[v2] Rewrite mixed source/assembler disassembly

Hi.

Here's v2 of this patch:

https://sourceware.org/ml/gdb-patches/2015-07/msg00546.html

I found at least one MI-based frontend that invokes
the CLI disassemble and then parses the output.
Arguably I could break it, but it's a headache I'd
rather avoid.

Ironically, MI output contains everything that's needed
so this should work with real MI frontends just fine.
They just need to pass a new mode, 4 or 5, to get the new output.
An alternative would be to change the behaviour of -data-disassemble
modes 1 and 3, and still not change the behaviour of disassemble /m.
To not force the change in behaviour on MI frontends I went with
adding new modes.

Regression tested on amd64-linux, and hand-tested with TUI.
I haven't regression tested this with any MI clients yet but I
haven't changed anything so fingers crossed.
If there are no objections I will check this in after I'm
happy with some MI-client testing.

2015-08-03  Doug Evans  <xdje42@gmail.com>

	* NEWS: Document new /s modifier for the disassemble command.
	* cli/cli-cmds.c (disassemble_command): Add support for /s.
	(_initialize_cli_cmds): Update online docs of disassemble command.
	* disasm.c: #include "source.h".
	(struct deprecated_dis_line_entry): Renamed from dis_line_entry.
	All uses updated.
	(dis_line_entry): New struct.
	(hash_dis_line_entry, eq_dis_line_entry): New functions.
	(allocate_dis_line_table): New functions.
	(maybe_add_dis_line_entry, line_has_code_p): New functions.
	(dump_insns): New arg end_pc.  All callers updated.
	(do_mixed_source_and_assembly_deprecated): Renamed from
	do_mixed_source_and_assembly.  All callers updated.
	(do_mixed_source_and_assembly): New function.
	(gdb_disassembly): Handle /s (DISASSEMBLY_SOURCE).
	* disasm.h (DISASSEMBLY_SOURCE_DEPRECATED): Renamed from
	DISASSEMBLY_SOURCE.  All uses updated.
	(DISASSEMBLY_SOURCE): New macro.
	* mi/mi-cmd-disas.c (mi_cmd_disassemble): New modes 4,5.

	doc/
	* gdb.texinfo (Machine Code): Update docs for mixed source/assembly
	disassembly.
	(GDB/MI Data Manipulation): Update docs for new disassembly modes.

	testsuite/
	* gdb.mi/mi-disassemble.exp: Update.
	* gdb.base/disasm-optim.S: New file.
	* gdb.base/disasm-optim.c: New file.
	* gdb.base/disasm-optim.h: New file.
	* gdb.base/disasm-optim.exp: New file.

Doug Evans <xdje42@gmail.com> writes:
> Hi.
>
> Here's v2 of this patch:
>
> https://sourceware.org/ml/gdb-patches/2015-07/msg00546.html
>
> I found at least one MI-based frontend that invokes
> the CLI disassemble and then parses the output.
> Arguably I could break it, but it's a headache I'd
> rather avoid.
>
> Ironically, MI output contains everything that's needed
> so this should work with real MI frontends just fine.
> They just need to pass a new mode, 4 or 5, to get the new output.
> An alternative would be to change the behaviour of -data-disassemble
> modes 1 and 3, and still not change the behaviour of disassemble /m.
> To not force the change in behaviour on MI frontends I went with
> adding new modes.
>
> Regression tested on amd64-linux, and hand-tested with TUI.
> I haven't regression tested this with any MI clients yet but I
> haven't changed anything so fingers crossed.
> If there are no objections I will check this in after I'm
> happy with some MI-client testing.

Hi.

doc ping.


>
> 2015-08-03  Doug Evans  <xdje42@gmail.com>
>
> 	* NEWS: Document new /s modifier for the disassemble command.
> 	* cli/cli-cmds.c (disassemble_command): Add support for /s.
> 	(_initialize_cli_cmds): Update online docs of disassemble command.
> 	* disasm.c: #include "source.h".
> 	(struct deprecated_dis_line_entry): Renamed from dis_line_entry.
> 	All uses updated.
> 	(dis_line_entry): New struct.
> 	(hash_dis_line_entry, eq_dis_line_entry): New functions.
> 	(allocate_dis_line_table): New functions.
> 	(maybe_add_dis_line_entry, line_has_code_p): New functions.
> 	(dump_insns): New arg end_pc.  All callers updated.
> 	(do_mixed_source_and_assembly_deprecated): Renamed from
> 	do_mixed_source_and_assembly.  All callers updated.
> 	(do_mixed_source_and_assembly): New function.
> 	(gdb_disassembly): Handle /s (DISASSEMBLY_SOURCE).
> 	* disasm.h (DISASSEMBLY_SOURCE_DEPRECATED): Renamed from
> 	DISASSEMBLY_SOURCE.  All uses updated.
> 	(DISASSEMBLY_SOURCE): New macro.
> 	* mi/mi-cmd-disas.c (mi_cmd_disassemble): New modes 4,5.
>
> 	doc/
> 	* gdb.texinfo (Machine Code): Update docs for mixed source/assembly
> 	disassembly.
> 	(GDB/MI Data Manipulation): Update docs for new disassembly modes.
>
> 	testsuite/
> 	* gdb.mi/mi-disassemble.exp: Update.
> 	* gdb.base/disasm-optim.S: New file.
> 	* gdb.base/disasm-optim.c: New file.
> 	* gdb.base/disasm-optim.h: New file.
> 	* gdb.base/disasm-optim.exp: New file.
>
> diff --git a/gdb/NEWS b/gdb/NEWS
> index 7ce9758..d0c865a 100644
> --- a/gdb/NEWS
> +++ b/gdb/NEWS
> @@ -5,6 +5,13 @@
>  
>  * Support for tracepoints on aarch64-linux was added in GDBserver.
>  
> +* The "disassemble" command accepts a new modifier: /s.
> +  It prints mixed source+disassembly like /m with two differences:
> +  - disassembled instructions are now printed in program order, and
> +  - and source for all relevant files is now printed.
> +  The "/m" option is now considered deprecated: its "source-centric"
> +  output hasn't proved useful in practice.
> +
>  *** Changes in GDB 7.10
>  
>  * Support for process record-replay and reverse debugging on aarch64*-linux*
> diff --git a/gdb/cli/cli-cmds.c b/gdb/cli/cli-cmds.c
> index 2ec2dd3..54b06fd 100644
> --- a/gdb/cli/cli-cmds.c
> +++ b/gdb/cli/cli-cmds.c
> @@ -1144,16 +1144,26 @@ disassemble_current_function (int flags)
>  /* Dump a specified section of assembly code.
>  
>     Usage:
> -     disassemble [/mr]
> +     disassemble [/mrs]
>         - dump the assembly code for the function of the current pc
> -     disassemble [/mr] addr
> +     disassemble [/mrs] addr
>         - dump the assembly code for the function at ADDR
> -     disassemble [/mr] low,high
> -     disassemble [/mr] low,+length
> +     disassemble [/mrs] low,high
> +     disassemble [/mrs] low,+length
>         - dump the assembly code in the range [LOW,HIGH), or [LOW,LOW+length)
>  
> -   A /m modifier will include source code with the assembly.
> -   A /r modifier will include raw instructions in hex with the assembly.  */
> +   A /m modifier will include source code with the assembly in a
> +   "source centric" view.  This view lists only the file of the first insn,
> +   even if other source files are involved (e.g., inlined functions), and
> +   the output is in source order, even with optimized code.  This view is
> +   considered deprecated as it hasn't been useful in practice.
> +
> +   A /r modifier will include raw instructions in hex with the assembly.
> +
> +   A /s modifier will include source code with the assembly, like /m, with
> +   two important differences:
> +   1) The output is still in pc address order.
> +   2) File names and contents for all relevant source files are displayed.  */
>  
>  static void
>  disassemble_command (char *arg, int from_tty)
> @@ -1181,11 +1191,14 @@ disassemble_command (char *arg, int from_tty)
>  	  switch (*p++)
>  	    {
>  	    case 'm':
> -	      flags |= DISASSEMBLY_SOURCE;
> +	      flags |= DISASSEMBLY_SOURCE_DEPRECATED;
>  	      break;
>  	    case 'r':
>  	      flags |= DISASSEMBLY_RAW_INSN;
>  	      break;
> +	    case 's':
> +	      flags |= DISASSEMBLY_SOURCE;
> +	      break;
>  	    default:
>  	      error (_("Invalid disassembly modifier."));
>  	    }
> @@ -1194,6 +1207,10 @@ disassemble_command (char *arg, int from_tty)
>        p = skip_spaces_const (p);
>      }
>  
> +  if ((flags & (DISASSEMBLY_SOURCE_DEPRECATED | DISASSEMBLY_SOURCE))
> +      == (DISASSEMBLY_SOURCE_DEPRECATED | DISASSEMBLY_SOURCE))
> +    error (_("Cannot specify both /m and /s."));
> +
>    if (! p || ! *p)
>      {
>        flags |= DISASSEMBLY_OMIT_FNAME;
> @@ -1854,8 +1871,21 @@ the other arg."));
>    c = add_com ("disassemble", class_vars, disassemble_command, _("\
>  Disassemble a specified section of memory.\n\
>  Default is the function surrounding the pc of the selected frame.\n\
> +\n\
>  With a /m modifier, source lines are included (if available).\n\
> +This view is \"source centric\": the output is in source line order,\n\
> +regardless of any optimization that is present.  Only the main source file\n\
> +is displayed, not those of, e.g., any inlined functions.\n\
> +This modifier hasn't proved useful in practice and is deprecated\n\
> +in favor of /s.\n\
> +\n\
> +With a /s modifier, source lines are included (if available).\n\
> +This differs from /m in two important respects:\n\
> +- the output is still in pc address order, and\n\
> +- file names and contents for all relevant source files are displayed.\n\
> +\n\
>  With a /r modifier, raw instructions in hex are included.\n\
> +\n\
>  With a single argument, the function surrounding that address is dumped.\n\
>  Two arguments (separated by a comma) are taken as a range of memory to dump,\n\
>    in the form of \"start,end\", or \"start,+length\".\n\
> diff --git a/gdb/disasm.c b/gdb/disasm.c
> index 483df01..2b65c6a 100644
> --- a/gdb/disasm.c
> +++ b/gdb/disasm.c
> @@ -24,23 +24,101 @@
>  #include "disasm.h"
>  #include "gdbcore.h"
>  #include "dis-asm.h"
> +#include "source.h"
>  
>  /* Disassemble functions.
>     FIXME: We should get rid of all the duplicate code in gdb that does
>     the same thing: disassemble_command() and the gdbtk variation.  */
>  
> -/* This Structure is used to store line number information.
> +/* This structure is used to store line number information for the
> +   deprecated /m option.
>     We need a different sort of line table from the normal one cuz we can't
>     depend upon implicit line-end pc's for lines to do the
>     reordering in this function.  */
>  
> -struct dis_line_entry
> +struct deprecated_dis_line_entry
>  {
>    int line;
>    CORE_ADDR start_pc;
>    CORE_ADDR end_pc;
>  };
>  
> +/* This Structure is used to store line number information.
> +   We need a different sort of line table from the normal one cuz we can't
> +   depend upon implicit line-end pc's for lines to do the
> +   reordering in this function.  */
> +
> +struct dis_line_entry
> +{
> +  struct symtab *symtab;
> +  int line;
> +};
> +
> +/* Hash function for dis_line_entry.  */
> +
> +static hashval_t
> +hash_dis_line_entry (const void *item)
> +{
> +  const struct dis_line_entry *dle = item;
> +
> +  return htab_hash_pointer (dle->symtab) + dle->line;
> +}
> +
> +/* Equal function for dis_line_entry.  */
> +
> +static int
> +eq_dis_line_entry (const void *item_lhs, const void *item_rhs)
> +{
> +  const struct dis_line_entry *lhs = item_lhs;
> +  const struct dis_line_entry *rhs = item_rhs;
> +
> +  return (lhs->symtab == rhs->symtab
> +	  && lhs->line == rhs->line);
> +}
> +
> +/* Create the table to manage lines for mixed source/disassembly.  */
> +
> +static htab_t
> +allocate_dis_line_table (void)
> +{
> +  return htab_create_alloc (41,
> +			    hash_dis_line_entry, eq_dis_line_entry,
> +			    xfree, xcalloc, xfree);
> +}
> +
> +/* Add DLE to TABLE.
> +   Returns 1 if added, 0 if already present.  */
> +
> +static void
> +maybe_add_dis_line_entry (htab_t table, struct symtab *symtab, int line)
> +{
> +  void **slot;
> +  struct dis_line_entry dle, *dlep;
> +
> +  dle.symtab = symtab;
> +  dle.line = line;
> +  slot = htab_find_slot (table, &dle, INSERT);
> +  if (*slot == NULL)
> +    {
> +      dlep = XNEW (struct dis_line_entry);
> +      dlep->symtab = symtab;
> +      dlep->line = line;
> +      *slot = dlep;
> +    }
> +}
> +
> +/* Return non-zero if SYMTAB, LINE are in TABLE.  */
> +
> +static int
> +line_has_code_p (htab_t table, struct symtab *symtab, int line)
> +{
> +  struct dis_line_entry dle;
> +
> +  dle.symtab = symtab;
> +  dle.line = line;
> +  return htab_find (table, &dle) != NULL;
> +}
> +
>  /* Like target_read_memory, but slightly different parameters.  */
>  static int
>  dis_asm_read_memory (bfd_vma memaddr, gdb_byte *myaddr, unsigned int len,
> @@ -69,11 +147,11 @@ dis_asm_print_address (bfd_vma addr, struct disassemble_info *info)
>  static int
>  compare_lines (const void *mle1p, const void *mle2p)
>  {
> -  struct dis_line_entry *mle1, *mle2;
> +  struct deprecated_dis_line_entry *mle1, *mle2;
>    int val;
>  
> -  mle1 = (struct dis_line_entry *) mle1p;
> -  mle2 = (struct dis_line_entry *) mle2p;
> +  mle1 = (struct deprecated_dis_line_entry *) mle1p;
> +  mle2 = (struct deprecated_dis_line_entry *) mle2p;
>  
>    /* End of sequence markers have a line number of 0 but don't want to
>       be sorted to the head of the list, instead sort by PC.  */
> @@ -96,7 +174,8 @@ static int
>  dump_insns (struct gdbarch *gdbarch, struct ui_out *uiout,
>  	    struct disassemble_info * di,
>  	    CORE_ADDR low, CORE_ADDR high,
> -	    int how_many, int flags, struct ui_file *stb)
> +	    int how_many, int flags, struct ui_file *stb,
> +	    CORE_ADDR *end_pc)
>  {
>    int num_displayed = 0;
>    CORE_ADDR pc;
> @@ -182,24 +261,30 @@ dump_insns (struct gdbarch *gdbarch, struct ui_out *uiout,
>        do_cleanups (ui_out_chain);
>        ui_out_text (uiout, "\n");
>      }
> +
> +  if (end_pc != NULL)
> +    *end_pc = pc;
>    return num_displayed;
>  }
>  
>  /* The idea here is to present a source-O-centric view of a
>     function to the user.  This means that things are presented
>     in source order, with (possibly) out of order assembly
> -   immediately following.  */
> +   immediately following.
> +
> +   N.B. This view is deprecated.  */
>  
>  static void
> -do_mixed_source_and_assembly (struct gdbarch *gdbarch, struct ui_out *uiout,
> -			      struct disassemble_info *di, int nlines,
> -			      struct linetable_entry *le,
> -			      CORE_ADDR low, CORE_ADDR high,
> -			      struct symtab *symtab,
> -			      int how_many, int flags, struct ui_file *stb)
> +do_mixed_source_and_assembly_deprecated
> +  (struct gdbarch *gdbarch, struct ui_out *uiout,
> +   struct disassemble_info *di, struct symtab *symtab,
> +   CORE_ADDR low, CORE_ADDR high,
> +   int how_many, int flags, struct ui_file *stb)
>  {
>    int newlines = 0;
> -  struct dis_line_entry *mle;
> +  int nlines;
> +  struct linetable_entry *le;
> +  struct deprecated_dis_line_entry *mle;
>    struct symtab_and_line sal;
>    int i;
>    int out_of_order = 0;
> @@ -210,11 +295,16 @@ do_mixed_source_and_assembly (struct gdbarch *gdbarch, struct ui_out *uiout,
>    struct cleanup *ui_out_tuple_chain = make_cleanup (null_cleanup, 0);
>    struct cleanup *ui_out_list_chain = make_cleanup (null_cleanup, 0);
>  
> +  gdb_assert (symtab != NULL && SYMTAB_LINETABLE (symtab) != NULL);
> +
> +  nlines = SYMTAB_LINETABLE (symtab)->nitems;
> +  le = SYMTAB_LINETABLE (symtab)->item;
> +
>    if (flags & DISASSEMBLY_FILENAME)
>      psl_flags |= PRINT_SOURCE_LINES_FILENAME;
>  
> -  mle = (struct dis_line_entry *) alloca (nlines
> -					  * sizeof (struct dis_line_entry));
> +  mle = (struct deprecated_dis_line_entry *)
> +    alloca (nlines * sizeof (struct deprecated_dis_line_entry));
>  
>    /* Copy linetable entries for this function into our data
>       structure, creating end_pc's and setting out_of_order as
> @@ -255,11 +345,11 @@ do_mixed_source_and_assembly (struct gdbarch *gdbarch, struct ui_out *uiout,
>        newlines++;
>      }
>  
> -  /* Now, sort mle by line #s (and, then by addresses within
> -     lines).  */
> +  /* Now, sort mle by line #s (and, then by addresses within lines).  */
>  
>    if (out_of_order)
> -    qsort (mle, newlines, sizeof (struct dis_line_entry), compare_lines);
> +    qsort (mle, newlines, sizeof (struct deprecated_dis_line_entry),
> +	   compare_lines);
>  
>    /* Now, for each line entry, emit the specified lines (unless
>       they have been emitted before), followed by the assembly code
> @@ -324,7 +414,7 @@ do_mixed_source_and_assembly (struct gdbarch *gdbarch, struct ui_out *uiout,
>  
>        num_displayed += dump_insns (gdbarch, uiout, di,
>  				   mle[i].start_pc, mle[i].end_pc,
> -				   how_many, flags, stb);
> +				   how_many, flags, stb, NULL);
>  
>        /* When we've reached the end of the mle array, or we've seen the last
>           assembly range for this source line, close out the list/tuple.  */
> @@ -342,6 +432,260 @@ do_mixed_source_and_assembly (struct gdbarch *gdbarch, struct ui_out *uiout,
>    do_cleanups (ui_out_chain);
>  }
>  
> +/* The idea here is to present a source-O-centric view of a
> +   function to the user.  This means that things are presented
> +   in source order, with (possibly) out of order assembly
> +   immediately following.  */
> +
> +static void
> +do_mixed_source_and_assembly (struct gdbarch *gdbarch, struct ui_out *uiout,
> +			      struct disassemble_info *di,
> +			      struct symtab *main_symtab,
> +			      CORE_ADDR low, CORE_ADDR high,
> +			      int how_many, int flags, struct ui_file *stb)
> +{
> +  int newlines = 0;
> +  const struct linetable_entry *le, *first_le;
> +  struct symtab_and_line sal;
> +  int i, nlines;
> +  int out_of_order = 0;
> +  int next_line = 0;
> +  int num_displayed = 0;
> +  enum print_source_lines_flags psl_flags = 0;
> +  struct cleanup *cleanups;
> +  struct cleanup *ui_out_chain;
> +  struct cleanup *ui_out_tuple_chain;
> +  struct cleanup *ui_out_list_chain;
> +  CORE_ADDR pc;
> +  struct symtab *last_symtab;
> +  int last_line;
> +  htab_t dis_line_table;
> +
> +  gdb_assert (main_symtab != NULL && SYMTAB_LINETABLE (main_symtab) != NULL);
> +
> +  /* First pass: collect the list of all source files and lines.
> +     We do this so that we can only print lines containing code once.
> +     We try to print the source text leading up to the next instruction,
> +     but if that text is for code that will be disassembled later, then
> +     we'll want to defer printing it until later with its associated code.  */
> +
> +  dis_line_table = allocate_dis_line_table ();
> +  cleanups = make_cleanup_htab_delete (dis_line_table);
> +
> +  pc = low;
> +
> +  /* The prologue may be empty, but there may still be a line number entry
> +     for the opening brace which is distinct from the first line of code.
> +     If the prologue has been eliminated find_pc_line may return the source
> +     line after the opening brace.  We still want to print this opening brace.
> +     first_le is used to implement this.  */
> +
> +  nlines = SYMTAB_LINETABLE (main_symtab)->nitems;
> +  le = SYMTAB_LINETABLE (main_symtab)->item;
> +  first_le = NULL;
> +
> +  /* Skip all the preceding functions.  */
> +  for (i = 0; i < nlines && le[i].pc < low; i++)
> +    continue;
> +
> +  if (i < nlines && le[i].pc < high)
> +    first_le = &le[i];
> +
> +  /* Add lines for every pc value.  */
> +  while (pc < high)
> +    {
> +      struct symtab_and_line sal;
> +      int length;
> +
> +      sal = find_pc_line (pc, 0);
> +      length = gdb_insn_length (gdbarch, pc);
> +      pc += length;
> +
> +      if (sal.symtab != NULL)
> +	maybe_add_dis_line_entry (dis_line_table, sal.symtab, sal.line);
> +    }
> +
> +  /* Second pass: print the disassembly.
> +
> +     Output format, from an MI perspective:
> +       The result is a ui_out list, field name "asm_insns", where elements have
> +       name "src_and_asm_line".
> +       Each element is a tuple of source line specs (field names line, file,
> +       fullname), and field "line_asm_insn" which contains the disassembly.
> +       Field "line_asm_insn" is a list of tuples: address, func-name, offset,
> +       opcodes, inst.
> +
> +     CLI output works on top of this because MI ignores ui_out_text output,
> +     which is where we put file name and source line contents output.
> +
> +     Cleanup usage:
> +     cleanups:
> +       For things created at the beginning of this function and need to be
> +       kept until the end of this function.
> +     ui_out_chain
> +       Handles the outer "asm_insns" list.
> +     ui_out_tuple_chain
> +       The tuples for each group of consecutive disassemblies.
> +     ui_out_list_chain
> +       List of consecutive source lines or disassembled insns.  */
> +
> +  if (flags & DISASSEMBLY_FILENAME)
> +    psl_flags |= PRINT_SOURCE_LINES_FILENAME;
> +
> +  ui_out_chain = make_cleanup_ui_out_list_begin_end (uiout, "asm_insns");
> +
> +  ui_out_tuple_chain = NULL;
> +  ui_out_list_chain = NULL;
> +
> +  last_symtab = NULL;
> +  last_line = 0;
> +  pc = low;
> +
> +  while (pc < high)
> +    {
> +      struct linetable_entry *le = NULL;
> +      struct symtab_and_line sal;
> +      CORE_ADDR end_pc;
> +      int start_preceding_line_to_display = 0;
> +      int end_preceding_line_to_display = 0;
> +      int new_source_line = 0;
> +
> +      sal = find_pc_line (pc, 0);
> +
> +      if (sal.symtab != last_symtab)
> +	{
> +	  /* New source file.  */
> +	  new_source_line = 1;
> +
> +	  /* If this is the first line of output, check for any preceding
> +	     lines.  */
> +	  if (last_line == 0
> +	      && first_le != NULL
> +	      && first_le->line < sal.line)
> +	    {
> +	      start_preceding_line_to_display = first_le->line;
> +	      end_preceding_line_to_display = sal.line;
> +	    }
> +	}
> +      else
> +	{
> +	  /* Same source file as last time.  */
> +	  if (sal.symtab != NULL)
> +	    {
> +	      if (sal.line > last_line + 1 && last_line != 0)
> +		{
> +		  int l;
> +
> +		  /* Several preceding source lines.  Print the trailing ones
> +		     not associated with code that we'll print later.  */
> +		  for (l = sal.line - 1; l > last_line; --l)
> +		    {
> +		      if (line_has_code_p (dis_line_table, sal.symtab, l))
> +			break;
> +		    }
> +		  if (l < sal.line - 1)
> +		    {
> +		      start_preceding_line_to_display = l + 1;
> +		      end_preceding_line_to_display = sal.line;
> +		    }
> +		}
> +	      if (sal.line != last_line)
> +		new_source_line = 1;
> +	      else
> +		{
> +		  /* Same source line as last time.  This can happen, depending
> +		     on the debug info.  */
> +		}
> +	    }
> +	}
> +
> +      if (new_source_line)
> +	{
> +	  /* Skip the newline if this is the first instruction.  */
> +	  if (pc > low)
> +	    ui_out_text (uiout, "\n");
> +	  if (ui_out_tuple_chain != NULL)
> +	    {
> +	      gdb_assert (ui_out_list_chain != NULL);
> +	      do_cleanups (ui_out_list_chain);
> +	      do_cleanups (ui_out_tuple_chain);
> +	    }
> +	  if (sal.symtab != last_symtab
> +	      && !(flags & DISASSEMBLY_FILENAME))
> +	    {
> +	      /* Remember MI ignores ui_out_text.
> +		 We don't have to do anything here for MI because MI
> +		 output includes the source specs for each line.  */
> +	      if (sal.symtab != NULL)
> +		{
> +		  ui_out_text (uiout,
> +			       symtab_to_filename_for_display (sal.symtab));
> +		}
> +	      else
> +		ui_out_text (uiout, "unknown");
> +	      ui_out_text (uiout, ":\n");
> +	    }
> +	  if (start_preceding_line_to_display > 0)
> +	    {
> +	      /* Several source lines w/o asm instructions associated.
> +		 We need to preserve the structure of the output, so output
> +		 a bunch of line tuples with no asm entries.  */
> +	      int l;
> +	      struct cleanup *ui_out_list_chain_line;
> +	      struct cleanup *ui_out_tuple_chain_line;
> +
> +	      gdb_assert (sal.symtab != NULL);
> +	      for (l = start_preceding_line_to_display;
> +		   l < end_preceding_line_to_display;
> +		   ++l)
> +		{
> +		  ui_out_tuple_chain_line
> +		    = make_cleanup_ui_out_tuple_begin_end (uiout,
> +							   "src_and_asm_line");
> +		  print_source_lines (sal.symtab, l, l + 1, psl_flags);
> +		  ui_out_list_chain_line
> +		    = make_cleanup_ui_out_list_begin_end (uiout,
> +							  "line_asm_insn");
> +		  do_cleanups (ui_out_list_chain_line);
> +		  do_cleanups (ui_out_tuple_chain_line);
> +		}
> +	    }
> +	  ui_out_tuple_chain
> +	    = make_cleanup_ui_out_tuple_begin_end (uiout, "src_and_asm_line");
> +	  if (sal.symtab != NULL)
> +	    print_source_lines (sal.symtab, sal.line, sal.line + 1, psl_flags);
> +	  else
> +	    ui_out_text (uiout, _("--- no source info for this pc ---\n"));
> +	  ui_out_list_chain
> +	    = make_cleanup_ui_out_list_begin_end (uiout, "line_asm_insn");
> +	}
> +      else
> +	{
> +	  /* Here we're appending instructions to an existing line.
> +	     By construction the very first insn will have a symtab
> +	     and follow the new_source_line path above.  */
> +	  gdb_assert (ui_out_tuple_chain != NULL);
> +	  gdb_assert (ui_out_list_chain != NULL);
> +	}
> +
> +      if (sal.end != 0)
> +	end_pc = min (sal.end, high);
> +      else
> +	end_pc = pc + 1;
> +      num_displayed += dump_insns (gdbarch, uiout, di, pc, end_pc,
> +				   how_many, flags, stb, &end_pc);
> +      pc = end_pc;
> +
> +      if (how_many >= 0 && num_displayed >= how_many)
> +	break;
> +
> +      last_symtab = sal.symtab;
> +      last_line = sal.line;
> +    }
> +
> +  do_cleanups (ui_out_chain);
> +  do_cleanups (cleanups);
> +}
>  
>  static void
>  do_assembly_only (struct gdbarch *gdbarch, struct ui_out *uiout,
> @@ -355,7 +699,7 @@ do_assembly_only (struct gdbarch *gdbarch, struct ui_out *uiout,
>    ui_out_chain = make_cleanup_ui_out_list_begin_end (uiout, "asm_insns");
>  
>    num_displayed = dump_insns (gdbarch, uiout, di, low, high, how_many,
> -                              flags, stb);
> +                              flags, stb, NULL);
>  
>    do_cleanups (ui_out_chain);
>  }
> @@ -418,19 +762,19 @@ gdb_disassembly (struct gdbarch *gdbarch, struct ui_out *uiout,
>    symtab = find_pc_line_symtab (low);
>  
>    if (symtab != NULL && SYMTAB_LINETABLE (symtab) != NULL)
> -    {
> -      /* Convert the linetable to a bunch of my_line_entry's.  */
> -      le = SYMTAB_LINETABLE (symtab)->item;
> -      nlines = SYMTAB_LINETABLE (symtab)->nitems;
> -    }
> +    nlines = SYMTAB_LINETABLE (symtab)->nitems;
>  
> -  if (!(flags & DISASSEMBLY_SOURCE) || nlines <= 0
> -      || symtab == NULL || SYMTAB_LINETABLE (symtab) == NULL)
> +  if (!(flags & (DISASSEMBLY_SOURCE_DEPRECATED | DISASSEMBLY_SOURCE))
> +      || nlines <= 0)
>      do_assembly_only (gdbarch, uiout, &di, low, high, how_many, flags, stb);
>  
>    else if (flags & DISASSEMBLY_SOURCE)
> -    do_mixed_source_and_assembly (gdbarch, uiout, &di, nlines, le, low,
> -				  high, symtab, how_many, flags, stb);
> +    do_mixed_source_and_assembly (gdbarch, uiout, &di, symtab, low, high,
> +				  how_many, flags, stb);
> +
> +  else if (flags & DISASSEMBLY_SOURCE_DEPRECATED)
> +    do_mixed_source_and_assembly_deprecated (gdbarch, uiout, &di, symtab,
> +					     low, high, how_many, flags, stb);
>  
>    do_cleanups (cleanups);
>    gdb_flush (gdb_stdout);
> diff --git a/gdb/disasm.h b/gdb/disasm.h
> index a91211e..7e6f1a2 100644
> --- a/gdb/disasm.h
> +++ b/gdb/disasm.h
> @@ -21,11 +21,12 @@
>  
>  #include "dis-asm.h"
>  
> -#define DISASSEMBLY_SOURCE	(0x1 << 0)
> +#define DISASSEMBLY_SOURCE_DEPRECATED (0x1 << 0)
>  #define DISASSEMBLY_RAW_INSN	(0x1 << 1)
>  #define DISASSEMBLY_OMIT_FNAME	(0x1 << 2)
>  #define DISASSEMBLY_FILENAME	(0x1 << 3)
>  #define DISASSEMBLY_OMIT_PC	(0x1 << 4)
> +#define DISASSEMBLY_SOURCE	(0x1 << 5)
>  
>  struct gdbarch;
>  struct ui_out;
> diff --git a/gdb/doc/gdb.texinfo b/gdb/doc/gdb.texinfo
> index 9e2ecd1..612a5af 100644
> --- a/gdb/doc/gdb.texinfo
> +++ b/gdb/doc/gdb.texinfo
> @@ -8028,11 +8028,12 @@ Variables}).
>  @cindex listing machine instructions
>  @item disassemble
>  @itemx disassemble /m
> +@itemx disassemble /s
>  @itemx disassemble /r
>  This specialized command dumps a range of memory as machine
>  instructions.  It can also print mixed source+disassembly by specifying
> -the @code{/m} modifier and print the raw instructions in hex as well as
> -in symbolic form by specifying the @code{/r}.
> +the @code{/m} or @code{/s} modifier and print the raw instructions in hex
> +as well as in symbolic form by specifying the @code{/r} modifier.
>  The default memory range is the function surrounding the
>  program counter of the selected frame.  A single argument to this
>  command is a program counter value; @value{GDBN} dumps the function
> @@ -8076,8 +8077,9 @@ Dump of assembler code from 0x32c4 to 0x32e4:
>  End of assembler dump.
>  @end smallexample
>  
> -Here is an example showing mixed source+assembly for Intel x86, when the
> -program is stopped just after function prologue:
> +Here is an example showing mixed source+assembly for Intel x86
> +with @code{/m} or @code{/s}, when the program is stopped just after
> +function prologue in a non-optimized function with no inline code.
>  
>  @smallexample
>  (@value{GDBP}) disas /m main
> @@ -8102,6 +8104,96 @@ Dump of assembler code for function main:
>  End of assembler dump.
>  @end smallexample
>  
> +The @code{/m} option is deprecated as its output is not useful when
> +there is either inlined code or re-ordered code.
> +The @code{/s} option is the preferred choice.
> +Here is an example for AMD x86-64 showing the difference between
> +@code{/m} output and @code{/s} output.
> +This example has one inline function defined in a header file,
> +and the code is compiled with @samp{-O2} optimization.
> +Note how the @code{/m} output is missing the disassembly of
> +several instructions that are present in the @code{/s} output.
> +
> +@file{foo.h}:
> +
> +@smallexample
> +int
> +foo (int a)
> +@{
> +  if (a < 0)
> +    return a * 2;
> +  if (a == 0)
> +    return 1;
> +  return a + 10;
> +@}
> +@end smallexample
> +
> +@file{foo.c}:
> +
> +@smallexample
> +#include "foo.h"
> +volatile int x, y;
> +int
> +main ()
> +@{
> +  x = foo (y);
> +  return 0;
> +@}
> +@end smallexample
> +
> +@smallexample
> +(@value{GDBP}) disas /m main
> +Dump of assembler code for function main:
> +5	@{
> +
> +6	  x = foo (y);
> +   0x0000000000400400 <+0>:	mov    0x200c2e(%rip),%eax # 0x601034 <y>
> +   0x0000000000400417 <+23>:	mov    %eax,0x200c13(%rip) # 0x601030 <x>
> +
> +7	  return 0;
> +8	@}
> +   0x000000000040041d <+29>:	xor    %eax,%eax
> +   0x000000000040041f <+31>:	retq
> +   0x0000000000400420 <+32>:	add    %eax,%eax
> +   0x0000000000400422 <+34>:	jmp    0x400417 <main+23>
> +
> +End of assembler dump.
> +(@value{GDBP}) disas /s main
> +Dump of assembler code for function main:
> +foo.c:
> +5	@{
> +6	  x = foo (y);
> +   0x0000000000400400 <+0>:	mov    0x200c2e(%rip),%eax # 0x601034 <y>
> +
> +foo.h:
> +4	  if (a < 0)
> +   0x0000000000400406 <+6>:	test   %eax,%eax
> +   0x0000000000400408 <+8>:	js     0x400420 <main+32>
> +
> +6	  if (a == 0)
> +7	    return 1;
> +8	  return a + 10;
> +   0x000000000040040a <+10>:	lea    0xa(%rax),%edx
> +   0x000000000040040d <+13>:	test   %eax,%eax
> +   0x000000000040040f <+15>:	mov    $0x1,%eax
> +   0x0000000000400414 <+20>:	cmovne %edx,%eax
> +
> +foo.c:
> +6	  x = foo (y);
> +   0x0000000000400417 <+23>:	mov    %eax,0x200c13(%rip) # 0x601030 <x>
> +
> +7	  return 0;
> +8	@}
> +   0x000000000040041d <+29>:	xor    %eax,%eax
> +   0x000000000040041f <+31>:	retq
> +
> +foo.h:
> +5	    return a * 2;
> +   0x0000000000400420 <+32>:	add    %eax,%eax
> +   0x0000000000400422 <+34>:	jmp    0x400417 <main+23>
> +End of assembler dump.
> +@end smallexample
> +
>  Here is another example showing raw instructions in hex for AMD x86-64,
>  
>  @smallexample
> @@ -29705,9 +29797,20 @@ displayed; if @var{lines} is higher than the number of lines between
>  @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
>  are displayed.
>  @item @var{mode}
> -is either 0 (meaning only disassembly), 1 (meaning mixed source and
> -disassembly), 2 (meaning disassembly with raw opcodes), or 3 (meaning
> -mixed source and disassembly with raw opcodes).
> +is one of:
> +@itemize @bullet
> +@item 0 disassembly only
> +@item 1 mixed source and disassembly (deprecated)
> +@item 2 disassembly with raw opcodes
> +@item 3 mixed source and disassembly with raw opcodes (deprecated)
> +@item 4 mixed source and disassembly
> +@item 5 mixed source and disassembly with raw opcodes
> +@end itemize
> +
> +Modes 1 and 3 are deprecated.  The output is ``source centric''
> +which hasn't proved useful in practice.
> +@xref{Machine Code}, for a discussion of the difference between
> +@code{/m} and @code{/s} output of the @code{disassemble} command.
>  @end table
>  
>  @subsubheading Result
> @@ -29733,12 +29836,12 @@ The decimal offset in bytes from the start of @samp{func-name}.
>  The text disassembly for this @samp{address}.
>  
>  @item opcodes
> -This field is only present for mode 2.  This contains the raw opcode
> +This field is only present for modes 2, 3 and 5.  This contains the raw opcode
>  bytes for the @samp{inst} field.
>  
>  @end table
>  
> -For modes 1 and 3 the @samp{asm_insns} list contains tuples named
> +For modes 1, 3, 4 and 5 the @samp{asm_insns} list contains tuples named
>  @samp{src_and_asm_line}, each of which has the following fields:
>  
>  @table @code
> diff --git a/gdb/mi/mi-cmd-disas.c b/gdb/mi/mi-cmd-disas.c
> index 01cad2d..250124d 100644
> --- a/gdb/mi/mi-cmd-disas.c
> +++ b/gdb/mi/mi-cmd-disas.c
> @@ -44,10 +44,11 @@
>     always required:
>  
>     MODE: 0 -- disassembly.
> -         1 -- disassembly and source.
> +         1 -- disassembly and source (with deprecated source-centric view).
>           2 -- disassembly and opcodes.
> -         3 -- disassembly, source and opcodes.
> -*/
> +         3 -- disassembly, source-centric and opcodes.
> +         4 -- disassembly, and source (with pc-centric view).
> +         5 -- disassembly, source (pc-centric) and opcodes.  */
>  
>  void
>  mi_cmd_disassemble (char *command, char **argv, int argc)
> @@ -141,16 +142,34 @@ mi_cmd_disassemble (char *command, char **argv, int argc)
>  	     "[-n howmany]] [-s startaddr -e endaddr] [--] mode."));
>  
>    mode = atoi (argv[0]);
> -  if (mode < 0 || mode > 3)
> -    error (_("-data-disassemble: Mode argument must be 0, 1, 2, or 3."));
> +  if (mode < 0 || mode > 5)
> +    error (_("-data-disassemble: Mode argument must be in the range 0-5."));
>  
>    /* Convert the mode into a set of disassembly flags.  */
>  
> -  disasm_flags = 0;
> -  if (mode & 0x1)
> -    disasm_flags |= DISASSEMBLY_SOURCE;
> -  if (mode & 0x2)
> -    disasm_flags |= DISASSEMBLY_RAW_INSN;
> +  disasm_flags = 0;  /* Initialize here for -Wall.  */
> +  switch (mode)
> +    {
> +    case 0:
> +      break;
> +    case 1:
> +      disasm_flags |= DISASSEMBLY_SOURCE_DEPRECATED;
> +      break;
> +    case 2:
> +      disasm_flags |= DISASSEMBLY_RAW_INSN;
> +      break;
> +    case 3:
> +      disasm_flags |= DISASSEMBLY_SOURCE_DEPRECATED | DISASSEMBLY_RAW_INSN;
> +      break;
> +    case 4:
> +      disasm_flags |= DISASSEMBLY_SOURCE;
> +      break;
> +    case 5:
> +      disasm_flags |= DISASSEMBLY_SOURCE | DISASSEMBLY_RAW_INSN;
> +      break;
> +    default:
> +      gdb_assert_not_reached ("bad disassembly mode");
> +    }
>  
>    /* We must get the function beginning and end where line_num is
>       contained.  */
> diff --git a/gdb/record.c b/gdb/record.c
> index a64543a..ad83a29 100644
> --- a/gdb/record.c
> +++ b/gdb/record.c
> @@ -458,7 +458,7 @@ get_insn_history_modifiers (char **arg)
>  	  switch (*args)
>  	    {
>  	    case 'm':
> -	      modifiers |= DISASSEMBLY_SOURCE;
> +	      modifiers |= DISASSEMBLY_SOURCE_DEPRECATED;
>  	      modifiers |= DISASSEMBLY_FILENAME;
>  	      break;
>  	    case 'r':
> diff --git a/gdb/testsuite/gdb.base/disasm-optim.S b/gdb/testsuite/gdb.base/disasm-optim.S
> new file mode 100644
> index 0000000..0856731
> --- /dev/null
> +++ b/gdb/testsuite/gdb.base/disasm-optim.S
> @@ -0,0 +1,352 @@
> +/* This testcase is part of GDB, the GNU debugger.
> +
> +   Copyright (C) 2015 Free Software Foundation, Inc.
> +
> +   This program is free software; you can redistribute it and/or modify
> +   it under the terms of the GNU General Public License as published by
> +   the Free Software Foundation; either version 3 of the License, or
> +   (at your option) any later version.
> +
> +   This program is distributed in the hope that it will be useful,
> +   but WITHOUT ANY WARRANTY; without even the implied warranty of
> +   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> +   GNU General Public License for more details.
> +
> +   You should have received a copy of the GNU General Public License
> +   along with this program.  If not, see <http://www.gnu.org/licenses/>.
> +
> +   This file was created with gcc -O2 -g -S -fverbose-asm -dA disasm-optim.c
> +   and then cleaning up the output.  */
> +
> +	.file	"disasm-optim.c"
> +
> +	.text
> +	.p2align 4,,15
> +	.globl	main
> +	.type	main, @function
> +main:
> +.LFB1:
> +	.file 1 "disasm-optim.c"
> +	# disasm-optim.c:24
> +	.loc 1 24 0
> +	.cfi_startproc
> +	# disasm-optim.c:25
> +	.loc 1 25 0
> +	movl	y(%rip), %eax
> +.LVL0:
> +.LBB4:
> +.LBB5:
> +	.file 2 "disasm-optim.h"
> +	# disasm-optim.h:21
> +	.loc 2 21 0
> +	testl	%eax, %eax
> +	js	.L6
> +	# disasm-optim.h:25
> +	.loc 2 25 0
> +	leal	10(%rax), %edx
> +	testl	%eax, %eax
> +	movl	$1, %eax
> +.LVL1:
> +	cmovne	%edx, %eax
> +.LVL2:
> +.L3:
> +.LBE5:
> +.LBE4:
> +	# disasm-optim.c:25
> +	.loc 1 25 0
> +	movl	%eax, x(%rip)
> +	# disasm-optim.c:27
> +	.loc 1 27 0
> +	xorl	%eax, %eax
> +	ret
> +.LVL3:
> +.L6:
> +.LBB7:
> +.LBB6:
> +	# disasm-optim.h:22
> +	.loc 2 22 0
> +	addl	%eax, %eax
> +.LVL4:
> +	jmp	.L3
> +.LBE6:
> +.LBE7:
> +	.cfi_endproc
> +.LFE1:
> +	.size	main, .-main
> +	.comm	y,4,4
> +	.comm	x,4,4
> +	.text
> +.Letext0:
> +	.section	.debug_info,"",@progbits
> +.Ldebug_info0:
> +	.long	0xb1	# Length of Compilation Unit Info
> +	.value	0x4	# DWARF version number
> +	.long	.Ldebug_abbrev0	# Offset Into Abbrev. Section
> +	.byte	0x8	# Pointer Size (in bytes)
> +	.uleb128 0x1	# (DIE (0xb) DW_TAG_compile_unit)
> +	.long	.LASF0	# DW_AT_producer: "GNU C 4.9.2 20150212 (Red Hat 4.9.2-6) -mtune=generic -march=x86-64 -g -O2"
> +	.byte	0x1	# DW_AT_language
> +	.long	.LASF1	# DW_AT_name: "disasm-optim.c"
> +	.long	.LASF2	# DW_AT_comp_dir: "/main/disassemble3/gdb/testsuite/gdb.base"
> +	.long	.Ldebug_ranges0+0x30	# DW_AT_ranges
> +	.quad	0	# DW_AT_low_pc
> +	.long	.Ldebug_line0	# DW_AT_stmt_list
> +	.uleb128 0x2	# (DIE (0x29) DW_TAG_subprogram)
> +			# DW_AT_external
> +	.ascii "foo\0"	# DW_AT_name
> +	.byte	0x2	# DW_AT_decl_file (disasm-optim.h)
> +	.byte	0x13	# DW_AT_decl_line
> +			# DW_AT_prototyped
> +	.long	0x43	# DW_AT_type
> +	.byte	0x3	# DW_AT_inline
> +	.long	0x43	# DW_AT_sibling
> +	.uleb128 0x3	# (DIE (0x39) DW_TAG_formal_parameter)
> +	.ascii "a\0"	# DW_AT_name
> +	.byte	0x2	# DW_AT_decl_file (disasm-optim.h)
> +	.byte	0x13	# DW_AT_decl_line
> +	.long	0x43	# DW_AT_type
> +	.byte	0	# end of children of DIE 0x29
> +	.uleb128 0x4	# (DIE (0x43) DW_TAG_base_type)
> +	.byte	0x4	# DW_AT_byte_size
> +	.byte	0x5	# DW_AT_encoding
> +	.ascii "int\0"	# DW_AT_name
> +	.uleb128 0x5	# (DIE (0x4a) DW_TAG_subprogram)
> +			# DW_AT_external
> +	.long	.LASF3	# DW_AT_name: "main"
> +	.byte	0x1	# DW_AT_decl_file (disasm-optim.c)
> +	.byte	0x17	# DW_AT_decl_line
> +	.long	0x43	# DW_AT_type
> +	.quad	.LFB1	# DW_AT_low_pc
> +	.quad	.LFE1-.LFB1	# DW_AT_high_pc
> +	.uleb128 0x1	# DW_AT_frame_base
> +	.byte	0x9c	# DW_OP_call_frame_cfa
> +			# DW_AT_GNU_all_call_sites
> +	.long	0x89	# DW_AT_sibling
> +	.uleb128 0x6	# (DIE (0x6b) DW_TAG_inlined_subroutine)
> +	.long	0x29	# DW_AT_abstract_origin
> +	.quad	.LBB4	# DW_AT_entry_pc
> +	.long	.Ldebug_ranges0+0	# DW_AT_ranges
> +	.byte	0x1	# DW_AT_call_file (disasm-optim.c)
> +	.byte	0x19	# DW_AT_call_line
> +	.uleb128 0x7	# (DIE (0x7e) DW_TAG_formal_parameter)
> +	.long	0x39	# DW_AT_abstract_origin
> +	.long	.LLST0	# DW_AT_location
> +	.byte	0	# end of children of DIE 0x6b
> +	.byte	0	# end of children of DIE 0x4a
> +	.uleb128 0x8	# (DIE (0x89) DW_TAG_variable)
> +	.ascii "x\0"	# DW_AT_name
> +	.byte	0x1	# DW_AT_decl_file (disasm-optim.c)
> +	.byte	0x14	# DW_AT_decl_line
> +	.long	0x9c	# DW_AT_type
> +			# DW_AT_external
> +	.uleb128 0x9	# DW_AT_location
> +	.byte	0x3	# DW_OP_addr
> +	.quad	x
> +	.uleb128 0x9	# (DIE (0x9c) DW_TAG_volatile_type)
> +	.long	0x43	# DW_AT_type
> +	.uleb128 0x8	# (DIE (0xa1) DW_TAG_variable)
> +	.ascii "y\0"	# DW_AT_name
> +	.byte	0x1	# DW_AT_decl_file (disasm-optim.c)
> +	.byte	0x14	# DW_AT_decl_line
> +	.long	0x9c	# DW_AT_type
> +			# DW_AT_external
> +	.uleb128 0x9	# DW_AT_location
> +	.byte	0x3	# DW_OP_addr
> +	.quad	y
> +	.byte	0	# end of children of DIE 0xb
> +	.section	.debug_abbrev,"",@progbits
> +.Ldebug_abbrev0:
> +	.uleb128 0x1	# (abbrev code)
> +	.uleb128 0x11	# (TAG: DW_TAG_compile_unit)
> +	.byte	0x1	# DW_children_yes
> +	.uleb128 0x25	# (DW_AT_producer)
> +	.uleb128 0xe	# (DW_FORM_strp)
> +	.uleb128 0x13	# (DW_AT_language)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.uleb128 0x3	# (DW_AT_name)
> +	.uleb128 0xe	# (DW_FORM_strp)
> +	.uleb128 0x1b	# (DW_AT_comp_dir)
> +	.uleb128 0xe	# (DW_FORM_strp)
> +	.uleb128 0x55	# (DW_AT_ranges)
> +	.uleb128 0x17	# (DW_FORM_sec_offset)
> +	.uleb128 0x11	# (DW_AT_low_pc)
> +	.uleb128 0x1	# (DW_FORM_addr)
> +	.uleb128 0x10	# (DW_AT_stmt_list)
> +	.uleb128 0x17	# (DW_FORM_sec_offset)
> +	.byte	0
> +	.byte	0
> +	.uleb128 0x2	# (abbrev code)
> +	.uleb128 0x2e	# (TAG: DW_TAG_subprogram)
> +	.byte	0x1	# DW_children_yes
> +	.uleb128 0x3f	# (DW_AT_external)
> +	.uleb128 0x19	# (DW_FORM_flag_present)
> +	.uleb128 0x3	# (DW_AT_name)
> +	.uleb128 0x8	# (DW_FORM_string)
> +	.uleb128 0x3a	# (DW_AT_decl_file)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.uleb128 0x3b	# (DW_AT_decl_line)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.uleb128 0x27	# (DW_AT_prototyped)
> +	.uleb128 0x19	# (DW_FORM_flag_present)
> +	.uleb128 0x49	# (DW_AT_type)
> +	.uleb128 0x13	# (DW_FORM_ref4)
> +	.uleb128 0x20	# (DW_AT_inline)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.uleb128 0x1	# (DW_AT_sibling)
> +	.uleb128 0x13	# (DW_FORM_ref4)
> +	.byte	0
> +	.byte	0
> +	.uleb128 0x3	# (abbrev code)
> +	.uleb128 0x5	# (TAG: DW_TAG_formal_parameter)
> +	.byte	0	# DW_children_no
> +	.uleb128 0x3	# (DW_AT_name)
> +	.uleb128 0x8	# (DW_FORM_string)
> +	.uleb128 0x3a	# (DW_AT_decl_file)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.uleb128 0x3b	# (DW_AT_decl_line)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.uleb128 0x49	# (DW_AT_type)
> +	.uleb128 0x13	# (DW_FORM_ref4)
> +	.byte	0
> +	.byte	0
> +	.uleb128 0x4	# (abbrev code)
> +	.uleb128 0x24	# (TAG: DW_TAG_base_type)
> +	.byte	0	# DW_children_no
> +	.uleb128 0xb	# (DW_AT_byte_size)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.uleb128 0x3e	# (DW_AT_encoding)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.uleb128 0x3	# (DW_AT_name)
> +	.uleb128 0x8	# (DW_FORM_string)
> +	.byte	0
> +	.byte	0
> +	.uleb128 0x5	# (abbrev code)
> +	.uleb128 0x2e	# (TAG: DW_TAG_subprogram)
> +	.byte	0x1	# DW_children_yes
> +	.uleb128 0x3f	# (DW_AT_external)
> +	.uleb128 0x19	# (DW_FORM_flag_present)
> +	.uleb128 0x3	# (DW_AT_name)
> +	.uleb128 0xe	# (DW_FORM_strp)
> +	.uleb128 0x3a	# (DW_AT_decl_file)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.uleb128 0x3b	# (DW_AT_decl_line)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.uleb128 0x49	# (DW_AT_type)
> +	.uleb128 0x13	# (DW_FORM_ref4)
> +	.uleb128 0x11	# (DW_AT_low_pc)
> +	.uleb128 0x1	# (DW_FORM_addr)
> +	.uleb128 0x12	# (DW_AT_high_pc)
> +	.uleb128 0x7	# (DW_FORM_data8)
> +	.uleb128 0x40	# (DW_AT_frame_base)
> +	.uleb128 0x18	# (DW_FORM_exprloc)
> +	.uleb128 0x2117	# (DW_AT_GNU_all_call_sites)
> +	.uleb128 0x19	# (DW_FORM_flag_present)
> +	.uleb128 0x1	# (DW_AT_sibling)
> +	.uleb128 0x13	# (DW_FORM_ref4)
> +	.byte	0
> +	.byte	0
> +	.uleb128 0x6	# (abbrev code)
> +	.uleb128 0x1d	# (TAG: DW_TAG_inlined_subroutine)
> +	.byte	0x1	# DW_children_yes
> +	.uleb128 0x31	# (DW_AT_abstract_origin)
> +	.uleb128 0x13	# (DW_FORM_ref4)
> +	.uleb128 0x52	# (DW_AT_entry_pc)
> +	.uleb128 0x1	# (DW_FORM_addr)
> +	.uleb128 0x55	# (DW_AT_ranges)
> +	.uleb128 0x17	# (DW_FORM_sec_offset)
> +	.uleb128 0x58	# (DW_AT_call_file)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.uleb128 0x59	# (DW_AT_call_line)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.byte	0
> +	.byte	0
> +	.uleb128 0x7	# (abbrev code)
> +	.uleb128 0x5	# (TAG: DW_TAG_formal_parameter)
> +	.byte	0	# DW_children_no
> +	.uleb128 0x31	# (DW_AT_abstract_origin)
> +	.uleb128 0x13	# (DW_FORM_ref4)
> +	.uleb128 0x2	# (DW_AT_location)
> +	.uleb128 0x17	# (DW_FORM_sec_offset)
> +	.byte	0
> +	.byte	0
> +	.uleb128 0x8	# (abbrev code)
> +	.uleb128 0x34	# (TAG: DW_TAG_variable)
> +	.byte	0	# DW_children_no
> +	.uleb128 0x3	# (DW_AT_name)
> +	.uleb128 0x8	# (DW_FORM_string)
> +	.uleb128 0x3a	# (DW_AT_decl_file)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.uleb128 0x3b	# (DW_AT_decl_line)
> +	.uleb128 0xb	# (DW_FORM_data1)
> +	.uleb128 0x49	# (DW_AT_type)
> +	.uleb128 0x13	# (DW_FORM_ref4)
> +	.uleb128 0x3f	# (DW_AT_external)
> +	.uleb128 0x19	# (DW_FORM_flag_present)
> +	.uleb128 0x2	# (DW_AT_location)
> +	.uleb128 0x18	# (DW_FORM_exprloc)
> +	.byte	0
> +	.byte	0
> +	.uleb128 0x9	# (abbrev code)
> +	.uleb128 0x35	# (TAG: DW_TAG_volatile_type)
> +	.byte	0	# DW_children_no
> +	.uleb128 0x49	# (DW_AT_type)
> +	.uleb128 0x13	# (DW_FORM_ref4)
> +	.byte	0
> +	.byte	0
> +	.byte	0
> +	.section	.debug_loc,"",@progbits
> +.Ldebug_loc0:
> +.LLST0:
> +	.quad	.LVL0	# Location list begin address (*.LLST0)
> +	.quad	.LVL1	# Location list end address (*.LLST0)
> +	.value	0x1	# Location expression size
> +	.byte	0x50	# DW_OP_reg0
> +	.quad	.LVL1	# Location list begin address (*.LLST0)
> +	.quad	.LVL2	# Location list end address (*.LLST0)
> +	.value	0x3	# Location expression size
> +	.byte	0x71	# DW_OP_breg1
> +	.sleb128 -10
> +	.byte	0x9f	# DW_OP_stack_value
> +	.quad	.LVL3	# Location list begin address (*.LLST0)
> +	.quad	.LVL4	# Location list end address (*.LLST0)
> +	.value	0x1	# Location expression size
> +	.byte	0x50	# DW_OP_reg0
> +	.quad	0	# Location list terminator begin (*.LLST0)
> +	.quad	0	# Location list terminator end (*.LLST0)
> +	.section	.debug_aranges,"",@progbits
> +	.long	0x2c	# Length of Address Ranges Info
> +	.value	0x2	# DWARF Version
> +	.long	.Ldebug_info0	# Offset of Compilation Unit Info
> +	.byte	0x8	# Size of Address
> +	.byte	0	# Size of Segment Descriptor
> +	.value	0	# Pad to 16 byte boundary
> +	.value	0
> +	.quad	.LFB1	# Address
> +	.quad	.LFE1-.LFB1	# Length
> +	.quad	0
> +	.quad	0
> +	.section	.debug_ranges,"",@progbits
> +.Ldebug_ranges0:
> +	.quad	.LBB4	# Offset 0
> +	.quad	.LBE4
> +	.quad	.LBB7
> +	.quad	.LBE7
> +	.quad	0
> +	.quad	0
> +	.quad	.LFB1	# Offset 0x30
> +	.quad	.LFE1
> +	.quad	0
> +	.quad	0
> +	.section	.debug_line,"",@progbits
> +.Ldebug_line0:
> +	.section	.debug_str,"MS",@progbits,1
> +.LASF2:
> +	.string	"/main/disassemble3/gdb/testsuite/gdb.base"
> +.LASF1:
> +	.string	"disasm-optim.c"
> +.LASF3:
> +	.string	"main"
> +.LASF0:
> +	.string	"GNU C 4.9.2 20150212 (Red Hat 4.9.2-6) -mtune=generic -march=x86-64 -g -O2"
> +	.ident	"GCC: (GNU) 4.9.2 20150212 (Red Hat 4.9.2-6)"
> +	.section	.note.GNU-stack,"",@progbits
> diff --git a/gdb/testsuite/gdb.base/disasm-optim.c b/gdb/testsuite/gdb.base/disasm-optim.c
> new file mode 100644
> index 0000000..710c8a7
> --- /dev/null
> +++ b/gdb/testsuite/gdb.base/disasm-optim.c
> @@ -0,0 +1,27 @@
> +/* This testcase is part of GDB, the GNU debugger.
> +
> +   Copyright (C) 2015 Free Software Foundation, Inc.
> +
> +   This program is free software; you can redistribute it and/or modify
> +   it under the terms of the GNU General Public License as published by
> +   the Free Software Foundation; either version 3 of the License, or
> +   (at your option) any later version.
> +
> +   This program is distributed in the hope that it will be useful,
> +   but WITHOUT ANY WARRANTY; without even the implied warranty of
> +   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> +   GNU General Public License for more details.
> +
> +   You should have received a copy of the GNU General Public License
> +   along with this program.  If not, see <http://www.gnu.org/licenses/>.  */
> +
> +#include "disasm-optim.h"
> +
> +volatile int x, y;
> +
> +int
> +main ()
> +{
> +  x = foo (y);
> +  return 0;
> +}
> diff --git a/gdb/testsuite/gdb.base/disasm-optim.exp b/gdb/testsuite/gdb.base/disasm-optim.exp
> new file mode 100644
> index 0000000..ef6b7be
> --- /dev/null
> +++ b/gdb/testsuite/gdb.base/disasm-optim.exp
> @@ -0,0 +1,40 @@
> +# Copyright (C) 2015 Free Software Foundation, Inc.
> +
> +# This program is free software; you can redistribute it and/or modify
> +# it under the terms of the GNU General Public License as published by
> +# the Free Software Foundation; either version 3 of the License, or
> +# (at your option) any later version.
> +#
> +# This program is distributed in the hope that it will be useful,
> +# but WITHOUT ANY WARRANTY; without even the implied warranty of
> +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> +# GNU General Public License for more details.
> +#
> +# You should have received a copy of the GNU General Public License
> +# along with this program.  If not, see <http://www.gnu.org/licenses/>.
> +
> +# This test exercises disassemble /s with optimized and inlined code.
> +
> +if { ![istarget "x86_64-*-linux*"] } {
> +    continue
> +}
> +
> +standard_testfile .S
> +
> +if { [prepare_for_testing ${testfile}.exp $testfile ${testfile}.S {nodebug}] } {
> +    return -1
> +}
> +
> +if ![runto_main] {
> +    return -1
> +}
> +
> +gdb_test_sequence "disassemble /s main" \
> +    "Disassemble main with source" {
> +	"Dump of assembler code for function main:"
> +	"disasm-optim\\.c:\r\n24"
> +	"disasm-optim\\.h:\r\n21"
> +	"disasm-optim\\.c:\r\n25"
> +	"disasm-optim\\.h:\r\n22"
> +	"End of assembler dump\\."
> +    }
> diff --git a/gdb/testsuite/gdb.base/disasm-optim.h b/gdb/testsuite/gdb.base/disasm-optim.h
> new file mode 100644
> index 0000000..8229d13
> --- /dev/null
> +++ b/gdb/testsuite/gdb.base/disasm-optim.h
> @@ -0,0 +1,26 @@
> +/* This testcase is part of GDB, the GNU debugger.
> +
> +   Copyright (C) 2015 Free Software Foundation, Inc.
> +
> +   This program is free software; you can redistribute it and/or modify
> +   it under the terms of the GNU General Public License as published by
> +   the Free Software Foundation; either version 3 of the License, or
> +   (at your option) any later version.
> +
> +   This program is distributed in the hope that it will be useful,
> +   but WITHOUT ANY WARRANTY; without even the implied warranty of
> +   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> +   GNU General Public License for more details.
> +
> +   You should have received a copy of the GNU General Public License
> +   along with this program.  If not, see <http://www.gnu.org/licenses/>.  */
> +
> +extern inline int
> +foo (int a)
> +{
> +  if (a < 0)
> +    return a * 2;
> +  if (a == 0)
> +    return 1;
> +  return a + 10;
> +}
> diff --git a/gdb/testsuite/gdb.mi/mi-disassemble.exp b/gdb/testsuite/gdb.mi/mi-disassemble.exp
> index 4f5a352..f97c471 100644
> --- a/gdb/testsuite/gdb.mi/mi-disassemble.exp
> +++ b/gdb/testsuite/gdb.mi/mi-disassemble.exp
> @@ -223,7 +223,7 @@ proc test_disassembly_bogus_args {} {
>               "data-disassemble mix different args"
>  
>      mi_gdb_test "789-data-disassemble -f basics.c -l $line_main_body -- 9" \
> -             "789\\^error,msg=\"-data-disassemble: Mode argument must be 0, 1, 2, or 3.\"" \
> +             "789\\^error,msg=\"-data-disassemble: Mode argument must be in the range 0-5.\"" \
>               "data-disassemble wrong mode arg"
>  
>  }

> From: Doug Evans <xdje42@gmail.com>
> Cc: gdb-patches@sourceware.org
> Date: Wed, 12 Aug 2015 20:39:03 -0700
> 
> doc ping.

Doc OK.

Thanks.

On Thu, Aug 13, 2015 at 7:13 AM, Eli Zaretskii <eliz@gnu.org> wrote:
>> From: Doug Evans <xdje42@gmail.com>
>> Cc: gdb-patches@sourceware.org
>> Date: Wed, 12 Aug 2015 20:39:03 -0700
>>
>> doc ping.
>
> Doc OK.
>
> Thanks.

Committed.