sigcdump 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: c7ecae7f81f6ea8257c8cede3efa1fc30328ece87d99f0e6e7e949c6bc1d7381
+  data.tar.gz: 16414972dc9b8cd428a7218eb9de2c29494bc892a65965970e28e2972f837b39
+SHA512:
+  metadata.gz: 06272e5057743388ba089fbc5db399e77697162b01b0bf65a59869d75e0a33a2edca6ab2c84163694091bb12c09461ce0ad30bbf88fcaf794e3321280985d376
+  data.tar.gz: 33419cd6d433b20c8951063dc755ac06773965a03e0119910f8e54b7c8bcd30da33a754f9c17fd7cbc44aa674c62610776c65d044d298fd3b65d7cfdd2f75f08

data/.gitignore

@@ -0,0 +1,14 @@
+/.bundle/
+/.yardoc
+/_yardoc/
+/coverage/
+/doc/
+/pkg/
+/spec/reports/
+/tmp/
+*.bundle
+*.so
+*.o
+*.a
+mkmf.log
+.ruby-version

data/.gitmodules

@@ -0,0 +1,3 @@
+[submodule "ext/sigcdump/ruby"]
+	path = ext/sigcdump/ruby
+	url = https://github.com/ruby/ruby.git

data/Gemfile

@@ -0,0 +1,8 @@
+source "https://rubygems.org"
+
+git_source(:github) {|repo_name| "https://github.com/#{repo_name}" }
+
+# Specify your gem's dependencies in sigcdump.gemspec
+gemspec
+
+gem 'pry'

data/Gemfile.lock

@@ -0,0 +1,29 @@
+PATH
+  remote: .
+  specs:
+    sigcdump (0.1.0)
+
+GEM
+  remote: https://rubygems.org/
+  specs:
+    coderay (1.1.2)
+    method_source (0.9.2)
+    pry (0.12.2)
+      coderay (~> 1.1.0)
+      method_source (~> 0.9.0)
+    rake (13.0.1)
+    rake-compiler (1.0.8)
+      rake
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  bundler
+  pry
+  rake
+  rake-compiler
+  sigcdump!
+
+BUNDLED WITH
+   1.17.2

data/LICENSE.txt

@@ -0,0 +1,56 @@
+Ruby is copyrighted free software by Yukihiro Matsumoto <matz@netlab.jp>.
+You can redistribute it and/or modify it under either the terms of the
+2-clause BSDL (see the file BSDL), or the conditions below:
+
+1. You may make and give away verbatim copies of the source form of the
+   software without restriction, provided that you duplicate all of the
+   original copyright notices and associated disclaimers.
+
+2. You may modify your copy of the software in any way, provided that
+   you do at least ONE of the following:
+
+   a. place your modifications in the Public Domain or otherwise
+      make them Freely Available, such as by posting said
+      modifications to Usenet or an equivalent medium, or by allowing
+      the author to include your modifications in the software.
+
+   b. use the modified software only within your corporation or
+      organization.
+
+   c. give non-standard binaries non-standard names, with
+      instructions on where to get the original software distribution.
+
+   d. make other distribution arrangements with the author.
+
+3. You may distribute the software in object code or binary form,
+   provided that you do at least ONE of the following:
+
+   a. distribute the binaries and library files of the software,
+      together with instructions (in the manual page or equivalent)
+      on where to get the original distribution.
+
+   b. accompany the distribution with the machine-readable source of
+      the software.
+
+   c. give non-standard binaries non-standard names, with
+      instructions on where to get the original software distribution.
+
+   d. make other distribution arrangements with the author.
+
+4. You may modify and include the part of the software into any other
+   software (possibly commercial).  But some files in the distribution
+   are not written by the author, so that they are not under these terms.
+
+   For the list of those files and their copying conditions, see the
+   file LEGAL.
+
+5. The scripts and library files supplied as input to or produced as
+   output from the software do not automatically fall under the
+   copyright of the software, but belong to whomever generated them,
+   and may be sold commercially, and may be aggregated with this
+   software.
+
+6. THIS SOFTWARE IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR
+   IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED
+   WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+   PURPOSE.

data/README.md

@@ -0,0 +1,37 @@
+# Sigcdump
+
+[Sigdump](https://github.com/frsyuki/sigdump) for C backtrace.
+
+## Installation
+
+Add this line to your application's Gemfile:
+
+```ruby
+gem 'sigcdump'
+```
+
+And then execute:
+
+    $ bundle
+
+NOTE: Unlike sigdump, this gem supports only CRuby.
+
+## Usage
+
+```rb
+require 'sigcdump'
+Sigcdump.setup('SIGCONT')
+```
+
+or:
+
+```rb
+# this calls Sigcdump.setup('SIGCONT')
+require 'sigcdump/setup'
+```
+
+Then `kill -TTIN <pid>` will show a C backtrace of a CRuby process.
+
+## License
+
+The gem is available as open source under the terms of the [Ruby License](./LICENSE.txt).

data/Rakefile

@@ -0,0 +1,10 @@
+require "bundler/gem_tasks"
+require "rake/extensiontask"
+
+task :build => :compile
+
+Rake::ExtensionTask.new("sigcdump") do |ext|
+  ext.lib_dir = "lib/sigcdump"
+end
+
+task :default => [:clobber, :compile, :spec]

data/bin/console

@@ -0,0 +1,11 @@
+#!/usr/bin/env ruby
+
+require "bundler/setup"
+require "sigcdump"
+
+# You can add fixtures and/or initialization code here to make experimenting
+# with your gem easier. You can also use a different console, if you like.
+
+# (If you use this, don't forget to add pry to your Gemfile!)
+require "pry"
+Pry.start

data/bin/setup

@@ -0,0 +1,8 @@
+#!/usr/bin/env bash
+set -euo pipefail
+IFS=$'\n\t'
+set -vx
+
+bundle install
+
+# Do any other automated setup that you need to do here

data/ext/sigcdump/extconf.rb

@@ -0,0 +1,16 @@
+require 'mkmf'
+
+ruby_dir = File.expand_path('./ruby', __dir__)
+$INCFLAGS << " -I#{ruby_dir}"
+
+$srcs = %w[
+  sigcdump.c
+]
+%w[
+  addr2line.c
+].each do |c|
+  FileUtils.ln_s(File.join(ruby_dir, c), c, force: true)
+  $srcs << c
+end
+
+create_makefile('sigcdump/sigcdump')

data/ext/sigcdump/ruby/addr2line.c

@@ -0,0 +1,2663 @@
+/**********************************************************************
+
+  addr2line.c -
+
+  $Author$
+
+  Copyright (C) 2010 Shinichiro Hamaji
+
+**********************************************************************/
+
+#if defined(__clang__)
+#pragma clang diagnostic ignored "-Wgnu-empty-initializer"
+#pragma clang diagnostic ignored "-Wgcc-compat"
+#endif
+
+#include "ruby/config.h"
+#include "ruby/defines.h"
+#include "ruby/missing.h"
+#include "addr2line.h"
+
+#include <stdio.h>
+#include <errno.h>
+
+#ifdef HAVE_LIBPROC_H
+#include <libproc.h>
+#endif
+
+#ifdef HAVE_STDBOOL_H
+#include <stdbool.h>
+#else
+#include "missing/stdbool.h"
+#endif
+
+#if defined(USE_ELF) || defined(HAVE_MACH_O_LOADER_H)
+
+#include <fcntl.h>
+#include <limits.h>
+#include <stdio.h>
+#include <stdint.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/mman.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <unistd.h>
+
+/* Make alloca work the best possible way.  */
+#ifdef __GNUC__
+# ifndef alloca
+#  define alloca __builtin_alloca
+# endif
+#else
+# ifdef HAVE_ALLOCA_H
+#  include <alloca.h>
+# else
+#  ifdef _AIX
+#pragma alloca
+#  else
+#   ifndef alloca		/* predefined by HP cc +Olibcalls */
+void *alloca();
+#   endif
+#  endif /* AIX */
+# endif	/* HAVE_ALLOCA_H */
+#endif /* __GNUC__ */
+
+#ifdef HAVE_DLADDR
+# include <dlfcn.h>
+#endif
+
+#ifdef HAVE_MACH_O_LOADER_H
+# include <crt_externs.h>
+# include <mach-o/fat.h>
+# include <mach-o/loader.h>
+# include <mach-o/nlist.h>
+# include <mach-o/stab.h>
+#endif
+
+#ifdef USE_ELF
+# ifdef __OpenBSD__
+#  include <elf_abi.h>
+# else
+#  include <elf.h>
+# endif
+
+#ifndef ElfW
+# if SIZEOF_VOIDP == 8
+#  define ElfW(x) Elf64##_##x
+# else
+#  define ElfW(x) Elf32##_##x
+# endif
+#endif
+#ifndef ELF_ST_TYPE
+# if SIZEOF_VOIDP == 8
+#  define ELF_ST_TYPE ELF64_ST_TYPE
+# else
+#  define ELF_ST_TYPE ELF32_ST_TYPE
+# endif
+#endif
+#endif
+
+#ifdef SHF_COMPRESSED
+# if defined(ELFCOMPRESS_ZLIB) && defined(HAVE_LIBZ)
+   /* FreeBSD 11.0 lacks ELFCOMPRESS_ZLIB */
+#  include <zlib.h>
+#  define SUPPORT_COMPRESSED_DEBUG_LINE
+# endif
+#else /* compatibility with glibc < 2.22 */
+# define SHF_COMPRESSED 0
+#endif
+
+#ifndef PATH_MAX
+#define PATH_MAX 4096
+#endif
+
+#define DW_LNS_copy                     0x01
+#define DW_LNS_advance_pc               0x02
+#define DW_LNS_advance_line             0x03
+#define DW_LNS_set_file                 0x04
+#define DW_LNS_set_column               0x05
+#define DW_LNS_negate_stmt              0x06
+#define DW_LNS_set_basic_block          0x07
+#define DW_LNS_const_add_pc             0x08
+#define DW_LNS_fixed_advance_pc         0x09
+#define DW_LNS_set_prologue_end         0x0a /* DWARF3 */
+#define DW_LNS_set_epilogue_begin       0x0b /* DWARF3 */
+#define DW_LNS_set_isa                  0x0c /* DWARF3 */
+
+/* Line number extended opcode name. */
+#define DW_LNE_end_sequence             0x01
+#define DW_LNE_set_address              0x02
+#define DW_LNE_define_file              0x03
+#define DW_LNE_set_discriminator        0x04  /* DWARF4 */
+
+PRINTF_ARGS(static int kprintf(const char *fmt, ...), 1, 2);
+
+typedef struct line_info {
+    const char *dirname;
+    const char *filename;
+    const char *path; /* object path */
+    int line;
+
+    uintptr_t base_addr;
+    uintptr_t saddr;
+    const char *sname; /* function name */
+
+    struct line_info *next;
+} line_info_t;
+
+struct dwarf_section {
+    char *ptr;
+    size_t size;
+    uint64_t flags;
+};
+
+typedef struct obj_info {
+    const char *path; /* object path */
+    char *mapped;
+    size_t mapped_size;
+    void *uncompressed;
+    uintptr_t base_addr;
+    uintptr_t vmaddr;
+    struct dwarf_section debug_abbrev;
+    struct dwarf_section debug_info;
+    struct dwarf_section debug_line;
+    struct dwarf_section debug_ranges;
+    struct dwarf_section debug_str;
+    struct obj_info *next;
+} obj_info_t;
+
+#define DWARF_SECTION_COUNT 5
+
+static struct dwarf_section *
+obj_dwarf_section_at(obj_info_t *obj, int n)
+{
+    struct dwarf_section *ary[] = {
+        &obj->debug_abbrev,
+        &obj->debug_info,
+        &obj->debug_line,
+        &obj->debug_ranges,
+        &obj->debug_str
+    };
+    if (n < 0 || DWARF_SECTION_COUNT <= n) {
+        abort();
+    }
+    return ary[n];
+}
+
+struct debug_section_definition {
+    const char *name;
+    struct dwarf_section *dwarf;
+};
+
+/* Avoid consuming stack as this module may be used from signal handler */
+static char binary_filename[PATH_MAX + 1];
+
+static unsigned long
+uleb128(char **p)
+{
+    unsigned long r = 0;
+    int s = 0;
+    for (;;) {
+	unsigned char b = *(unsigned char *)(*p)++;
+	if (b < 0x80) {
+	    r += (unsigned long)b << s;
+	    break;
+	}
+	r += (b & 0x7f) << s;
+	s += 7;
+    }
+    return r;
+}
+
+static long
+sleb128(char **p)
+{
+    long r = 0;
+    int s = 0;
+    for (;;) {
+	unsigned char b = *(unsigned char *)(*p)++;
+	if (b < 0x80) {
+	    if (b & 0x40) {
+		r -= (0x80 - b) << s;
+	    }
+	    else {
+		r += (b & 0x3f) << s;
+	    }
+	    break;
+	}
+	r += (b & 0x7f) << s;
+	s += 7;
+    }
+    return r;
+}
+
+static const char *
+get_nth_dirname(unsigned long dir, char *p)
+{
+    if (!dir--) {
+	return "";
+    }
+    while (dir--) {
+	while (*p) p++;
+	p++;
+	if (!*p) {
+	    kprintf("Unexpected directory number %lu in %s\n",
+		    dir, binary_filename);
+	    return "";
+	}
+    }
+    return p;
+}
+
+static void
+fill_filename(int file, char *include_directories, char *filenames, line_info_t *line, obj_info_t *obj)
+{
+    int i;
+    char *p = filenames;
+    char *filename;
+    unsigned long dir;
+    for (i = 1; i <= file; i++) {
+	filename = p;
+	if (!*p) {
+	    /* Need to output binary file name? */
+	    kprintf("Unexpected file number %d in %s at %tx\n",
+		    file, binary_filename, filenames - obj->mapped);
+	    return;
+	}
+	while (*p) p++;
+	p++;
+	dir = uleb128(&p);
+	/* last modified. */
+	uleb128(&p);
+	/* size of the file. */
+	uleb128(&p);
+
+	if (i == file) {
+	    line->filename = filename;
+	    line->dirname = get_nth_dirname(dir, include_directories);
+	}
+    }
+}
+
+static void
+fill_line(int num_traces, void **traces, uintptr_t addr, int file, int line,
+	  char *include_directories, char *filenames,
+	  obj_info_t *obj, line_info_t *lines, int offset)
+{
+    int i;
+    addr += obj->base_addr - obj->vmaddr;
+    for (i = offset; i < num_traces; i++) {
+	uintptr_t a = (uintptr_t)traces[i];
+	/* We assume one line code doesn't result >100 bytes of native code.
+       We may want more reliable way eventually... */
+	if (addr < a && a < addr + 100) {
+	    fill_filename(file, include_directories, filenames, &lines[i], obj);
+	    lines[i].line = line;
+	}
+    }
+}
+
+struct LineNumberProgramHeader {
+    uint64_t unit_length;
+    uint16_t version;
+    uint8_t format; /* 4 or 8 */
+    uint64_t header_length;
+    uint8_t minimum_instruction_length;
+    uint8_t maximum_operations_per_instruction;
+    uint8_t default_is_stmt;
+    int8_t line_base;
+    uint8_t line_range;
+    uint8_t opcode_base;
+    /* uint8_t standard_opcode_lengths[opcode_base-1]; */
+    const char *include_directories;
+    const char *filenames;
+    const char *cu_start;
+    const char *cu_end;
+};
+
+static int
+parse_debug_line_header(const char **pp, struct LineNumberProgramHeader *header)
+{
+    const char *p = *pp;
+    header->unit_length = *(uint32_t *)p;
+    p += sizeof(uint32_t);
+
+    header->format = 4;
+    if (header->unit_length == 0xffffffff) {
+	header->unit_length = *(uint64_t *)p;
+	p += sizeof(uint64_t);
+        header->format = 8;
+    }
+
+    header->cu_end = p + header->unit_length;
+
+    header->version = *(uint16_t *)p;
+    p += sizeof(uint16_t);
+    if (header->version > 4) return -1;
+
+    header->header_length = header->format == 4 ? *(uint32_t *)p : *(uint64_t *)p;
+    p += header->format;
+    header->cu_start = p + header->header_length;
+
+    header->minimum_instruction_length = *(uint8_t *)p++;
+
+    if (header->version >= 4) {
+        /* maximum_operations_per_instruction = *(uint8_t *)p; */
+        if (*p != 1) return -1; /* For non-VLIW architectures, this field is 1 */
+        p++;
+    }
+
+    header->default_is_stmt = *(uint8_t *)p++;
+    header->line_base = *(int8_t *)p++;
+    header->line_range = *(uint8_t *)p++;
+    header->opcode_base = *(uint8_t *)p++;
+    /* header->standard_opcode_lengths = (uint8_t *)p - 1; */
+    p += header->opcode_base - 1;
+
+    header->include_directories = p;
+
+    /* temporary measure for compress-debug-sections */
+    if (p >= header->cu_end) return -1;
+
+    /* skip include directories */
+    while (*p) {
+	p = memchr(p, '\0', header->cu_end - p);
+	if (!p) return -1;
+	p++;
+    }
+    p++;
+
+    header->filenames = p;
+
+    *pp = header->cu_start;
+
+    return 0;
+}
+
+static int
+parse_debug_line_cu(int num_traces, void **traces, char **debug_line,
+		obj_info_t *obj, line_info_t *lines, int offset)
+{
+    const char *p = (const char *)*debug_line;
+    struct LineNumberProgramHeader header;
+
+    /* The registers. */
+    unsigned long addr = 0;
+    unsigned int file = 1;
+    unsigned int line = 1;
+    /* unsigned int column = 0; */
+    int is_stmt;
+    /* int basic_block = 0; */
+    /* int end_sequence = 0; */
+    /* int prologue_end = 0; */
+    /* int epilogue_begin = 0; */
+    /* unsigned int isa = 0; */
+
+    if (parse_debug_line_header(&p, &header))
+        return -1;
+    is_stmt = header.default_is_stmt;
+
+#define FILL_LINE()						    \
+    do {							    \
+	fill_line(num_traces, traces, addr, file, line,		    \
+                  (char *)header.include_directories,               \
+                  (char *)header.filenames,                         \
+		  obj, lines, offset);				    \
+	/*basic_block = prologue_end = epilogue_begin = 0;*/	    \
+    } while (0)
+
+    while (p < header.cu_end) {
+	unsigned long a;
+	unsigned char op = *p++;
+	switch (op) {
+	case DW_LNS_copy:
+	    FILL_LINE();
+	    break;
+	case DW_LNS_advance_pc:
+	    a = uleb128((char **)&p);
+	    addr += a;
+	    break;
+	case DW_LNS_advance_line: {
+	    long a = sleb128((char **)&p);
+	    line += a;
+	    break;
+	}
+	case DW_LNS_set_file:
+	    file = (unsigned int)uleb128((char **)&p);
+	    break;
+	case DW_LNS_set_column:
+	    /*column = (unsigned int)*/(void)uleb128((char **)&p);
+	    break;
+	case DW_LNS_negate_stmt:
+	    is_stmt = !is_stmt;
+	    break;
+	case DW_LNS_set_basic_block:
+	    /*basic_block = 1; */
+	    break;
+	case DW_LNS_const_add_pc:
+	    a = ((255UL - header.opcode_base) / header.line_range) *
+		header.minimum_instruction_length;
+	    addr += a;
+	    break;
+	case DW_LNS_fixed_advance_pc:
+	    a = *(unsigned char *)p++;
+	    addr += a;
+	    break;
+	case DW_LNS_set_prologue_end:
+	    /* prologue_end = 1; */
+	    break;
+	case DW_LNS_set_epilogue_begin:
+	    /* epilogue_begin = 1; */
+	    break;
+	case DW_LNS_set_isa:
+	    /* isa = (unsigned int)*/(void)uleb128((char **)&p);
+	    break;
+	case 0:
+	    a = *(unsigned char *)p++;
+	    op = *p++;
+	    switch (op) {
+	    case DW_LNE_end_sequence:
+		/* end_sequence = 1; */
+		FILL_LINE();
+		addr = 0;
+		file = 1;
+		line = 1;
+		/* column = 0; */
+		is_stmt = header.default_is_stmt;
+		/* end_sequence = 0; */
+		/* isa = 0; */
+		break;
+	    case DW_LNE_set_address:
+		addr = *(unsigned long *)p;
+		p += sizeof(unsigned long);
+		break;
+	    case DW_LNE_define_file:
+		kprintf("Unsupported operation in %s\n",
+			binary_filename);
+		break;
+	    case DW_LNE_set_discriminator:
+		/* TODO:currently ignore */
+		uleb128((char **)&p);
+		break;
+	    default:
+		kprintf("Unknown extended opcode: %d in %s\n",
+			op, binary_filename);
+	    }
+	    break;
+	default: {
+            uint8_t adjusted_opcode = op - header.opcode_base;
+            uint8_t operation_advance = adjusted_opcode / header.line_range;
+            /* NOTE: this code doesn't support VLIW */
+            addr += operation_advance * header.minimum_instruction_length;
+            line += header.line_base + (adjusted_opcode % header.line_range);
+	    FILL_LINE();
+	}
+	}
+    }
+    *debug_line = (char *)p;
+    return 0;
+}
+
+static int
+parse_debug_line(int num_traces, void **traces,
+		 char *debug_line, unsigned long size,
+		 obj_info_t *obj, line_info_t *lines, int offset)
+{
+    char *debug_line_end = debug_line + size;
+    while (debug_line < debug_line_end) {
+	if (parse_debug_line_cu(num_traces, traces, &debug_line, obj, lines, offset))
+	    return -1;
+    }
+    if (debug_line != debug_line_end) {
+	kprintf("Unexpected size of .debug_line in %s\n",
+		binary_filename);
+    }
+    return 0;
+}
+
+/* read file and fill lines */
+static uintptr_t
+fill_lines(int num_traces, void **traces, int check_debuglink,
+	   obj_info_t **objp, line_info_t *lines, int offset);
+
+static void
+append_obj(obj_info_t **objp)
+{
+    obj_info_t *newobj = calloc(1, sizeof(obj_info_t));
+    if (*objp) (*objp)->next = newobj;
+    *objp = newobj;
+}
+
+#ifdef USE_ELF
+static void
+follow_debuglink(const char *debuglink, int num_traces, void **traces,
+		 obj_info_t **objp, line_info_t *lines, int offset)
+{
+    /* Ideally we should check 4 paths to follow gnu_debuglink,
+       but we handle only one case for now as this format is used
+       by some linux distributions. See GDB's info for detail. */
+    static const char global_debug_dir[] = "/usr/lib/debug";
+    const size_t global_debug_dir_len = sizeof(global_debug_dir) - 1;
+    char *p;
+    obj_info_t *o1 = *objp, *o2;
+    size_t len;
+
+    p = strrchr(binary_filename, '/');
+    if (!p) {
+	return;
+    }
+    p[1] = '\0';
+
+    len = strlen(binary_filename);
+    if (len >= PATH_MAX - global_debug_dir_len)
+	len = PATH_MAX - global_debug_dir_len - 1;
+    memmove(binary_filename + global_debug_dir_len, binary_filename, len);
+    memcpy(binary_filename, global_debug_dir, global_debug_dir_len);
+    len += global_debug_dir_len;
+    strlcpy(binary_filename + len, debuglink, PATH_MAX - len);
+
+    append_obj(objp);
+    o2 = *objp;
+    o2->base_addr = o1->base_addr;
+    o2->path = o1->path;
+    fill_lines(num_traces, traces, 0, objp, lines, offset);
+}
+#endif
+
+enum
+{
+    DW_TAG_compile_unit = 0x11,
+    DW_TAG_inlined_subroutine = 0x1d,
+    DW_TAG_subprogram = 0x2e,
+};
+
+/* Attributes encodings */
+enum
+{
+    DW_AT_sibling = 0x01,
+    DW_AT_location = 0x02,
+    DW_AT_name = 0x03,
+    /* Reserved 0x04 */
+    /* Reserved 0x05 */
+    /* Reserved 0x06 */
+    /* Reserved 0x07 */
+    /* Reserved 0x08 */
+    DW_AT_ordering = 0x09,
+    /* Reserved 0x0a */
+    DW_AT_byte_size = 0x0b,
+    /* Reserved 0x0c */
+    DW_AT_bit_size = 0x0d,
+    /* Reserved 0x0e */
+    /* Reserved 0x0f */
+    DW_AT_stmt_list = 0x10,
+    DW_AT_low_pc = 0x11,
+    DW_AT_high_pc = 0x12,
+    DW_AT_language = 0x13,
+    /* Reserved 0x14 */
+    DW_AT_discr = 0x15,
+    DW_AT_discr_value = 0x16,
+    DW_AT_visibility = 0x17,
+    DW_AT_import = 0x18,
+    DW_AT_string_length = 0x19,
+    DW_AT_common_reference = 0x1a,
+    DW_AT_comp_dir = 0x1b,
+    DW_AT_const_value = 0x1c,
+    DW_AT_containing_type = 0x1d,
+    DW_AT_default_value = 0x1e,
+    /* Reserved 0x1f */
+    DW_AT_inline = 0x20,
+    DW_AT_is_optional = 0x21,
+    DW_AT_lower_bound = 0x22,
+    /* Reserved 0x23 */
+    /* Reserved 0x24 */
+    DW_AT_producer = 0x25,
+    /* Reserved 0x26 */
+    DW_AT_prototyped = 0x27,
+    /* Reserved 0x28 */
+    /* Reserved 0x29 */
+    DW_AT_return_addr = 0x2a,
+    /* Reserved 0x2b */
+    DW_AT_start_scope = 0x2c,
+    /* Reserved 0x2d */
+    DW_AT_bit_stride = 0x2e,
+    DW_AT_upper_bound = 0x2f,
+    /* Reserved 0x30 */
+    DW_AT_abstract_origin = 0x31,
+    DW_AT_accessibility = 0x32,
+    DW_AT_address_class = 0x33,
+    DW_AT_artificial = 0x34,
+    DW_AT_base_types = 0x35,
+    DW_AT_calling_convention = 0x36,
+    DW_AT_count = 0x37,
+    DW_AT_data_member_location = 0x38,
+    DW_AT_decl_column = 0x39,
+    DW_AT_decl_file = 0x3a,
+    DW_AT_decl_line = 0x3b,
+    DW_AT_declaration = 0x3c,
+    DW_AT_discr_list = 0x3d,
+    DW_AT_encoding = 0x3e,
+    DW_AT_external = 0x3f,
+    DW_AT_frame_base = 0x40,
+    DW_AT_friend = 0x41,
+    DW_AT_identifier_case = 0x42,
+    /* Reserved 0x43 */
+    DW_AT_namelist_item = 0x44,
+    DW_AT_priority = 0x45,
+    DW_AT_segment = 0x46,
+    DW_AT_specification = 0x47,
+    DW_AT_static_link = 0x48,
+    DW_AT_type = 0x49,
+    DW_AT_use_location = 0x4a,
+    DW_AT_variable_parameter = 0x4b,
+    DW_AT_virtuality = 0x4c,
+    DW_AT_vtable_elem_location = 0x4d,
+    DW_AT_allocated = 0x4e,
+    DW_AT_associated = 0x4f,
+    DW_AT_data_location = 0x50,
+    DW_AT_byte_stride = 0x51,
+    DW_AT_entry_pc = 0x52,
+    DW_AT_use_UTF8 = 0x53,
+    DW_AT_extension = 0x54,
+    DW_AT_ranges = 0x55,
+    DW_AT_trampoline = 0x56,
+    DW_AT_call_column = 0x57,
+    DW_AT_call_file = 0x58,
+    DW_AT_call_line = 0x59,
+    DW_AT_description = 0x5a,
+    DW_AT_binary_scale = 0x5b,
+    DW_AT_decimal_scale = 0x5c,
+    DW_AT_small = 0x5d,
+    DW_AT_decimal_sign = 0x5e,
+    DW_AT_digit_count = 0x5f,
+    DW_AT_picture_string = 0x60,
+    DW_AT_mutable = 0x61,
+    DW_AT_threads_scaled = 0x62,
+    DW_AT_explicit = 0x63,
+    DW_AT_object_pointer = 0x64,
+    DW_AT_endianity = 0x65,
+    DW_AT_elemental = 0x66,
+    DW_AT_pure = 0x67,
+    DW_AT_recursive = 0x68,
+    DW_AT_signature = 0x69,
+    DW_AT_main_subprogram = 0x6a,
+    DW_AT_data_bit_offset = 0x6b,
+    DW_AT_const_expr = 0x6c,
+    DW_AT_enum_class = 0x6d,
+    DW_AT_linkage_name = 0x6e,
+    DW_AT_string_length_bit_size = 0x6f,
+    DW_AT_string_length_byte_size = 0x70,
+    DW_AT_rank = 0x71,
+    DW_AT_str_offsets_base = 0x72,
+    DW_AT_addr_base = 0x73,
+    DW_AT_rnglists_base = 0x74,
+    /* Reserved 0x75 */
+    DW_AT_dwo_name = 0x76,
+    DW_AT_reference = 0x77,
+    DW_AT_rvalue_reference = 0x78,
+    DW_AT_macros = 0x79,
+    DW_AT_call_all_calls = 0x7a,
+    DW_AT_call_all_source_calls = 0x7b,
+    DW_AT_call_all_tail_calls = 0x7c,
+    DW_AT_call_return_pc = 0x7d,
+    DW_AT_call_value = 0x7e,
+    DW_AT_call_origin = 0x7f,
+    DW_AT_call_parameter = 0x80,
+    DW_AT_call_pc = 0x81,
+    DW_AT_call_tail_call = 0x82,
+    DW_AT_call_target = 0x83,
+    DW_AT_call_target_clobbered = 0x84,
+    DW_AT_call_data_location = 0x85,
+    DW_AT_call_data_value = 0x86,
+    DW_AT_noreturn = 0x87,
+    DW_AT_alignment = 0x88,
+    DW_AT_export_symbols = 0x89,
+    DW_AT_deleted = 0x8a,
+    DW_AT_defaulted = 0x8b,
+    DW_AT_loclists_base = 0x8c,
+    DW_AT_lo_user = 0x2000,
+    DW_AT_hi_user = 0x3fff
+};
+
+/* Attribute form encodings */
+enum
+{
+    DW_FORM_addr = 0x01,
+    /* Reserved 0x02 */
+    DW_FORM_block2 = 0x03,
+    DW_FORM_block4 = 0x04,
+    DW_FORM_data2 = 0x05,
+    DW_FORM_data4 = 0x06,
+    DW_FORM_data8 = 0x07,
+    DW_FORM_string = 0x08,
+    DW_FORM_block = 0x09,
+    DW_FORM_block1 = 0x0a,
+    DW_FORM_data1 = 0x0b,
+    DW_FORM_flag = 0x0c,
+    DW_FORM_sdata = 0x0d,
+    DW_FORM_strp = 0x0e,
+    DW_FORM_udata = 0x0f,
+    DW_FORM_ref_addr = 0x10,
+    DW_FORM_ref1 = 0x11,
+    DW_FORM_ref2 = 0x12,
+    DW_FORM_ref4 = 0x13,
+    DW_FORM_ref8 = 0x14,
+    DW_FORM_ref_udata = 0x15,
+    DW_FORM_indirect = 0x16,
+    DW_FORM_sec_offset = 0x17,
+    DW_FORM_exprloc = 0x18,
+    DW_FORM_flag_present = 0x19,
+    DW_FORM_strx = 0x1a,
+    DW_FORM_addrx = 0x1b,
+    DW_FORM_ref_sup4 = 0x1c,
+    DW_FORM_strp_sup = 0x1d,
+    DW_FORM_data16 = 0x1e,
+    DW_FORM_line_strp = 0x1f,
+    DW_FORM_ref_sig8 = 0x20,
+    DW_FORM_implicit_const = 0x21,
+    DW_FORM_loclistx = 0x22,
+    DW_FORM_rnglistx = 0x23,
+    DW_FORM_ref_sup8 = 0x24,
+    DW_FORM_strx1 = 0x25,
+    DW_FORM_strx2 = 0x26,
+    DW_FORM_strx3 = 0x27,
+    DW_FORM_strx4 = 0x28,
+    DW_FORM_addrx1 = 0x29,
+    DW_FORM_addrx2 = 0x2a,
+    DW_FORM_addrx3 = 0x2b,
+    DW_FORM_addrx4 = 0x2c
+};
+
+enum {
+    VAL_none = 0,
+    VAL_cstr = 1,
+    VAL_data = 2,
+    VAL_uint = 3,
+    VAL_int = 4
+};
+
+# define ABBREV_TABLE_SIZE 256
+typedef struct {
+    obj_info_t *obj;
+    char *file;
+    char *current_cu;
+    uint64_t current_low_pc;
+    char *debug_line_cu_end;
+    char *debug_line_files;
+    char *debug_line_directories;
+    char *p;
+    char *cu_end;
+    char *pend;
+    char *q0;
+    char *q;
+    int format; // 4 or 8
+    uint8_t address_size;
+    int level;
+    char *abbrev_table[ABBREV_TABLE_SIZE];
+} DebugInfoReader;
+
+typedef struct {
+    ptrdiff_t pos;
+    int tag;
+    int has_children;
+} DIE;
+
+typedef struct {
+    union {
+        char *ptr;
+        uint64_t uint64;
+        int64_t int64;
+    } as;
+    uint64_t off;
+    uint64_t at;
+    uint64_t form;
+    size_t size;
+    int type;
+} DebugInfoValue;
+
+/* TODO: Big Endian */
+#define MERGE_2INTS(a,b,sz) (((uint64_t)(b)<<sz)|(a))
+
+static uint16_t
+get_uint16(const uint8_t *p)
+{
+    return (uint16_t)MERGE_2INTS(p[0],p[1],8);
+}
+
+static uint32_t
+get_uint32(const uint8_t *p)
+{
+    return (uint32_t)MERGE_2INTS(get_uint16(p),get_uint16(p+2),16);
+}
+
+static uint64_t
+get_uint64(const uint8_t *p)
+{
+    return MERGE_2INTS(get_uint32(p),get_uint32(p+4),32);
+}
+
+static uint8_t
+read_uint8(char **ptr)
+{
+    const unsigned char *p = (const unsigned char *)*ptr;
+    *ptr = (char *)(p + 1);
+    return *p;
+}
+
+static uint16_t
+read_uint16(char **ptr)
+{
+    const unsigned char *p = (const unsigned char *)*ptr;
+    *ptr = (char *)(p + 2);
+    return get_uint16(p);
+}
+
+static uint32_t
+read_uint24(char **ptr)
+{
+    const unsigned char *p = (const unsigned char *)*ptr;
+    *ptr = (char *)(p + 3);
+    return (*p << 16) | get_uint16(p+1);
+}
+
+static uint32_t
+read_uint32(char **ptr)
+{
+    const unsigned char *p = (const unsigned char *)*ptr;
+    *ptr = (char *)(p + 4);
+    return get_uint32(p);
+}
+
+static uint64_t
+read_uint64(char **ptr)
+{
+    const unsigned char *p = (const unsigned char *)*ptr;
+    *ptr = (char *)(p + 8);
+    return get_uint64(p);
+}
+
+static uintptr_t
+read_uintptr(char **ptr)
+{
+    const unsigned char *p = (const unsigned char *)*ptr;
+    *ptr = (char *)(p + SIZEOF_VOIDP);
+#if SIZEOF_VOIDP == 8
+    return get_uint64(p);
+#else
+    return get_uint32(p);
+#endif
+}
+
+static uint64_t
+read_uint(DebugInfoReader *reader)
+{
+    if (reader->format == 4) {
+        return read_uint32(&reader->p);
+    } else { /* 64 bit */
+        return read_uint64(&reader->p);
+    }
+}
+
+static uint64_t
+read_uleb128(DebugInfoReader *reader)
+{
+    return uleb128(&reader->p);
+}
+
+static int64_t
+read_sleb128(DebugInfoReader *reader)
+{
+    return sleb128(&reader->p);
+}
+
+static void
+debug_info_reader_init(DebugInfoReader *reader, obj_info_t *obj)
+{
+    reader->file = obj->mapped;
+    reader->obj = obj;
+    reader->p = obj->debug_info.ptr;
+    reader->pend = obj->debug_info.ptr + obj->debug_info.size;
+    reader->debug_line_cu_end = obj->debug_line.ptr;
+}
+
+static void
+di_read_debug_abbrev_cu(DebugInfoReader *reader)
+{
+    uint64_t prev = 0;
+    char *p = reader->q0;
+    for (;;) {
+        uint64_t abbrev_number = uleb128(&p);
+        if (abbrev_number <= prev) break;
+        if (abbrev_number < ABBREV_TABLE_SIZE) {
+            reader->abbrev_table[abbrev_number] = p;
+        }
+        prev = abbrev_number;
+        uleb128(&p); /* tag */
+        p++; /* has_children */
+        /* skip content */
+        for (;;) {
+            uint64_t at = uleb128(&p);
+            uint64_t form = uleb128(&p);
+            if (!at && !form) break;
+        }
+    }
+}
+
+static int
+di_read_debug_line_cu(DebugInfoReader *reader)
+{
+    const char *p;
+    struct LineNumberProgramHeader header;
+
+    p = (const char *)reader->debug_line_cu_end;
+    if (parse_debug_line_header(&p, &header))
+        return -1;
+
+    reader->debug_line_cu_end = (char *)header.cu_end;
+    reader->debug_line_directories = (char *)header.include_directories;
+    reader->debug_line_files = (char *)header.filenames;
+
+    return 0;
+}
+
+static void
+set_uint_value(DebugInfoValue *v, uint64_t n)
+{
+    v->as.uint64 = n;
+    v->type = VAL_uint;
+}
+
+static void
+set_int_value(DebugInfoValue *v, int64_t n)
+{
+    v->as.int64 = n;
+    v->type = VAL_int;
+}
+
+static void
+set_cstr_value(DebugInfoValue *v, char *s)
+{
+    v->as.ptr = s;
+    v->off = 0;
+    v->type = VAL_cstr;
+}
+
+static void
+set_cstrp_value(DebugInfoValue *v, char *s, uint64_t off)
+{
+    v->as.ptr = s;
+    v->off = off;
+    v->type = VAL_cstr;
+}
+
+static void
+set_data_value(DebugInfoValue *v, char *s)
+{
+    v->as.ptr = s;
+    v->type = VAL_data;
+}
+
+static const char *
+get_cstr_value(DebugInfoValue *v)
+{
+    if (v->as.ptr) {
+        return v->as.ptr + v->off;
+    } else {
+        return NULL;
+    }
+}
+
+static void
+debug_info_reader_read_value(DebugInfoReader *reader, uint64_t form, DebugInfoValue *v)
+{
+    switch (form) {
+      case DW_FORM_addr:
+        if (reader->address_size == 4) {
+            set_uint_value(v, read_uint32(&reader->p));
+        } else if (reader->address_size == 8) {
+            set_uint_value(v, read_uint64(&reader->p));
+        } else {
+            fprintf(stderr,"unknown address_size:%d", reader->address_size);
+            abort();
+        }
+        break;
+      case DW_FORM_block2:
+        v->size = read_uint16(&reader->p);
+        set_data_value(v, reader->p);
+        reader->p += v->size;
+        break;
+      case DW_FORM_block4:
+        v->size = read_uint32(&reader->p);
+        set_data_value(v, reader->p);
+        reader->p += v->size;
+        break;
+      case DW_FORM_data2:
+        set_uint_value(v, read_uint16(&reader->p));
+        break;
+      case DW_FORM_data4:
+        set_uint_value(v, read_uint32(&reader->p));
+        break;
+      case DW_FORM_data8:
+        set_uint_value(v, read_uint64(&reader->p));
+        break;
+      case DW_FORM_string:
+        v->size = strlen(reader->p);
+        set_cstr_value(v, reader->p);
+        reader->p += v->size + 1;
+        break;
+      case DW_FORM_block:
+        v->size = uleb128(&reader->p);
+        set_data_value(v, reader->p);
+        reader->p += v->size;
+        break;
+      case DW_FORM_block1:
+        v->size = read_uint8(&reader->p);
+        set_data_value(v, reader->p);
+        reader->p += v->size;
+        break;
+      case DW_FORM_data1:
+        set_uint_value(v, read_uint8(&reader->p));
+        break;
+      case DW_FORM_flag:
+        set_uint_value(v, read_uint8(&reader->p));
+        break;
+      case DW_FORM_sdata:
+        set_int_value(v, read_sleb128(reader));
+        break;
+      case DW_FORM_strp:
+        set_cstrp_value(v, reader->obj->debug_str.ptr, read_uint(reader));
+        break;
+      case DW_FORM_udata:
+        set_uint_value(v, read_uleb128(reader));
+        break;
+      case DW_FORM_ref_addr:
+        if (reader->address_size == 4) {
+            set_uint_value(v, read_uint32(&reader->p));
+        } else if (reader->address_size == 8) {
+            set_uint_value(v, read_uint64(&reader->p));
+        } else {
+            fprintf(stderr,"unknown address_size:%d", reader->address_size);
+            abort();
+        }
+        break;
+      case DW_FORM_ref1:
+        set_uint_value(v, read_uint8(&reader->p));
+        break;
+      case DW_FORM_ref2:
+        set_uint_value(v, read_uint16(&reader->p));
+        break;
+      case DW_FORM_ref4:
+        set_uint_value(v, read_uint32(&reader->p));
+        break;
+      case DW_FORM_ref8:
+        set_uint_value(v, read_uint64(&reader->p));
+        break;
+      case DW_FORM_ref_udata:
+        set_uint_value(v, uleb128(&reader->p));
+        break;
+      case DW_FORM_indirect:
+        /* TODO: read the referred value */
+        set_uint_value(v, uleb128(&reader->p));
+        break;
+      case DW_FORM_sec_offset:
+        set_uint_value(v, read_uint(reader)); /* offset */
+        /* addrptr: debug_addr */
+        /* lineptr: debug_line */
+        /* loclist: debug_loclists */
+        /* loclistptr: debug_loclists */
+        /* macptr: debug_macro */
+        /* rnglist: debug_rnglists */
+        /* rnglistptr: debug_rnglists */
+        /* stroffsetsptr: debug_str_offsets */
+        break;
+      case DW_FORM_exprloc:
+        v->size = (size_t)read_uleb128(reader);
+        set_data_value(v, reader->p);
+        reader->p += v->size;
+        break;
+      case DW_FORM_flag_present:
+        set_uint_value(v, 1);
+        break;
+      case DW_FORM_strx:
+        set_uint_value(v, uleb128(&reader->p));
+        break;
+      case DW_FORM_addrx:
+        /* TODO: read .debug_addr */
+        set_uint_value(v, uleb128(&reader->p));
+        break;
+      case DW_FORM_ref_sup4:
+        set_uint_value(v, read_uint32(&reader->p));
+        break;
+      case DW_FORM_strp_sup:
+        set_uint_value(v, read_uint(reader));
+        /* *p = reader->sup_file + reader->sup_str->sh_offset + ret; */
+        break;
+      case DW_FORM_data16:
+        v->size = 16;
+        set_data_value(v, reader->p);
+        reader->p += v->size;
+        break;
+      case DW_FORM_line_strp:
+        set_uint_value(v, read_uint(reader));
+        /* *p = reader->file + reader->line->sh_offset + ret; */
+        break;
+      case DW_FORM_ref_sig8:
+        set_uint_value(v, read_uint64(&reader->p));
+        break;
+      case DW_FORM_implicit_const:
+        set_int_value(v, sleb128(&reader->q));
+        break;
+      case DW_FORM_loclistx:
+        set_uint_value(v, read_uleb128(reader));
+        break;
+      case DW_FORM_rnglistx:
+        set_uint_value(v, read_uleb128(reader));
+        break;
+      case DW_FORM_ref_sup8:
+        set_uint_value(v, read_uint64(&reader->p));
+        break;
+      case DW_FORM_strx1:
+        set_uint_value(v, read_uint8(&reader->p));
+        break;
+      case DW_FORM_strx2:
+        set_uint_value(v, read_uint16(&reader->p));
+        break;
+      case DW_FORM_strx3:
+        set_uint_value(v, read_uint24(&reader->p));
+        break;
+      case DW_FORM_strx4:
+        set_uint_value(v, read_uint32(&reader->p));
+        break;
+      case DW_FORM_addrx1:
+        set_uint_value(v, read_uint8(&reader->p));
+        break;
+      case DW_FORM_addrx2:
+        set_uint_value(v, read_uint16(&reader->p));
+        break;
+      case DW_FORM_addrx3:
+        set_uint_value(v, read_uint24(&reader->p));
+        break;
+      case DW_FORM_addrx4:
+        set_uint_value(v, read_uint32(&reader->p));
+        break;
+      case 0:
+        goto fail;
+        break;
+    }
+    return;
+
+  fail:
+    fprintf(stderr, "%d: unsupported form: %#"PRIx64"\n", __LINE__, form);
+    exit(1);
+}
+
+/* find abbrev in current compilation unit */
+static char *
+di_find_abbrev(DebugInfoReader *reader, uint64_t abbrev_number)
+{
+    char *p;
+    if (abbrev_number < ABBREV_TABLE_SIZE) {
+        return reader->abbrev_table[abbrev_number];
+    }
+    p = reader->abbrev_table[ABBREV_TABLE_SIZE-1];
+    /* skip 255th record */
+    uleb128(&p); /* tag */
+    p++; /* has_children */
+    /* skip content */
+    for (;;) {
+        uint64_t at = uleb128(&p);
+        uint64_t form = uleb128(&p);
+        if (!at && !form) break;
+    }
+    for (uint64_t n = uleb128(&p); abbrev_number != n; n = uleb128(&p)) {
+        if (n == 0) {
+            fprintf(stderr,"%d: Abbrev Number %"PRId64" not found\n",__LINE__, abbrev_number);
+            exit(1);
+        }
+        uleb128(&p); /* tag */
+        p++; /* has_children */
+        /* skip content */
+        for (;;) {
+            uint64_t at = uleb128(&p);
+            uint64_t form = uleb128(&p);
+            if (!at && !form) break;
+        }
+    }
+    return p;
+}
+
+#if 0
+static void
+hexdump0(const unsigned char *p, size_t n)
+{
+    size_t i;
+    fprintf(stderr, "     0  1  2  3  4  5  6  7  8  9  A  B  C  D  E  F\n");
+    for (i=0; i < n; i++){
+        switch (i & 15) {
+          case 0:
+            fprintf(stderr, "%02zd: %02X ", i/16, p[i]);
+            break;
+          case 15:
+            fprintf(stderr, "%02X\n", p[i]);
+            break;
+          default:
+            fprintf(stderr, "%02X ", p[i]);
+            break;
+        }
+    }
+    if ((i & 15) != 15) {
+        fprintf(stderr, "\n");
+    }
+}
+#define hexdump(p,n) hexdump0((const unsigned char *)p, n)
+
+static void
+div_inspect(DebugInfoValue *v)
+{
+    switch (v->type) {
+      case VAL_uint:
+        fprintf(stderr,"%d: type:%d size:%zx v:%lx\n",__LINE__,v->type,v->size,v->as.uint64);
+        break;
+      case VAL_int:
+        fprintf(stderr,"%d: type:%d size:%zx v:%ld\n",__LINE__,v->type,v->size,(int64_t)v->as.uint64);
+        break;
+      case VAL_cstr:
+        fprintf(stderr,"%d: type:%d size:%zx v:'%s'\n",__LINE__,v->type,v->size,v->as.ptr);
+        break;
+      case VAL_data:
+        fprintf(stderr,"%d: type:%d size:%zx v:\n",__LINE__,v->type,v->size);
+        hexdump(v->as.ptr, 16);
+        break;
+    }
+}
+#endif
+
+static DIE *
+di_read_die(DebugInfoReader *reader, DIE *die)
+{
+    uint64_t abbrev_number = uleb128(&reader->p);
+    if (abbrev_number == 0) {
+        reader->level--;
+        return NULL;
+    }
+
+    reader->q = di_find_abbrev(reader, abbrev_number);
+
+    die->pos = reader->p - reader->obj->debug_info.ptr - 1;
+    die->tag = (int)uleb128(&reader->q); /* tag */
+    die->has_children = *reader->q++; /* has_children */
+    if (die->has_children) {
+        reader->level++;
+    }
+    return die;
+}
+
+static DebugInfoValue *
+di_read_record(DebugInfoReader *reader, DebugInfoValue *vp)
+{
+    uint64_t at = uleb128(&reader->q);
+    uint64_t form = uleb128(&reader->q);
+    if (!at || !form) return NULL;
+    vp->at = at;
+    vp->form = form;
+    debug_info_reader_read_value(reader, form, vp);
+    return vp;
+}
+
+static void
+di_skip_records(DebugInfoReader *reader)
+{
+    for (;;) {
+        DebugInfoValue v = {{}};
+        uint64_t at = uleb128(&reader->q);
+        uint64_t form = uleb128(&reader->q);
+        if (!at || !form) return;
+        debug_info_reader_read_value(reader, form, &v);
+    }
+}
+
+typedef struct {
+    uint64_t low_pc;
+    uint64_t high_pc;
+    uint64_t ranges;
+    bool low_pc_set;
+    bool high_pc_set;
+    bool ranges_set;
+} ranges_t;
+
+static void
+ranges_set(ranges_t *ptr, DebugInfoValue *v)
+{
+    switch (v->at) {
+      case DW_AT_low_pc:
+        ptr->low_pc = v->as.uint64;
+        ptr->low_pc_set = true;
+        break;
+      case DW_AT_high_pc:
+        if (v->form == DW_FORM_addr) {
+            ptr->high_pc = v->as.uint64;
+        }
+        else {
+            ptr->high_pc = ptr->low_pc + v->as.uint64;
+        }
+        ptr->high_pc_set = true;
+        break;
+      case DW_AT_ranges:
+        ptr->ranges = v->as.uint64;
+        ptr->ranges_set = true;
+        break;
+    }
+}
+
+static uintptr_t
+ranges_include(DebugInfoReader *reader, ranges_t *ptr, uint64_t addr)
+{
+    if (ptr->high_pc_set) {
+        if (ptr->ranges_set || !ptr->low_pc_set) {
+            exit(1);
+        }
+        if (ptr->low_pc <= addr && addr <= ptr->high_pc) {
+            return (uintptr_t)ptr->low_pc;
+        }
+    }
+    else if (ptr->ranges_set) {
+        /* TODO: support base address selection entry */
+        char *p = reader->obj->debug_ranges.ptr + ptr->ranges;
+        uint64_t base = ptr->low_pc_set ? ptr->low_pc : reader->current_low_pc;
+        for (;;) {
+            uintptr_t from = read_uintptr(&p);
+            uintptr_t to = read_uintptr(&p);
+            if (!from && !to) break;
+            if (from == UINTPTR_MAX) {
+                /* base address selection entry */
+                base = to;
+            }
+            else if (base + from <= addr && addr < base + to) {
+                return from;
+            }
+        }
+    }
+    else if (ptr->low_pc_set) {
+        if (ptr->low_pc == addr) {
+            return (uintptr_t)ptr->low_pc;
+        }
+    }
+    return false;
+}
+
+#if 0
+static void
+ranges_inspect(DebugInfoReader *reader, ranges_t *ptr)
+{
+    if (ptr->high_pc_set) {
+        if (ptr->ranges_set || !ptr->low_pc_set) {
+            fprintf(stderr,"low_pc_set:%d high_pc_set:%d ranges_set:%d\n",ptr->low_pc_set,ptr->high_pc_set,ptr->ranges_set);
+            exit(1);
+        }
+        fprintf(stderr,"low_pc:%"PRIx64" high_pc:%"PRIx64"\n",ptr->low_pc,ptr->high_pc);
+    }
+    else if (ptr->ranges_set) {
+        char *p = reader->obj->debug_ranges.ptr + ptr->ranges;
+        fprintf(stderr,"low_pc:%"PRIx64" ranges:%"PRIx64" %lx ",ptr->low_pc,ptr->ranges, p-reader->obj->mapped);
+        for (;;) {
+            uintptr_t from = read_uintptr(&p);
+            uintptr_t to = read_uintptr(&p);
+            if (!from && !to) break;
+            fprintf(stderr,"%"PRIx64"-%"PRIx64" ",ptr->low_pc+from,ptr->low_pc+to);
+        }
+        fprintf(stderr,"\n");
+    }
+    else if (ptr->low_pc_set) {
+        fprintf(stderr,"low_pc:%"PRIx64"\n",ptr->low_pc);
+    }
+    else {
+        fprintf(stderr,"empty\n");
+    }
+}
+#endif
+
+static int
+di_read_cu(DebugInfoReader *reader)
+{
+    uint64_t unit_length;
+    uint16_t version;
+    uint64_t debug_abbrev_offset;
+    reader->format = 4;
+    reader->current_cu = reader->p;
+    unit_length = read_uint32(&reader->p);
+    if (unit_length == 0xffffffff) {
+        unit_length = read_uint64(&reader->p);
+        reader->format = 8;
+    }
+    reader->cu_end = reader->p + unit_length;
+    version = read_uint16(&reader->p);
+    if (version > 5) {
+        return -1;
+    }
+    else if (version == 5) {
+        /* unit_type = */ read_uint8(&reader->p);
+        reader->address_size = read_uint8(&reader->p);
+        debug_abbrev_offset = read_uint(reader);
+    }
+    else {
+        debug_abbrev_offset = read_uint(reader);
+        reader->address_size = read_uint8(&reader->p);
+    }
+    reader->q0 = reader->obj->debug_abbrev.ptr + debug_abbrev_offset;
+
+    reader->level = 0;
+    di_read_debug_abbrev_cu(reader);
+    if (di_read_debug_line_cu(reader)) return -1;
+
+#if defined(__GNUC__) && !defined(__clang__) && !defined(__INTEL_COMPILER_BUILD_DATE)
+    /* Though DWARF specifies "the applicable base address defaults to the base
+       address of the compilation unit", but GCC seems to use zero as default */
+#else
+    do {
+        DIE die;
+
+        if (!di_read_die(reader, &die)) continue;
+
+        if (die.tag != DW_TAG_compile_unit) {
+            di_skip_records(reader);
+            break;
+        }
+
+        /* enumerate abbrev */
+        for (;;) {
+            DebugInfoValue v = {{}};
+            if (!di_read_record(reader, &v)) break;
+            switch (v.at) {
+              case DW_AT_low_pc:
+                reader->current_low_pc = v.as.uint64;
+                break;
+            }
+        }
+    } while (0);
+#endif
+    return 0;
+}
+
+static void
+read_abstract_origin(DebugInfoReader *reader, uint64_t abstract_origin, line_info_t *line)
+{
+    char *p = reader->p;
+    char *q = reader->q;
+    int level = reader->level;
+    DIE die;
+
+    reader->p = reader->current_cu + abstract_origin;
+    if (!di_read_die(reader, &die)) goto finish;
+
+    /* enumerate abbrev */
+    for (;;) {
+        DebugInfoValue v = {{}};
+        if (!di_read_record(reader, &v)) break;
+        switch (v.at) {
+          case DW_AT_name:
+            line->sname = get_cstr_value(&v);
+            break;
+        }
+    }
+
+  finish:
+    reader->p = p;
+    reader->q = q;
+    reader->level = level;
+}
+
+static void
+debug_info_read(DebugInfoReader *reader, int num_traces, void **traces,
+         line_info_t *lines, int offset) {
+    while (reader->p < reader->cu_end) {
+        DIE die;
+        ranges_t ranges = {};
+        line_info_t line = {};
+
+        if (!di_read_die(reader, &die)) continue;
+        /* fprintf(stderr,"%d:%tx: <%d>\n",__LINE__,die.pos,reader->level,die.tag); */
+
+        if (die.tag != DW_TAG_subprogram && die.tag != DW_TAG_inlined_subroutine) {
+          skip_die:
+            di_skip_records(reader);
+            continue;
+        }
+
+        /* enumerate abbrev */
+        for (;;) {
+            DebugInfoValue v = {{}};
+            /* ptrdiff_t pos = reader->p - reader->p0; */
+            if (!di_read_record(reader, &v)) break;
+            /* fprintf(stderr,"\n%d:%tx: AT:%lx FORM:%lx\n",__LINE__,pos,v.at,v.form); */
+            /* div_inspect(&v); */
+            switch (v.at) {
+              case DW_AT_name:
+                line.sname = get_cstr_value(&v);
+                break;
+              case DW_AT_call_file:
+                fill_filename((int)v.as.uint64, reader->debug_line_directories, reader->debug_line_files, &line, reader->obj);
+                break;
+              case DW_AT_call_line:
+                line.line = (int)v.as.uint64;
+                break;
+              case DW_AT_low_pc:
+              case DW_AT_high_pc:
+              case DW_AT_ranges:
+                ranges_set(&ranges, &v);
+                break;
+              case DW_AT_declaration:
+                goto skip_die;
+              case DW_AT_inline:
+                /* 1 or 3 */
+                break; /* goto skip_die; */
+              case DW_AT_abstract_origin:
+                read_abstract_origin(reader, v.as.uint64, &line);
+                break; /* goto skip_die; */
+            }
+        }
+        /* ranges_inspect(reader, &ranges); */
+        /* fprintf(stderr,"%d:%tx: %x ",__LINE__,diepos,die.tag); */
+        for (int i=offset; i < num_traces; i++) {
+            uintptr_t addr = (uintptr_t)traces[i];
+            uintptr_t offset = addr - reader->obj->base_addr + reader->obj->vmaddr;
+            uintptr_t saddr = ranges_include(reader, &ranges, offset);
+            if (saddr) {
+                /* fprintf(stderr, "%d:%tx: %d %lx->%lx %x %s: %s/%s %d %s %s %s\n",__LINE__,die.pos, i,addr,offset, die.tag,line.sname,line.dirname,line.filename,line.line,reader->obj->path,line.sname,lines[i].sname); */
+                if (lines[i].sname) {
+                    line_info_t *lp = malloc(sizeof(line_info_t));
+                    memcpy(lp, &lines[i], sizeof(line_info_t));
+                    lines[i].next = lp;
+                    lp->dirname = line.dirname;
+                    lp->filename = line.filename;
+                    lp->line = line.line;
+                    lp->saddr = 0;
+                }
+                lines[i].path = reader->obj->path;
+                lines[i].base_addr = line.base_addr;
+                lines[i].sname = line.sname;
+                lines[i].saddr = saddr + reader->obj->base_addr - reader->obj->vmaddr;
+            }
+        }
+    }
+}
+
+#ifdef USE_ELF
+static unsigned long
+uncompress_debug_section(ElfW(Shdr) *shdr, char *file, char **ptr)
+{
+#ifdef SUPPORT_COMPRESSED_DEBUG_LINE
+    ElfW(Chdr) *chdr = (ElfW(Chdr) *)(file + shdr->sh_offset);
+    unsigned long destsize = chdr->ch_size;
+    int ret = 0;
+
+    if (chdr->ch_type != ELFCOMPRESS_ZLIB) {
+	/* unsupported compression type */
+	return 0;
+    }
+
+    *ptr = malloc(destsize);
+    if (!*ptr) return 0;
+    ret = uncompress((Bytef *)*ptr, &destsize,
+	    (const Bytef*)chdr + sizeof(ElfW(Chdr)),
+            shdr->sh_size - sizeof(ElfW(Chdr)));
+    if (ret != Z_OK) goto fail;
+    return destsize;
+
+fail:
+    free(*ptr);
+#endif
+    return 0;
+}
+
+/* read file and fill lines */
+static uintptr_t
+fill_lines(int num_traces, void **traces, int check_debuglink,
+	   obj_info_t **objp, line_info_t *lines, int offset)
+{
+    int i, j;
+    char *shstr;
+    ElfW(Ehdr) *ehdr;
+    ElfW(Shdr) *shdr, *shstr_shdr;
+    ElfW(Shdr) *gnu_debuglink_shdr = NULL;
+    int fd;
+    off_t filesize;
+    char *file;
+    ElfW(Shdr) *symtab_shdr = NULL, *strtab_shdr = NULL;
+    ElfW(Shdr) *dynsym_shdr = NULL, *dynstr_shdr = NULL;
+    obj_info_t *obj = *objp;
+    uintptr_t dladdr_fbase = 0;
+
+    fd = open(binary_filename, O_RDONLY);
+    if (fd < 0) {
+	goto fail;
+    }
+    filesize = lseek(fd, 0, SEEK_END);
+    if (filesize < 0) {
+	int e = errno;
+	close(fd);
+	kprintf("lseek: %s\n", strerror(e));
+	goto fail;
+    }
+#if SIZEOF_OFF_T > SIZEOF_SIZE_T
+    if (filesize > (off_t)SIZE_MAX) {
+	close(fd);
+	kprintf("Too large file %s\n", binary_filename);
+	goto fail;
+    }
+#endif
+    lseek(fd, 0, SEEK_SET);
+    /* async-signal unsafe */
+    file = (char *)mmap(NULL, (size_t)filesize, PROT_READ, MAP_SHARED, fd, 0);
+    if (file == MAP_FAILED) {
+	int e = errno;
+	close(fd);
+	kprintf("mmap: %s\n", strerror(e));
+	goto fail;
+    }
+    close(fd);
+
+    ehdr = (ElfW(Ehdr) *)file;
+    if (memcmp(ehdr->e_ident, "\177ELF", 4) != 0) {
+	/*
+	 * Huh? Maybe filename was overridden by setproctitle() and
+	 * it match non-elf file.
+	 */
+	goto fail;
+    }
+    obj->mapped = file;
+    obj->mapped_size = (size_t)filesize;
+
+    shdr = (ElfW(Shdr) *)(file + ehdr->e_shoff);
+
+    shstr_shdr = shdr + ehdr->e_shstrndx;
+    shstr = file + shstr_shdr->sh_offset;
+
+    for (i = 0; i < ehdr->e_shnum; i++) {
+        char *section_name = shstr + shdr[i].sh_name;
+	switch (shdr[i].sh_type) {
+	  case SHT_STRTAB:
+	    if (!strcmp(section_name, ".strtab")) {
+		strtab_shdr = shdr + i;
+	    }
+	    else if (!strcmp(section_name, ".dynstr")) {
+		dynstr_shdr = shdr + i;
+	    }
+	    break;
+	  case SHT_SYMTAB:
+	    /* if (!strcmp(section_name, ".symtab")) */
+	    symtab_shdr = shdr + i;
+	    break;
+	  case SHT_DYNSYM:
+	    /* if (!strcmp(section_name, ".dynsym")) */
+	    dynsym_shdr = shdr + i;
+	    break;
+	  case SHT_PROGBITS:
+	    if (!strcmp(section_name, ".gnu_debuglink")) {
+		gnu_debuglink_shdr = shdr + i;
+	    }
+            else {
+                const char *debug_section_names[] = {
+                    ".debug_abbrev",
+                    ".debug_info",
+                    ".debug_line",
+                    ".debug_ranges",
+                    ".debug_str"
+                };
+
+                for (j=0; j < DWARF_SECTION_COUNT; j++) {
+                    struct dwarf_section *s = obj_dwarf_section_at(obj, j);
+
+                    if (strcmp(section_name, debug_section_names[j]) != 0)
+                        continue;
+
+                    s->ptr = file + shdr[i].sh_offset;
+                    s->size = shdr[i].sh_size;
+                    s->flags = shdr[i].sh_flags;
+                    if (s->flags & SHF_COMPRESSED) {
+                        s->size = uncompress_debug_section(&shdr[i], file, &s->ptr);
+                        if (!s->size) goto fail;
+                    }
+                    break;
+                }
+            }
+	    break;
+	}
+    }
+
+    if (offset == -1) {
+	/* main executable */
+	offset = 0;
+	if (dynsym_shdr && dynstr_shdr) {
+	    char *strtab = file + dynstr_shdr->sh_offset;
+	    ElfW(Sym) *symtab = (ElfW(Sym) *)(file + dynsym_shdr->sh_offset);
+	    int symtab_count = (int)(dynsym_shdr->sh_size / sizeof(ElfW(Sym)));
+            void *handle = dlopen(NULL, RTLD_NOW|RTLD_LOCAL);
+            if (handle) {
+                for (j = 0; j < symtab_count; j++) {
+                    ElfW(Sym) *sym = &symtab[j];
+                    Dl_info info;
+                    void *s;
+                    if (ELF_ST_TYPE(sym->st_info) != STT_FUNC || sym->st_size == 0) continue;
+                    s = dlsym(handle, strtab + sym->st_name);
+                    if (s && dladdr(s, &info)) {
+                        obj->base_addr = dladdr_fbase;
+                        dladdr_fbase = (uintptr_t)info.dli_fbase;
+                        break;
+                    }
+                }
+                dlclose(handle);
+            }
+	    if (ehdr->e_type == ET_EXEC) {
+		obj->base_addr = 0;
+	    }
+	    else {
+		/* PIE (position-independent executable) */
+		obj->base_addr = dladdr_fbase;
+	    }
+	}
+    }
+
+    if (obj->debug_info.ptr && obj->debug_abbrev.ptr) {
+        DebugInfoReader reader;
+        debug_info_reader_init(&reader, obj);
+        i = 0;
+        while (reader.p < reader.pend) {
+            /* fprintf(stderr, "%d:%tx: CU[%d]\n", __LINE__, reader.p - reader.obj->debug_info.ptr, i++); */
+            if (di_read_cu(&reader)) goto use_symtab;
+            debug_info_read(&reader, num_traces, traces, lines, offset);
+        }
+    }
+    else {
+        /* This file doesn't have dwarf, use symtab or dynsym */
+use_symtab:
+        if (!symtab_shdr) {
+            /* This file doesn't have symtab, use dynsym instead */
+            symtab_shdr = dynsym_shdr;
+            strtab_shdr = dynstr_shdr;
+        }
+
+        if (symtab_shdr && strtab_shdr) {
+            char *strtab = file + strtab_shdr->sh_offset;
+            ElfW(Sym) *symtab = (ElfW(Sym) *)(file + symtab_shdr->sh_offset);
+            int symtab_count = (int)(symtab_shdr->sh_size / sizeof(ElfW(Sym)));
+            for (j = 0; j < symtab_count; j++) {
+                ElfW(Sym) *sym = &symtab[j];
+                uintptr_t saddr = (uintptr_t)sym->st_value + obj->base_addr;
+                if (ELF_ST_TYPE(sym->st_info) != STT_FUNC) continue;
+                for (i = offset; i < num_traces; i++) {
+                    uintptr_t d = (uintptr_t)traces[i] - saddr;
+                    if (lines[i].line > 0 || d > (uintptr_t)sym->st_size)
+                        continue;
+                    /* fill symbol name and addr from .symtab */
+                    if (!lines[i].sname) lines[i].sname = strtab + sym->st_name;
+                    lines[i].saddr = saddr;
+                    lines[i].path  = obj->path;
+                    lines[i].base_addr = obj->base_addr;
+                }
+            }
+        }
+    }
+
+    if (!obj->debug_line.ptr) {
+	/* This file doesn't have .debug_line section,
+	   let's check .gnu_debuglink section instead. */
+	if (gnu_debuglink_shdr && check_debuglink) {
+	    follow_debuglink(file + gnu_debuglink_shdr->sh_offset,
+			     num_traces, traces,
+			     objp, lines, offset);
+	}
+	goto finish;
+    }
+
+    if (parse_debug_line(num_traces, traces,
+            obj->debug_line.ptr,
+            obj->debug_line.size,
+            obj, lines, offset) == -1)
+        goto fail;
+
+finish:
+    return dladdr_fbase;
+fail:
+    return (uintptr_t)-1;
+}
+#else /* Mach-O */
+/* read file and fill lines */
+static uintptr_t
+fill_lines(int num_traces, void **traces, int check_debuglink,
+        obj_info_t **objp, line_info_t *lines, int offset)
+{
+# ifdef __LP64__
+#  define LP(x) x##_64
+# else
+#  define LP(x) x
+# endif
+    int fd;
+    off_t filesize;
+    char *file, *p = NULL;
+    obj_info_t *obj = *objp;
+    struct LP(mach_header) *header;
+    uintptr_t dladdr_fbase = 0;
+
+    {
+        char *s = binary_filename;
+        char *base = strrchr(binary_filename, '/')+1;
+        size_t max = PATH_MAX;
+        size_t size = strlen(binary_filename);
+        size_t basesize = size - (base - binary_filename);
+        s += size;
+        max -= size;
+        p = s;
+        size = strlcpy(s, ".dSYM/Contents/Resources/DWARF/", max);
+        if (size == 0) goto fail;
+        s += size;
+        max -= size;
+        if (max <= basesize) goto fail;
+        memcpy(s, base, basesize);
+        s[basesize] = 0;
+
+        fd = open(binary_filename, O_RDONLY);
+        if (fd < 0) {
+            *p = 0; /* binary_filename becomes original file name */
+            fd = open(binary_filename, O_RDONLY);
+            if (fd < 0) {
+                goto fail;
+            }
+        }
+    }
+
+    filesize = lseek(fd, 0, SEEK_END);
+    if (filesize < 0) {
+        int e = errno;
+        close(fd);
+        kprintf("lseek: %s\n", strerror(e));
+        goto fail;
+    }
+#if SIZEOF_OFF_T > SIZEOF_SIZE_T
+    if (filesize > (off_t)SIZE_MAX) {
+        close(fd);
+        kprintf("Too large file %s\n", binary_filename);
+        goto fail;
+    }
+#endif
+    lseek(fd, 0, SEEK_SET);
+    /* async-signal unsafe */
+    file = (char *)mmap(NULL, (size_t)filesize, PROT_READ, MAP_SHARED, fd, 0);
+    if (file == MAP_FAILED) {
+        int e = errno;
+        close(fd);
+        kprintf("mmap: %s\n", strerror(e));
+        goto fail;
+    }
+    close(fd);
+
+    obj->mapped = file;
+    obj->mapped_size = (size_t)filesize;
+
+    header = (struct LP(mach_header) *)file;
+    if (header->magic == LP(MH_MAGIC)) {
+        /* non universal binary */
+        p = file;
+    }
+    else if (header->magic == FAT_CIGAM) {
+        struct LP(mach_header) *mhp = _NSGetMachExecuteHeader();
+        struct fat_header *fat = (struct fat_header *)file;
+        char *q = file + sizeof(*fat);
+        uint32_t nfat_arch = __builtin_bswap32(fat->nfat_arch);
+        /* fprintf(stderr,"%d: fat:%s %d\n",__LINE__, binary_filename,nfat_arch); */
+        for (uint32_t i = 0; i < nfat_arch; i++) {
+            struct fat_arch *arch = (struct fat_arch *)q;
+            cpu_type_t cputype = __builtin_bswap32(arch->cputype);
+            cpu_subtype_t cpusubtype = __builtin_bswap32(arch->cpusubtype);
+            uint32_t offset = __builtin_bswap32(arch->offset);
+            /* fprintf(stderr,"%d: fat %d %x/%x %x/%x\n",__LINE__, i, mhp->cputype,mhp->cpusubtype, cputype,cpusubtype); */
+            if (mhp->cputype == cputype &&
+                    (cpu_subtype_t)(mhp->cpusubtype & ~CPU_SUBTYPE_MASK) == cpusubtype) {
+                p = file + offset;
+                file = p;
+                header = (struct LP(mach_header) *)p;
+                if (header->magic == LP(MH_MAGIC)) {
+                    goto found_mach_header;
+                }
+                break;
+            }
+            q += sizeof(*arch);
+        }
+        kprintf("'%s' is not a Mach-O universal binary file!\n",binary_filename);
+        close(fd);
+        goto fail;
+    }
+    else {
+        kprintf("'%s' is not a "
+# ifdef __LP64__
+                "64"
+# else
+                "32"
+# endif
+                "-bit Mach-O file!\n",binary_filename);
+        close(fd);
+        goto fail;
+    }
+found_mach_header:
+    p += sizeof(*header);
+
+    for (uint32_t i = 0; i < (uint32_t)header->ncmds; i++) {
+        struct load_command *lcmd = (struct load_command *)p;
+        switch (lcmd->cmd) {
+          case LP(LC_SEGMENT):
+            {
+                static const char *debug_section_names[] = {
+                    "__debug_abbrev",
+                    "__debug_info",
+                    "__debug_line",
+                    "__debug_ranges",
+                    "__debug_str"
+                };
+                struct LP(segment_command) *scmd = (struct LP(segment_command) *)lcmd;
+                if (strcmp(scmd->segname, "__TEXT") == 0) {
+                    obj->vmaddr = scmd->vmaddr;
+                }
+                else if (strcmp(scmd->segname, "__DWARF") == 0) {
+                    p += sizeof(struct LP(segment_command));
+                    for (uint64_t i = 0; i < scmd->nsects; i++) {
+                        struct LP(section) *sect = (struct LP(section) *)p;
+                        p += sizeof(struct LP(section));
+                        for (int j=0; j < DWARF_SECTION_COUNT; j++) {
+                            struct dwarf_section *s = obj_dwarf_section_at(obj, j);
+
+                            if (strcmp(sect->sectname, debug_section_names[j]) != 0)
+                                continue;
+
+                            s->ptr = file + sect->offset;
+                            s->size = sect->size;
+                            s->flags = sect->flags;
+                            if (s->flags & SHF_COMPRESSED) {
+                                goto fail;
+                            }
+                            break;
+                        }
+                    }
+                }
+            }
+            break;
+
+          case LC_SYMTAB:
+            {
+                struct symtab_command *cmd = (struct symtab_command *)lcmd;
+                struct LP(nlist) *nl = (struct LP(nlist) *)(file + cmd->symoff);
+                char *strtab = file + cmd->stroff, *sname = 0;
+                uint32_t j;
+                uintptr_t saddr = 0;
+                /* kprintf("[%2d]: %x/symtab %p\n", i, cmd->cmd, p); */
+                for (j = 0; j < cmd->nsyms; j++) {
+                    uintptr_t symsize, d;
+                    struct LP(nlist) *e = &nl[j];
+                        /* kprintf("[%2d][%4d]: %02x/%x/%x: %s %llx\n", i, j, e->n_type,e->n_sect,e->n_desc,strtab+e->n_un.n_strx,e->n_value); */
+                    if (e->n_type != N_FUN) continue;
+                    if (e->n_sect) {
+                        saddr = (uintptr_t)e->n_value + obj->base_addr - obj->vmaddr;
+                        sname = strtab + e->n_un.n_strx;
+                        /* kprintf("[%2d][%4d]: %02x/%x/%x: %s %llx\n", i, j, e->n_type,e->n_sect,e->n_desc,strtab+e->n_un.n_strx,e->n_value); */
+                        continue;
+                    }
+                    for (int k = offset; k < num_traces; k++) {
+                        d = (uintptr_t)traces[k] - saddr;
+                        symsize = e->n_value;
+                        /* kprintf("%lx %lx %lx\n",saddr,symsize,traces[k]); */
+                        if (lines[k].line > 0 || d > (uintptr_t)symsize)
+                            continue;
+                        /* fill symbol name and addr from .symtab */
+                        if (!lines[k].sname) lines[k].sname = sname;
+                        lines[k].saddr = saddr;
+                        lines[k].path  = obj->path;
+                        lines[k].base_addr = obj->base_addr;
+                    }
+                }
+            }
+        }
+        p += lcmd->cmdsize;
+    }
+
+    if (obj->debug_info.ptr && obj->debug_abbrev.ptr) {
+        DebugInfoReader reader;
+        debug_info_reader_init(&reader, obj);
+        while (reader.p < reader.pend) {
+            if (di_read_cu(&reader)) goto fail;
+            debug_info_read(&reader, num_traces, traces, lines, offset);
+        }
+    }
+
+    if (parse_debug_line(num_traces, traces,
+            obj->debug_line.ptr,
+            obj->debug_line.size,
+            obj, lines, offset) == -1)
+        goto fail;
+
+    return dladdr_fbase;
+fail:
+    return (uintptr_t)-1;
+}
+#endif
+
+#define HAVE_MAIN_EXE_PATH
+#if defined(__FreeBSD__)
+# include <sys/sysctl.h>
+#endif
+/* ssize_t main_exe_path(void)
+ *
+ * store the path of the main executable to `binary_filename`,
+ * and returns strlen(binary_filename).
+ * it is NUL terminated.
+ */
+#if defined(__linux__)
+static ssize_t
+main_exe_path(void)
+{
+# define PROC_SELF_EXE "/proc/self/exe"
+    ssize_t len = readlink(PROC_SELF_EXE, binary_filename, PATH_MAX);
+    if (len < 0) return 0;
+    binary_filename[len] = 0;
+    return len;
+}
+#elif defined(__FreeBSD__)
+static ssize_t
+main_exe_path(void)
+{
+    int mib[4] = {CTL_KERN, KERN_PROC, KERN_PROC_PATHNAME, -1};
+    size_t len = PATH_MAX;
+    int err = sysctl(mib, 4, binary_filename, &len, NULL, 0);
+    if (err) {
+	kprintf("Can't get the path of ruby");
+	return -1;
+    }
+    len--; /* sysctl sets strlen+1 */
+    return len;
+}
+#elif defined(HAVE_LIBPROC_H)
+static ssize_t
+main_exe_path(void)
+{
+    int len = proc_pidpath(getpid(), binary_filename, PATH_MAX);
+    if (len == 0) return 0;
+    binary_filename[len] = 0;
+    return len;
+}
+#else
+#undef HAVE_MAIN_EXE_PATH
+#endif
+
+static void
+print_line0(line_info_t *line, void *address)
+{
+    uintptr_t addr = (uintptr_t)address;
+    uintptr_t d = addr - line->saddr;
+    if (!address) {
+        /* inlined */
+        if (line->dirname && line->dirname[0]) {
+            kprintf("%s(%s) %s/%s:%d\n", line->path, line->sname, line->dirname, line->filename, line->line);
+        }
+        else {
+            kprintf("%s(%s) %s:%d\n", line->path, line->sname, line->filename, line->line);
+        }
+    }
+    else if (!line->path) {
+        kprintf("[0x%"PRIxPTR"]\n", addr);
+    }
+    else if (!line->saddr || !line->sname) {
+        kprintf("%s(0x%"PRIxPTR") [0x%"PRIxPTR"]\n", line->path, addr-line->base_addr, addr);
+    }
+    else if (line->line <= 0) {
+        kprintf("%s(%s+0x%"PRIxPTR") [0x%"PRIxPTR"]\n", line->path, line->sname,
+                d, addr);
+    }
+    else if (!line->filename) {
+        kprintf("%s(%s+0x%"PRIxPTR") [0x%"PRIxPTR"] ???:%d\n", line->path, line->sname,
+                d, addr, line->line);
+    }
+    else if (line->dirname && line->dirname[0]) {
+        kprintf("%s(%s+0x%"PRIxPTR") [0x%"PRIxPTR"] %s/%s:%d\n", line->path, line->sname,
+                d, addr, line->dirname, line->filename, line->line);
+    }
+    else {
+        kprintf("%s(%s+0x%"PRIxPTR") [0x%"PRIxPTR"] %s:%d\n", line->path, line->sname,
+                d, addr, line->filename, line->line);
+    }
+}
+
+static void
+print_line(line_info_t *line, void *address)
+{
+    print_line0(line, address);
+    if (line->next) {
+        print_line(line->next, NULL);
+    }
+}
+
+void
+rb_dump_backtrace_with_lines(int num_traces, void **traces)
+{
+    int i;
+    /* async-signal unsafe */
+    line_info_t *lines = (line_info_t *)calloc(num_traces, sizeof(line_info_t));
+    obj_info_t *obj = NULL;
+    /* 2 is NULL + main executable */
+    void **dladdr_fbases = (void **)calloc(num_traces+2, sizeof(void *));
+#ifdef HAVE_MAIN_EXE_PATH
+    char *main_path = NULL; /* used on printing backtrace */
+    ssize_t len;
+    if ((len = main_exe_path()) > 0) {
+	main_path = (char *)alloca(len + 1);
+	if (main_path) {
+	    uintptr_t addr;
+	    memcpy(main_path, binary_filename, len+1);
+	    append_obj(&obj);
+	    obj->path = main_path;
+	    addr = fill_lines(num_traces, traces, 1, &obj, lines, -1);
+	    if (addr != (uintptr_t)-1) {
+		dladdr_fbases[0] = (void *)addr;
+	    }
+	}
+    }
+#endif
+
+    /* fill source lines by reading dwarf */
+    for (i = 0; i < num_traces; i++) {
+	Dl_info info;
+	if (lines[i].line) continue;
+	if (dladdr(traces[i], &info)) {
+	    const char *path;
+	    void **p;
+
+	    /* skip symbols which is in already checked objects */
+	    /* if the binary is strip-ed, this may effect */
+	    for (p=dladdr_fbases; *p; p++) {
+		if (*p == info.dli_fbase) {
+		    lines[i].path = info.dli_fname;
+		    lines[i].sname = info.dli_sname;
+		    goto next_line;
+		}
+	    }
+	    *p = info.dli_fbase;
+
+	    append_obj(&obj);
+	    obj->base_addr = (uintptr_t)info.dli_fbase;
+	    path = info.dli_fname;
+	    obj->path = path;
+	    lines[i].path = path;
+            lines[i].sname = info.dli_sname;
+            lines[i].saddr = (uintptr_t)info.dli_saddr;
+	    strlcpy(binary_filename, path, PATH_MAX);
+	    if (fill_lines(num_traces, traces, 1, &obj, lines, i) == (uintptr_t)-1)
+		break;
+	}
+next_line:
+	continue;
+    }
+
+    /* output */
+    for (i = 0; i < num_traces; i++) {
+        print_line(&lines[i], traces[i]);
+
+	/* FreeBSD's backtrace may show _start and so on */
+	if (lines[i].sname && strcmp("main", lines[i].sname) == 0)
+	    break;
+    }
+
+    /* free */
+    while (obj) {
+	obj_info_t *o = obj;
+        for (i=0; i < DWARF_SECTION_COUNT; i++) {
+            struct dwarf_section *s = obj_dwarf_section_at(obj, i);
+            if (s->flags & SHF_COMPRESSED) {
+                free(s->ptr);
+            }
+        }
+	if (obj->mapped_size) {
+	    munmap(obj->mapped, obj->mapped_size);
+	}
+	obj = o->next;
+	free(o);
+    }
+    for (i = 0; i < num_traces; i++) {
+        line_info_t *line = lines[i].next;
+        while (line) {
+            line_info_t *l = line;
+            line = line->next;
+            free(l);
+        }
+    }
+    free(lines);
+    free(dladdr_fbases);
+}
+
+/* From FreeBSD's lib/libstand/printf.c */
+/*-
+ * Copyright (c) 1986, 1988, 1991, 1993
+ *	The Regents of the University of California.  All rights reserved.
+ * (c) UNIX System Laboratories, Inc.
+ * All or some portions of this file are derived from material licensed
+ * to the University of California by American Telephone and Telegraph
+ * Co. or Unix System Laboratories, Inc. and are reproduced herein with
+ * the permission of UNIX System Laboratories, Inc.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 4. Neither the name of the University nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ *	@(#)subr_prf.c	8.3 (Berkeley) 1/21/94
+ */
+
+#include <stdarg.h>
+#define MAXNBUF (sizeof(intmax_t) * CHAR_BIT + 1)
+static inline int toupper(int c) { return ('A' <= c && c <= 'Z') ? (c&0x5f) : c; }
+#define    hex2ascii(hex)  (hex2ascii_data[hex])
+static const char hex2ascii_data[] = "0123456789abcdefghijklmnopqrstuvwxyz";
+static inline int imax(int a, int b) { return (a > b ? a : b); }
+static int kvprintf(char const *fmt, void (*func)(int), void *arg, int radix, va_list ap);
+
+static void putce(int c)
+{
+    char s[1];
+    ssize_t ret;
+
+    s[0] = (char)c;
+    ret = write(2, s, 1);
+    (void)ret;
+}
+
+static int
+kprintf(const char *fmt, ...)
+{
+	va_list ap;
+	int retval;
+
+	va_start(ap, fmt);
+	retval = kvprintf(fmt, putce, NULL, 10, ap);
+	va_end(ap);
+	return retval;
+}
+
+/*
+ * Put a NUL-terminated ASCII number (base <= 36) in a buffer in reverse
+ * order; return an optional length and a pointer to the last character
+ * written in the buffer (i.e., the first character of the string).
+ * The buffer pointed to by `nbuf' must have length >= MAXNBUF.
+ */
+static char *
+ksprintn(char *nbuf, uintmax_t num, int base, int *lenp, int upper)
+{
+	char *p, c;
+
+	p = nbuf;
+	*p = '\0';
+	do {
+		c = hex2ascii(num % base);
+		*++p = upper ? toupper(c) : c;
+	} while (num /= base);
+	if (lenp)
+		*lenp = (int)(p - nbuf);
+	return (p);
+}
+
+/*
+ * Scaled down version of printf(3).
+ *
+ * Two additional formats:
+ *
+ * The format %b is supported to decode error registers.
+ * Its usage is:
+ *
+ *	printf("reg=%b\n", regval, "<base><arg>*");
+ *
+ * where <base> is the output base expressed as a control character, e.g.
+ * \10 gives octal; \20 gives hex.  Each arg is a sequence of characters,
+ * the first of which gives the bit number to be inspected (origin 1), and
+ * the next characters (up to a control character, i.e. a character <= 32),
+ * give the name of the register.  Thus:
+ *
+ *	kvprintf("reg=%b\n", 3, "\10\2BITTWO\1BITONE\n");
+ *
+ * would produce output:
+ *
+ *	reg=3<BITTWO,BITONE>
+ *
+ * XXX:  %D  -- Hexdump, takes pointer and separator string:
+ *		("%6D", ptr, ":")   -> XX:XX:XX:XX:XX:XX
+ *		("%*D", len, ptr, " " -> XX XX XX XX ...
+ */
+static int
+kvprintf(char const *fmt, void (*func)(int), void *arg, int radix, va_list ap)
+{
+#define PCHAR(c) {int cc=(c); if (func) (*func)(cc); else *d++ = cc; retval++; }
+	char nbuf[MAXNBUF];
+	char *d;
+	const char *p, *percent, *q;
+	unsigned char *up;
+	int ch, n;
+	uintmax_t num;
+	int base, lflag, qflag, tmp, width, ladjust, sharpflag, neg, sign, dot;
+	int cflag, hflag, jflag, tflag, zflag;
+	int dwidth, upper;
+	char padc;
+	int stop = 0, retval = 0;
+
+	num = 0;
+	if (!func)
+		d = (char *) arg;
+	else
+		d = NULL;
+
+	if (fmt == NULL)
+		fmt = "(fmt null)\n";
+
+	if (radix < 2 || radix > 36)
+		radix = 10;
+
+	for (;;) {
+		padc = ' ';
+		width = 0;
+		while ((ch = (unsigned char)*fmt++) != '%' || stop) {
+			if (ch == '\0')
+				return (retval);
+			PCHAR(ch);
+		}
+		percent = fmt - 1;
+		qflag = 0; lflag = 0; ladjust = 0; sharpflag = 0; neg = 0;
+		sign = 0; dot = 0; dwidth = 0; upper = 0;
+		cflag = 0; hflag = 0; jflag = 0; tflag = 0; zflag = 0;
+reswitch:	switch (ch = (unsigned char)*fmt++) {
+		case '.':
+			dot = 1;
+			goto reswitch;
+		case '#':
+			sharpflag = 1;
+			goto reswitch;
+		case '+':
+			sign = 1;
+			goto reswitch;
+		case '-':
+			ladjust = 1;
+			goto reswitch;
+		case '%':
+			PCHAR(ch);
+			break;
+		case '*':
+			if (!dot) {
+				width = va_arg(ap, int);
+				if (width < 0) {
+					ladjust = !ladjust;
+					width = -width;
+				}
+			} else {
+				dwidth = va_arg(ap, int);
+			}
+			goto reswitch;
+		case '0':
+			if (!dot) {
+				padc = '0';
+				goto reswitch;
+			}
+		case '1': case '2': case '3': case '4':
+		case '5': case '6': case '7': case '8': case '9':
+				for (n = 0;; ++fmt) {
+					n = n * 10 + ch - '0';
+					ch = *fmt;
+					if (ch < '0' || ch > '9')
+						break;
+				}
+			if (dot)
+				dwidth = n;
+			else
+				width = n;
+			goto reswitch;
+		case 'b':
+			num = (unsigned int)va_arg(ap, int);
+			p = va_arg(ap, char *);
+			for (q = ksprintn(nbuf, num, *p++, NULL, 0); *q;)
+				PCHAR(*q--);
+
+			if (num == 0)
+				break;
+
+			for (tmp = 0; *p;) {
+				n = *p++;
+				if (num & (1 << (n - 1))) {
+					PCHAR(tmp ? ',' : '<');
+					for (; (n = *p) > ' '; ++p)
+						PCHAR(n);
+					tmp = 1;
+				} else
+					for (; *p > ' '; ++p)
+						continue;
+			}
+			if (tmp)
+				PCHAR('>');
+			break;
+		case 'c':
+			PCHAR(va_arg(ap, int));
+			break;
+		case 'D':
+			up = va_arg(ap, unsigned char *);
+			p = va_arg(ap, char *);
+			if (!width)
+				width = 16;
+			while(width--) {
+				PCHAR(hex2ascii(*up >> 4));
+				PCHAR(hex2ascii(*up & 0x0f));
+				up++;
+				if (width)
+					for (q=p;*q;q++)
+						PCHAR(*q);
+			}
+			break;
+		case 'd':
+		case 'i':
+			base = 10;
+			sign = 1;
+			goto handle_sign;
+		case 'h':
+			if (hflag) {
+				hflag = 0;
+				cflag = 1;
+			} else
+				hflag = 1;
+			goto reswitch;
+		case 'j':
+			jflag = 1;
+			goto reswitch;
+		case 'l':
+			if (lflag) {
+				lflag = 0;
+				qflag = 1;
+			} else
+				lflag = 1;
+			goto reswitch;
+		case 'n':
+			if (jflag)
+				*(va_arg(ap, intmax_t *)) = retval;
+			else if (qflag)
+				*(va_arg(ap, int64_t *)) = retval;
+			else if (lflag)
+				*(va_arg(ap, long *)) = retval;
+			else if (zflag)
+				*(va_arg(ap, size_t *)) = retval;
+			else if (hflag)
+				*(va_arg(ap, short *)) = retval;
+			else if (cflag)
+				*(va_arg(ap, char *)) = retval;
+			else
+				*(va_arg(ap, int *)) = retval;
+			break;
+		case 'o':
+			base = 8;
+			goto handle_nosign;
+		case 'p':
+			base = 16;
+			sharpflag = (width == 0);
+			sign = 0;
+			num = (uintptr_t)va_arg(ap, void *);
+			goto number;
+		case 'q':
+			qflag = 1;
+			goto reswitch;
+		case 'r':
+			base = radix;
+			if (sign)
+				goto handle_sign;
+			goto handle_nosign;
+		case 's':
+			p = va_arg(ap, char *);
+			if (p == NULL)
+				p = "(null)";
+			if (!dot)
+				n = (int)strlen (p);
+			else
+				for (n = 0; n < dwidth && p[n]; n++)
+					continue;
+
+			width -= n;
+
+			if (!ladjust && width > 0)
+				while (width--)
+					PCHAR(padc);
+			while (n--)
+				PCHAR(*p++);
+			if (ladjust && width > 0)
+				while (width--)
+					PCHAR(padc);
+			break;
+		case 't':
+			tflag = 1;
+			goto reswitch;
+		case 'u':
+			base = 10;
+			goto handle_nosign;
+		case 'X':
+			upper = 1;
+		case 'x':
+			base = 16;
+			goto handle_nosign;
+		case 'y':
+			base = 16;
+			sign = 1;
+			goto handle_sign;
+		case 'z':
+			zflag = 1;
+			goto reswitch;
+handle_nosign:
+			sign = 0;
+			if (jflag)
+				num = va_arg(ap, uintmax_t);
+			else if (qflag)
+				num = va_arg(ap, uint64_t);
+			else if (tflag)
+				num = va_arg(ap, ptrdiff_t);
+			else if (lflag)
+				num = va_arg(ap, unsigned long);
+			else if (zflag)
+				num = va_arg(ap, size_t);
+			else if (hflag)
+				num = (unsigned short)va_arg(ap, int);
+			else if (cflag)
+				num = (unsigned char)va_arg(ap, int);
+			else
+				num = va_arg(ap, unsigned int);
+			goto number;
+handle_sign:
+			if (jflag)
+				num = va_arg(ap, intmax_t);
+			else if (qflag)
+				num = va_arg(ap, int64_t);
+			else if (tflag)
+				num = va_arg(ap, ptrdiff_t);
+			else if (lflag)
+				num = va_arg(ap, long);
+			else if (zflag)
+				num = va_arg(ap, ssize_t);
+			else if (hflag)
+				num = (short)va_arg(ap, int);
+			else if (cflag)
+				num = (char)va_arg(ap, int);
+			else
+				num = va_arg(ap, int);
+number:
+			if (sign && (intmax_t)num < 0) {
+				neg = 1;
+				num = -(intmax_t)num;
+			}
+			p = ksprintn(nbuf, num, base, &n, upper);
+			tmp = 0;
+			if (sharpflag && num != 0) {
+				if (base == 8)
+					tmp++;
+				else if (base == 16)
+					tmp += 2;
+			}
+			if (neg)
+				tmp++;
+
+			if (!ladjust && padc == '0')
+				dwidth = width - tmp;
+			width -= tmp + imax(dwidth, n);
+			dwidth -= n;
+			if (!ladjust)
+				while (width-- > 0)
+					PCHAR(' ');
+			if (neg)
+				PCHAR('-');
+			if (sharpflag && num != 0) {
+				if (base == 8) {
+					PCHAR('0');
+				} else if (base == 16) {
+					PCHAR('0');
+					PCHAR('x');
+				}
+			}
+			while (dwidth-- > 0)
+				PCHAR('0');
+
+			while (*p)
+				PCHAR(*p--);
+
+			if (ladjust)
+				while (width-- > 0)
+					PCHAR(' ');
+
+			break;
+		default:
+			while (percent < fmt)
+				PCHAR(*percent++);
+			/*
+			 * Since we ignore an formatting argument it is no
+			 * longer safe to obey the remaining formatting
+			 * arguments as the arguments will no longer match
+			 * the format specs.
+			 */
+			stop = 1;
+			break;
+		}
+	}
+#undef PCHAR
+}
+#else /* defined(USE_ELF) */
+#error not supported
+#endif