[v2] Rewrite mixed source/assembler disassembly
Commit Message
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.
Comments
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.
@@ -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*
@@ -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\
@@ -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);
@@ -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;
@@ -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
@@ -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. */
@@ -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':
new file mode 100644
@@ -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
new file mode 100644
@@ -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;
+}
new file mode 100644
@@ -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\\."
+ }
new file mode 100644
@@ -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;
+}
@@ -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"
}