public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Andrew Burgess <aburgess@redhat.com>
To: gdb-patches@sourceware.org
Cc: Andrew Burgess <aburgess@redhat.com>
Subject: [PATCHv3 10/13] gdb: add timeouts for inferior function calls
Date: Tue, 31 Jan 2023 17:27:15 +0000	[thread overview]
Message-ID: <d625aad4ffa546eb471cb4d79becd3852a03864d.1675185990.git.aburgess@redhat.com> (raw)
In-Reply-To: <cover.1675185990.git.aburgess@redhat.com>

In the previous commits I have been working on improving inferior
function call support.  One thing that worries me about using inferior
function calls from a conditional breakpoint is: what happens if the
inferior function call fails?

If the failure is obvious, e.g. the thread performing the call
crashes, or hits a breakpoint, then this case is already well handled,
and the error is reported to the user.

But what if the thread performing the inferior call just deadlocks?
If the user made the call from a 'print' or 'call' command, then the
user might have some expectation of when the function call should
complete, and, when this time limit is exceeded, the user
will (hopefully) interrupt GDB and regain control of the debug
session.

But, when the inferior function call is from a breakpoint condition it
is much harder to understand that GDB is deadlocked within an inferior
call.  Maybe the breakpoint hasn't been hit yet?  Or maybe the
condition was always false?  Or maybe GDB is deadlocked in an inferior
call?  The only way to know for sure is to periodically interrupt GDB,
check on all the threads, and then continue.

Additionally, the focus of the previous commit was inferior function
calls, from a conditional breakpoint, in a multi-threaded inferior.
This opens up a whole new set of potential failure conditions.  For
example, what if the function called relies on interaction with some
other thread, and the other thread crashes?  Or hits a breakpoint?
Given how inferior function calls work (in a synchronous manor), a
stop event in some other thread is going to be ignored while the
inferior function call is being executed as part of a breakpoint
condition, and this means that GDB could get stuck waiting for the
original condition thread, which will now never complete.

In this commit I propose a solution to this problem.  A timeout.  For
targets that support async-mode we can install an event-loop timer
before starting the inferior function call.  When the timer expires we
will stop the thread performing the inferior function call.  With this
mechanism in place a user can be sure that any inferior call they make
will either complete, or timeout eventually.

Adding a timer like this is obviously a change in behaviour for the
more common 'call' and 'print' uses of inferior function calls, so, in
this patch, I propose having two different timers.  One I call the
'direct-call-timeout', which is used for 'call' and 'print' commands.
This timeout is by default set to unlimited, which, not surprisingly,
means there is no timeout in place.

A second timer, which I've called 'indirect-call-timeout', is used for
inferior function calls from breakpoint conditions.  This timeout has
a default value of 30 seconds.  This is a reasonably long time to
wait, and hopefully should be enough in most cases to allow the
inferior call to complete.  An inferior call that takes more than 30
seconds, which is installed on a breakpoint condition is really going
to slow down the debug session, so hopefully this is not a common use
case.

The user is, of course, free to reduce, or increase the timeout value,
and can always use Ctrl-c to interrupt an inferior function call, but
this timeout will ensure that GDB will stop at some point.

The new commands added by this commit are:

  set direct-call-timeout SECONDS
  show direct-call-timeout
  set indirect-call-timeout SECONDS
  show indirect-call-timeout

These new timeouts do depend on async-mode, so, if async-mode is
disabled (maint set target-async off), or not supported (e.g. target
sim), then the timeout is treated as unlimited (that is, no timeout is
set).

For targets that "fake" non-async mode, e.g. Linux native, where
non-async mode is really just async mode, but then we park the target
in a sissuspend, we could easily fix things so that the timeouts still
work, however, for targets that really are not async aware, like the
simulator, fixing things so that timeouts work correctly would be a
much bigger task - that effort would be better spent just making the
target async-aware.  And so, I'm happy for now that this feature will
only work on async targets.

The two new show commands will display slightly different text if the
current target is a non-async target, which should allow users to
understand what's going on.

There's a somewhat random test adjustment needed in gdb.base/help.exp,
the test uses a regexp with the apropos command, and expects to find a
single result.  Turns out the new settings I added also matched the
regexp, which broke the test.  I've updated the regexp a little to
exclude my new settings.

In infcall.c you'll notice the thread_info::stop_requested flag being
set when a timeout occurs.  This flag setting is not required as part
of this commit, but will be needed in a later commit.  However, it
seemed like setting this flag fitted better with this commit, which is
why the change is added here.
---
 gdb/NEWS                                      |  18 ++
 gdb/doc/gdb.texinfo                           |  66 ++++++
 gdb/infcall.c                                 | 221 +++++++++++++++++-
 gdb/testsuite/gdb.base/help.exp               |   2 +-
 gdb/testsuite/gdb.base/infcall-timeout.c      |  36 +++
 gdb/testsuite/gdb.base/infcall-timeout.exp    |  82 +++++++
 .../infcall-from-bp-cond-timeout.c            | 169 ++++++++++++++
 .../infcall-from-bp-cond-timeout.exp          | 156 +++++++++++++
 8 files changed, 745 insertions(+), 5 deletions(-)
 create mode 100644 gdb/testsuite/gdb.base/infcall-timeout.c
 create mode 100644 gdb/testsuite/gdb.base/infcall-timeout.exp
 create mode 100644 gdb/testsuite/gdb.threads/infcall-from-bp-cond-timeout.c
 create mode 100644 gdb/testsuite/gdb.threads/infcall-from-bp-cond-timeout.exp

diff --git a/gdb/NEWS b/gdb/NEWS
index 445d28efed9..835bf7f8680 100644
--- a/gdb/NEWS
+++ b/gdb/NEWS
@@ -38,6 +38,24 @@ maintenance print record-instruction [ N ]
   prints how GDB would undo the N-th previous instruction, and if N is
   positive, it prints how GDB will redo the N-th following instruction.
 
+set direct-call-timeout SECONDS
+show direct-call-timeout
+set indirect-call-timeout SECONDS
+show indirect-call-timeout
+  These new settings can be used to limit how long GDB will wait for
+  an inferior function call to complete.  The direct timeout is used
+  for inferior function calls from e.g. 'call' and 'print' commands,
+  while the indirect timeout is used for inferior function calls from
+  within a conditional breakpoint expression.
+
+  The default for the direct timeout is unlimited, while the default
+  for the indirect timeout is 30 seconds.
+
+  These timeouts will only have an effect for targets that are
+  operating in async mode.  For non-async targets the timeouts are
+  ignored, GDB will wait indefinitely for an inferior function to
+  complete, unless interrupted by the user using Ctrl-C.
+
 * MI changes
 
 ** mi now reports 'no-history' as a stop reason when hitting the end of the
diff --git a/gdb/doc/gdb.texinfo b/gdb/doc/gdb.texinfo
index dd8267f448f..f5d460eb9e5 100644
--- a/gdb/doc/gdb.texinfo
+++ b/gdb/doc/gdb.texinfo
@@ -20808,6 +20808,72 @@
 @code{step}, etc).  In this case, when the inferior finally returns to
 the dummy-frame, @value{GDBN} will once again halt the inferior.
 
+On targets that support asynchronous execution (@pxref{Background
+Execution}) @value{GDBN} can place a timeout on any functions called
+from @value{GDBN}.  If the timeout expires and the function call is
+still ongoing, then @value{GDBN} will interrupt the program.
+
+For targets that don't support asynchronous execution
+(@pxref{Background Execution}) then timeouts for functions called from
+@value{GDBN} are not supported, the timeout settings described below
+will be treated as @code{unlimited}, meaning @value{GDBN} will wait
+indefinitely for function call to complete, unless interrupted by the
+user using @kbd{Ctrl-C}.
+
+@table @code
+@item set direct-call-timeout @var{seconds}
+@kindex set direct-call-timeout
+@cindex timeout for called functions
+Set the timeout used when calling functions in the program to
+@var{seconds}, which should be an integer greater than zero, or the
+special value @code{unlimited}, which indicates no timeout should be
+used.  The default for this setting is @code{unlimited}.
+
+This setting is used when the user calls a function directly from the
+command prompt, for example with a @code{call} or @code{print}
+command.
+
+This setting only works for targets that support asynchronous
+execution (@pxref{Background Execution}), for any other target the
+setting is treated as @code{unlimited}.
+
+@item show direct-call-timeout
+@kindex show direct-call-timeout
+@cindex timeout for called functions
+Show the timeout used when calling functions in the program with a
+@code{call} or @code{print} command.
+@end table
+
+It is also possible to call functions within the program from the
+condition of a conditional breakpoint (@pxref{Conditions, ,Break
+Conditions}).  A different setting controls the timeout used for
+function calls made from a breakpoint condition.
+
+@table @code
+@item set indirect-call-timeout @var{seconds}
+@kindex set indirect-call-timeout
+@cindex timeout for called functions
+Set the timeout used when calling functions in the program from a
+breakpoint or watchpoint condition to @var{seconds}, which should be
+an integer greater than zero, or the special value @code{unlimited},
+which indicates no timeout should be used.  The default for this
+setting is @code{30} seconds.
+
+This setting only works for targets that support asynchronous
+execution (@pxref{Background Execution}), for any other target the
+setting is treated as @code{unlimited}.
+
+If a function called from a breakpoint or watchpoint condition times
+out, then @value{GDBN} will stop at the point where the timeout
+occurred.  The breakpoint condition evaluation will be abandoned.
+
+@item show indirect-call-timeout
+@kindex show indirect-call-timeout
+@cindex timeout for called functions
+Show the timeout used when calling functions in the program from a
+breakpoint or watchpoint condition.
+@end table
+
 @subsection Calling functions with no debug info
 
 @cindex no debug info functions
diff --git a/gdb/infcall.c b/gdb/infcall.c
index 34838dea1bd..92f49a64e3e 100644
--- a/gdb/infcall.c
+++ b/gdb/infcall.c
@@ -95,6 +95,53 @@ show_may_call_functions_p (struct ui_file *file, int from_tty,
 	      value);
 }
 
+/* A timeout (in seconds) for direct inferior calls.  A direct inferior
+   call is one the user triggers from the prompt, e.g. with a 'call' or
+   'print' command.  Compare with the definition of indirect calls below.  */
+
+static unsigned int direct_call_timeout = UINT_MAX;
+
+/* Implement 'show direct-call-timeout'.  */
+
+static void
+show_direct_call_timeout (struct ui_file *file, int from_tty,
+			  struct cmd_list_element *c, const char *value)
+{
+  if (target_has_execution () && !target_can_async_p ())
+    gdb_printf (file, _("Current target does not support async mode, timeout "
+			"for direct inferior calls is \"unlimited\".\n"));
+  else if (direct_call_timeout == UINT_MAX)
+    gdb_printf (file, _("Timeout for direct inferior function calls "
+			"is \"unlimited\".\n"));
+  else
+    gdb_printf (file, _("Timeout for direct inferior function calls "
+			"is \"%s seconds\".\n"), value);
+}
+
+/* A timeout (in seconds) for indirect inferior calls.  An indirect inferior
+   call is one that originates from within GDB, for example, when
+   evaluating an expression for a conditional breakpoint.  Compare with
+   the definition of direct calls above.  */
+
+static unsigned int indirect_call_timeout = 30;
+
+/* Implement 'show indirect-call-timeout'.  */
+
+static void
+show_indirect_call_timeout (struct ui_file *file, int from_tty,
+			  struct cmd_list_element *c, const char *value)
+{
+  if (target_has_execution () && !target_can_async_p ())
+    gdb_printf (file, _("Current target does not support async mode, timeout "
+			"for indirect inferior calls is \"unlimited\".\n"));
+  else if (indirect_call_timeout == UINT_MAX)
+    gdb_printf (file, _("Timeout for indirect inferior function calls "
+			"is \"unlimited\".\n"));
+  else
+    gdb_printf (file, _("Timeout for indirect inferior function calls "
+			"is \"%s seconds\".\n"), value);
+}
+
 /* How you should pass arguments to a function depends on whether it
    was defined in K&R style or prototype style.  If you define a
    function using the K&R syntax that takes a `float' argument, then
@@ -589,6 +636,86 @@ call_thread_fsm::should_notify_stop ()
   return true;
 }
 
+/* A class to control creation of a timer that will interrupt a thread
+   during an inferior call.  */
+struct infcall_timer_controller
+{
+  /* Setup an event-loop timer that will interrupt PTID if the inferior
+     call takes too long.  DIRECT_CALL_P is true when this inferior call is
+     a result of the user using a 'print' or 'call' command, and false when
+     this inferior call is a result of e.g. a conditional breakpoint
+     expression, this is used to select which timeout to use.  */
+  infcall_timer_controller (thread_info *thr, bool direct_call_p)
+    : m_thread (thr)
+  {
+    unsigned int timeout
+      = direct_call_p ? direct_call_timeout : indirect_call_timeout;
+    if (timeout < UINT_MAX && target_can_async_p ())
+      {
+	int ms = timeout * 1000;
+	int id = create_timer (ms, infcall_timer_controller::timed_out, this);
+	m_timer_id.emplace (id);
+	infcall_debug_printf ("Setting up infcall timeout timer for "
+			      "ptid %s: %d milliseconds",
+			      m_thread->ptid.to_string ().c_str (), ms);
+      }
+  }
+
+  /* Destructor.  Ensure that the timer is removed from the event loop.  */
+  ~infcall_timer_controller ()
+  {
+    /* If the timer has already triggered, then it will have already been
+       deleted from the event loop.  If the timer has not triggered, then
+       delete it now.  */
+    if (m_timer_id.has_value () && !m_triggered)
+      delete_timer (*m_timer_id);
+
+    /* Just for clarity, discard the timer id now.  */
+    m_timer_id.reset ();
+  }
+
+  /* Return true if there was a timer in place, and the timer triggered,
+     otherwise, return false.  */
+  bool triggered_p ()
+  {
+    gdb_assert (!m_triggered || m_timer_id.has_value ());
+    return m_triggered;
+  }
+
+private:
+  /* The thread we should interrupt.  */
+  thread_info *m_thread;
+
+  /* Set true when the timer is triggered.  */
+  bool m_triggered = false;
+
+  /* Given a value when a timer is in place.  */
+  gdb::optional<int> m_timer_id;
+
+  /* Callback for the timer, forwards to ::trigger below.  */
+  static void
+  timed_out (gdb_client_data context)
+  {
+    infcall_timer_controller *ctrl
+      = static_cast<infcall_timer_controller *> (context);
+    ctrl->trigger ();
+  }
+
+  /* Called when the timer goes off.  Stop thread m_thread.  */
+  void
+  trigger ()
+  {
+    m_triggered = true;
+
+    scoped_disable_commit_resumed disable_commit_resumed ("infcall timeout");
+
+    infcall_debug_printf ("Stopping thread %s",
+			  m_thread->ptid.to_string ().c_str ());
+    target_stop (m_thread->ptid);
+    m_thread->stop_requested = true;
+  }
+};
+
 /* Subroutine of call_function_by_hand to simplify it.
    Start up the inferior and wait for it to stop.
    Return the exception if there's an error, or an exception with
@@ -599,13 +726,15 @@ call_thread_fsm::should_notify_stop ()
 
 static struct gdb_exception
 run_inferior_call (std::unique_ptr<call_thread_fsm> sm,
-		   struct thread_info *call_thread, CORE_ADDR real_pc)
+		   struct thread_info *call_thread, CORE_ADDR real_pc,
+		   bool *timed_out_p)
 {
   INFCALL_SCOPED_DEBUG_ENTER_EXIT;
 
   struct gdb_exception caught_error;
   ptid_t call_thread_ptid = call_thread->ptid;
   int was_running = call_thread->state == THREAD_RUNNING;
+  *timed_out_p = false;
 
   infcall_debug_printf ("call function at %s in thread %s, was_running = %d",
 			core_addr_to_string (real_pc),
@@ -617,6 +746,16 @@ run_inferior_call (std::unique_ptr<call_thread_fsm> sm,
   scoped_restore restore_in_infcall
     = make_scoped_restore (&call_thread->control.in_infcall, 1);
 
+  /* If the thread making the inferior call stops with a time out then the
+     stop_requested flag will be set.  However, we don't want changes to
+     this flag to leak back to our caller, we might be here to handle an
+     inferior call from a breakpoint condition, so leaving this flag set
+     would appear that the breakpoint stop was actually a requested stop,
+     which is not true, and will cause GDB to print extra messages to the
+     output.  */
+  scoped_restore restore_stop_requested
+    = make_scoped_restore (&call_thread->stop_requested, false);
+
   clear_proceed_status (0);
 
   /* Associate the FSM with the thread after clear_proceed_status
@@ -650,11 +789,23 @@ run_inferior_call (std::unique_ptr<call_thread_fsm> sm,
       infrun_debug_show_threads ("non-exited threads after proceed for inferior-call",
 				 all_non_exited_threads ());
 
+      /* Setup a timer (if possible, and if the settings allow) to prevent
+	 the inferior call running forever.  */
+      bool direct_call_p = !call_thread->control.in_cond_eval;
+      infcall_timer_controller infcall_timer (call_thread, direct_call_p);
+
       /* Inferior function calls are always synchronous, even if the
 	 target supports asynchronous execution.  */
       wait_sync_command_done ();
 
-      infcall_debug_printf ("inferior call completed successfully");
+      /* If the timer triggered then the inferior call failed.  */
+      if (infcall_timer.triggered_p ())
+	{
+	  infcall_debug_printf ("inferior call timed out");
+	  *timed_out_p = true;
+	}
+      else
+	infcall_debug_printf ("inferior call completed successfully");
     }
   catch (gdb_exception &e)
     {
@@ -1308,6 +1459,10 @@ call_function_by_hand_dummy (struct value *function,
   scoped_restore restore_stopped_by_random_signal
     = make_scoped_restore (&stopped_by_random_signal, 0);
 
+  /* Set to true by the call to run_inferior_call below if the inferior
+     call is artificially interrupted by GDB due to taking too long.  */
+  bool timed_out_p = false;
+
   /* - SNIP - SNIP - SNIP - SNIP - SNIP - SNIP - SNIP - SNIP - SNIP -
      If you're looking to implement asynchronous dummy-frames, then
      just below is the place to chop this function in two..  */
@@ -1334,7 +1489,8 @@ call_function_by_hand_dummy (struct value *function,
 			      struct_addr);
     {
       std::unique_ptr<call_thread_fsm> sm_up (sm);
-      e = run_inferior_call (std::move (sm_up), call_thread.get (), real_pc);
+      e = run_inferior_call (std::move (sm_up), call_thread.get (), real_pc,
+			     &timed_out_p);
     }
 
     if (e.reason < 0)
@@ -1486,7 +1642,10 @@ When the function is done executing, GDB will silently stop."),
       std::string name = get_function_name (funaddr, name_buf,
 					    sizeof (name_buf));
 
-      if (stopped_by_random_signal)
+      /* If the inferior call timed out then it will have been interrupted
+	 by a signal, but we want to report this differently to the user,
+	 which is done later in this function.  */
+      if (stopped_by_random_signal && !timed_out_p)
 	{
 	  /* We stopped inside the FUNCTION because of a random
 	     signal.  Further execution of the FUNCTION is not
@@ -1530,6 +1689,36 @@ GDB remains in the frame where the signal was received.\n\
 To change this behavior use \"set unwindonsignal on\".\n\
 Evaluation of the expression containing the function\n\
 (%s) will be abandoned.\n\
+When the function is done executing, GDB will silently stop."),
+		     name.c_str ());
+	    }
+	}
+
+      if (timed_out_p)
+	{
+	  /* A timeout results in a signal being sent to the inferior.  */
+	  gdb_assert (stopped_by_random_signal);
+
+	  /* Indentation is weird here.  A later patch is going to move the
+	    following block into an if/else, so I'm leaving the indentation
+	    here to minimise the later patch.
+
+	    Also, the error message used below refers to 'set
+	    unwind-on-timeout' which doesn't exist yet.  This will be added
+	    in a later commit, I'm leaving this in for now to minimise the
+	    churn caused by the commit that adds unwind-on-timeout.  */
+	    {
+	      /* The user wants to stay in the frame where we stopped
+		 (default).  Discard inferior status, we're not at the same
+		 point we started at.  */
+	      discard_infcall_control_state (inf_status.release ());
+
+	      error (_("\
+The program being debugged timed out while in a function called from GDB.\n\
+GDB remains in the frame where the timeout occurred.\n\
+To change this behavior use \"set unwind-on-timeout on\".\n\
+Evaluation of the expression containing the function\n\
+(%s) will be abandoned.\n\
 When the function is done executing, GDB will silently stop."),
 		     name.c_str ());
 	    }
@@ -1643,6 +1832,30 @@ The default is to unwind the frame."),
 			   show_unwind_on_terminating_exception_p,
 			   &setlist, &showlist);
 
+  add_setshow_uinteger_cmd ("direct-call-timeout", no_class,
+			    &direct_call_timeout, _("\
+Set the timeout, for direct calls to inferior function calls."), _("\
+Show the timeout, for direct calls to inferior function calls."), _("\
+If running on a target that supports, and is running in, async mode\n\
+then this timeout is used for any inferior function calls triggered\n\
+directly from the prompt, i.e. from a 'call' or 'print' command.  The\n\
+timeout is specified in seconds."),
+			    nullptr,
+			    show_direct_call_timeout,
+			    &setlist, &showlist);
+
+  add_setshow_uinteger_cmd ("indirect-call-timeout", no_class,
+			    &indirect_call_timeout, _("\
+Set the timeout, for indirect calls to inferior function calls."), _("\
+Show the timeout, for indirect calls to inferior function calls."), _("\
+If running on a target that supports, and is running in, async mode\n\
+then this timeout is used for any inferior function calls triggered\n\
+indirectly, i.e. being made as part of a breakpoint, or watchpoint,\n\
+condition expression.  The timeout is specified in seconds."),
+			    nullptr,
+			    show_indirect_call_timeout,
+			    &setlist, &showlist);
+
   add_setshow_boolean_cmd
     ("infcall", class_maintenance, &debug_infcall,
      _("Set inferior call debugging."),
diff --git a/gdb/testsuite/gdb.base/help.exp b/gdb/testsuite/gdb.base/help.exp
index 87919a819ab..504bf90cc15 100644
--- a/gdb/testsuite/gdb.base/help.exp
+++ b/gdb/testsuite/gdb.base/help.exp
@@ -121,7 +121,7 @@ gdb_test "help info bogus-gdb-command" "Undefined info command: \"bogus-gdb-comm
 gdb_test "help gotcha" "Undefined command: \"gotcha\"\.  Try \"help\"\."
 
 # Test apropos regex.
-gdb_test "apropos \\\(print\[\^\[ bsiedf\\\".-\]\\\)" "handle -- Specify how to handle signals\."
+gdb_test "apropos \\\(print\[\^\[ bsiedf\\\"'.-\]\\\)" "handle -- Specify how to handle signals\."
 # Test apropos >1 word string.
 gdb_test "apropos handle signal" "handle -- Specify how to handle signals\."
 # Test apropos apropos.
diff --git a/gdb/testsuite/gdb.base/infcall-timeout.c b/gdb/testsuite/gdb.base/infcall-timeout.c
new file mode 100644
index 00000000000..895e8a36d59
--- /dev/null
+++ b/gdb/testsuite/gdb.base/infcall-timeout.c
@@ -0,0 +1,36 @@
+/* Copyright 2022 Free Software Foundation, Inc.
+
+   This file is part of GDB.
+
+   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 <unistd.h>
+
+/* This function is called from GDB.  */
+int
+function_that_never_returns ()
+{
+  while (1)
+    sleep (1);
+
+  return 0;
+}
+
+int
+main ()
+{
+  alarm (300);
+
+  return 0;
+}
diff --git a/gdb/testsuite/gdb.base/infcall-timeout.exp b/gdb/testsuite/gdb.base/infcall-timeout.exp
new file mode 100644
index 00000000000..a5b0111ed04
--- /dev/null
+++ b/gdb/testsuite/gdb.base/infcall-timeout.exp
@@ -0,0 +1,82 @@
+# Copyright 2022 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/>.
+
+# Test GDB's direct-call-timeout setting, that is, ensure that if an
+# inferior function call, invoked from e.g. a 'print' command, takes
+# too long, then GDB can interrupt it, and return control to the user.
+
+standard_testfile
+
+if { [build_executable "failed to prepare" ${binfile} "${srcfile}" \
+	  {debug}] == -1 } {
+    return
+}
+
+# Start GDB according to TARGET_ASYNC and TARGET_NON_STOP, then adjust
+# the direct-call-timeout, and make an inferior function call that
+# will never return.  GDB should eventually timeout and stop the
+# inferior.
+proc_with_prefix run_test { target_async target_non_stop } {
+    save_vars { ::GDBFLAGS } {
+	append ::GDBFLAGS \
+	    " -ex \"maint set target-non-stop $target_non_stop\""
+	append ::GDBFLAGS \
+	    " -ex \"maintenance set target-async ${target_async}\""
+
+	clean_restart ${::binfile}
+    }
+
+    if {![runto_main]} {
+	fail "run to main"
+	return
+    }
+
+    gdb_test_no_output "set direct-call-timeout 5"
+
+    # When non-stop mode is off we get slightly different output from GDB.
+    if { [gdb_is_remote_or_extended_remote_target] && $target_non_stop == "off" } {
+	set stopped_line_pattern "Program received signal SIGINT, Interrupt\\."
+    } else {
+	set stopped_line_pattern "Program stopped\\."
+    }
+
+    gdb_test "print function_that_never_returns ()" \
+	[multi_line \
+	     $stopped_line_pattern \
+	     ".*" \
+	     "The program being debugged timed out while in a function called from GDB\\." \
+	     "GDB remains in the frame where the timeout occurred\\." \
+	     "To change this behavior use \"set unwind-on-timeout on\"\\." \
+	     "Evaluation of the expression containing the function" \
+	     "\\(function_that_never_returns\\) will be abandoned\\." \
+	     "When the function is done executing, GDB will silently stop\\."]
+
+    gdb_test "bt" ".* function_that_never_returns .*<function called from gdb>.*"
+}
+
+foreach_with_prefix target_async { "on" "off" } {
+
+    if { $target_async == "off" } {
+	# GDB can't timeout while waiting for a thread if the target
+	# runs with async-mode turned off; once the target is running
+	# GDB is effectively blocked until the target stops for some
+	# reason.
+	continue
+    }
+
+    foreach_with_prefix target_non_stop { "on" "off" } {
+	run_test $target_async $target_non_stop
+    }
+}
diff --git a/gdb/testsuite/gdb.threads/infcall-from-bp-cond-timeout.c b/gdb/testsuite/gdb.threads/infcall-from-bp-cond-timeout.c
new file mode 100644
index 00000000000..3bd91d7377d
--- /dev/null
+++ b/gdb/testsuite/gdb.threads/infcall-from-bp-cond-timeout.c
@@ -0,0 +1,169 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+   Copyright 2022 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 <stdio.h>
+#include <pthread.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <errno.h>
+#include <semaphore.h>
+
+#define NUM_THREADS 5
+
+/* Semaphores, used to track when threads have started, and to control
+   when the threads finish.  */
+sem_t startup_semaphore;
+sem_t finish_semaphore;
+sem_t thread_1_semaphore;
+sem_t thread_2_semaphore;
+
+/* Mutex to control when the first worker thread hit a breakpoint
+   location.  */
+pthread_mutex_t mutex = PTHREAD_MUTEX_INITIALIZER;
+
+/* Global variable to poke, just so threads have something to do.  */
+volatile int global_var = 0;
+
+int
+condition_func ()
+{
+  /* Let thread 2 run.  */
+  if (sem_post (&thread_2_semaphore) != 0)
+    abort ();
+
+  /* Wait for thread 2 to complete its actions.  */
+  if (sem_wait (&thread_1_semaphore) != 0)
+    abort ();
+
+  return 1;
+}
+
+void
+do_segfault ()
+{
+  volatile int *p = 0;
+  *p = 0;	/* Segfault here.  */
+}
+
+void *
+worker_func (void *arg)
+{
+  int tid = *((int *) arg);
+
+  /* Let the main thread know that this worker has started.  */
+  if (sem_post (&startup_semaphore) != 0)
+    abort ();
+
+  switch (tid)
+    {
+    case 0:
+      /* Wait for MUTEX to become available, then pass through the
+	 conditional breakpoint location.  */
+      if (pthread_mutex_lock (&mutex) != 0)
+	abort ();
+      global_var = 99;	/* Conditional breakpoint here.  */
+      if (pthread_mutex_unlock (&mutex) != 0)
+	abort ();
+      break;
+
+    case 1:
+      if (sem_wait (&thread_2_semaphore) != 0)
+	abort ();
+      do_segfault ();
+      if (sem_post (&thread_1_semaphore) != 0)
+	abort ();
+
+      /* Fall through.  */
+    default:
+      /* Wait until we are allowed to finish.  */
+      if (sem_wait (&finish_semaphore) != 0)
+	abort ();
+      break;
+    }
+}
+
+void
+stop_marker ()
+{
+  global_var = 99;	/* Stop marker.  */
+}
+
+/* The main program entry point.  */
+
+int
+main ()
+{
+  pthread_t threads[NUM_THREADS];
+  int args[NUM_THREADS];
+  void *retval;
+
+  /* An alarm, just in case the thread deadlocks.  */
+  alarm (300);
+
+  /* Semaphore initialization.  */
+  if (sem_init (&startup_semaphore, 0, 0) != 0)
+    abort ();
+  if (sem_init (&finish_semaphore, 0, 0) != 0)
+    abort ();
+  if (sem_init (&thread_1_semaphore, 0, 0) != 0)
+    abort ();
+  if (sem_init (&thread_2_semaphore, 0, 0) != 0)
+    abort ();
+
+  /* Lock MUTEX, this prevents the first worker thread from rushing ahead.  */
+  if (pthread_mutex_lock (&mutex) != 0)
+    abort ();
+
+  /* Worker thread creation.  */
+  for (int i = 0; i < NUM_THREADS; i++)
+    {
+      args[i] = i;
+      pthread_create (&threads[i], NULL, worker_func, &args[i]);
+    }
+
+  /* Wait for every thread to start.  */
+  for (int i = 0; i < NUM_THREADS; i++)
+    {
+      if (sem_wait (&startup_semaphore) != 0)
+	abort ();
+    }
+
+  /* Unlock the first thread so it can proceed.  */
+  if (pthread_mutex_unlock (&mutex) != 0)
+    abort ();
+
+  /* Wait for the first thread only.  */
+  pthread_join (threads[0], &retval);
+
+  /* Now post FINISH_SEMAPHORE to allow all the other threads to finish.  */
+  for (int i = 1; i < NUM_THREADS; i++)
+    sem_post (&finish_semaphore);
+
+  /* Now wait for the remaining threads to complete.  */
+  for (int i = 1; i < NUM_THREADS; i++)
+    pthread_join (threads[i], &retval);
+
+  /* Semaphore cleanup.  */
+  sem_destroy (&finish_semaphore);
+  sem_destroy (&startup_semaphore);
+  sem_destroy (&thread_1_semaphore);
+  sem_destroy (&thread_2_semaphore);
+
+  stop_marker ();
+
+  return 0;
+}
diff --git a/gdb/testsuite/gdb.threads/infcall-from-bp-cond-timeout.exp b/gdb/testsuite/gdb.threads/infcall-from-bp-cond-timeout.exp
new file mode 100644
index 00000000000..3341ff33f19
--- /dev/null
+++ b/gdb/testsuite/gdb.threads/infcall-from-bp-cond-timeout.exp
@@ -0,0 +1,156 @@
+# Copyright 2020 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/>.
+
+# Tests inferior calls executed from a breakpoint condition in
+# a multi-threaded program.
+#
+# This test has the inferior function call timeout, and checks how GDB
+# handles this situation.
+
+standard_testfile
+
+if { [build_executable "failed to prepare" ${binfile} "${srcfile}" \
+	  {debug pthreads}] } {
+    return
+}
+
+set cond_bp_line [gdb_get_line_number "Conditional breakpoint here"]
+set final_bp_line [gdb_get_line_number "Stop marker"]
+set segfault_line [gdb_get_line_number "Segfault here"]
+
+# Setup GDB based on TARGET_ASYNC and TARGET_NON_STOP.  Setup some
+# breakpoints in the inferior, one of which has an inferior call
+# within its condition.
+#
+# Continue GDB, the breakpoint with inferior call will be hit, but the
+# inferior call will never return.  We expect GDB to timeout.
+#
+# The reason that the inferior call never completes is that a second
+# thread, on which the inferior call relies, either hits a breakpoint
+# (when OTHER_THREAD_BP is true), or crashes (when OTHER_THREAD_BP is
+# false).
+proc run_test { target_async target_non_stop other_thread_bp } {
+    save_vars { ::GDBFLAGS } {
+	append ::GDBFLAGS " -ex \"maint set target-non-stop $target_non_stop\""
+	append ::GDBFLAGS " -ex \"maintenance set target-async ${target_async}\""
+
+	clean_restart ${::binfile}
+    }
+
+    if {![runto_main]} {
+	fail "run to main"
+	return
+    }
+
+    # The default timeout for indirect inferior calls (e.g. inferior
+    # calls for conditional breakpoint expressions) is pretty high.
+    # We don't want the test to take too long, so reduce this.
+    #
+    # However, the test relies on a second thread hitting some event
+    # (either a breakpoint or signal) before this timeout expires.
+    #
+    # There is a chance that on a really slow system this might not
+    # happen, in which case the test might fail.
+    #
+    # However, we still allocate 5 seconds, which feels like it should
+    # be enough time in most cases, but maybe we need to do something
+    # smarter here?  Possibly we could have some initial run where the
+    # inferior doesn't timeout, but does do the same interaction
+    # between threads, we could time that, and use that as the basis
+    # for this timeout.  For now though, we just hope 5 seconds is
+    # enough.
+    gdb_test_no_output "set indirect-call-timeout 5"
+
+    gdb_breakpoint \
+	"${::srcfile}:${::cond_bp_line} if (condition_func ())"
+    set bp_num [get_integer_valueof "\$bpnum" "*UNKNOWN*" \
+		    "get number for conditional breakpoint"]
+
+    gdb_breakpoint "${::srcfile}:${::final_bp_line}"
+    set final_bp_num [get_integer_valueof "\$bpnum" "*UNKNOWN*" \
+			  "get number for final breakpoint"]
+
+    # The thread performing an inferior call relies on a second
+    # thread.  The second thread will segfault unless it hits a
+    # breakpoint first.  In either case the initial thread will not
+    # complete its inferior call.
+    if { $other_thread_bp } {
+	gdb_breakpoint "${::srcfile}:${::segfault_line}"
+	set segfault_bp_num [get_integer_valueof "\$bpnum" "*UNKNOWN*" \
+				 "get number for segfault breakpoint"]
+    }
+
+    # When non-stop mode is off we get slightly different output from GDB.
+    if { [gdb_is_remote_or_extended_remote_target] && $target_non_stop == "off" } {
+	set stopped_line_pattern "Thread ${::decimal} \"\[^\r\n\"\]+\" received signal SIGINT, Interrupt\\."
+    } else {
+	set stopped_line_pattern "Thread ${::decimal} \"\[^\r\n\"\]+\" stopped\\."
+    }
+
+    gdb_test "continue" \
+	[multi_line \
+	     $stopped_line_pattern \
+	     ".*" \
+	     "Error in testing condition for breakpoint ${bp_num}:" \
+	     "The program being debugged timed out while in a function called from GDB\\." \
+	     "GDB remains in the frame where the timeout occurred\\." \
+	     "To change this behavior use \"set unwind-on-timeout on\"\\." \
+	     "Evaluation of the expression containing the function" \
+	     "\\(condition_func\\) will be abandoned\\." \
+	     "When the function is done executing, GDB will silently stop\\."] \
+	"expected timeout waiting for inferior call to complete"
+
+    # Remember that other thread that either crashed (with a segfault)
+    # or hit a breakpoint?  Now that the inferior call has timed out,
+    # if we try to resume then we should see the pending event from
+    # that other thread.
+    if { $other_thread_bp } {
+	gdb_test "continue" \
+	    [multi_line \
+		 "Continuing\\." \
+		 ".*" \
+		 "" \
+		 "Thread ${::decimal} \"\[^\"\r\n\]+\" hit Breakpoint ${segfault_bp_num}, do_segfault \[^\r\n\]+:${::segfault_line}" \
+		 "${::decimal}\\s+\[^\r\n\]+Segfault here\[^\r\n\]+"] \
+	    "hit the segfault breakpoint"
+    } else {
+	gdb_test "continue" \
+	    [multi_line \
+		 "Continuing\\." \
+		 ".*" \
+		 "Thread ${::decimal} \"infcall-from-bp\" received signal SIGSEGV, Segmentation fault\\." \
+		 "\\\[Switching to Thread \[^\r\n\]+\\\]" \
+		 "${::hex} in do_segfault \\(\\) at \[^\r\n\]+:${::segfault_line}" \
+		 "${::decimal}\\s+\[^\r\n\]+Segfault here\[^\r\n\]+"] \
+	    "hit the segfault"
+    }
+}
+
+foreach_with_prefix target_async {"on" "off" } {
+
+    if { $target_async == "off" } {
+	# GDB can't timeout while waiting for a thread if the target
+	# runs with async-mode turned off; once the target is running
+	# GDB is effectively blocked until the target stops for some
+	# reason.
+	continue
+    }
+
+    foreach_with_prefix target_non_stop {"off" "on"} {
+	foreach_with_prefix other_thread_bp { true false } {
+	    run_test $target_async $target_non_stop $other_thread_bp
+	}
+    }
+}
-- 
2.25.4


  parent reply	other threads:[~2023-01-31 17:49 UTC|newest]

Thread overview: 202+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-21  8:43 [PATCH 00/12] Infcalls from B/P conditions in multi-threaded inferiors Andrew Burgess
2022-10-21  8:43 ` [PATCH 01/12] gdb: int to bool conversion for normal_stop Andrew Burgess
2022-11-04 12:20   ` Lancelot SIX
2023-01-13 16:35     ` Andrew Burgess
2022-10-21  8:43 ` [PATCH 02/12] gdb/infrun: add debug print in print_signal_received_reason Andrew Burgess
2023-01-13 16:38   ` Andrew Burgess
2022-10-21  8:43 ` [PATCH 03/12] gdb: include breakpoint number in testing condition error message Andrew Burgess
2022-10-21  8:43 ` [PATCH 04/12] gdbserver: add comments to read_inferior_memory function Andrew Burgess
2023-01-13 16:42   ` Andrew Burgess
2022-10-21  8:43 ` [PATCH 05/12] gdbserver: allows agent_mem_read to return an error code Andrew Burgess
2022-10-21  8:43 ` [PATCH 06/12] gdbserver: allow agent expressions to fail with invalid memory access Andrew Burgess
2022-10-21  8:43 ` [PATCH 07/12] gdb: avoid repeated signal reporting during failed conditional breakpoint Andrew Burgess
2022-10-21  8:43 ` [PATCH 08/12] gdb: don't always print breakpoint location after failed condition check Andrew Burgess
2022-10-21  8:43 ` [PATCH 09/12] Revert "gdb: remove unnecessary parameter wait_ptid from do_target_wait" Andrew Burgess
2022-10-21  8:43 ` [PATCH 10/12] gdb: fix b/p conditions with infcalls in multi-threaded inferiors Andrew Burgess
2022-10-21  8:43 ` [PATCH 11/12] gdb: add timeouts for inferior function calls Andrew Burgess
2022-10-21 11:08   ` Eli Zaretskii
2023-01-14 11:00     ` Andrew Burgess
2023-01-14 11:48       ` Eli Zaretskii
2023-01-16 17:22         ` Andrew Burgess
2023-01-16 17:27           ` Eli Zaretskii
2022-11-04 23:17   ` Lancelot SIX
2023-01-13 16:49     ` Andrew Burgess
2023-01-16  9:44       ` Lancelot SIX
2022-10-21  8:43 ` [PATCH 12/12] gdb/remote: avoid SIGINT after calling remote_target::stop Andrew Burgess
2023-01-18 16:17 ` [PATCHv2 00/13] Infcalls from B/P conditions in multi-threaded inferiors Andrew Burgess
2023-01-18 16:17   ` [PATCHv2 01/13] gdb/doc: extended documentation for inferior function calls Andrew Burgess
2023-01-18 17:20     ` Eli Zaretskii
2023-03-16 17:15       ` Andrew Burgess
2023-01-19  9:00     ` Aktemur, Tankut Baris
2023-01-18 16:17   ` [PATCHv2 02/13] gdb/doc: extend the documentation for conditional breakpoints Andrew Burgess
2023-01-18 17:22     ` Eli Zaretskii
2023-01-19  9:04     ` Aktemur, Tankut Baris
2023-01-19 10:07       ` Eli Zaretskii
2023-01-18 16:17   ` [PATCHv2 03/13] gdb: include breakpoint number in testing condition error message Andrew Burgess
2023-01-19  9:54     ` Aktemur, Tankut Baris
2023-01-19 10:54     ` Aktemur, Tankut Baris
2023-01-19 11:34       ` Eli Zaretskii
2023-01-20  9:46         ` Aktemur, Tankut Baris
2023-01-25 16:49           ` Andrew Burgess
2023-01-25 17:09             ` Eli Zaretskii
2023-01-18 16:18   ` [PATCHv2 04/13] gdbserver: allows agent_mem_read to return an error code Andrew Burgess
2023-01-19  9:59     ` Aktemur, Tankut Baris
2023-01-18 16:18   ` [PATCHv2 05/13] gdbserver: allow agent expressions to fail with invalid memory access Andrew Burgess
2023-01-19 10:13     ` Aktemur, Tankut Baris
2023-01-18 16:18   ` [PATCHv2 06/13] gdb: avoid repeated signal reporting during failed conditional breakpoint Andrew Burgess
2023-01-19 10:33     ` Aktemur, Tankut Baris
2023-01-18 16:18   ` [PATCHv2 07/13] gdb: don't always print breakpoint location after failed condition check Andrew Burgess
2023-01-19 10:49     ` Aktemur, Tankut Baris
2023-01-18 16:18   ` [PATCHv2 08/13] Revert "gdb: remove unnecessary parameter wait_ptid from do_target_wait" Andrew Burgess
2023-01-19 11:05     ` Aktemur, Tankut Baris
2023-01-18 16:18   ` [PATCHv2 09/13] gdb: fix b/p conditions with infcalls in multi-threaded inferiors Andrew Burgess
2023-01-20  7:13     ` Aktemur, Tankut Baris
2023-01-18 16:18   ` [PATCHv2 10/13] gdb: add timeouts for inferior function calls Andrew Burgess
2023-01-18 17:30     ` Eli Zaretskii
2023-01-20  8:50     ` Aktemur, Tankut Baris
2023-01-18 16:18   ` [PATCHv2 11/13] gdb/remote: avoid SIGINT after calling remote_target::stop Andrew Burgess
2023-01-20  9:14     ` Aktemur, Tankut Baris
2023-01-18 16:18   ` [PATCHv2 12/13] gdb: introduce unwind-on-timeout setting Andrew Burgess
2023-01-18 17:33     ` Eli Zaretskii
2023-01-20  9:26     ` Aktemur, Tankut Baris
2023-01-18 16:18   ` [PATCHv2 13/13] gdb: rename unwindonsignal to unwind-on-signal Andrew Burgess
2023-01-18 17:35     ` Eli Zaretskii
2023-01-20  9:34   ` [PATCHv2 00/13] Infcalls from B/P conditions in multi-threaded inferiors Aktemur, Tankut Baris
2023-01-25 15:53     ` Andrew Burgess
2023-02-16 11:09       ` Aktemur, Tankut Baris
2023-01-31 17:27   ` [PATCHv3 " Andrew Burgess
2023-01-31 17:27     ` [PATCHv3 01/13] gdb/doc: extended documentation for inferior function calls Andrew Burgess
2023-01-31 17:27     ` [PATCHv3 02/13] gdb/doc: extend the documentation for conditional breakpoints Andrew Burgess
2023-01-31 18:07       ` Eli Zaretskii
2023-02-01 17:47         ` Andrew Burgess
2023-02-01 18:25           ` Eli Zaretskii
2023-02-02 13:34             ` Andrew Burgess
2023-01-31 17:27     ` [PATCHv3 03/13] gdb: include breakpoint number in testing condition error message Andrew Burgess
2023-02-16 10:15       ` Aktemur, Tankut Baris
2023-01-31 17:27     ` [PATCHv3 04/13] gdbserver: allows agent_mem_read to return an error code Andrew Burgess
2023-01-31 17:27     ` [PATCHv3 05/13] gdbserver: allow agent expressions to fail with invalid memory access Andrew Burgess
2023-02-16 10:29       ` Aktemur, Tankut Baris
2023-01-31 17:27     ` [PATCHv3 06/13] gdb: avoid repeated signal reporting during failed conditional breakpoint Andrew Burgess
2023-01-31 17:27     ` [PATCHv3 07/13] gdb: don't always print breakpoint location after failed condition check Andrew Burgess
2023-01-31 17:27     ` [PATCHv3 08/13] Revert "gdb: remove unnecessary parameter wait_ptid from do_target_wait" Andrew Burgess
2023-01-31 17:27     ` [PATCHv3 09/13] gdb: fix b/p conditions with infcalls in multi-threaded inferiors Andrew Burgess
2023-02-16 10:47       ` Aktemur, Tankut Baris
2023-01-31 17:27     ` Andrew Burgess [this message]
2023-01-31 18:11       ` [PATCHv3 10/13] gdb: add timeouts for inferior function calls Eli Zaretskii
2023-02-01 17:50         ` Andrew Burgess
2023-02-01 18:29           ` Eli Zaretskii
2023-02-16 10:53       ` Aktemur, Tankut Baris
2023-01-31 17:27     ` [PATCHv3 11/13] gdb/remote: avoid SIGINT after calling remote_target::stop Andrew Burgess
2023-01-31 17:27     ` [PATCHv3 12/13] gdb: introduce unwind-on-timeout setting Andrew Burgess
2023-01-31 18:09       ` Eli Zaretskii
2023-02-16 11:01       ` Aktemur, Tankut Baris
2023-01-31 17:27     ` [PATCHv3 13/13] gdb: rename unwindonsignal to unwind-on-signal Andrew Burgess
2023-01-31 18:12       ` Eli Zaretskii
2023-02-28 16:42     ` [PATCHv4 00/12] Infcalls from B/P conditions in multi-threaded inferiors Andrew Burgess
2023-02-28 16:42       ` [PATCHv4 01/12] gdb/doc: extended documentation for inferior function calls Andrew Burgess
2024-03-21  9:03         ` Tom de Vries
2024-03-21  9:11           ` Tom de Vries
2023-02-28 16:42       ` [PATCHv4 02/12] gdb: include breakpoint number in testing condition error message Andrew Burgess
2023-02-28 16:42       ` [PATCHv4 03/12] gdbserver: allows agent_mem_read to return an error code Andrew Burgess
2023-02-28 16:42       ` [PATCHv4 04/12] gdbserver: allow agent expressions to fail with invalid memory access Andrew Burgess
2023-02-28 16:42       ` [PATCHv4 05/12] gdb: avoid repeated signal reporting during failed conditional breakpoint Andrew Burgess
2023-02-28 16:42       ` [PATCHv4 06/12] gdb: don't always print breakpoint location after failed condition check Andrew Burgess
2023-02-28 16:42       ` [PATCHv4 07/12] Revert "gdb: remove unnecessary parameter wait_ptid from do_target_wait" Andrew Burgess
2023-02-28 16:42       ` [PATCHv4 08/12] gdb: fix b/p conditions with infcalls in multi-threaded inferiors Andrew Burgess
2023-02-28 16:42       ` [PATCHv4 09/12] gdb: add timeouts for inferior function calls Andrew Burgess
2023-02-28 16:42       ` [PATCHv4 10/12] gdb/remote: avoid SIGINT after calling remote_target::stop Andrew Burgess
2023-02-28 16:42       ` [PATCHv4 11/12] gdb: introduce unwind-on-timeout setting Andrew Burgess
2023-02-28 16:42       ` [PATCHv4 12/12] gdb: rename unwindonsignal to unwind-on-signal Andrew Burgess
2023-03-16 17:36       ` [PATCHv5 00/11] Infcalls from B/P conditions in multi-threaded inferiors Andrew Burgess
2023-03-16 17:36         ` [PATCHv5 01/11] gdb: include breakpoint number in testing condition error message Andrew Burgess
2023-04-03 13:50           ` Andrew Burgess
2023-07-07 12:08           ` Pedro Alves
2023-07-07 15:43             ` Andrew Burgess
2023-07-07 16:19               ` Pedro Alves
2023-07-10 10:30                 ` Andrew Burgess
2023-03-16 17:36         ` [PATCHv5 02/11] gdbserver: allows agent_mem_read to return an error code Andrew Burgess
2023-04-03 13:50           ` Andrew Burgess
2023-03-16 17:36         ` [PATCHv5 03/11] gdbserver: allow agent expressions to fail with invalid memory access Andrew Burgess
2023-04-03 13:50           ` Andrew Burgess
2023-07-07 12:25           ` Pedro Alves
2023-07-07 16:28             ` Andrew Burgess
2023-07-07 17:26               ` Pedro Alves
2023-07-07 21:19                 ` Andrew Burgess
2023-07-10 10:32                 ` Andrew Burgess
2023-07-10 10:44                   ` Pedro Alves
2023-07-10 13:44                     ` Andrew Burgess
2023-03-16 17:36         ` [PATCHv5 04/11] gdb: avoid repeated signal reporting during failed conditional breakpoint Andrew Burgess
2023-04-03 13:50           ` Andrew Burgess
2023-03-16 17:37         ` [PATCHv5 05/11] gdb: don't always print breakpoint location after failed condition check Andrew Burgess
2023-04-03 13:51           ` Andrew Burgess
2023-07-07 15:20           ` Pedro Alves
2023-07-07 15:24             ` Pedro Alves
2023-07-07 21:18               ` Andrew Burgess
2023-07-11 12:06                 ` Pedro Alves
2023-07-14 12:17                   ` Andrew Burgess
2023-07-17 17:17                     ` Pedro Alves
2023-08-03 13:57                       ` Andrew Burgess
2023-03-16 17:37         ` [PATCHv5 06/11] Revert "gdb: remove unnecessary parameter wait_ptid from do_target_wait" Andrew Burgess
2023-03-16 17:37         ` [PATCHv5 07/11] gdb: fix b/p conditions with infcalls in multi-threaded inferiors Andrew Burgess
2023-03-16 17:37         ` [PATCHv5 08/11] gdb: add timeouts for inferior function calls Andrew Burgess
2023-03-16 17:37         ` [PATCHv5 09/11] gdb/remote: avoid SIGINT after calling remote_target::stop Andrew Burgess
2023-03-16 17:37         ` [PATCHv5 10/11] gdb: introduce unwind-on-timeout setting Andrew Burgess
2023-03-16 17:37         ` [PATCHv5 11/11] gdb: rename unwindonsignal to unwind-on-signal Andrew Burgess
2023-04-03 14:01         ` [PATCHv6 0/6] Infcalls from B/P conditions in multi-threaded inferiors Andrew Burgess
2023-04-03 14:01           ` [PATCHv6 1/6] Revert "gdb: remove unnecessary parameter wait_ptid from do_target_wait" Andrew Burgess
2023-04-03 14:01           ` [PATCHv6 2/6] gdb: fix b/p conditions with infcalls in multi-threaded inferiors Andrew Burgess
2023-04-03 14:01           ` [PATCHv6 3/6] gdb: add timeouts for inferior function calls Andrew Burgess
2023-07-11 14:23             ` Pedro Alves
2023-07-14 15:20               ` Andrew Burgess
2023-07-14 19:52                 ` Andrew Burgess
2023-04-03 14:01           ` [PATCHv6 4/6] gdb/remote: avoid SIGINT after calling remote_target::stop Andrew Burgess
2023-04-03 14:01           ` [PATCHv6 5/6] gdb: introduce unwind-on-timeout setting Andrew Burgess
2023-04-03 14:01           ` [PATCHv6 6/6] gdb: rename unwindonsignal to unwind-on-signal Andrew Burgess
2023-05-15 19:22           ` [PATCHv7 0/6] Infcalls from B/P conditions in multi-threaded inferiors Andrew Burgess
2023-05-15 19:22             ` [PATCHv7 1/6] Revert "gdb: remove unnecessary parameter wait_ptid from do_target_wait" Andrew Burgess
2023-05-16 15:08               ` Aktemur, Tankut Baris
2023-05-15 19:22             ` [PATCHv7 2/6] gdb: fix b/p conditions with infcalls in multi-threaded inferiors Andrew Burgess
2023-05-16 15:09               ` Aktemur, Tankut Baris
2023-06-05 13:53                 ` Andrew Burgess
2023-05-15 19:22             ` [PATCHv7 3/6] gdb: add timeouts for inferior function calls Andrew Burgess
2023-05-16 15:42               ` Aktemur, Tankut Baris
2023-06-05 13:54                 ` Andrew Burgess
2023-05-15 19:22             ` [PATCHv7 4/6] gdb/remote: avoid SIGINT after calling remote_target::stop Andrew Burgess
2023-05-16 16:00               ` Aktemur, Tankut Baris
2023-06-05 13:55                 ` Andrew Burgess
2023-05-15 19:22             ` [PATCHv7 5/6] gdb: introduce unwind-on-timeout setting Andrew Burgess
2023-05-15 19:22             ` [PATCHv7 6/6] gdb: rename unwindonsignal to unwind-on-signal Andrew Burgess
2023-06-07 10:01             ` [PATCHv8 0/6] Infcalls from B/P conditions in multi-threaded inferiors Andrew Burgess
2023-06-07 10:01               ` [PATCHv8 1/6] Revert "gdb: remove unnecessary parameter wait_ptid from do_target_wait" Andrew Burgess
2023-06-07 10:01               ` [PATCHv8 2/6] gdb: fix b/p conditions with infcalls in multi-threaded inferiors Andrew Burgess
2023-06-07 10:01               ` [PATCHv8 3/6] gdb: add timeouts for inferior function calls Andrew Burgess
2023-06-07 10:01               ` [PATCHv8 4/6] gdb/remote: avoid SIGINT after calling remote_target::stop Andrew Burgess
2023-07-07 17:18                 ` Pedro Alves
2023-07-10 20:04                   ` Andrew Burgess
2023-06-07 10:01               ` [PATCHv8 5/6] gdb: introduce unwind-on-timeout setting Andrew Burgess
2023-06-07 10:01               ` [PATCHv8 6/6] gdb: rename unwindonsignal to unwind-on-signal Andrew Burgess
2023-06-07 12:41                 ` Eli Zaretskii
2023-06-07 14:29                   ` Andrew Burgess
2023-06-07 15:31                     ` Eli Zaretskii
2023-07-04 11:20               ` [PATCHv8 0/6] Infcalls from B/P conditions in multi-threaded inferiors Andrew Burgess
2023-12-02 10:52               ` [PATCHv9 0/5] " Andrew Burgess
2023-12-02 10:52                 ` [PATCHv9 1/5] Revert "gdb: remove unnecessary parameter wait_ptid from do_target_wait" Andrew Burgess
2023-12-02 10:52                 ` [PATCHv9 2/5] gdb: fix b/p conditions with infcalls in multi-threaded inferiors Andrew Burgess
2023-12-02 10:52                 ` [PATCHv9 3/5] gdb: add timeouts for inferior function calls Andrew Burgess
2023-12-02 10:52                 ` [PATCHv9 4/5] gdb: introduce unwind-on-timeout setting Andrew Burgess
2023-12-02 10:52                 ` [PATCHv9 5/5] gdb: rename unwindonsignal to unwind-on-signal Andrew Burgess
2024-01-02 15:57                 ` [PATCHv10 0/5] Infcalls from B/P conditions in multi-threaded inferiors Andrew Burgess
2024-01-02 15:57                   ` [PATCHv10 1/5] Revert "gdb: remove unnecessary parameter wait_ptid from do_target_wait" Andrew Burgess
2024-01-02 15:57                   ` [PATCHv10 2/5] gdb: fix b/p conditions with infcalls in multi-threaded inferiors Andrew Burgess
2024-01-02 15:57                   ` [PATCHv10 3/5] gdb: add timeouts for inferior function calls Andrew Burgess
2024-01-02 15:57                   ` [PATCHv10 4/5] gdb: introduce unwind-on-timeout setting Andrew Burgess
2024-01-02 15:57                   ` [PATCHv10 5/5] gdb: rename unwindonsignal to unwind-on-signal Andrew Burgess
2024-03-05 15:40                   ` [PATCHv11 0/5] Infcalls from B/P conditions in multi-threaded inferiors Andrew Burgess
2024-03-05 15:40                     ` [PATCHv11 1/5] Revert "gdb: remove unnecessary parameter wait_ptid from do_target_wait" Andrew Burgess
2024-03-05 15:40                     ` [PATCHv11 2/5] gdb: fix b/p conditions with infcalls in multi-threaded inferiors Andrew Burgess
2024-03-05 15:40                     ` [PATCHv11 3/5] gdb: add timeouts for inferior function calls Andrew Burgess
2024-03-05 15:40                     ` [PATCHv11 4/5] gdb: introduce unwind-on-timeout setting Andrew Burgess
2024-03-05 15:40                     ` [PATCHv11 5/5] gdb: rename unwindonsignal to unwind-on-signal Andrew Burgess
2024-03-14 16:08                     ` [PATCHv11 0/5] Infcalls from B/P conditions in multi-threaded inferiors Keith Seitz
2024-03-15 13:26                     ` Luis Machado
2024-03-25 17:47                     ` Andrew Burgess

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=d625aad4ffa546eb471cb4d79becd3852a03864d.1675185990.git.aburgess@redhat.com \
    --to=aburgess@redhat.com \
    --cc=gdb-patches@sourceware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).