public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: "Andrew Burgess (Code Review)" <gerrit@gnutoolchain-gerrit.osci.io>
To: gdb-patches@sourceware.org
Subject: [review v2] gdb: Enable stdin on exception in execute_gdb_command
Date: Fri, 06 Dec 2019 10:46:00 -0000	[thread overview]
Message-ID: <20191206104616.1920128174@gnutoolchain-gerrit.osci.io> (raw)
In-Reply-To: <gerrit.1574471097000.I1cfc36ee9f8484cc1ed82be9be338353db6bc080@gnutoolchain-gerrit.osci.io>

Change URL: https://gnutoolchain-gerrit.osci.io/r/c/binutils-gdb/+/701
......................................................................

gdb: Enable stdin on exception in execute_gdb_command

This is an update of this patch:

  https://sourceware.org/ml/gdb-patches/2018-09/msg00884.html

This patch attempts to address PR gdb/23718 by re-enabling stdin
whenever an exception is caught during gdb.execute().

When Python gdb.execute() is called, an exception could occur (e.g. the
target disappearing), which is then converted into a Python exception.  If
stdin was disabled before the exception is caught, it is not re-enabled,
because the exception doesn't propagate to the top level of the event loop,
whose catch block would otherwise enable it.

The result is that when execution of a Python script completes, GDB does
not prompt or accept input, and is effectively hung.

This change rectifies the issue by re-enabling stdin in the catch block of
execute_gdb_command, prior to converting the exception to a Python
exception.

Since this patch was originally posted I've added a test, and also I
converted the code to re-enable stdin from this:

  SWITCH_THRU_ALL_UIS ()
    {
      async_enable_stdin ();
    }

to simply this:

  async_enable_stdin ();

My reasoning is that we only need the SWITCH_THRU_ALL_UIS if, at the time
the exception is caught, the current_ui might be different than at the time
we called async_disable_stdin.  Within python's execute_gdb_command I think
it should be impossible to switch current_ui, so the SWITCH_THRU_ALL_UIS
isn't needed.

gdb/ChangeLog:

	PR gdb/23718
	* gdb/python/python.c (execute_gdb_command): Call
	async_enable_stdin in catch block.

gdb/testsuite/ChangeLog:

        PR gdb/23718
	* gdb.server/server-kill-python.exp: New file.

Change-Id: I1cfc36ee9f8484cc1ed82be9be338353db6bc080
---
M gdb/ChangeLog
M gdb/python/python.c
M gdb/testsuite/ChangeLog
A gdb/testsuite/gdb.server/server-kill-python.exp
4 files changed, 98 insertions(+), 0 deletions(-)



diff --git a/gdb/ChangeLog b/gdb/ChangeLog
index 3b4afc8..e68591b 100644
--- a/gdb/ChangeLog
+++ b/gdb/ChangeLog
@@ -1,3 +1,9 @@
+2019-12-06  Graham Markall  <graham.markall@embecosm.com>
+
+	PR gdb/23718
+	* gdb/python/python.c (execute_gdb_command): Call
+	async_enable_stdin in catch block.
+
 2019-12-06  Andrew Burgess  <andrew.burgess@embecosm.com>
 
 	* event-loop.c (start_event_loop): Wrap async_enable_stdin with
diff --git a/gdb/python/python.c b/gdb/python/python.c
index fad54e9..65ccc40 100644
--- a/gdb/python/python.c
+++ b/gdb/python/python.c
@@ -620,6 +620,12 @@
     }
   catch (const gdb_exception &except)
     {
+      /* If an exception occurred then we won't hit normal_stop (), or have
+	 an exception reach the top level of the event loop, which are the
+	 two usual places in which stdin would be re-enabled. So, before we
+	 convert the exception and continue back in Python, we should
+	 re-enable stdin here.  */
+      async_enable_stdin ();
       GDB_PY_HANDLE_EXCEPTION (except);
     }
 
diff --git a/gdb/testsuite/ChangeLog b/gdb/testsuite/ChangeLog
index d3666d9..5b19ddd 100644
--- a/gdb/testsuite/ChangeLog
+++ b/gdb/testsuite/ChangeLog
@@ -1,5 +1,10 @@
 2019-12-06  Andrew Burgess  <andrew.burgess@embecosm.com>
 
+	PR gdb/23718
+	* gdb.server/server-kill-python.exp: New file.
+
+2019-12-06  Andrew Burgess  <andrew.burgess@embecosm.com>
+
 	* gdb.server/multi-ui-errors.c: New file.
 	* gdb.server/multi-ui-errors.exp: New file.
 
diff --git a/gdb/testsuite/gdb.server/server-kill-python.exp b/gdb/testsuite/gdb.server/server-kill-python.exp
new file mode 100644
index 0000000..0130459
--- /dev/null
+++ b/gdb/testsuite/gdb.server/server-kill-python.exp
@@ -0,0 +1,81 @@
+# This testcase is part of GDB, the GNU debugger.
+#
+# Copyright 2019 Free Software Foundation, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+# This test script exposes a bug where, if gdbserver dies while GDB is
+# sourcing a python command like 'gdb.execute ("continue")', then GDB
+# will deadlock.
+
+load_lib gdbserver-support.exp
+
+standard_testfile multi-ui-errors.c
+
+if {[skip_gdbserver_tests]} {
+    return 0
+}
+
+if {[build_executable "failed to prepare" ${testfile} \
+	 ${srcfile}] == -1} {
+    return -1
+}
+
+# Start gdbserver.
+set res [gdbserver_spawn "${binfile}"]
+set gdbserver_protocol [lindex $res 0]
+set gdbserver_gdbport [lindex $res 1]
+
+# Generate a python script we will later source.
+set file1 [standard_output_file file1.py]
+set fd [open "$file1" w]
+puts $fd \
+"import gdb
+
+def do_gdb_stuff ():
+    gdb.execute ('target $gdbserver_protocol $gdbserver_gdbport')
+    gdb.execute ('continue')
+
+do_gdb_stuff()"
+close $fd
+
+# Figure out the pid for the gdbserver, and arrange to kill it in a
+# short while.
+set gdbserver_pid [exp_pid -i $server_spawn_id]
+after 1000 { remote_exec target "kill -9 $gdbserver_pid" }
+
+# Now start GDB, sourcing the python command file we generated above.
+# Set the height and width so we don't end up at a paging prompt.
+if {[gdb_spawn_with_cmdline_opts \
+	 "-quiet -iex \"set height 0\" -iex \"set width 0\" -ex \"source $file1\""] != 0} {
+    fail "spawn"
+    return
+}
+
+# Expect that we get back to a GDB prompt.  We can't use
+# gdb_test_multiple here as we don't need to send a command to GDB;
+# the script we source at startup issues a command for us.  Here we
+# really are waiting for GDB to give us back a prompt.
+set testname "landed at prompt after gdbserver dies"
+gdb_expect 10 {
+    -re "$gdb_prompt $" {
+	pass $testname
+    }
+    timeout {
+	fail "$testname (timeout)"
+    }
+}
+
+# Run a simple command to ensure we can interact with GDB.
+gdb_test "echo hello\\n" "hello" "can we interact with gdb"

-- 
Gerrit-Project: binutils-gdb
Gerrit-Branch: master
Gerrit-Change-Id: I1cfc36ee9f8484cc1ed82be9be338353db6bc080
Gerrit-Change-Number: 701
Gerrit-PatchSet: 2
Gerrit-Owner: Andrew Burgess <andrew.burgess@embecosm.com>
Gerrit-MessageType: newpatchset

  reply	other threads:[~2019-12-06 10:46 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-23  1:05 [review] " Andrew Burgess (Code Review)
2019-12-06 10:46 ` Andrew Burgess (Code Review) [this message]
2019-12-23  1:46 ` [review v2] " Andrew Burgess (Code Review)

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=20191206104616.1920128174@gnutoolchain-gerrit.osci.io \
    --to=gerrit@gnutoolchain-gerrit.osci.io \
    --cc=andrew.burgess@embecosm.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).