[RFC 04/13] objtool: Print symbol during disassembly

Alexandre Chartre posted 13 patches 8 months ago
There is a newer version of this series
[RFC 04/13] objtool: Print symbol during disassembly
Posted by Alexandre Chartre 8 months ago
Print symbols referenced during disassembly instead of just printing
raw addresses. Also handle address relocation.

Signed-off-by: Alexandre Chartre <alexandre.chartre@oracle.com>
---
 tools/objtool/check.c                 |   9 ---
 tools/objtool/disas.c                 | 101 ++++++++++++++++++++++++++
 tools/objtool/include/objtool/check.h |   9 +++
 3 files changed, 110 insertions(+), 9 deletions(-)

diff --git a/tools/objtool/check.c b/tools/objtool/check.c
index 9cfac23185b8..ee613f03e57d 100644
--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -131,15 +131,6 @@ static struct instruction *prev_insn_same_sym(struct objtool_file *file,
 	for (insn = next_insn_same_sec(file, insn); insn;		\
 	     insn = next_insn_same_sec(file, insn))
 
-static inline struct symbol *insn_call_dest(struct instruction *insn)
-{
-	if (insn->type == INSN_JUMP_DYNAMIC ||
-	    insn->type == INSN_CALL_DYNAMIC)
-		return NULL;
-
-	return insn->_call_dest;
-}
-
 static inline struct reloc *insn_jump_table(struct instruction *insn)
 {
 	if (insn->type == INSN_JUMP_DYNAMIC ||
diff --git a/tools/objtool/disas.c b/tools/objtool/disas.c
index f2eb1050ce11..83fe2c018c4b 100644
--- a/tools/objtool/disas.c
+++ b/tools/objtool/disas.c
@@ -13,10 +13,108 @@
 
 struct disas_context {
 	struct objtool_file *file;
+	struct instruction *insn;
 	disassembler_ftype disassembler;
 	struct disassemble_info info;
 };
 
+#define DINFO_FPRINTF(dinfo, ...)	\
+	((*(dinfo)->fprintf_func)((dinfo)->stream, __VA_ARGS__))
+
+static void disas_print_address(bfd_vma addr, struct disassemble_info *dinfo)
+{
+	struct disas_context *dctx = dinfo->application_data;
+	struct instruction *insn = dctx->insn;
+	struct objtool_file *file = dctx->file;
+	struct symbol *call_dest, *sym;
+	struct instruction *jump_dest;
+	struct section *sec;
+	struct reloc *reloc;
+	bool is_reloc;
+	s64 offset;
+
+	/*
+	 * If the instruction is a call/jump and it references a
+	 * destination then this is likely the address we are looking
+	 * up. So check it first.
+	 */
+	jump_dest = insn->jump_dest;
+	if (jump_dest && jump_dest->offset == addr) {
+		DINFO_FPRINTF(dinfo, "%lx <%s+0x%lx>", addr,
+			      jump_dest->sym->name,
+			      jump_dest->offset - jump_dest->sym->offset);
+		return;
+	}
+
+	/*
+	 * Assume the address is a relocation if it points to the next
+	 * instruction.
+	 */
+	is_reloc = (addr == insn->offset + insn->len);
+
+	/*
+	 * The call destination offset can be the address we are looking
+	 * up, or 0 if there is a relocation.
+	 */
+	call_dest = insn_call_dest(insn);
+	if (call_dest) {
+		if (call_dest->offset == addr) {
+			DINFO_FPRINTF(dinfo, "%lx <%s>", addr, call_dest->name);
+			return;
+		}
+		if (call_dest->offset == 0 && is_reloc) {
+			DINFO_FPRINTF(dinfo, "%s", call_dest->name);
+			return;
+		}
+	}
+
+	if (!is_reloc) {
+		DINFO_FPRINTF(dinfo, "0x%lx", addr);
+		return;
+	}
+
+	/*
+	 * If this is a relocation, check if we have relocation information
+	 * for this instruction.
+	 */
+	reloc = find_reloc_by_dest_range(file->elf, insn->sec,
+					 insn->offset, insn->len);
+	if (!reloc) {
+		DINFO_FPRINTF(dinfo, "0x%lx", addr);
+		return;
+	}
+
+	if (reloc_type(reloc) == R_X86_64_PC32 ||
+	    reloc_type(reloc) == R_X86_64_PLT32)
+		offset = arch_dest_reloc_offset(reloc_addend(reloc));
+	else
+		offset = reloc_addend(reloc);
+
+	/*
+	 * If the relocation symbol is a section name (for example ".bss")
+	 * then we try to further resolve the name.
+	 */
+	sec = find_section_by_name(file->elf, reloc->sym->name);
+	if (sec) {
+		sym = find_symbol_containing(sec, offset);
+		if (sym) {
+			if (sym->offset == offset)
+				DINFO_FPRINTF(dinfo, "%s+0x%lx = %s",
+					     reloc->sym->name, offset, sym->name);
+			else
+				DINFO_FPRINTF(dinfo, "%s+0x%lx = %s+0x%lx",
+					      reloc->sym->name, offset,
+					      sym->name, offset - sym->offset);
+			return;
+		}
+	}
+
+	if (offset)
+		DINFO_FPRINTF(dinfo, "%s+0x%lx", reloc->sym->name, offset);
+	else
+		DINFO_FPRINTF(dinfo, "%s", reloc->sym->name);
+}
+
 /*
  * Initialize disassemble info arch, mach (32 or 64-bit) and options.
  */
@@ -65,6 +163,7 @@ struct disas_context *disas_context_create(struct objtool_file *file)
 				     fprintf_styled);
 
 	dinfo->read_memory_func = buffer_read_memory;
+	dinfo->print_address_func = disas_print_address;
 	dinfo->application_data = dctx;
 
 	/*
@@ -117,6 +216,8 @@ static size_t disas_insn(struct disas_context *dctx,
 	disassembler_ftype disasm = dctx->disassembler;
 	struct disassemble_info *dinfo = &dctx->info;
 
+	dctx->insn = insn;
+
 	/*
 	 * Set the disassembler buffer to read data from the section
 	 * containing the instruction to disassemble.
diff --git a/tools/objtool/include/objtool/check.h b/tools/objtool/include/objtool/check.h
index 5290ac1ebbc1..4adbcd760c6f 100644
--- a/tools/objtool/include/objtool/check.h
+++ b/tools/objtool/include/objtool/check.h
@@ -115,6 +115,15 @@ static inline bool is_jump(struct instruction *insn)
 	return is_static_jump(insn) || is_dynamic_jump(insn);
 }
 
+static inline struct symbol *insn_call_dest(struct instruction *insn)
+{
+	if (insn->type == INSN_JUMP_DYNAMIC ||
+	    insn->type == INSN_CALL_DYNAMIC)
+		return NULL;
+
+	return insn->_call_dest;
+}
+
 struct instruction *find_insn(struct objtool_file *file,
 			      struct section *sec, unsigned long offset);
 
-- 
2.43.5
Re: [RFC 04/13] objtool: Print symbol during disassembly
Posted by Josh Poimboeuf 8 months ago
On Fri, Jun 06, 2025 at 05:34:31PM +0200, Alexandre Chartre wrote:
> +static void disas_print_address(bfd_vma addr, struct disassemble_info *dinfo)
> +{
> +	struct disas_context *dctx = dinfo->application_data;
> +	struct instruction *insn = dctx->insn;
> +	struct objtool_file *file = dctx->file;
> +	struct symbol *call_dest, *sym;
> +	struct instruction *jump_dest;
> +	struct section *sec;
> +	struct reloc *reloc;
> +	bool is_reloc;
> +	s64 offset;
> +
> +	/*
> +	 * If the instruction is a call/jump and it references a
> +	 * destination then this is likely the address we are looking
> +	 * up. So check it first.
> +	 */
> +	jump_dest = insn->jump_dest;
> +	if (jump_dest && jump_dest->offset == addr) {
> +		DINFO_FPRINTF(dinfo, "%lx <%s+0x%lx>", addr,
> +			      jump_dest->sym->name,
> +			      jump_dest->offset - jump_dest->sym->offset);
> +		return;
> +	}

IIRC, there may be a few cases where an instruction's 'sym' field can be
NULL, might want to check for !jump_dest->sym here.

> +	/*
> +	 * If this is a relocation, check if we have relocation information
> +	 * for this instruction.
> +	 */
> +	reloc = find_reloc_by_dest_range(file->elf, insn->sec,
> +					 insn->offset, insn->len);
> +	if (!reloc) {
> +		DINFO_FPRINTF(dinfo, "0x%lx", addr);
> +		return;
> +	}
> +
> +	if (reloc_type(reloc) == R_X86_64_PC32 ||
> +	    reloc_type(reloc) == R_X86_64_PLT32)

Can use arch_pc_relative_reloc() here.

> +		offset = arch_dest_reloc_offset(reloc_addend(reloc));
> +	else
> +		offset = reloc_addend(reloc);
> +
> +	/*
> +	 * If the relocation symbol is a section name (for example ".bss")
> +	 * then we try to further resolve the name.
> +	 */

This can be checked with reloc->sym->type == STT_SECTION.

> +	sec = find_section_by_name(file->elf, reloc->sym->name);
> +	if (sec) {
> +		sym = find_symbol_containing(sec, offset);
> +		if (sym) {
> +			if (sym->offset == offset)
> +				DINFO_FPRINTF(dinfo, "%s+0x%lx = %s",
> +					     reloc->sym->name, offset, sym->name);
> +			else
> +				DINFO_FPRINTF(dinfo, "%s+0x%lx = %s+0x%lx",
> +					      reloc->sym->name, offset,
> +					      sym->name, offset - sym->offset);
> +			return;
> +		}
> +	}
> +
> +	if (offset)
> +		DINFO_FPRINTF(dinfo, "%s+0x%lx", reloc->sym->name, offset);
> +	else
> +		DINFO_FPRINTF(dinfo, "%s", reloc->sym->name);

We have offstr() which does similar things.  You might be able to get
away with replacing the above hunk with something like:

	DINFO_FPRINTF(dinfo, "%s", offstr(reloc->sym->sec, sym->offset + offset));

-- 
Josh