This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Darwin 2/4] Do not crash (failed assertion) after PT_KILL ptrace error


This is something I noticed while reading the code.  Putting an
assertion that the PT_KILL ptrace call never returns an error
is too strong.  It might not be a debugger bug that caused the
PT_KILL ptrace operation to fail, so a failed-assertion "crash"
would not be justified.  It also seems easy enough to continue
and get ready for the next debugging session.  So this patch
changes the assertion into a warning.

This patch also tries to handle the case where ptrace return -1,
but left errno set to zero.  According to the ptrace man page,
it is possible for some ptrace operations to return -1 in non-error
situations, and to detect those situations, it explains that errno
should be set prior to calling ptrace, and then checked again after.

gdb/ChangeLog:

        * darwin-nat.c (darwin_ptrace): Add documentation.
        Set errno to zero before calling ptrace.  If ptrace returns
        -1 and errno is zero, then change then return zero.
        (darwin_kill_inferior): Issue a warning instead of triggering
        a failed assertion when the PT_KILL ptrace operations returned
        nonzero.

Tested on x86_64-darwin.  Checked in.

---
 gdb/ChangeLog    |    9 +++++++++
 gdb/darwin-nat.c |   17 ++++++++++++++++-
 2 files changed, 25 insertions(+), 1 deletions(-)

diff --git a/gdb/ChangeLog b/gdb/ChangeLog
index 5be8066..ce522fb 100644
--- a/gdb/ChangeLog
+++ b/gdb/ChangeLog
@@ -1,5 +1,14 @@
 2011-07-01  Joel Brobecker  <brobecker@adacore.com>
 
+	* darwin-nat.c (darwin_ptrace): Add documentation.
+	Set errno to zero before calling ptrace.  If ptrace returns
+	-1 and errno is zero, then change then return zero.
+	(darwin_kill_inferior): Issue a warning instead of triggering
+	a failed assertion when the PT_KILL ptrace operations returned
+	nonzero.
+
+2011-07-01  Joel Brobecker  <brobecker@adacore.com>
+
 	* darwin-nat.c (darwin_detach): Call darwin_resume_inferior
 	only when inf->private->no_ptrace.
 
diff --git a/gdb/darwin-nat.c b/gdb/darwin-nat.c
index fc5263a..27c6e2c 100644
--- a/gdb/darwin-nat.c
+++ b/gdb/darwin-nat.c
@@ -233,13 +233,25 @@ unparse_exception_type (unsigned int i)
     }
 }
 
+/* Set errno to zero, and then call ptrace with the given arguments.
+   If inferior debugging traces are on, then also print a debug
+   trace.
+
+   The returned value is the same as the value returned by ptrace,
+   except in the case where that value is -1 but errno is zero.
+   This case is documented to be a non-error situation, so we
+   return zero in that case. */
+
 static int
 darwin_ptrace (const char *name,
 	       int request, int pid, PTRACE_TYPE_ARG3 arg3, int arg4)
 {
   int ret;
 
+  errno = 0;
   ret = ptrace (request, pid, (caddr_t) arg3, arg4);
+  if (ret == -1 && errno == 0)
+    ret = 0;
 
   inferior_debug (4, _("ptrace (%s, %d, 0x%x, %d): %d (%s)\n"),
                   name, pid, arg3, arg4, ret,
@@ -1301,7 +1313,10 @@ darwin_kill_inferior (struct target_ops *ops)
       darwin_stop_inferior (inf);
 
       res = PTRACE (PT_KILL, inf->pid, 0, 0);
-      gdb_assert (res == 0);
+      if (res != 0)
+        warning (_("Failed to kill inferior: ptrace returned %d "
+	           "[%s] (pid=%d)"),
+		 res, safe_strerror (errno), inf->pid);
 
       darwin_reply_to_all_pending_messages (inf);
 
-- 
1.7.1


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]