[PATCH 6/7] binutils/readelf: build against msgpack, dump NT_AMDGPU_METADATA note contents

[Simon Marchi <simon.marchi@polymtl.ca>]

From: Simon Marchi <simon.marchi@efficios.com>

[This patch might be a bit more controversial, given its adds an
optional dependency on a third party library.  If this is really a
problems, we can drop it from the series.  The consequence being that
there will be no human readable output for the AMDGPU ELF notes.]

The AMDGPU HSA OS ABI (code object v3 and above) defines the
NT_AMDGPU_METADATA ELF note [1].  The content is a msgpack object
describing, among other things, the kernels present in the code object
and how to call them.

I think it would be useful for readelf to be able to display the content
of those notes.  msgpack is a structured format, a bit like JSON, except
not text-based.  It is therefore possible to dump the contents in
human-readable form without knowledge of the specific layout of the
note.

Add configury to binutils to optionally check for the msgpack C library
[2].  Add There is a new --with{,out}-msgpack configure flag, and the actual
library lookup is done using pkg-config.

If msgpack support is enabled, dumping a NT_AMDGPU_METADATA note looks
like:

    $ readelf --notes amdgpu-code-object
    Displaying notes found in: .note
      Owner                Data size        Description
      AMDGPU               0x0000040d       NT_AMDGPU_METADATA (code object metadata)
        {
          "amdhsa.kernels": [
            {
              ".args": [
                {
                  ".address_space": "global",
                  ".name": "out.coerce",
                  ".offset": 0,
                  ".size": 8,
                  ".value_kind": "global_buffer",
                },
      <snip>

If msgpack support is disabled, dump the contents as hex, as is done
with notes that are not handled in a special way.  This allows one to
decode the contents manually (maybe using a command-line msgpack
decoder) if really needed.

[1] https://llvm.org/docs/AMDGPUUsage.html#code-object-metadata
[2] https://github.com/msgpack/msgpack-c/tree/c_master

binutils/ChangeLog:

	* Makefile.am (readelf_CFLAGS): New.
	(readelf_LDADD): Add MSGPACK_LIBS.
	* Makefile.in: Re-generate.
	* config.in: Re-generate.
	* configure: Re-generate.
	* configure.ac: Add --with-msgpack flag and check for msgpack
	using pkg-config.
	* readelf.c: Include msgpack.h if HAVE_MSGPACK.
	(print_note_contents_hex): New.
	(print_indents): New.
	(dump_msgpack_obj): New.
	(dump_msgpack): New.
	(print_amdgpu_note): New.
	(process_note): Handle NT_AMDGPU_METADATA note contents.
	Use print_note_contents_hex.

Change-Id: Ia60a654e620bc32dfdb1bccd845594e2af328b84
---
 binutils/Makefile.am  |  14 ++-
 binutils/Makefile.in  |  11 ++-
 binutils/config.in    |   3 +
 binutils/configure    | 137 +++++++++++++++++++++++++++++-
 binutils/configure.ac |  17 ++++
 binutils/readelf.c    | 193 +++++++++++++++++++++++++++++++++++++++---
 6 files changed, 356 insertions(+), 19 deletions(-)

diff --git a/binutils/Makefile.am b/binutils/Makefile.am
index 5b13af29cc38..751fbacce127 100644
--- a/binutils/Makefile.am
+++ b/binutils/Makefile.am
@@ -256,7 +256,7 @@ objcopy_SOURCES = objcopy.c not-strip.c rename.c $(WRITE_DEBUG_SRCS) $(BULIBS)
 strings_SOURCES = strings.c $(BULIBS)
 
 readelf_SOURCES = readelf.c version.c unwind-ia64.c dwarf.c demanguse.c $(ELFLIBS)
-readelf_LDADD   = $(LIBCTF_NOBFD) $(LIBINTL) $(LIBIBERTY) $(ZLIB) $(DEBUGINFOD_LIBS)
+readelf_LDADD   = $(LIBCTF_NOBFD) $(LIBINTL) $(LIBIBERTY) $(ZLIB) $(DEBUGINFOD_LIBS) $(MSGPACK_LIBS)
 
 elfedit_SOURCES = elfedit.c version.c $(ELFLIBS)
 elfedit_LDADD = $(LIBINTL) $(LIBIBERTY)
@@ -381,6 +381,18 @@ endif
 	$(AM_V_CC)$(COMPILE) $(DEBUGINFOD_CFLAGS) -c -o $@ $(srcdir)/dwarf.c
 endif
 
+readelf.@OBJEXT@: readelf.c
+if am__fastdepCC
+	$(AM_V_CC)$(COMPILE) -MT $@ -MD -MP -MF $(DEPDIR)/$*.Tpo $(MSGPACK_CFLAGS) -c -o $@ $(srcdir)/readelf.c
+	$(AM_V_at)mv -f $(DEPDIR)/$*.Tpo $(DEPDIR)/$*.Po
+else
+if AMDEP
+	source='readelf.c' object='$@' libtool=no @AMDEPBACKSLASH@
+	DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+endif
+	$(AM_V_CC)$(COMPILE) $(MSGPACK_CFLAGS) -c -o $@ $(srcdir)/readelf.c
+endif
+
 sysroff.@OBJEXT@: sysroff.c
 if am__fastdepCC
 	$(AM_V_CC)$(COMPILE) -MT $@ -MD -MP -MF $(DEPDIR)/$*.Tpo -c -o $@ `test -f sysroff.c || echo $(srcdir)/`sysroff.c $(NO_WERROR)
diff --git a/binutils/Makefile.in b/binutils/Makefile.in
index 87f78628380b..78d32b350e38 100644
--- a/binutils/Makefile.in
+++ b/binutils/Makefile.in
@@ -537,6 +537,8 @@ MKDIR_P = @MKDIR_P@
 MKINSTALLDIRS = @MKINSTALLDIRS@
 MSGFMT = @MSGFMT@
 MSGMERGE = @MSGMERGE@
+MSGPACK_CFLAGS = @MSGPACK_CFLAGS@
+MSGPACK_LIBS = @MSGPACK_LIBS@
 NM = @NM@
 NMEDIT = @NMEDIT@
 NO_WERROR = @NO_WERROR@
@@ -785,7 +787,7 @@ size_SOURCES = size.c $(BULIBS)
 objcopy_SOURCES = objcopy.c not-strip.c rename.c $(WRITE_DEBUG_SRCS) $(BULIBS)
 strings_SOURCES = strings.c $(BULIBS)
 readelf_SOURCES = readelf.c version.c unwind-ia64.c dwarf.c demanguse.c $(ELFLIBS)
-readelf_LDADD = $(LIBCTF_NOBFD) $(LIBINTL) $(LIBIBERTY) $(ZLIB) $(DEBUGINFOD_LIBS)
+readelf_LDADD = $(LIBCTF_NOBFD) $(LIBINTL) $(LIBIBERTY) $(ZLIB) $(DEBUGINFOD_LIBS) $(MSGPACK_LIBS)
 elfedit_SOURCES = elfedit.c version.c $(ELFLIBS)
 elfedit_LDADD = $(LIBINTL) $(LIBIBERTY)
 strip_new_SOURCES = objcopy.c is-strip.c rename.c $(WRITE_DEBUG_SRCS) $(BULIBS)
@@ -1919,6 +1921,13 @@ dwarf.@OBJEXT@: dwarf.c
 @AMDEP_TRUE@@am__fastdepCC_FALSE@	DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
 @am__fastdepCC_FALSE@	$(AM_V_CC)$(COMPILE) $(DEBUGINFOD_CFLAGS) -c -o $@ $(srcdir)/dwarf.c
 
+readelf.@OBJEXT@: readelf.c
+@am__fastdepCC_TRUE@	$(AM_V_CC)$(COMPILE) -MT $@ -MD -MP -MF $(DEPDIR)/$*.Tpo $(MSGPACK_CFLAGS) -c -o $@ $(srcdir)/readelf.c
+@am__fastdepCC_TRUE@	$(AM_V_at)mv -f $(DEPDIR)/$*.Tpo $(DEPDIR)/$*.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@	source='readelf.c' object='$@' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@	DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@	$(AM_V_CC)$(COMPILE) $(MSGPACK_CFLAGS) -c -o $@ $(srcdir)/readelf.c
+
 sysroff.@OBJEXT@: sysroff.c
 @am__fastdepCC_TRUE@	$(AM_V_CC)$(COMPILE) -MT $@ -MD -MP -MF $(DEPDIR)/$*.Tpo -c -o $@ `test -f sysroff.c || echo $(srcdir)/`sysroff.c $(NO_WERROR)
 @am__fastdepCC_TRUE@	$(AM_V_at)mv -f $(DEPDIR)/$*.Tpo $(DEPDIR)/$*.Po
diff --git a/binutils/config.in b/binutils/config.in
index e6fa66fc41b6..81bd143140e0 100644
--- a/binutils/config.in
+++ b/binutils/config.in
@@ -104,6 +104,9 @@
 /* Define to 1 if you have a working `mmap' system call. */
 #undef HAVE_MMAP
 
+/* Define to 1 if msgpack is available. */
+#undef HAVE_MSGPACK
+
 /* Define to 1 if you have the `sbrk' function. */
 #undef HAVE_SBRK
 
diff --git a/binutils/configure b/binutils/configure
index 19d82badc638..f3ad831ad381 100755
--- a/binutils/configure
+++ b/binutils/configure
@@ -648,6 +648,8 @@ BUILD_DLLTOOL
 BUILD_SRCONV
 LTLIBICONV
 LIBICONV
+MSGPACK_LIBS
+MSGPACK_CFLAGS
 zlibinc
 zlibdir
 DEMANGLER_NAME
@@ -830,6 +832,7 @@ enable_build_warnings
 enable_nls
 enable_maintainer_mode
 with_system_zlib
+with_msgpack
 enable_rpath
 with_libiconv_prefix
 with_libiconv_type
@@ -849,7 +852,9 @@ PKG_CONFIG_LIBDIR
 DEBUGINFOD_CFLAGS
 DEBUGINFOD_LIBS
 YACC
-YFLAGS'
+YFLAGS
+MSGPACK_CFLAGS
+MSGPACK_LIBS'
 
 
 # Initialize some variables set by options.
@@ -1512,6 +1517,7 @@ Optional Packages:
   --with-debuginfod       Enable debuginfo lookups with debuginfod
                           (auto/yes/no)
   --with-system-zlib      use installed libz
+  --with-msgpack          Enable msgpack support (auto/yes/no)
   --with-gnu-ld           assume the C compiler uses GNU ld default=no
   --with-libiconv-prefix[=DIR]  search for libiconv in DIR/include and DIR/lib
   --without-libiconv-prefix     don't search for libiconv in includedir and libdir
@@ -1541,6 +1547,10 @@ Some influential environment variables:
   YFLAGS      The list of arguments that will be passed by default to $YACC.
               This script will default YFLAGS to the empty string to avoid a
               default value of `-d' given by some make applications.
+  MSGPACK_CFLAGS
+              C compiler flags for MSGPACK, overriding pkg-config
+  MSGPACK_LIBS
+              linker flags for MSGPACK, overriding pkg-config
 
 Use these variables to override the choices made by `configure' or to help
 it to find libraries and programs with nonstandard names/locations.
@@ -10971,7 +10981,7 @@ else
   lt_dlunknown=0; lt_dlno_uscore=1; lt_dlneed_uscore=2
   lt_status=$lt_dlunknown
   cat > conftest.$ac_ext <<_LT_EOF
-#line 10974 "configure"
+#line 10984 "configure"
 #include "confdefs.h"
 
 #if HAVE_DLFCN_H
@@ -11077,7 +11087,7 @@ else
   lt_dlunknown=0; lt_dlno_uscore=1; lt_dlneed_uscore=2
   lt_status=$lt_dlunknown
   cat > conftest.$ac_ext <<_LT_EOF
-#line 11080 "configure"
+#line 11090 "configure"
 #include "confdefs.h"
 
 #if HAVE_DLFCN_H
@@ -13723,6 +13733,127 @@ $as_echo "#define USE_BINARY_FOPEN 1" >>confdefs.h
  ;;
 esac
 
+# Support for the msgpack C library.
+
+# Check whether --with-msgpack was given.
+if test "${with_msgpack+set}" = set; then :
+  withval=$with_msgpack;
+else
+  with_msgpack=auto
+fi
+
+
+
+if test "$with_msgpack" != no; then
+
+pkg_failed=no
+{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for msgpack" >&5
+$as_echo_n "checking for msgpack... " >&6; }
+
+if test -n "$MSGPACK_CFLAGS"; then
+    pkg_cv_MSGPACK_CFLAGS="$MSGPACK_CFLAGS"
+ elif test -n "$PKG_CONFIG"; then
+    if test -n "$PKG_CONFIG" && \
+    { { $as_echo "$as_me:${as_lineno-$LINENO}: \$PKG_CONFIG --exists --print-errors \"msgpack\""; } >&5
+  ($PKG_CONFIG --exists --print-errors "msgpack") 2>&5
+  ac_status=$?
+  $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5
+  test $ac_status = 0; }; then
+  pkg_cv_MSGPACK_CFLAGS=`$PKG_CONFIG --cflags "msgpack" 2>/dev/null`
+		      test "x$?" != "x0" && pkg_failed=yes
+else
+  pkg_failed=yes
+fi
+ else
+    pkg_failed=untried
+fi
+if test -n "$MSGPACK_LIBS"; then
+    pkg_cv_MSGPACK_LIBS="$MSGPACK_LIBS"
+ elif test -n "$PKG_CONFIG"; then
+    if test -n "$PKG_CONFIG" && \
+    { { $as_echo "$as_me:${as_lineno-$LINENO}: \$PKG_CONFIG --exists --print-errors \"msgpack\""; } >&5
+  ($PKG_CONFIG --exists --print-errors "msgpack") 2>&5
+  ac_status=$?
+  $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5
+  test $ac_status = 0; }; then
+  pkg_cv_MSGPACK_LIBS=`$PKG_CONFIG --libs "msgpack" 2>/dev/null`
+		      test "x$?" != "x0" && pkg_failed=yes
+else
+  pkg_failed=yes
+fi
+ else
+    pkg_failed=untried
+fi
+
+if test $pkg_failed = no; then
+  pkg_save_LDFLAGS="$LDFLAGS"
+  LDFLAGS="$LDFLAGS $pkg_cv_MSGPACK_LIBS"
+  cat confdefs.h - <<_ACEOF >conftest.$ac_ext
+/* end confdefs.h.  */
+
+int
+main ()
+{
+
+  ;
+  return 0;
+}
+_ACEOF
+if ac_fn_c_try_link "$LINENO"; then :
+
+else
+  pkg_failed=yes
+fi
+rm -f core conftest.err conftest.$ac_objext \
+    conftest$ac_exeext conftest.$ac_ext
+  LDFLAGS=$pkg_save_LDFLAGS
+fi
+
+
+
+if test $pkg_failed = yes; then
+        { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5
+$as_echo "no" >&6; }
+
+if $PKG_CONFIG --atleast-pkgconfig-version 0.20; then
+        _pkg_short_errors_supported=yes
+else
+        _pkg_short_errors_supported=no
+fi
+        if test $_pkg_short_errors_supported = yes; then
+	        MSGPACK_PKG_ERRORS=`$PKG_CONFIG --short-errors --print-errors --cflags --libs "msgpack" 2>&1`
+        else
+	        MSGPACK_PKG_ERRORS=`$PKG_CONFIG --print-errors --cflags --libs "msgpack" 2>&1`
+        fi
+	# Put the nasty error message in config.log where it belongs
+	echo "$MSGPACK_PKG_ERRORS" >&5
+
+
+      if test "$with_msgpack" = yes; then
+	as_fn_error $? "--with-msgpack was given, but msgpack is missing or unusable." "$LINENO" 5
+      fi
+
+elif test $pkg_failed = untried; then
+        { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5
+$as_echo "no" >&6; }
+
+      if test "$with_msgpack" = yes; then
+	as_fn_error $? "--with-msgpack was given, but msgpack is missing or unusable." "$LINENO" 5
+      fi
+
+else
+	MSGPACK_CFLAGS=$pkg_cv_MSGPACK_CFLAGS
+	MSGPACK_LIBS=$pkg_cv_MSGPACK_LIBS
+        { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5
+$as_echo "yes" >&6; }
+
+
+$as_echo "#define HAVE_MSGPACK 1" >>confdefs.h
+
+
+fi
+fi
+
 # target-specific stuff:
 
 # Canonicalize the secondary target names.
diff --git a/binutils/configure.ac b/binutils/configure.ac
index 7b70e7652435..e204a201a0a9 100644
--- a/binutils/configure.ac
+++ b/binutils/configure.ac
@@ -275,6 +275,23 @@ AM_ZLIB
 
 BFD_BINARY_FOPEN
 
+# Support for the msgpack C library.
+AC_ARG_WITH([msgpack],
+	    AC_HELP_STRING([--with-msgpack], [Enable msgpack support (auto/yes/no)]),
+	    [],
+	    [with_msgpack=auto])
+
+
+if test "$with_msgpack" != no; then
+  PKG_CHECK_MODULES(MSGPACK, msgpack, [
+    AC_DEFINE([HAVE_MSGPACK], [1], [Define to 1 if msgpack is available.])
+  ], [
+      if test "$with_msgpack" = yes; then
+	AC_MSG_ERROR([--with-msgpack was given, but msgpack is missing or unusable.])
+      fi
+  ])
+fi
+
 # target-specific stuff:
 
 # Canonicalize the secondary target names.
diff --git a/binutils/readelf.c b/binutils/readelf.c
index 91515bdf0faa..ff07112eb843 100644
--- a/binutils/readelf.c
+++ b/binutils/readelf.c
@@ -46,6 +46,10 @@
 #include <zlib.h>
 #include <wchar.h>
 
+#if defined HAVE_MSGPACK
+#include <msgpack.h>
+#endif
+
 #if __GNUC__ >= 2
 /* Define BFD64 here, even if our default architecture is 32 bit ELF
    as this will allow us to read in and parse 64bit and 32bit ELF files.
@@ -21307,6 +21311,177 @@ print_gnu_build_attribute_name (Elf_Internal_Note * pnote)
   return true;
 }
 
+/* Print the contents of PNOTE as hex.  */
+
+static void
+print_note_contents_hex (Elf_Internal_Note *pnote)
+{
+  if (pnote->descsz)
+    {
+      unsigned long i;
+
+      printf (_("   description data: "));
+      for (i = 0; i < pnote->descsz; i++)
+	printf ("%02x ", pnote->descdata[i] & 0xff);
+      if (!do_wide)
+	printf ("\n");
+    }
+
+  if (do_wide)
+    printf ("\n");
+}
+
+#if defined HAVE_MSGPACK
+
+static void
+print_indents (int n)
+{
+  printf ("    ");
+
+  for (int i = 0; i < n; i++)
+    printf ("  ");
+}
+
+/* Print OBJ in human-readable form.  */
+
+static void
+dump_msgpack_obj (const msgpack_object *obj, int indent)
+{
+  switch (obj->type)
+    {
+    case MSGPACK_OBJECT_NIL:
+      printf ("(nil)");
+      break;
+
+    case MSGPACK_OBJECT_BOOLEAN:
+      printf ("%s", obj->via.boolean ? "true" : "false");
+      break;
+
+    case MSGPACK_OBJECT_POSITIVE_INTEGER:
+      printf ("%" PRIu64, obj->via.u64);
+      break;
+
+    case MSGPACK_OBJECT_NEGATIVE_INTEGER:
+      printf ("%" PRIi64, obj->via.i64);
+      break;
+
+    case MSGPACK_OBJECT_FLOAT32:
+    case MSGPACK_OBJECT_FLOAT64:
+      printf ("%f", obj->via.f64);
+      break;
+
+    case MSGPACK_OBJECT_STR:
+      printf ("\"%.*s\"", obj->via.str.size, obj->via.str.ptr);
+      break;
+
+    case MSGPACK_OBJECT_ARRAY:
+      {
+	const msgpack_object_array *array = &obj->via.array;
+
+	printf ("[\n");
+	++indent;
+
+	for (uint32_t i = 0; i < array->size; ++i)
+	  {
+	    const msgpack_object *item = &array->ptr[i];
+
+	    print_indents (indent);
+	    dump_msgpack_obj (item, indent);
+	    printf (",\n");
+	  }
+
+	--indent;
+	print_indents (indent);
+	printf ("]");
+	break;
+      }
+      break;
+
+    case MSGPACK_OBJECT_MAP:
+      {
+	const msgpack_object_map *map = &obj->via.map;
+
+	printf ("{\n");
+	++indent;
+
+	for (uint32_t i = 0; i < map->size; ++i)
+	  {
+	    const msgpack_object_kv *kv = &map->ptr[i];
+	    const msgpack_object *key = &kv->key;
+	    const msgpack_object *val = &kv->val;
+
+	    print_indents (indent);
+	    dump_msgpack_obj (key, indent);
+	    printf (": ");
+	    dump_msgpack_obj (val, indent);
+
+	    printf (",\n");
+	  }
+
+	--indent;
+	print_indents (indent);
+	printf ("}");
+
+	break;
+      }
+
+    case MSGPACK_OBJECT_BIN:
+      printf ("(bin)");
+      break;
+
+    case MSGPACK_OBJECT_EXT:
+      printf ("(ext)");
+      break;
+    }
+}
+
+static void
+dump_msgpack (const msgpack_unpacked *msg)
+{
+  print_indents (0);
+  dump_msgpack_obj (&msg->data, 0);
+  printf ("\n");
+}
+
+#endif /* defined HAVE_MSGPACK */
+
+static bool
+print_amdgpu_note (Elf_Internal_Note *pnote)
+{
+#if defined HAVE_MSGPACK
+  /* If msgpack is available, decode and dump the note's content.  */
+  bool ret;
+  msgpack_unpacked msg;
+  msgpack_unpack_return msgpack_ret;
+
+  assert (pnote->type == NT_AMDGPU_METADATA);
+
+  msgpack_unpacked_init (&msg);
+  msgpack_ret = msgpack_unpack_next (&msg, pnote->descdata, pnote->descsz,
+				     NULL);
+
+  switch (msgpack_ret)
+    {
+    case MSGPACK_UNPACK_SUCCESS:
+      dump_msgpack (&msg);
+      ret = true;
+      break;
+
+    default:
+      error (_("failed to unpack msgpack contents in NT_AMDGPU_METADATA note"));
+      ret = false;
+      break;
+    }
+
+  msgpack_unpacked_destroy (&msg);
+  return ret;
+#else
+  /* msgpack is not available, dump contents as hex.  */
+  print_note_contents_hex (pnote);
+  return true;
+#endif
+}
+
 /* Note that by the ELF standard, the name field is already null byte
    terminated, and namesz includes the terminating null byte.
    I.E. the value of namesz for the name "FSF" is 4.
@@ -21404,21 +21579,11 @@ process_note (Elf_Internal_Note *  pnote,
 	   && (pnote->type == NT_GNU_BUILD_ATTRIBUTE_OPEN
 	       || pnote->type == NT_GNU_BUILD_ATTRIBUTE_FUNC))
     return print_gnu_build_attribute_description (pnote, filedata);
+  else if (startswith (pnote->namedata, "AMDGPU")
+	   && pnote->type == NT_AMDGPU_METADATA)
+    return print_amdgpu_note (pnote);
 
-  if (pnote->descsz)
-    {
-      unsigned long i;
-
-      printf (_("   description data: "));
-      for (i = 0; i < pnote->descsz; i++)
-	printf ("%02x ", pnote->descdata[i] & 0xff);
-      if (!do_wide)
-	printf ("\n");
-    }
-
-  if (do_wide)
-    printf ("\n");
-
+  print_note_contents_hex (pnote);
   return true;
 }
 
-- 
2.35.1