloc(sec->sh.sh_size / sec->sh.sh_entsize, sizeof(*reloc));
+		if (!sec->reloc_data) {
+			perror("calloc");
+			return -1;
+		}
 		for (i = 0; i < sec->sh.sh_size / sec->sh.sh_entsize; i++) {
-			reloc = malloc(sizeof(*reloc));
-			if (!reloc) {
-				perror("malloc");
-				return -1;
-			}
-			memset(reloc, 0, sizeof(*reloc));
+			reloc = &sec->reloc_data[i];
 			switch (sec->sh.sh_type) {
 			case SHT_REL:
 				if (read_rel_reloc(sec, i, reloc, &symndx))
@@ -1453,16 +1453,16 @@ void elf_close(struct elf *elf)
 		list_for_each_entry_safe(sym, tmpsym, &sec->symbol_list, list) {
 			list_del(&sym->list);
 			hash_del(&sym->hash);
-			free(sym);
 		}
 		list_for_each_entry_safe(reloc, tmpreloc, &sec->reloc_list, list) {
 			list_del(&reloc->list);
 			hash_del(&reloc->hash);
-			free(reloc);
 		}
 		list_del(&sec->list);
-		free(sec);
+		free(sec->reloc_data);
 	}
 
+	free(elf->symbol_data);
+	free(elf->section_data);
 	free(elf);
 }

tools/objtool/include/objtool/arch.h

@@ -62,9 +62,9 @@ struct op_src {
 };
 
 struct stack_op {
+	struct stack_op *next;
 	struct op_dest dest;
 	struct op_src src;
-	struct list_head list;
 };
 
 struct instruction;
@@ -75,9 +75,7 @@ void arch_initial_func_cfi_state(struct cfi_init_state *state);
 
 int arch_decode_instruction(struct objtool_file *file, const struct section *sec,
 			    unsigned long offset, unsigned int maxlen,
-			    unsigned int *len, enum insn_type *type,
-			    unsigned long *immediate,
-			    struct list_head *ops_list);
+			    struct instruction *insn);
 
 bool arch_callee_saved_reg(unsigned char reg);
 

tools/objtool/include/objtool/builtin.h

@@ -7,8 +7,6 @@
 
 #include <subcmd/parse-options.h>
 
-extern const struct option check_options[];
-
 struct opts {
 	/* actions: */
 	bool dump_orc;

tools/objtool/include/objtool/cfi.h

@@ -34,6 +34,7 @@ struct cfi_state {
 	unsigned char type;
 	bool bp_scratch;
 	bool drap;
+	bool signal;
 	bool end;
 };
 

tools/objtool/include/objtool/check.h

@@ -27,7 +27,7 @@ struct alt_group {
 	struct alt_group *orig_group;
 
 	/* First and last instructions in the group */
-	struct instruction *first_insn, *last_insn;
+	struct instruction *first_insn, *last_insn, *nop;
 
 	/*
 	 * Byte-offset-addressed len-sized array of pointers to CFI structs.
@@ -36,39 +36,46 @@ struct alt_group {
 	struct cfi_state **cfi;
 };
 
+#define INSN_CHUNK_BITS		8
+#define INSN_CHUNK_SIZE		(1 << INSN_CHUNK_BITS)
+#define INSN_CHUNK_MAX		(INSN_CHUNK_SIZE - 1)
+
 struct instruction {
-	struct list_head list;
 	struct hlist_node hash;
 	struct list_head call_node;
 	struct section *sec;
 	unsigned long offset;
-	unsigned int len;
-	enum insn_type type;
 	unsigned long immediate;
 
-	u16 dead_end		: 1,
-	   ignore		: 1,
-	   ignore_alts		: 1,
-	   hint			: 1,
-	   save			: 1,
-	   restore		: 1,
-	   retpoline_safe	: 1,
-	   noendbr		: 1,
-	   entry		: 1;
-		/* 7 bit hole */
-
+	u8 len;
+	u8 prev_len;
+	u8 type;
 	s8 instr;
-	u8 visited;
+
+	u32 idx			: INSN_CHUNK_BITS,
+	    dead_end		: 1,
+	    ignore		: 1,
+	    ignore_alts		: 1,
+	    hint		: 1,
+	    save		: 1,
+	    restore		: 1,
+	    retpoline_safe	: 1,
+	    noendbr		: 1,
+	    entry		: 1,
+	    visited		: 4,
+	    no_reloc		: 1;
+		/* 10 bit hole */
 
 	struct alt_group *alt_group;
-	struct symbol *call_dest;
 	struct instruction *jump_dest;
 	struct instruction *first_jump_src;
-	struct reloc *jump_table;
-	struct reloc *reloc;
-	struct list_head alts;
+	union {
+		struct symbol *_call_dest;
+		struct reloc *_jump_table;
+	};
+	struct alternative *alts;
 	struct symbol *sym;
-	struct list_head stack_ops;
+	struct stack_op *stack_ops;
 	struct cfi_state *cfi;
 };
 
@@ -107,13 +114,11 @@ static inline bool is_jump(struct instruction *insn)
 struct instruction *find_insn(struct objtool_file *file,
 			      struct section *sec, unsigned long offset);
 
-#define for_each_insn(file, insn)					\
-	list_for_each_entry(insn, &file->insn_list, list)
+struct instruction *next_insn_same_sec(struct objtool_file *file, struct instruction *insn);
 
-#define sec_for_each_insn(file, sec, insn)				\
-	for (insn = find_insn(file, sec, 0);				\
-	     insn && &insn->list != &file->insn_list &&			\
-			insn->sec == sec;				\
-	     insn = list_next_entry(insn, list))
+#define sec_for_each_insn(file, _sec, insn)				\
+	for (insn = find_insn(file, _sec, 0);				\
+	     insn && insn->sec == _sec;					\
+	     insn = next_insn_same_sec(file, insn))
 
 #endif /* _CHECK_H */

tools/objtool/include/objtool/elf.h

@@ -39,6 +39,7 @@ struct section {
 	char *name;
 	int idx;
 	bool changed, text, rodata, noinstr, init, truncate;
+	struct reloc *reloc_data;
 };
 
 struct symbol {
@@ -49,12 +50,11 @@ struct symbol {
 	GElf_Sym sym;
 	struct section *sec;
 	char *name;
-	unsigned int idx;
-	unsigned char bind, type;
+	unsigned int idx, len;
 	unsigned long offset;
-	unsigned int len;
 	unsigned long __subtree_last;
 	struct symbol *pfunc, *cfunc, *alias;
+	unsigned char bind, type;
 	u8 uaccess_safe      : 1;
 	u8 static_call_tramp : 1;
 	u8 retpoline_thunk   : 1;
@@ -104,6 +104,9 @@ struct elf {
 	struct hlist_head *section_hash;
 	struct hlist_head *section_name_hash;
 	struct hlist_head *reloc_hash;
+
+	struct section *section_data;
+	struct symbol *symbol_data;
 };
 
 #define OFFSET_STRIDE_BITS	4

tools/objtool/include/objtool/objtool.h

@@ -21,7 +21,6 @@ struct pv_state {
 
 struct objtool_file {
 	struct elf *elf;
-	struct list_head insn_list;
 	DECLARE_HASHTABLE(insn_hash, 20);
 	struct list_head retpoline_call_list;
 	struct list_head return_thunk_list;

tools/objtool/include/objtool/special.h

@@ -19,6 +19,7 @@ struct special_alt {
 	bool skip_orig;
 	bool skip_alt;
 	bool jump_or_nop;
+	u8 key_addend;
 
 	struct section *orig_sec;
 	unsigned long orig_off;
@@ -27,7 +28,6 @@ struct special_alt {
 	unsigned long new_off;
 
 	unsigned int orig_len, new_len; /* group only */
-	u8 key_addend;
 };
 
 int special_get_alts(struct elf *elf, struct list_head *alts);

tools/objtool/objtool.c

@@ -99,7 +99,6 @@ struct objtool_file *objtool_open_read(const char *_objname)
 		return NULL;
 	}
 
-	INIT_LIST_HEAD(&file.insn_list);
 	hash_init(file.insn_hash);
 	INIT_LIST_HEAD(&file.retpoline_call_list);
 	INIT_LIST_HEAD(&file.return_thunk_list);

tools/objtool/orc_dump.c

@@ -211,8 +211,8 @@ int orc_dump(const char *_objname)
 
 		print_reg(orc[i].bp_reg, bswap_if_needed(&dummy_elf, orc[i].bp_offset));
 
-		printf(" type:%s end:%d\n",
-		       orc_type_name(orc[i].type), orc[i].end);
+		printf(" type:%s signal:%d end:%d\n",
+		       orc_type_name(orc[i].type), orc[i].signal, orc[i].end);
 	}
 
 	elf_end(elf);

tools/objtool/orc_gen.c

@@ -27,6 +27,7 @@ static int init_orc_entry(struct orc_entry *orc, struct cfi_state *cfi,
 	}
 
 	orc->end = cfi->end;
+	orc->signal = cfi->signal;
 
 	if (cfi->cfa.base == CFI_UNDEFINED) {
 		orc->sp_reg = ORC_REG_UNDEFINED;

tools/objtool/special.c

@@ -26,7 +26,7 @@ struct special_entry {
 	unsigned char key; /* jump_label key */
 };
 
-struct special_entry entries[] = {
+static const struct special_entry entries[] = {
 	{
 		.sec = ".altinstructions",
 		.group = true,
@@ -65,7 +65,7 @@ static void reloc_to_sec_off(struct reloc *reloc, struct section **sec,
 	*off = reloc->sym->offset + reloc->addend;
 }
 
-static int get_alt_entry(struct elf *elf, struct special_entry *entry,
+static int get_alt_entry(struct elf *elf, const struct special_entry *entry,
 			 struct section *sec, int idx,
 			 struct special_alt *alt)
 {
@@ -139,7 +139,7 @@ static int get_alt_entry(struct elf *elf, struct special_entry *entry,
  */
 int special_get_alts(struct elf *elf, struct list_head *alts)
 {
-	struct special_entry *entry;
+	const struct special_entry *entry;
 	struct section *sec;
 	unsigned int nr_entries;
 	struct special_alt *alt;