binutils-gdb.git
10 years agoGDB testsuite: Fix warnings with -std=gnu11
Andreas Arnez [Tue, 28 Oct 2014 12:39:22 +0000 (12:39 +0000)]
GDB testsuite: Fix warnings with -std=gnu11

Since upstream GCC has changed the default C language dialect to
'gnu11', it yields multiple warnings in the GDB testsuite for missing
function return types and implicit function declarations.  This patch
attempts to fix these.

gdb/testsuite/ChangeLog:

* gdb.ada/cond_lang/foo.c (callme): Add return type.
* gdb.base/call-sc.c (zed): Likewise.
* gdb.base/checkpoint.c (main): Likewise.
* gdb.base/dump.c (main): Likewise.
* gdb.base/gcore.c (main): Likewise.
* gdb.base/huge.c (main): Likewise.
* gdb.base/multi-forks.c (main): Likewise.
* gdb.base/pr10179-a.c (main): Likewise.
* gdb.base/savedregs.c (main): Likewise.
* gdb.base/sigaltstack.c (main): Likewise.
* gdb.base/siginfo.c (main): Likewise.
* gdb.base/structs.c (zed): Likewise.
* gdb.mi/mi-stack.c (callee3, callee2, callee1, main): Likewise.
* gdb.mi/mi-syn-frame.c (main): Likewise.
* gdb.mi/until.c (foo, main): Likewise.
* gdb.base/global-var-nested-by-dso.c (b_main, c_main): Declare.
* gdb.base/solib-weak.c (foo): Declare.
* gdb.base/attach-twice.c: Include stdio.h.
* gdb.base/weaklib1.c: Likewise.
* gdb.base/weaklib2.c: Likewise.
* gdb.base/catch-signal-fork.c: Include stdio.h and sys/wait.h.
* gdb.mi/mi-condbreak-call-thr-state-mt.c: Include stdio.h and
unistd.h.
* gdb.base/attach-pie-misread.c: Include stdlib.h.
* gdb.mi/mi-exit-code.c: Likewise.
* gdb.base/break-interp-lib.c: Include string.h.
* gdb.base/coremaker.c: Likewise.
* gdb.base/testenv.c: Likewise.
* gdb.python/py-finish-breakpoint.c: Likewise.
* gdb.base/inferior-died.c: Include sys/wait.h.
* gdb.base/fileio.c: Include time.h.
* gdb.base/async-shell.c: Include unistd.h.
* gdb.base/dprintf-non-stop.c: Likewise.
* gdb.base/info-os.c: Likewise.
* gdb.mi/mi-console.c: Likewise.
* gdb.mi/watch-nonstop.c: Likewise.
* gdb.python/py-events.c: Likewise.
* gdb.base/async.c (baz): Move up before its invocation.
* gdb.base/code_elim2.c (my_global_func): Likewise.
* gdb.base/skip-solib-lib.c (multiply): Likewise.
* gdb.base/advance.c (func2): Likewise.

10 years agoGDB testsuite: drop non-prototype C function header variants
Andreas Arnez [Fri, 24 Oct 2014 12:07:53 +0000 (12:07 +0000)]
GDB testsuite: drop non-prototype C function header variants

Remove many old-style function header variants in C source files of
the GDB test suite, using the 'unifdef' tool with '-DPROTOTYPES=1'.

gdb/testsuite/ChangeLog:

* gdb.base/annota1.c: Remove #ifdef PROTOTYPES, keep prototyped
variant.
* gdb.base/annota3.c: Likewise.
* gdb.base/async.c: Likewise.
* gdb.base/average.c: Likewise.
* gdb.base/call-ar-st.c: Likewise.
* gdb.base/call-rt-st.c: Likewise.
* gdb.base/call-sc.c: Likewise.
* gdb.base/call-strs.c: Likewise.
* gdb.base/ending-run.c: Likewise.
* gdb.base/execd-prog.c: Likewise.
* gdb.base/exprs.c: Likewise.
* gdb.base/foll-exec.c: Likewise.
* gdb.base/foll-fork.c: Likewise.
* gdb.base/foll-vfork.c: Likewise.
* gdb.base/funcargs.c: Likewise.
* gdb.base/gcore.c: Likewise.
* gdb.base/jump.c: Likewise.
* gdb.base/langs0.c: Likewise.
* gdb.base/langs1.c: Likewise.
* gdb.base/langs2.c: Likewise.
* gdb.base/mips_pro.c: Likewise.
* gdb.base/nodebug.c: Likewise.
* gdb.base/opaque0.c: Likewise.
* gdb.base/opaque1.c: Likewise.
* gdb.base/recurse.c: Likewise.
* gdb.base/run.c: Likewise.
* gdb.base/scope0.c: Likewise.
* gdb.base/scope1.c: Likewise.
* gdb.base/setshow.c: Likewise.
* gdb.base/setvar.c: Likewise.
* gdb.base/shmain.c: Likewise.
* gdb.base/shr1.c: Likewise.
* gdb.base/shr2.c: Likewise.
* gdb.base/sigall.c: Likewise.
* gdb.base/signals.c: Likewise.
* gdb.base/so-indr-cl.c: Likewise.
* gdb.base/solib2.c: Likewise.
* gdb.base/structs.c: Likewise.
* gdb.base/sum.c: Likewise.
* gdb.base/vforked-prog.c: Likewise.
* gdb.base/watchpoint.c: Likewise.
* gdb.reverse/shr2.c: Likewise.
* gdb.reverse/until-reverse.c: Likewise.
* gdb.reverse/ur1.c: Likewise.
* gdb.reverse/watch-reverse.c: Likewise.

10 years agoDrop non-prototype C function header variants: 'sepdebug' test case
Andreas Arnez [Thu, 23 Oct 2014 11:02:31 +0000 (11:02 +0000)]
Drop non-prototype C function header variants: 'sepdebug' test case

Remove old-style function header variants from sepdebug.c.  Eliminate
references to the removed locations "breakpoint 9" and "breakpoint 13"
from sepdebug.exp.

gdb/testsuite/ChangeLog:

* gdb.base/sepdebug.c: Remove #ifdef PROTOTYPES, keep prototyped
variant.
* gdb.base/sepdebug.exp: Drop references to removed code.

10 years agoDrop non-prototype C function header variants: 'list' test case
Andreas Arnez [Wed, 22 Oct 2014 17:13:34 +0000 (17:13 +0000)]
Drop non-prototype C function header variants: 'list' test case

Remove old-style function header variants from list0.h and list1.c.
Fill the removed lines with comments or empty lines, such that the
line numbering is undisturbed.  Changes to the line numbering would
require heavy adjustments to list.exp, where many line numbers are
hard-coded, as well as a fair amount of knowledge about the source
code in and around certain lines.  Thus the dependency on the line
numbering can not be eliminated so easily, and it may not even be a
useful goal for a "list" test case.  Another option might be to adjust
the literal line numbers in list.exp, but even that is not as
straightforward as it may seem, since the test case expects certain
source lines to be exactly n lines apart.

gdb/testsuite/ChangeLog:

* gdb.base/list0.h: Remove #ifdef PROTOTYPES, keep prototyped
variant.  Preserve original line numbering.
* gdb.base/list1.c: Likewise.

10 years agoDrop non-prototype C function header variants: 'break' test case
Andreas Arnez [Wed, 22 Oct 2014 09:24:25 +0000 (09:24 +0000)]
Drop non-prototype C function header variants: 'break' test case

Remove old-style function headers from break.c and break1.c.  Adjust
break.exp accordingly; in particular eliminate references to the
removed locations "breakpoint 9, 13, and 16" from break.exp.

gdb/testsuite/ChangeLog:

* gdb.base/break.c: Remove #ifdef PROTOTYPES, keep prototyped
variant.
* gdb.base/break1.c: Likewise.
* gdb.base/break.exp: Drop references to removed code.

10 years agoDrop non-prototype C function header variants: solib1.c
Andreas Arnez [Thu, 23 Oct 2014 14:27:57 +0000 (14:27 +0000)]
Drop non-prototype C function header variants: solib1.c

Clean up solib1.c by removing the #ifdef PROTOTYPES conditional.

gdb/testsuite/ChangeLog:

* gdb.base/solib1.c: Remove #ifdef PROTOTYPES, keep prototyped
variant.

10 years agocallfuncs.exp: Indent perform_all_tests()
Andreas Arnez [Tue, 4 Nov 2014 13:50:42 +0000 (13:50 +0000)]
callfuncs.exp: Indent perform_all_tests()

The previous patch did not indent perform_all_tests() correctly after
moving the main logic into it, to avoid obscuring the functional
changes.  This patch fixes the indentation.

gdb/testsuite/ChangeLog:

* gdb.base/callfuncs.exp (perform_all_tests): Re-indent.

10 years agoPerform all tests in callfuncs.exp with and without C function prototypes
Andreas Arnez [Tue, 4 Nov 2014 12:55:30 +0000 (12:55 +0000)]
Perform all tests in callfuncs.exp with and without C function prototypes

In callfuncs.exp, compile callfuncs.c with and without C function
header prototypes and execute all tests after each compilation.

gdb/testsuite/ChangeLog:

* gdb.base/callfuncs.exp: Remove 'prototypes' variable.  Move main
logic into perform_all_tests() and invoke it with and without
function header prototypes.
(do_function_calls): Remove conditional XFAIL for PR 5318.
(rerun_and_prepare): Remove duplicate code.
(perform_all_tests): New.  Main logic moved here.

10 years ago'callfuncs' test case: Fixes in conditionally compiled code
Andreas Arnez [Wed, 22 Oct 2014 17:46:29 +0000 (17:46 +0000)]
'callfuncs' test case: Fixes in conditionally compiled code

The C source file for the 'callfuncs' test case did not compile with
-DNO_PROTOTYPES or -DPROTOTYPES.  This patch fixes various syntax
errors under #ifdef NO_PROTOTYPES and a small typo under #ifdef
PROTOTYPES.

gdb/testsuite/ChangeLog:

* gdb.base/callfuncs.c (t_float_many_args): Fix syntax error in
code guarded by #ifdef NO_PROTOTYPES.
(t_double_many_args): Likewise.
(DEF_FUNC_MANY_ARGS_1): Likewise.
(DEF_FUNC_VALUES_1): Likewise.
(t_structs_ldc): Renamed from t_structs_fc in conditional code
guarded by #ifdef PROTOTYPES.

10 years agoEliminate literal line numbers in mi-console.exp
Andreas Arnez [Tue, 28 Oct 2014 15:19:08 +0000 (15:19 +0000)]
Eliminate literal line numbers in mi-console.exp

Remove the literal line number from a regexp in mi-console.exp.  Add
an appropriate eye-catcher to mi-console.c and refer to that instead.

gdb/testsuite/ChangeLog:

* gdb.mi/mi-console.c: Add eye-catcher.
* gdb.mi/mi-console.exp (semihosted_string): Refer to eye-catcher
instead of literal line number.

10 years agoEliminate literal line numbers in shlib-call.exp
Andreas Arnez [Fri, 24 Oct 2014 12:05:10 +0000 (12:05 +0000)]
Eliminate literal line numbers in shlib-call.exp

Remove the literal line number from a regexp in shlib-call.exp.  Add
an appropriate eye-catcher to shr2.c and refer to that instead.

gdb/testsuite/ChangeLog:

* gdb.base/shr2.c: Add eye-catcher.
* gdb.base/shlib-call.exp: Refer to eye-catcher instead of literal
line number.

10 years agoEliminate literal line numbers in jump.exp
Andreas Arnez [Fri, 24 Oct 2014 11:57:22 +0000 (11:57 +0000)]
Eliminate literal line numbers in jump.exp

Remove literal line numbers from the regexps in jump.exp.  Add
appropriate eye-catchers to jump.c and refer to those instead.

gdb/testsuite/ChangeLog:

* gdb.base/jump.c: Add eye-catchers.
* gdb.base/jump.exp: Refer to eye-catchers instead of literal line
numbers.

10 years agoEliminate literal line numbers in foll-exec.exp
Andreas Arnez [Fri, 24 Oct 2014 11:04:56 +0000 (11:04 +0000)]
Eliminate literal line numbers in foll-exec.exp

Remove literal line numbers from the regexps in foll-exec.exp.  Add
appropriate eye-catchers to foll-exec.c and execd-proc.c and refer to
those instead.

gdb/testsuite/ChangeLog:

* gdb.base/execd-prog.c: Add eye-catchers.
* gdb.base/foll-exec.c: Likewise.
* gdb.base/foll-exec.exp: Refer to eye-catchers instead of literal
line numbers.

10 years agoEliminate literal line numbers in ending-run.exp
Andreas Arnez [Fri, 24 Oct 2014 09:24:01 +0000 (09:24 +0000)]
Eliminate literal line numbers in ending-run.exp

Remove literal line numbers from the regexps in ending-run.exp.  Add
appropriate eye-catchers to ending-run.c and refer to those instead.

gdb/testsuite/ChangeLog:

* gdb.base/ending-run.c: Add eye-catchers.
* gdb.base/ending-run.exp: Refer to eye-catchers instead of
literal line numbers.

10 years agoEliminate literal line numbers in call-rt-st.exp
Andreas Arnez [Fri, 24 Oct 2014 09:15:10 +0000 (09:15 +0000)]
Eliminate literal line numbers in call-rt-st.exp

Remove literal line numbers from the regexps in call-rt-st.exp.  Add
appropriate eye-catchers to call-rt-st.c and refer to those instead.

gdb/testsuite/ChangeLog:

* gdb.base/call-rt-st.c: Add eye-catchers.
* gdb.base/call-rt-st.exp: Refer to eye-catchers instead of
literal line numbers.

10 years agoEliminate literal line numbers in call-ar-st.exp
Andreas Arnez [Thu, 23 Oct 2014 16:00:21 +0000 (16:00 +0000)]
Eliminate literal line numbers in call-ar-st.exp

Remove literal line numbers from the regexps in call-ar-st.exp.  Add
appropriate eye-catchers to call-ar-st.c and refer to those instead.

gdb/testsuite/ChangeLog:

* gdb.base/call-ar-st.c: Add eye-catchers.
* gdb.base/call-ar-st.exp: Refer to eye-catchers instead of
literal line numbers.

10 years agoEliminate literal line numbers in dbx.exp
Andreas Arnez [Wed, 29 Oct 2014 18:16:20 +0000 (18:16 +0000)]
Eliminate literal line numbers in dbx.exp

Remove literal line numbers from the commands and regexps in dbx.exp.
Add appropriate eye-catchers to average.c and sum.c and refer to those
instead.

gdb/testsuite/ChangeLog:

* gdb.base/average.c: Add eye-catchers.
* gdb.base/sum.c: Likewise.
* gdb.base/dbx.exp: Use eye-catchers to determine line numbers for
regexps dynamically.

10 years agoEliminate literal line numbers in so-impl-ld.exp
Andreas Arnez [Thu, 23 Oct 2014 13:56:52 +0000 (13:56 +0000)]
Eliminate literal line numbers in so-impl-ld.exp

Remove literal line numbers from the regexps in so-impl-ld.exp.  Add
appropriate eye-catchers to solib1.c and refer to those instead.

gdb/testsuite/ChangeLog:

* gdb.base/solib1.c: Add eye-catchers.
* gdb.base/so-impl-ld.exp: Match against eye-catchers instead of
literal line numbers.

10 years agoFix more memory faults uncovered by fuzzing various executables.
Nick Clifton [Wed, 12 Nov 2014 22:39:58 +0000 (22:39 +0000)]
Fix more memory faults uncovered by fuzzing various executables.

PR binutils/17512
* dwarf.c (read_and_display_attr_value): Check that we do not read
past end.
(display_debug_pubnames_worker): Add range checks.
(process_debug_info): Check for invalid pointer sizes.
(display_loc_list): Likewise.
(display_loc_list_dwo): Likewise.
(display_debug_ranges): Likewise.
(display_debug_aranges): Check for invalid address size.
(read_cie): Add range checks.  Replace call strchr with while loop.
* objdump.c (dump_dwarf): Replace abort with a warning message.
(print_section_stabs): Improve range checks.
* rdcoff.c (coff_get_slot): Use long for indx parameter type.
Add check for an excesively large index.
* rddbg.c (read_section_stabs_debugging_info): Zero terminate the
string table.  Avoid walking off the end of the stabs data.
* stabs.c (parse_stab_string): Add check for a NULL name.

PR binutils/17512
* coffcode.h (coff_slurp_line_table): Set the line number of
corrupt entries to -1.
(coff_slurp_symbol_table): Alway initialise the value of the
symbol.
* coffgen.c (coff_print_symbol): Check that the combined pointer
is valid.
(coff_print_symbol): Do not print negative line numbers.
* peXXigen.c (pe_print_idata): Add range checking displaying
member names.

10 years agoGDBserver: clean up 'cont_thread' handling
Pedro Alves [Wed, 12 Nov 2014 11:17:40 +0000 (11:17 +0000)]
GDBserver: clean up 'cont_thread' handling

As no place in the backends check cont_thread anymore, we can stop
setting and clearing it in places that resume the target and wait for
events.  Instead simply clear it whenever a new GDB connects.

gdb/gdbserver/
2014-11-12  Pedro Alves  <palves@redhat.com>

* server.c (cont_thread): Update comment.
(start_inferior, attach_inferior): No longer clear cont_thread.
(handle_v_cont): No longer set cont_thread.
(captured_main): Clear cont_thread each time a GDB connects.

10 years agoGDBserver: don't resume all threads if the Hc thread disapears
Pedro Alves [Wed, 12 Nov 2014 11:17:39 +0000 (11:17 +0000)]
GDBserver: don't resume all threads if the Hc thread disapears

There's code in linux_wait_1 that resumes all threads if the Hc thread
disappears.  It's the wrong thing to do, as GDB has told GDBserver to
resume only one thread, because e.g., the user has scheduler-locking
enabled, or because GDB was stepping the program over a breakpoint.
Resuming all threads behind GDB's back can't be good in either case.

The right thing to do is to detect that that the (only) resumed thread
is gone, and let GDB know about it.  The Linux backend is already
doing that nowadays, since:

 commit fa96cb382c12b099675c5cc238aaa7352a3fd3d7
 Author:     Pedro Alves <palves@redhat.com>
 AuthorDate: Thu Feb 27 14:30:08 2014 +0000

     Teach GDBserver's Linux backend about no unwaited-for children (TARGET_WAITKIND_NO_RESUMED).

The backend detects that all resumed threads have disappeared, and
returns TARGET_WAITKIND_NO_RESUMED to the core of GDBserver, which
then reports an error to GDB.

There's no need to frob the passed in ptid to wait for the continue
thread either -- linux_wait_for_event only returns events for resumed
threads.

The badness (of resuming threads) can actually be observed in the
testsuite, if we force-disable vCont support in GDBserver -- before
the patch, gdb.threads/no-unwaited-for-left.exp hangs if we disable
vCont:

 (gdb) continue
 Continuing.
 FAIL: gdb.threads/no-unwaited-for-left.exp: continue to breakpoint: break-here (timeout)
 ... more cascading timeouts ....

After the patch, gdb.threads/no-unwaited-for-left.exp behaves the same
with or without vCont support:

 (gdb) continue
 Continuing.
 [New Thread 32226]
 [Switching to Thread 32226]

 Breakpoint 2, thread_a (arg=0x0) at /home/pedro/gdb/mygit/build/../src/gdb/testsuite/gdb.threads/no-unwaited-for-left.c:28
 28   return 0; /* break-here */
 (gdb) PASS: gdb.threads/no-unwaited-for-left.exp: continue to breakpoint: break-here
...
 continue
 Continuing.
 warning: Remote failure reply: E.No unwaited-for children left.

 [Thread 32222] #1 stopped.
 (gdb) FAIL: gdb.threads/no-unwaited-for-left.exp: continue stops when the main thread exits

Overall, this is also good for getting rid of a RSP detail from the backend.

gdb/gdbserver/
2014-11-12  Pedro Alves  <palves@redhat.com>

* linux-low.c (linux_wait_1): Don't force a wait for the Hc
thread, and don't resume all threads if the Hc thread has exited.

10 years agoGDBserver: ctrl-c after leader has exited
Pedro Alves [Wed, 12 Nov 2014 11:17:39 +0000 (11:17 +0000)]
GDBserver: ctrl-c after leader has exited

The target->request_interrupt callback implements the handling for
ctrl-c.  User types ctrl-c in GDB, GDB sends a \003 to the remote
target, and the remote targets stops the program with a SIGINT, just
like if the user typed ctrl-c in GDBserver's terminal.

The trouble is that using kill_lwp(signal_pid, SIGINT) sends the
SIGINT directly to the program's main thread.  If that thread has
exited already, then that kill won't do anything.

Instead, send the SIGINT to the process group, just like GDB
does (see inf-ptrace.c:inf_ptrace_stop).

gdb.threads/leader-exit.exp is extended to cover the scenario.  It
fails against GDBserver before the patch.

Tested on x86_64 Fedora 20, native and GDBserver.

gdb/gdbserver/
2014-11-12  Pedro Alves  <palves@redhat.com>

* linux-low.c (linux_request_interrupt): Always send a SIGINT to
the process group instead of to a specific LWP.

gdb/testsuite/
2014-11-12  Pedro Alves  <palves@redhat.com>

* gdb.threads/leader-exit.exp: Test sending ctrl-c works after the
leader has exited.

10 years agoGarbage collect the infwait_state global
Pedro Alves [Wed, 12 Nov 2014 11:02:11 +0000 (11:02 +0000)]
Garbage collect the infwait_state global

No longer used since the non-continuable watchpoints handling rework.

gdb/
2014-11-12  Pedro Alves  <palves@redhat.com>

* infrun.c (enum infwait_states, infwait_state): Delete.

10 years agofix skipping permanent breakpoints
Pedro Alves [Wed, 12 Nov 2014 10:10:49 +0000 (10:10 +0000)]
fix skipping permanent breakpoints

The gdb.arch/i386-bp_permanent.exp test is currently failing an
assertion recently added:

 (gdb) stepi
 ../../src/gdb/infrun.c:2237: internal-error: resume: Assertion `sig != GDB_SIGNAL_0' failed.
 A problem internal to GDB has been detected,
 further debugging may prove unreliable.
 Quit this debugging session? (y or n)
 FAIL: gdb.arch/i386-bp_permanent.exp: Single stepping past permanent breakpoint. (GDB internal error)

The assertion expects that the only reason we currently need to step a
breakpoint instruction is when we have a signal to deliver.  But when
stepping a permanent breakpoint (with or without a signal) we also
reach this code.

The assertion is correct and the permanent breakpoints skipping code
is wrong.

Consider the case of the user doing "step/stepi" when stopped at a
permanent breakpoint.  GDB's `resume' calls the
gdbarch_skip_permanent_breakpoint hook and then happily continues
stepping:

  /* Normally, by the time we reach `resume', the breakpoints are either
     removed or inserted, as appropriate.  The exception is if we're sitting
     at a permanent breakpoint; we need to step over it, but permanent
     breakpoints can't be removed.  So we have to test for it here.  */
  if (breakpoint_here_p (aspace, pc) == permanent_breakpoint_here)
    {
      gdbarch_skip_permanent_breakpoint (gdbarch, regcache);
    }

But since gdbarch_skip_permanent_breakpoint already advanced the PC
manually, this ends up executing the instruction that is _after_ the
breakpoint instruction.  The user-visible result is that a single-step
steps two instructions.

The gdb.arch/i386-bp_permanent.exp test is actually ensuring that
that's indeed how things work.  It runs to an int3 instruction, does
"stepi", and checks that "leave" was executed with that "stepi".  Like
this:

 (gdb) b *0x0804848c
 Breakpoint 2 at 0x804848c
 (gdb) c
 Continuing.

 Breakpoint 2, 0x0804848c in standard ()
 (gdb) disassemble
 Dump of assembler code for function standard:
    0x08048488 <+0>:     push   %ebp
    0x08048489 <+1>:     mov    %esp,%ebp
    0x0804848b <+3>:     push   %edi
 => 0x0804848c <+4>:     int3
    0x0804848d <+5>:     leave
    0x0804848e <+6>:     ret
    0x0804848f <+7>:     nop
 (gdb) si
 0x0804848e in standard ()
 (gdb) disassemble
 Dump of assembler code for function standard:
    0x08048488 <+0>:     push   %ebp
    0x08048489 <+1>:     mov    %esp,%ebp
    0x0804848b <+3>:     push   %edi
    0x0804848c <+4>:     int3
    0x0804848d <+5>:     leave
 => 0x0804848e <+6>:     ret
    0x0804848f <+7>:     nop
 End of assembler dump.
 (gdb)

One would instead expect that a stepi at 0x0804848c stops at
0x0804848d, _before_ the "leave" is executed.  This commit changes GDB
this way.  Care is taken to make stepping into a signal handler when
the step starts at a permanent breakpoint instruction work correctly.

The patch adjusts gdb.arch/i386-bp_permanent.exp in this direction,
and also makes it work on x86_64 (currently it only works on i*86).

The patch also adds a new gdb.base/bp-permanent.exp test that
exercises many different code paths related to stepping permanent
breakpoints, including the stepping with signals cases.  The test uses
"hack/trick" to make it work on all (or most) platforms -- it doesn't
really hard code a breakpoint instruction.

Tested on x86_64 Fedora 20, native and gdbserver.

gdb/
2014-11-12  Pedro Alves  <palves@redhat.com>

* infrun.c (resume): Clear the thread's 'stepped_breakpoint' flag.
Rewrite stepping over a permanent breakpoint.
(thread_still_needs_step_over, proceed): Don't set
stepping_over_breakpoint for permanent breakpoints.
(handle_signal_stop): Don't clear stepped_breakpoint.  Also pull
single-step breakpoints out of the target on hardware step
targets.
(process_event_stop_test): If stepping a permanent breakpoint
doesn't hit the step-resume breakpoint, delete the step-resume
breakpoint.
(switch_back_to_stepped_thread): Also check if the stepped thread
has advanced already on hardware step targets.
(currently_stepping): Return true if the thread stepped a
breakpoint.

gdb/testsuite/
2014-11-12  Pedro Alves  <palves@redhat.com>

* gdb.arch/i386-bp_permanent.c: New file.
* gdb.arch/i386-bp_permanent.exp: Don't skip on x86_64.
(srcfile): Set to i386-bp_permanent.c.
(top level): Adjust to work in both 32-bit and 64-bit modes.  Test
that stepi does not execute the 'leave' instruction, instead of
testing it does execute.
* gdb.base/bp-permanent.c: New file.
* gdb.base/bp-permanent.exp: New file.

10 years agomake "permanent breakpoints" per location and disableable
Pedro Alves [Wed, 12 Nov 2014 10:10:49 +0000 (10:10 +0000)]
make "permanent breakpoints" per location and disableable

"permanent"-ness is currently a property of the breakpoint.  But, it
should actually be an implementation detail of a _location_.  Consider
this bit in infrun.c:

  /* Normally, by the time we reach `resume', the breakpoints are either
     removed or inserted, as appropriate.  The exception is if we're sitting
     at a permanent breakpoint; we need to step over it, but permanent
     breakpoints can't be removed.  So we have to test for it here.  */
  if (breakpoint_here_p (aspace, pc) == permanent_breakpoint_here)
    {
      if (gdbarch_skip_permanent_breakpoint_p (gdbarch))
gdbarch_skip_permanent_breakpoint (gdbarch, regcache);
      else
error (_("\
The program is stopped at a permanent breakpoint, but GDB does not know\n\
how to step past a permanent breakpoint on this architecture.  Try using\n\
a command like `return' or `jump' to continue execution."));
    }

This will wrongly skip a non-breakpoint instruction if we have a
multiple location breakpoint where the whole breakpoint was set to
"permanent" because one of the locations happened to be permanent,
even if the one GDB is resuming from is not.

Related, because the permanent breakpoints are only marked as such in
init_breakpoint_sal, we currently miss marking momentary breakpoints
as permanent.  A test added by a following patch trips on that.
Making permanent-ness be per-location, and marking locations as such
in add_location_to_breakpoint, the natural place to do this, fixes
this issue...

... and then exposes a latent issue with mark_breakpoints_out.  It's
clearing the inserted flag of permanent breakpoints.  This results in
assertions failing like this:

 Breakpoint 1, main () at testsuite/gdb.base/callexit.c:32
 32        return 0;
 (gdb) call callexit()
 [Inferior 1 (process 15849) exited normally]
 gdb/breakpoint.c:12854: internal-error: allegedly permanent breakpoint is not actually inserted
 A problem internal to GDB has been detected,
 further debugging may prove unreliable.

The call dummy breakpoint, which is a momentary breakpoint, is set on
top of a manually inserted breakpoint instruction, and so is now
rightfully marked as a permanent breakpoint.  See "Write a legitimate
instruction at the point where the infcall breakpoint is going to be
inserted." comment in infcall.c.

Re. make_breakpoint_permanent.  That's only called by solib-pa64.c.
Permanent breakpoints were actually originally invented for HP-UX [1].
I believe that that call (the only one in the tree) is unnecessary
nowadays, given that nowadays the core breakpoints code analyzes the
instruction under the breakpoint to automatically detect whether it's
setting a breakpoint on top of a breakpoint instruction in the
program.  I know close to nothing about HP-PA/HP-UX, though.

[1] https://sourceware.org/ml/gdb-patches/1999-q3/msg00245.html, and
    https://sourceware.org/ml/gdb-patches/1999-q3/msg00242.html

In addition to the per-location issue, "permanent breakpoints" are
currently always displayed as enabled=='n':

 (gdb) b main
 Breakpoint 3 at 0x40053c: file ../../../src/gdb/testsuite/gdb.arch/i386-permbkpt.S, line 29.
 (gdb) info breakpoints
 Num     Type           Disp Enb Address            What
 3       breakpoint     keep n   0x000000000040053c ../../../src/gdb/testsuite/gdb.arch/i386-permbkpt.S:29

But OTOH they're always enabled; there's no way to disable them...

In turn, this means that if one adds commands to such a breakpoint,
they're _always_ run:

 (gdb) start
 Starting program: /home/pedro/gdb/mygit/build/gdb/testsuite/gdb.arch/i386-permbkpt
 ...
 Temporary breakpoint 1, main () at ../../../src/gdb/testsuite/gdb.arch/i386-permbkpt.S:29
 29              int3
 (gdb) b main
 Breakpoint 2 at 0x40053c: file ../../../src/gdb/testsuite/gdb.arch/i386-permbkpt.S, line 29.
 (gdb) info breakpoints
 Num     Type           Disp Enb Address            What
 2       breakpoint     keep n   0x000000000040053c ../../../src/gdb/testsuite/gdb.arch/i386-permbkpt.S:29
 (gdb) commands
 Type commands for breakpoint(s) 2, one per line.
 End with a line saying just "end".
 >echo "hello!"
 >end
 (gdb) disable 2
 (gdb) start
 The program being debugged has been started already.
 Start it from the beginning? (y or n) y
 Temporary breakpoint 3 at 0x40053c: file ../../../src/gdb/testsuite/gdb.arch/i386-permbkpt.S, line 29.
 Starting program: /home/pedro/gdb/mygit/build/gdb/testsuite/gdb.arch/i386-permbkpt

 Breakpoint 2, main () at ../../../src/gdb/testsuite/gdb.arch/i386-permbkpt.S:29
 29              int3
 "hello!"(gdb)

IMO, one should be able to disable such a breakpoint, and GDB should
then behave just like if the user hadn't created the breakpoint in the
first place (that is, report a SIGTRAP).

By making permanent-ness a property of the location, and eliminating
the bp_permanent enum enable_state state ends up fixing that as well.

No tests are added for these changes yet; they'll be added in a follow
up patch, as skipping permanent breakpoints is currently broken and
trips on an assertion in infrun.

Tested on x86_64 Fedora 20, native and gdbserver.

gdb/ChangeLog:
2014-11-12  Pedro Alves  <palves@redhat.com>

Mark locations as permanent, not the whole breakpoint.
* breakpoint.c (remove_breakpoint_1, remove_breakpoint): Adjust.
(mark_breakpoints_out): Don't mark permanent breakpoints as
uninserted.
(breakpoint_init_inferior): Use mark_breakpoints_out.
(breakpoint_here_p): Adjust.
(bpstat_stop_status, describe_other_breakpoints): Remove handling
of permanent breakpoints.
(make_breakpoint_permanent): Mark each location as permanent,
instead of marking the breakpoint.
(add_location_to_breakpoint): If the location is permanent, mark
it as such, and as inserted.
(init_breakpoint_sal): Don't make the breakpoint permanent here.
(bp_location_compare, update_global_location_list): Adjust.
(update_breakpoint_locations): Don't make the breakpoint permanent
here.
(disable_breakpoint, enable_breakpoint_disp): Don't skip permanent
breakpoints.
* breakpoint.h (enum enable_state) <bp_permanent>: Delete field.
(struct bp_location) <permanent>: New field.
* guile/scm-breakpoint.c (bpscm_enable_state_to_string): Remove
reference to bp_permanent.

10 years agoadd a default method for gdbarch_skip_permanent_breakpoint
Pedro Alves [Wed, 12 Nov 2014 10:10:48 +0000 (10:10 +0000)]
add a default method for gdbarch_skip_permanent_breakpoint

breakpoint.c uses gdbarch_breakpoint_from_pc to determine whether a
breakpoint location points at a permanent breakpoint:

 static int
 bp_loc_is_permanent (struct bp_location *loc)
 {
 ...
   addr = loc->address;
   bpoint = gdbarch_breakpoint_from_pc (loc->gdbarch, &addr, &len);
 ...
  if (target_read_memory (loc->address, target_mem, len) == 0
      && memcmp (target_mem, bpoint, len) == 0)
    retval = 1;
 ...

So I think we should default the gdbarch_skip_permanent_breakpoint
hook to advancing the PC by the length of the breakpoint instruction,
as determined by gdbarch_breakpoint_from_pc.  I believe that simple
implementation does the right thing for most architectures.  If
there's an oddball architecture where that doesn't work, then it
should override the hook, just like it should be overriding the hook
if there was no default anyway.

The only two implementation of skip_permanent_breakpoint are
i386_skip_permanent_breakpoint, for x86, and
hppa_skip_permanent_breakpoint, for PA-RISC/HP-UX

The x86 implementation is trivial, and can clearly be replaced by the
new default.

I don't know about the HP-UX one though, I know almost nothing about
PA.  It may well be advancing the PC ends up being equivalent.
Otherwise, it must be that "jump $pc_after_bp" doesn't work either...

Tested on x86_64 Fedora 20 native and gdbserver.

gdb/
2014-11-12  Pedro Alves  <palves@redhat.com>

* arch-utils.c (default_skip_permanent_breakpoint): New function.
* arch-utils.h (default_skip_permanent_breakpoint): New
declaration.
* gdbarch.sh (skip_permanent_breakpoint): Now an 'f' function.
Install default_skip_permanent_breakpoint as default method.
* i386-tdep.c (i386_skip_permanent_breakpoint): Delete function.
(i386_gdbarch_init): Don't install it.
* infrun.c (resume): Assume there's always a
gdbarch_skip_permanent_breakpoint implementation.
* gdbarch.h, gdbarch.c: Regenerate.

10 years agoThrow away dodgy coff line number info earlier
Alan Modra [Wed, 12 Nov 2014 04:21:38 +0000 (14:51 +1030)]
Throw away dodgy coff line number info earlier

PR 17521
* coffcode.h (coff_slurp_line_table): Drop line number info
not preceded by a valid function entry.  Revert last change.

10 years agoFix z80-coff build breakage
Alan Modra [Wed, 12 Nov 2014 04:30:01 +0000 (15:00 +1030)]
Fix z80-coff build breakage

* config/tc-z80.c (parse_exp_not_indexed, parse_exp): Warning fixes.

10 years agoFix x86 non-ELF build breakage
Alan Modra [Wed, 12 Nov 2014 04:24:23 +0000 (14:54 +1030)]
Fix x86 non-ELF build breakage

PR ld/17482
* config/tc-i386.c (output_insn): Don't test x86_elf_abi when
not ELF.

10 years agodaily update
Alan Modra [Tue, 11 Nov 2014 23:00:56 +0000 (09:30 +1030)]
daily update

10 years agoFix invalid memory accesses for more corrupt binary files.
Nick Clifton [Tue, 11 Nov 2014 20:50:03 +0000 (20:50 +0000)]
Fix invalid memory accesses for more corrupt binary files.

PR binutils/17531
* binutils/readelf.c (dynamic_nent): Change type to size_t.
(slurp_rela_relocs): Use size_t type for nrelas.
(slurp_rel_relocs): Likewise.
(get_program_headers): Improve out of memory error message.
(get_32bit_section_headers): Likewise.
(get_32bit_section_headers): Likewise.
(get_64bit_section_headers): Likewise.
(get_32bit_elf_symbols): Likewise.
(get_64bit_elf_symbols): Likewise.
(process_section_groups): Likewise.
(get_32bit_dynamic_section): Likewise.
(get_64bit_dynamic_section): Likewise.
(process_dynamic_section): Likewise.
(process_version_sections): Likewise.
(get_symbol_index_type): Likewise.
(process_mips_specific): Likewise.
(process_corefile_note_segment): Likewise.
(process_version_sections): Use size_t type for total.
(get_dynamic_data): Change type of number parameter to size_t.
Improve out of memory error messages.
(process_symbol_table): Change type of nbuckets and nchains to
size_t.  Skip processing of sections headers if there are none.
Improve out of memory error messages.

10 years ago * avrtiny.sc: Apply avr.sc fixes.
Denis Chertykov [Tue, 11 Nov 2014 17:44:03 +0000 (20:44 +0300)]
* avrtiny.sc: Apply avr.sc fixes.

10 years agooops - missed the French translation.
Nick Clifton [Tue, 11 Nov 2014 16:57:28 +0000 (16:57 +0000)]
oops - missed the French translation.

10 years agoUpdated French and Ukranian translations supplied by the Translation Project.
Nick Clifton [Tue, 11 Nov 2014 16:54:22 +0000 (16:54 +0000)]
Updated French and Ukranian translations supplied by the Translation Project.

* po/uk.po: Updated Ukranian translation.

* po/fr.po: Updated French translation.

10 years ago[PATCH] treate -specs as both cflags & ldflags
Jiong Wang [Tue, 11 Nov 2014 16:41:37 +0000 (16:41 +0000)]
[PATCH] treate -specs as both cflags & ldflags

  ld/testsuite/
    * lib/ld-lib.exp (run_ld_link_exec_tests): Append board_cflags if gcc driver
    used as link tool.
    (run_cc_link_exec_tests): Likewise.

10 years agoPrevent a buffer overrun whilst displaying corrupt ARM tags.
Nick Clifton [Tue, 11 Nov 2014 16:15:47 +0000 (16:15 +0000)]
Prevent a buffer overrun whilst displaying corrupt ARM tags.

PR binutils/17531
* readelf.c (display_arm_attribute): Avoid reading off the end of
the buffer when processing a Tag_nodefaults.

10 years agoMore fixes for invalid memory accesses, uncovered by valgrind and binary fuzzers.
Nick Clifton [Tue, 11 Nov 2014 15:34:27 +0000 (15:34 +0000)]
More fixes for invalid memory accesses, uncovered by valgrind and binary fuzzers.

PR binutils/17512
* coffcode.h (coff_slurp_line_table): Initialise the parts of the
line number cache that would not be initialised by the copy from
the new line number table.
(coff_classify_symbol): Allow for _bfd_coff_internal_syment_name
returning NULL.
* coffgen.c (coff_get_normalized_symbols): Get the external
symbols before allocating space for the internal symbols, in case
the get fails.
* elf.c (_bfd_elf_slurp_version_tables): Only allocate a verref
array if one is needed.  Likewise with the verdef array.
* peXXigen.c (_bfd_XXi_swap_sym_in): Replace abort()'s with error
messages.
(_bfd_XXi_swap_aux_in): Make sure that all fields of the aux
structure are initialised.
(pe_print_edata): Avoid reading off the end of the data buffer.

10 years agoWarn users about mismatched PID namespaces
Daniel Colascione [Tue, 11 Nov 2014 14:18:23 +0000 (14:18 +0000)]
Warn users about mismatched PID namespaces

Linux supports multiple "PID namespaces".  Processes in different PID
namespaces have different views of the system process list.  Sometimes,
a single process can appear in more than one PID namespace, but with a
different PID in each.  When GDB and its target are in different PID
namespaces, various features can break due to the mismatch between
what the target believes its PID to be and what GDB believes its PID
to be.  The most visible broken functionality is thread enumeration
silently failing.

This patch explicitly warns users against trying to debug across PID
namespaces.

The patch introduced no new failures in my test suite run on an x86_64
installation of Ubuntu 14.10.  It doesn't include a test: writing an
automated test that exercises this code would be very involved because
CLONE_NEWNS requires CAP_SYS_ADMIN; the easier way to reproduce the
problem is to start a new lxc container.

gdb/
2014-11-11  Daniel Colascione  <dancol@dancol.org>

Warn about cross-PID-namespace debugging.
* nat/linux-procfs.h (linux_proc_pid_get_ns): New prototype.
* nat/linux-procfs.c (linux_proc_pid_get_ns): New function.
* linux-thread-db.c (check_pid_namespace_match): New function.
(thread_db_inferior_created): Call it.

10 years agocoff coff
Alan Modra [Tue, 11 Nov 2014 12:12:03 +0000 (22:42 +1030)]
coff coff

I missed this use of the loop induction variable outside the loop.

PR binutils/17512
* coffcode.h (coff_slurp_line_table): Use updated lineno_count
when building func_table.

10 years agoAvoid coff OOM
Alan Modra [Tue, 11 Nov 2014 11:06:37 +0000 (21:36 +1030)]
Avoid coff OOM

bfd_zalloc/bfd_zmalloc to fix uninitialized memory reads is too big a
hammer, when the size allocated depends on user input.  A typical
bfd_alloc, bfd_seek, bfd_bread sequence will give an error or warning
at the point the file read fails when some enormous item as described
by headers is not actually present in the file.  Nice operating system
allow memory overcommit.  But not if you write to the memory.  So
bfd_zalloc can cause an OOM, thrashing, or system hangs.

The patch also fixes a recently introduced endless loop on bad input.

PR binutils/17512
* coffcode.h (coff_slurp_line_table): Don't bfd_zalloc, just
memset the particular bits we need.  Update src after hitting loop
"continue".  Don't count lineno omitted due to invalid symbols in
nbr_func, and update lineno_count.  Init entire terminating
lineno.  Don't both allocating terminator in n_lineno_cache.
Redirect sym->lineno pointer to where n_lineno_cache will be
copied, and free n_lineno_cache.
* pe-mips.c (NUM_HOWTOS): Typo fix.

10 years agold -r abort in _bfd_elf_write_section_eh_frame
Alan Modra [Tue, 11 Nov 2014 09:43:03 +0000 (20:13 +1030)]
ld -r abort in _bfd_elf_write_section_eh_frame

Turning on .eh_frame processing for ld -r resulted in systemtap
tickling a ld bug.  Triggered by the zero terminator not being added
to .eh_frame in a separate file as it usually is (crtend.o), but
instead being present in the last .eh_frame section along with CIEs
and FDEs.  The 4-byte terminator makes the section size check fail
on 64-bit targets.

* elf-eh-frame (_bfd_elf_write_section_eh_frame): Adjust section
size check to account for possible zero terminator.

10 years agosymmisc.c: Remove trailing whitespace.
Doug Evans [Tue, 11 Nov 2014 01:19:57 +0000 (17:19 -0800)]
symmisc.c: Remove trailing whitespace.

gdb/ChangeLog:

* symmisc.c (print_objfile_statistics): Remove trailing whitespace.
(maintenance_info_symtabs, maintenance_check_symtabs): Ditto.

10 years agosource.c (select_source_symtab): Rewrite to use ALL_SYMTABS.
Doug Evans [Tue, 11 Nov 2014 00:49:44 +0000 (16:49 -0800)]
source.c (select_source_symtab): Rewrite to use ALL_SYMTABS.

gdb/ChangeLog:

* source.c (select_source_symtab): Rewrite to use ALL_SYMTABS.

10 years agoPR 17564: Fix objfile search order for static symbols.
Doug Evans [Mon, 10 Nov 2014 23:48:49 +0000 (15:48 -0800)]
PR 17564: Fix objfile search order for static symbols.

When searching static symbols, gdb would search over all
expanded symtabs of all objfiles, and if that fails only then
would it search all partial/gdb_index tables of all objfiles.
This means that the user could get a random instance of the
symbol depending on what symtabs have been previously expanded.
Now the search is consistent, searching each objfile completely
before proceeding to the next one.

gdb/ChangeLog:

PR symtab/17564
* symtab.c (lookup_symbol_in_all_objfiles): Delete.
(lookup_static_symbol): Move definition to new location and rewrite.
(lookup_symbol_in_objfile): New function.
(lookup_symbol_global_iterator_cb): Call it.

gdb/testsuite/ChangeLog:

PR symtab/17564
* gdb.base/symtab-search-order.exp: New file.
* gdb.base/symtab-search-order.c: New file.
* gdb.base/symtab-search-order-1.c: New file.
* gdb.base/symtab-search-order-shlib-1.c: New file.

10 years agodaily update
Alan Modra [Mon, 10 Nov 2014 23:00:33 +0000 (09:30 +1030)]
daily update

10 years agobfd/
Richard Sandiford [Mon, 10 Nov 2014 20:49:49 +0000 (20:49 +0000)]
bfd/
2014-11-10  James Cowgill  <James.Cowgill@imgtec.com>

* elfxx-mips.c (_bfd_mips_elf_section_processing): don't force small
data sections to be PROGBITS

10 years agoMore fixes for assertion failures and out-of-bounds reads by readelf.
Nick Clifton [Mon, 10 Nov 2014 16:32:32 +0000 (16:32 +0000)]
More fixes for assertion failures and out-of-bounds reads by readelf.

PR binutils/17531
* (ia64_process_unwind): Replace assertion with an error message.
Add range checking for group section indicies.
(hppa_process_unwind): Replace assertion with an error message.
(process_syminfo): Likewise.
(decode_arm_unwind_bytecode): Add range checking.
(dump_section_as_strings): Add more string range checking.
(display_tag_value): Likewise.
(display_arm_attribute): Likewise.
(display_gnu_attribute): Likewise.
(display_tic6x_attribute): Likewise.
(display_msp430x_attribute): Likewise.

10 years agoFix objcopy and strip so that they remove their temporary files even if an error...
Nick Clifton [Mon, 10 Nov 2014 14:28:43 +0000 (14:28 +0000)]
Fix objcopy and strip so that they remove their temporary files even if an error occurs.

PR binutils/17552
* (copy_archive): Clean up temporary files even if an error
occurs.

10 years agoMore fixes for problems exposed by valgrind and the address sanitizer
Nick Clifton [Mon, 10 Nov 2014 14:18:45 +0000 (14:18 +0000)]
More fixes for problems exposed by valgrind and the address sanitizer
when displaying the contents of corrupt files.

PR binutils/17521
* coff-i386.c (NUM_HOWTOS): New define.
(RTYPE2HOWTO): Use it.
(coff_i386_rtype_to_howto): Likewise.
(coff_i386_reloc_name_lookup): Likewise.
(CALC_ADDEND): Check that reloc r_type field is valid.
* coff-x86_64.c (NUM_HOWTOS): New define.
(RTYPE2HOWTO): Use it.
(coff_amd64_rtype_to_howto): Likewise.
(coff_amd64_reloc_name_lookup): Likewise.
(CALC_ADDEND): Check that reloc r_type field is valid.
* coffcode.h (coff_slurp_line_table): Check for symbol table
indexing underflow.
(coff_slurp_symbol_table): Use zalloc to ensure that all table
entries are initialised.
* coffgen.c (_bfd_coff_read_string_table): Initialise unused bits
in the string table.  Also ensure that the table is 0 terminated.
(coff_get_normalized_symtab): Check for symbol table indexing
underflow.
* opncls.c (bfd_alloc): Catch the case where a small negative size
can result in only 1 byte being allocated.
(bfd_alloc2): Use bfd_alloc.
* pe-mips.c (NUM_HOWTOS): New define.
(coff_mips_reloc_name_lookup): Use it.
(CALC_ADDEND): Check that reloc r_type field is valid.
* peXXigen.c (_bfd_XXi_swap_aouthdr_in): Initialise unused entries
in the DataDirectory.
(pe_print_idata): Avoid reading beyond the end of the data block
wen printing strings.
(pe_print_edata): Likewise.
Check for table indexing underflow.
* peicode.h (pe_mkobject): Initialise the pe_opthdr field.
(pe_bfd_object_p): Allocate and initialize enough space to hold a
PEAOUTHDR, even if the opt_hdr field specified less.

10 years agoWork around GCC bug 63748
Ulrich Weigand [Mon, 10 Nov 2014 14:11:44 +0000 (15:11 +0100)]
Work around GCC bug 63748

A recent change to eval.c triggered a GCC bug that causes a false positive
"may be used uninitialized" warning in evaluate_subexp_standard.  This seems
to be triggered by a specific CFG constructed via setjmp and gotos.

While the GCC bug is in the process of being fixed, there are released
compiler versions (in particular GCC 4.9) in the field that show this
problem.  In order to allow compiling GDB with one of those compilers,
this commit slightly reworks the CFG (in an equivalent way) of the
affected function, so that the GCC bug is no longer triggered.

gdb/ChangeLog:

* eval.c (evaluate_subexp_standard): Work around GCC bug 63748.

10 years agodaily update
Alan Modra [Sun, 9 Nov 2014 23:00:31 +0000 (09:30 +1030)]
daily update

10 years agodaily update
Alan Modra [Sat, 8 Nov 2014 23:00:33 +0000 (09:30 +1030)]
daily update

10 years agoCorrect buffer overrun test
Alan Modra [Sat, 8 Nov 2014 01:42:08 +0000 (12:12 +1030)]
Correct buffer overrun test

* peXXigen.c (pe_print_idata): Revert last patch, cast lhs instead.

10 years agoFix a typo in gas/ChangeLog
H.J. Lu [Sat, 8 Nov 2014 01:47:54 +0000 (17:47 -0800)]
Fix a typo in gas/ChangeLog

10 years agoFix race condition when using --threads with linker plugins.
Evgeniy Dushistov [Sat, 8 Nov 2014 00:12:58 +0000 (16:12 -0800)]
Fix race condition when using --threads with linker plugins.

2014-11-06  Evgeniy Dushistov  <dushistov@mail.ru>

gold/
* plugin.cc: use lock to searialize calls of Plugin_manager::claim_file
* plugin.h: add lock definition

10 years agodaily update
Alan Modra [Fri, 7 Nov 2014 23:01:06 +0000 (09:31 +1030)]
daily update

10 years agoCast time value to unsigned long to print
H.J. Lu [Fri, 7 Nov 2014 21:41:02 +0000 (13:41 -0800)]
Cast time value to unsigned long to print

* readelf.c (process_dynamic_section): Cast time value to unsigned
long to print.

10 years agoCast to unsigned long in range checks
H.J. Lu [Fri, 7 Nov 2014 21:39:15 +0000 (13:39 -0800)]
Cast to unsigned long in range checks

* peXXigen.c (pe_print_idata): Cast to unsigned long in range
checks.

10 years agoX32: Add REX prefix to encode R_X86_64_GOTTPOFF
H.J. Lu [Fri, 7 Nov 2014 20:22:53 +0000 (12:22 -0800)]
X32: Add REX prefix to encode R_X86_64_GOTTPOFF

Structions with R_X86_64_GOTTPOFF relocation must be encoded with REX
prefix even if it isn't required by destination register.  Otherwise
linker can't safely perform IE -> LE optimization.

bfd/

PR ld/17482
* elf64-x86-64.c (elf_x86_64_relocate_section): Update comments
for IE->LE transition.

gas/

PR ld/17482
* config/tc-i386.c (output_insn): Add a dummy REX_OPCODE prefix
for structions with R_X86_64_GOTTPOFF relocation for x32 if needed.

gas/testsuite/

PR ld/17482
* gas/i386/ilp32/x32-tls.d: New file.
* gas/i386/ilp32/x32-tls.s: Likewise.

ld/testsuite/

PR ld/17482
* ld-x86-64/tlsie4.dd: Updated.

10 years agogdb.base/sigstep.exp: xfail gdb/17511 on i?86 Linux
Pedro Alves [Fri, 7 Nov 2014 15:20:47 +0000 (15:20 +0000)]
gdb.base/sigstep.exp: xfail gdb/17511 on i?86 Linux

Running gdb.base/sigstep.exp with --target=i686-pc-linux-gnu on a
64-bit kernel naturally trips on PR gdb/17511 as well, given this is a
kernel bug.

I haven't really tested a real 32-bit kernel/machine, but given the
code in question in the kernel is shared between 32-bit and 64-bit,
I'm quite sure the bug triggers in those cases as well.

So, simply xfail i?86-*-linux* too.

gdb/testsuite/
2014-11-07  Pedro Alves  <palves@redhat.com>

PR gdb/17511
* gdb.base/sigstep.exp (in_handler_map) <si+advance>: xfail
i?86-*-linux*.

10 years agoRevert old nexti prologue check and eliminate in_prologue
Pedro Alves [Fri, 7 Nov 2014 13:53:01 +0000 (13:53 +0000)]
Revert old nexti prologue check and eliminate in_prologue

The in_prologue check in the nexti code is obsolete; this commit
removes that, and then removes the in_prologue function as nothing
else uses it.

Looking at the code in GDB that makes use in_prologue, all we find is
this one caller:

      if ((ecs->event_thread->control.step_over_calls == STEP_OVER_NONE)
  || ((ecs->event_thread->control.step_range_end == 1)
      && in_prologue (gdbarch, ecs->event_thread->prev_pc,
      ecs->stop_func_start)))
{
  /* I presume that step_over_calls is only 0 when we're
     supposed to be stepping at the assembly language level
     ("stepi").  Just stop.  */
  /* Also, maybe we just did a "nexti" inside a prolog, so we
     thought it was a subroutine call but it was not.  Stop as
     well.  FENN */
  /* And this works the same backward as frontward.  MVS */
  end_stepping_range (ecs);
  return;
}

This was added by:

 commit 100a02e1deec2f037a15cdf232f026dc79763bf8
 ...
     From Fernando Nasser:
     * infrun.c (handle_inferior_event): Handle "nexti" inside function
     prologues.

The mailing list thread is here:

  https://sourceware.org/ml/gdb-patches/2001-01/msg00047.html

Not much discussion there, and no test, but looking at the code around
what was patched in that revision, we see that the checks that detect
whether the program has just stepped into a subroutine didn't rely on
the unwinders at all back then.

From 'git show 100a02e1:gdb/infrun.c':

    if (stop_pc == ecs->stop_func_start         /* Quick test */
        || (in_prologue (stop_pc, ecs->stop_func_start) &&
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
            !IN_SOLIB_RETURN_TRAMPOLINE (stop_pc, ecs->stop_func_name))
        || IN_SOLIB_CALL_TRAMPOLINE (stop_pc, ecs->stop_func_name)
        || ecs->stop_func_name == 0)
      {
        /* It's a subroutine call.  */

        if ((step_over_calls == STEP_OVER_NONE)
            || ((step_range_end == 1)
                && in_prologue (prev_pc, ecs->stop_func_start)))
          {
            /* I presume that step_over_calls is only 0 when we're
               supposed to be stepping at the assembly language level
               ("stepi").  Just stop.  */
            /* Also, maybe we just did a "nexti" inside a prolog,
               so we thought it was a subroutine call but it was not.
               Stop as well.  FENN */
            stop_step = 1;
            print_stop_reason (END_STEPPING_RANGE, 0);
            stop_stepping (ecs);
            return;
          }

Stripping the IN_SOLIB_RETURN_TRAMPOLINE checks for simplicity, we had:

    if (stop_pc == ecs->stop_func_start         /* Quick test */
        || in_prologue (stop_pc, ecs->stop_func_start)
        || ecs->stop_func_name == 0)
      {
        /* It's a subroutine call.  */

That is, detecting a subroutine call was based on prologue detection
back then.  So the in_prologue check in the current tree only made
sense back then as it was undoing a bad decision the in_prologue check
that used to exist above did.

Today, the check for a subroutine call relies on frame ids instead,
which are stable throughout the function.  So we can just remove the
in_prologue check for nexti, and the whole in_prologue function along
with it.

Tested on x86_64 Fedora 20, and also by nexti-ing manually a prologue.

gdb/
2014-11-07  Pedro Alves  <palves@redhat.com>

* infrun.c (process_event_stop_test) <subroutine check>: Don't
check if we did a "nexti" inside a prologue.
* symtab.c (in_prologue): Delete function.
* symtab.h (in_prologue): Delete declaration.

10 years agoAdd more fixes for inavlid memory accesses triggered by corrupt files.
Nick Clifton [Fri, 7 Nov 2014 13:39:45 +0000 (13:39 +0000)]
Add more fixes for inavlid memory accesses triggered by corrupt files.

PR binutils/17531
* readelf.c (get_data): Avoid allocating memory when we know that
the read will fail.
(find_section_by_type): New function.
(get_unwind_section_word): Check for invalid symbol indicies.
Check for invalid reloc types.
(get_32bit_dynamic_section): Add range checks.
(get_64bit_dynamic_section): Add range checks.
(process_dynamic_section): Check for a corrupt time value.
(process_symbol_table): Add range checks.
(dump_section_as_strings): Add string length range checks.
(display_tag_value): Likewise.
(display_arm_attribute): Likewise.
(display_gnu_attribute): Likewise.
(display_tic6x_attribute): Likewise.
(display_msp430x_attribute): Likewise.
(process_mips_specific): Add range check.

10 years agotekhex architecure
Alan Modra [Fri, 7 Nov 2014 10:52:02 +0000 (21:22 +1030)]
tekhex architecure

is a don't care.

* tekhex.c (tekhex_set_arch_mach): Ignore unknown arch errors.

10 years agotekhex buffer management and symbol types
Alan Modra [Fri, 7 Nov 2014 09:59:43 +0000 (20:29 +1030)]
tekhex buffer management and symbol types

Dramatically reduces memory consumption and processing time for large
all-zero data segments.  Allows multiple symbol types attached to a
given segment to survive objcopy.

* tekhex.c (CHUNK_SPAN): Define.
(struct data_struct <chunk_init>): Use one byte per span, update
all code accessing this field.
(find_chunk): Add create param, don't create new entry unless set.
(insert_byte): Don't save zeros.
(first_phase): Set section SEC_CODE or SEC_DATA flag depending
on symbol type.  Create an alternate section if both types of
symbol are given.  Attach type '2' and '6' symbols to absolute
section.
(move_section_contents): Fix caching of chunk.  Don't create chunk
when reading, or for writing zeros.
(tekhex_set_section_contents): Don't create initial chunks.
(tekhex_write_object_contents): Use CHUNK_SPAN.

10 years agoaoutx.h tidy
Alan Modra [Fri, 7 Nov 2014 09:48:25 +0000 (20:18 +1030)]
aoutx.h tidy

Save a multiplication, and any concern that the buffer allocation
might be smaller than the amount read (as it could be if the header
size isn't a multiple of EXTERNAL_NLIST_SIZE).

* aoutx.h (aout_get_external_symbols): Tidy allocation of symbol buffer.

10 years agosymtab.h (lookup_global_symbol): Improve function comment.
Doug Evans [Fri, 7 Nov 2014 07:51:21 +0000 (23:51 -0800)]
symtab.h (lookup_global_symbol): Improve function comment.

gdb/ChangeLog:

* symtab.h (lookup_global_symbol): Improve function comment.

10 years agoRename lookup_symbol_global to lookup_global_symbol.
Doug Evans [Fri, 7 Nov 2014 07:48:18 +0000 (23:48 -0800)]
Rename lookup_symbol_global to lookup_global_symbol.

gdb/ChangeLog:

* symtab.c (lookup_global_symbol): Renamed from lookup_symbol_global.
All callers updated.
* symtab.h (lookup_global_symbol): Update decl.
(lookup_static_symbol): Move decl to better location.

10 years agosymtab.c (basic_lookup_symbol_nonlocal): Add comment.
Doug Evans [Fri, 7 Nov 2014 07:34:28 +0000 (23:34 -0800)]
symtab.c (basic_lookup_symbol_nonlocal): Add comment.

gdb/ChangeLog:

* symtab.c (basic_lookup_symbol_nonlocal): Add comment.

10 years agoRename some "aux" functions.
Doug Evans [Fri, 7 Nov 2014 07:29:49 +0000 (23:29 -0800)]
Rename some "aux" functions.

"aux" doesn't contribute anything to the name, and it makes the
reader wonder what it's supposed to mean.

gdb/ChangeLog:

* symtab.c (lookup_local_symbol): Renamed from lookup_symbol_aux_local.
All callers updated.
(lookup_symbol_in_all_objfiles): Renamed from
lookup_symbol_aux_symtabs.  All callers updated.
(lookup_symbol_via_quick_fns): Renamed from lookup_symbol_aux_quick.
All callers updated.
(lookup_symbol_in_objfile_symtabs): Renamed from
lookup_symbol_aux_objfile.  All callers updated.

10 years agolookup_symbol_in_block: Renamed from lookup_symbol_aux_block.
Doug Evans [Fri, 7 Nov 2014 07:04:15 +0000 (23:04 -0800)]
lookup_symbol_in_block: Renamed from lookup_symbol_aux_block.

gdb/ChangeLog:

* symtab.c (lookup_symbol_in_block): Renamed from
lookup_symbol_aux_block.  All callers updated.

10 years agoRename lookup_symbol_static to lookup_symbol_in_static_block,
Doug Evans [Fri, 7 Nov 2014 06:56:46 +0000 (22:56 -0800)]
Rename lookup_symbol_static to lookup_symbol_in_static_block,
and lookup_static_symbol_aux to lookup_static_symbol.

gdb/ChangeLog:

* symtab.c (lookup_static_symbol): Renamed from
lookup_static_symbol_aux.  All callers updated.
(lookup_symbol_in_static_block): Renamed from lookup_symbol_static.
All callers updated.

10 years agoNew macro ALL_BLOCK_SYMBOLS_WITH_NAME.
Doug Evans [Fri, 7 Nov 2014 06:50:12 +0000 (22:50 -0800)]
New macro ALL_BLOCK_SYMBOLS_WITH_NAME.

gdb/ChangeLog:

* block.h (ALL_BLOCK_SYMBOLS_WITH_NAME): New macro.
* block.c (block_lookup_symbol): Use it.
* cp-support.c (make_symbol_overload_list_block): Use it.
* symtab.c (iterate_over_symbols): Use it.

10 years agoMove lookup_block_symbol to block.c, rename to block_lookup_symbol.
Doug Evans [Fri, 7 Nov 2014 06:32:25 +0000 (22:32 -0800)]
Move lookup_block_symbol to block.c, rename to block_lookup_symbol.

There is another function, lookup_symbol_aux_block, and
the names lookup_block_symbol and lookup_symbol_aux_block don't
convey any real difference between them.

The difference is that lookup_block_symbol lives in the lower level
block API, and lookup_symbol_aux_block lives in the higher level symtab API.
This patch makes this distinction clear.

gdb/ChangeLog:

* symtab.c (lookup_block_symbol): Moved to ...
* block.c (block_lookup_symbol):  ... here and renamed.
All callers updated.
* block.h (block_lookup_symbol): Declare.
* symtab.h (lookup_block_symbol): Delete.

10 years agoRevert bfd_get_size checks
Alan Modra [Thu, 6 Nov 2014 23:07:39 +0000 (09:37 +1030)]
Revert bfd_get_size checks

* archive.c (_bfd_slurp_extended_name_table): Revert bfd_get_size check.
* coffcode.h (coff_set_alignment_hook): Likewise.
(coff_slurp_line_table): Likewise.
* coffgen.c (coff_get_normalized_symtab): Likewise.
(_bfd_coff_get_external_symbols): Likewise.
* elf.c (bfd_elf_get_str_section): Likewise.
* tekhex.c (first_phase): Likewise.

10 years agoUse ALL_PRIMARY_SYMTABS instead of ALL_SYMTABS in some places.
Doug Evans [Fri, 7 Nov 2014 01:27:55 +0000 (17:27 -0800)]
Use ALL_PRIMARY_SYMTABS instead of ALL_SYMTABS in some places.

Non-primary symtabs share the block vector with their primary symtabs.
In these cases there's no need to use ALL_SYMTABS.

gdb/ChangeLog:

* ada-lang.c (ada_make_symbol_completion_list): Use
ALL_PRIMARY_SYMTABS instead of ALL_SYMTABS.
* symtab.c (lookup_objfile_from_block): Ditto.

10 years agoForgotten ChangeLog entry for previous commit.
Doug Evans [Fri, 7 Nov 2014 01:24:12 +0000 (17:24 -0800)]
Forgotten ChangeLog entry for previous commit.

gdb/ChangeLog:

* gdbtypes.h (TYPE_CODE_CLASS): Delete.  All uses changed to use
TYPE_CODE_STRUCT.

10 years agoDelete TYPE_CODE_CLASS, it's just an alias of TYPE_CODE_STRUCT.
Doug Evans [Fri, 7 Nov 2014 01:19:06 +0000 (17:19 -0800)]
Delete TYPE_CODE_CLASS, it's just an alias of TYPE_CODE_STRUCT.

gdb/ChangeLog:

* gdbtypes.h (TYPE_CODE_CLASS): Delete.  All uses changed to use
TYPE_CODE_STRUCT.

10 years agosolib_global_lookup: Fetch arch from objfile, not target_gdbarch.
Doug Evans [Fri, 7 Nov 2014 01:10:51 +0000 (17:10 -0800)]
solib_global_lookup: Fetch arch from objfile, not target_gdbarch.

gdb/ChangeLog:

* objfiles.c (get_objfile_arch): Constify.
* objfiles.h (get_objfile_arch): Update prototype.
* solib.c (solib_global_lookup): Fetch arch from objfile,
not target_gdbarch.

10 years agodaily update
Alan Modra [Thu, 6 Nov 2014 23:01:04 +0000 (09:31 +1030)]
daily update

10 years agoAdd mach parameter to nios2_find_opcode_hash.
Sandra Loosemore [Thu, 6 Nov 2014 21:12:21 +0000 (13:12 -0800)]
Add mach parameter to nios2_find_opcode_hash.

2014-11-06  Sandra Loosemore  <sandra@codesourcery.com>

include/opcode/
* nios2.h (nios2_find_opcode_hash): Add mach parameter to
declaration.  Fix obsolete comment.

opcodes/
* nios2-dis.c (nios2_find_opcode_hash): Add mach parameter.
(nios2_disassemble): Adjust call to nios2_find_opcode_hash.

gas/
* config/tc-nios2.c (nios2_diagnose_overflow): Adjust call to
nios2_find_opcode_hash.

10 years agoUpdate Nios II prologue analysis to remove detection of long-obsolete
Sandra Loosemore [Thu, 6 Nov 2014 20:56:27 +0000 (12:56 -0800)]
Update Nios II prologue analysis to remove detection of long-obsolete
code sequences.

2014-11-06  Sandra Loosemore  <sandra@codesourcery.com>

gdb/
* nios2-tdep.c (wild_insn): Delete.
(profiler_insn, irqentry_insn): Delete.
(nios2_match_sequence): Delete.
(nios2_analyze_prologue): Update comments.  Remove matching
of obsolete profiler_insn and irqentry_insn sequences.

10 years agoPrevent archive memebers with illegal pathnames from being extracted from an archive.
Nick Clifton [Thu, 6 Nov 2014 14:49:10 +0000 (14:49 +0000)]
Prevent archive memebers with illegal pathnames from being extracted from an archive.

PR binutils/17552, binutils/17533
* bucomm.c (is_valid_archive_path): New function.  Returns false
for absolute pathnames and pathnames that include /../.
* bucomm.h (is_valid_archive_path): Add prototype.
* ar.c (extract_file): Use new function to check for valid
pathnames when extracting files from an archive.
* objcopy.c (copy_archive): Likewise.
* doc/binutils.texi: Update documentation to mention the
limitation on pathname of archive members.

10 years agoRevert changes in previous deltas that introduced new failures into
Nick Clifton [Thu, 6 Nov 2014 14:39:58 +0000 (14:39 +0000)]
Revert changes in previous deltas that introduced new failures into
the linker testsuite.

* aoutx.h (slurp_symbol_table): Revert previous delta.
(slurp_reloc_table): Likewise.
* compress.c (bfd_get_full_section_contents): Remove file size
test.
* coffgen.c (coff_get_normalized_symtab): Allow zero-sized symtabs
and do not complain about linker generated files.

10 years agobfd/elf-attrs.c: Fix possible infinite loop parsing attributes
Will Newton [Wed, 5 Nov 2014 15:24:30 +0000 (15:24 +0000)]
bfd/elf-attrs.c: Fix possible infinite loop parsing attributes

Handle the case of a zero length section or sub-section in
_bfd_elf_parse_attributes and in doing so prevent an infinite loop
in the parser.

bfd/ChangeLog:

2014-11-06  Will Newton  <will.newton@linaro.org>

* elf-attrs.c (_bfd_elf_parse_attributes): Handle zero
length sections and sub-sections.

10 years agodaily update
Alan Modra [Wed, 5 Nov 2014 23:00:53 +0000 (09:30 +1030)]
daily update

10 years agoMore fixes for memory problems uncovered by file fuzzers.
Nick Clifton [Wed, 5 Nov 2014 17:57:54 +0000 (17:57 +0000)]
More fixes for memory problems uncovered by file fuzzers.

PR binutils/17512
* coffcode.h (handle_COMDAT): Replace abort with BFD_ASSERT.
Replace another abort with an error message.
(coff_slurp_line_table): Add more range checking.
* peXXigen.c (pe_print_debugdata): Add range checking.

10 years agoMore fixes for reading corrupt ELF files.
Nick Clifton [Wed, 5 Nov 2014 16:19:03 +0000 (16:19 +0000)]
More fixes for reading corrupt ELF files.

PR binutils/15731
* readelf.c (printable_section_name): New function.
(printable_section_name_from_index): New function.
(dump_relocations): Use new function.
(process_program_headers, get_32bit_elf_symbols,
(get_64bit_elf_symbols, process_section_headers,
(process_section_groups, process_relocs, ia64_process_unwind,
(hppa_process_unwind, get_unwind_section_word, decode_arm_unwind,
(arm_process_unwind, process_version_sections,
(process_symbol_table, apply_relocations, get_section_contents,
(dump_section_as_strings, dump_section_as_bytes,
(display_debug_section, process_attributes, process_mips_specific,
(process_mips_specific process_gnu_liblist): Likewise.
(get_unwind_section_word): Check for a missing symbol table.
Replace aborts with error messages.
(arm_process_unwind): Check for a missing string table.
(process_attributes): Check for an attribute length that is too
small.
(process_mips_specific): Check for a corrupt GOT symbol offset.

10 years agoFix gnu11 fallout on Solaris 10+
Rainer Orth [Wed, 5 Nov 2014 14:07:38 +0000 (15:07 +0100)]
Fix gnu11 fallout on Solaris 10+

* sigsetmask.c (_POSIX_SOURCE): Remove.

10 years agoFix segfault when creating a dso with discarded .dynsym section.
Matthew Fortune [Wed, 5 Nov 2014 10:56:59 +0000 (10:56 +0000)]
Fix segfault when creating a dso with discarded .dynsym section.

bfd/

* elfxx-mips.c (_bfd_mips_elf_finish_dynamic_sections): Fix segfault
when creating a dso with discarded dynsym section.

10 years agoUpdate .MIPS.abiflags to support MIPS R6
Matthew Fortune [Tue, 4 Nov 2014 23:37:28 +0000 (23:37 +0000)]
Update .MIPS.abiflags to support MIPS R6

bfd/

* elfxx-mips.c (update_mips_abiflags_isa): Add E_MIPS_ARCH_32R6
and E_MIPS_ARCH_64R6 support.

ld/testsuite/

* ld-mips-elf/abiflags-strip10-ph.d: New file.
* ld-mips-elf/mips-eld.exp: Run the new test.

gas/

* config/tc-mips.c (mips_elf_final_processing): Add INSN_ISA32R6
and INSN_ISA64R6 support.

gas/testsuite/

* gas/mips/elf_arch_mips32r6.d: New file.
* gas/mips/elf_arch_mips64r6.d: New file.
* gas/mips/mips.exp: Run the new tests.

10 years ago2014-11-05 Jan-Benedict Glaw <jbglaw@lug-owl.de>
Jan-Benedict Glaw [Wed, 5 Nov 2014 09:08:37 +0000 (10:08 +0100)]
2014-11-05  Jan-Benedict Glaw  <jbglaw@lug-owl.de>

* readelf.c (process_mips_specific): Fix format string warning.

10 years agoMore fixes for processing corrupt files.
Nick Clifton [Wed, 5 Nov 2014 10:13:16 +0000 (10:13 +0000)]
More fixes for processing corrupt files.

PR binutils/17512
* coffcode.h (coff_set_alignment_hook): Warn if the file lies
about the number of relocations it contains.
(coff_sort_func_alent): Return 0 if the pointers are NULL.
(coff_slurp_line_table): Add more range checks.  Do not free new
tables created when sorting line numbers.
* peXXigen.c (pe_print_idata): Add range checks.
(pe_print_edata): Likewise.
(rsrc_print_resource_entries): Likewise.  Avoid printing control
characters.  Terminate priniting if corruption is detected.
(rsrc_print_resource_directory): Terminate printing if an unknown
directory type is encountered.
(pe_print_debugdata): Fix off-by-one error.
(rsrc_count_entries): Add range checking.
(rsrc_parse_entry): Likewise.

10 years agoCast result of obstack_next_free
Alan Modra [Tue, 4 Nov 2014 22:48:27 +0000 (09:18 +1030)]
Cast result of obstack_next_free

obstack_next_free is supposed to return a void*, rather than a char*
as it does currently.  Avoid warning on void* arithmetic when
obstack_next_free gets it proper return type.

* cp-valprint.c (cp_print_value_fields): Cast obstack_next_free
to char* before doing pointer arithmetic.

10 years agoUse obstack_blank_fast to shrink obstacks
Alan Modra [Tue, 4 Nov 2014 08:55:34 +0000 (19:25 +1030)]
Use obstack_blank_fast to shrink obstacks

obstack_blank isn't the correct macro to call for shrinking obstacks
since it does size checking.

* charset.c (convert_between_encodings): Shrink obstack using
obstack_blank_fast.
* minsyms.c (install_minimal_symbols): Likewise.

10 years agoUpdate description of ownership of files in include/
Doug Evans [Wed, 5 Nov 2014 00:14:14 +0000 (16:14 -0800)]
Update description of ownership of files in include/

10 years agoFix -Wshadow warning
Matthew Fortune [Tue, 4 Nov 2014 23:40:22 +0000 (23:40 +0000)]
Fix -Wshadow warning

binutils/

* readelf.c (process_mips_specific): Rename index to idx.

10 years agodaily update
Alan Modra [Tue, 4 Nov 2014 23:01:00 +0000 (09:31 +1030)]
daily update

10 years agotui: Fix newterm call for older ncurses
Simon Marchi [Tue, 4 Nov 2014 13:27:06 +0000 (08:27 -0500)]
tui: Fix newterm call for older ncurses

Older versions of ncurses' newterm can't take NULL for their ofp and ifp
parameters. Newer versions can, and they fall back on stdout/stdin if
that is the case.

This patch explicitly passes stdout/stdin to the call to newterm to
avoid segfaulting with older ncurses.

gdb/Changelog:

2014-11-04  Simon Marchi  <simon.marchi@ericsson.com>

* tui/tui.c (tui_enable): Pass stdout and stdin to newterm.

10 years agogarbage collect gdb/breakpoint.c:breakpoint_thread_match
Pedro Alves [Tue, 4 Nov 2014 18:42:28 +0000 (18:42 +0000)]
garbage collect gdb/breakpoint.c:breakpoint_thread_match

Used to be necessary for the thread-hop code, but that's gone now.
Nothing uses this anymore.

gdb/
2014-11-04  Pedro Alves  <palves@redhat.com>

* breakpoint.c (breakpoint_thread_match): Delete function.
* breakpoint.h (breakpoint_thread_match): Delete declaration.

10 years agoFix problem with linker created sections being mistakenly flagged
Nick Clifton [Tue, 4 Nov 2014 16:00:10 +0000 (16:00 +0000)]
Fix problem with linker created sections being mistakenly flagged
as corrupt by the new error checking code in bfd_get_full_section_contents.

PR binutils/17512
* compress.c (bfd_get_full_section_contents): Improve test for
linker created objects.