public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Simon Marchi <simon.marchi@polymtl.ca>
To: Pedro Alves <palves@redhat.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [PATCH] Add several "quit with live inferior" tests
Date: Thu, 12 Oct 2017 22:11:00 -0000	[thread overview]
Message-ID: <5ecf8183d97c76fc4ca009d6f3e10432@polymtl.ca> (raw)
In-Reply-To: <1507805601-22576-1-git-send-email-palves@redhat.com>

On 2017-10-12 06:53, Pedro Alves wrote:
> In my multi-target branch, I had managed to break GDB exiting
> successfuly in response to "quit" or SIGHUP/SIGTERM when:
> 
>  - you're debugging with "target extended-remote",
>  - have more than one inferior loaded in gdb, some running, and at
>    least one not running, and,
>  - quit gdb with the inferior that is not running yet selected.
> 
> The testsuite still passed cleanly anyway.  I only noticed because I
> was left with a bunch of core dumps in the gdb/testsuite/ directory --
> the testsuite infrastructure closes GDB's pty after running each
> testcase, which results in GDB getting a SIGHUP and should make GDB
> exit gracefully.  If GDB crashes at that point though, there's no
> indication about it in gdb.sum/gdb.log.
> 
> This commit adds a multitude of tests exercising quitting GDB with
> live inferiors, some of which would have caught the problem.

I think you accidentally a file (quit.c).  Should it be named the same 
as the exp file (quit-live.c)?

It's a quite complex test, but you wrote it very clearly, so it's easy 
to follow.  I just have minor comments.

> gdb/testsuite/ChangeLog:
> yyyy-mm-dd  Pedro Alves  <palves@redhat.com>
> 
> 	* gdb.base/quit-live.exp: New file.
> ---
>  gdb/testsuite/gdb.base/quit-live.exp | 180 
> +++++++++++++++++++++++++++++++++++
>  1 file changed, 180 insertions(+)
>  create mode 100644 gdb/testsuite/gdb.base/quit-live.exp
> 
> diff --git a/gdb/testsuite/gdb.base/quit-live.exp
> b/gdb/testsuite/gdb.base/quit-live.exp
> new file mode 100644
> index 0000000..ef235d9
> --- /dev/null
> +++ b/gdb/testsuite/gdb.base/quit-live.exp
> @@ -0,0 +1,180 @@
> +# Copyright (C) 2017 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 quitting GDB with live inferiors.
> +#
> +# Exercises combinations of:
> +#
> +# - quitting with "quit"command, or with SIGTERM/SIGHUP signals.

missing space

> +#
> +# - quitting with live inferior selected, or file_stratum inferior
> +#   selected.
> +#
> +# - quitting after "run", or after "attach".
> +#
> +# - quitting with local executable, or executable loaded from target
> +#   directly (via default "target:/" sysroot), or with no executable
> +#   loaded.
> +
> +# Note: sending an asynchronous SIGHUP with kill is not the exact same
> +# as closing GDB's input, and that resulting in SIGHUP.  However, it's
> +# still a good approximation, and it has the advantage that because
> +# GDB still has a terminal, internal errors (if any) are visible in
> +# gdb.sum/gdb.log.
> +
> +standard_testfile quit.c
> +
> +if {[build_executable "failed to build" $testfile $srcfile debug]} {
> +    return
> +}
> +
> +# Send signal SIG to GDB, and expect GDB to exit.
> +
> +proc test_quit_with_sig {sig} {
> +    set gdb_pid [exp_pid -i [board_info host fileid]]
> +    remote_exec host "kill -$sig ${gdb_pid}"
> +
> +    set test "quit with SIG$sig"
> +    # If GDB mishandles the signal and doesn't exit, this
> +    # should FAIL with timeout.  We don't expect a GDB prompt,
> +    # so we see one, we'll FAIL too.

"so if we see one" ?

In this case, does the test fail if there's any output (no necessarily a 
gdb_prompt)?

> +    gdb_test_multiple "" $test {
> +	eof {
> +	    pass $test
> +	}
> +    }
> +}
> +
> +# Call the "quit" command with an inferior live.
> +#
> +# APPEAR_HOW specifies how the running inferior appears in GDB.  Can
> +# be either:
> +#
> +# - "run"
> +#
> +#    Appear via the "run" command.
> +#
> +# - "attach"
> +#
> +#    Appear via the "attach" command.
> +#
> +# - "attach-nofile"
> +#
> +#    Appear via the "attach" command, but with no program preloaded in
> +#    GDB so that GDB reads the program directly from the target when
> +#    remote debugging (i.e., from the target:/ sysroot).  This makes
> +#    sure that GDB doesn't misbehave if it decides to close the
> +#    'target:/.../program' exec_file after closing the remote
> +#    connection.
> +#
> +# EXTRA_INFERIOR is a boolean that specifies whether the "quit"
> +# command is typed with an extra executable-only (before "run")
> +# inferior selected or whether it is typed when the live inferior is
> +# selected, with no extra inferior.
> +#
> +# QUIT_HOW specifies how to tell GDB to quit.  It can be either "quit"
> +# (for "quit" command), "sighup" or "sigterm" (for quitting with
> +# SIGHUP and SIGTERM signals, respectively).
> +
> +proc quit_with_live_inferior {appear_how extra_inferior quit_how} {
> +    global srcfile testfile binfile
> +    global gdb_spawn_id gdb_prompt
> +
> +    set test_spawn_id ""
> +
> +    if {$appear_how != "attach-nofile"} {
> +	clean_restart $binfile
> +    } else {
> +	clean_restart
> +    }
> +
> +    if {$appear_how == "run"} {
> +	if ![runto_main] then {
> +	    fail "can't run to main"
> +	    return
> +	}
> +    } elseif {$appear_how == "attach" || $appear_how == 
> "attach-nofile"} {
> +	set test_spawn_id [spawn_wait_for_attach $binfile]
> +	set testpid [spawn_id_get_pid $test_spawn_id]
> +
> +	if {[gdb_test "attach $testpid" \
> +		 "Attaching to .*process $testpid.*Reading symbols from.*" \
> +		 "attach"] != 0} {
> +	    kill_wait_spawned_process $test_spawn_id
> +	    return
> +	}
> +    } else {
> +	error "unhandled '\$appear_how': $appear_how"
> +    }
> +
> +    if {$extra_inferior} {
> +	gdb_test "add-inferior" "Added inferior 2*" \
> +	    "add empty inferior 2"
> +	gdb_test "inferior 2" "Switching to inferior 2.*" \
> +	    "switch to inferior 2"
> +    }
> +
> +    # Make regexp that matches the "quit" command's output.
> +    proc make_re {how} {
> +	multi_line \
> +	    "A debugging session is active.\[ \t\r\n\]*Inferior 1\[^\r\n\]*
> will be $how\." \
> +	    "" \
> +	    "Quit anyway\\? \\(y or n\\) $"
> +    }
> +
> +    if {$appear_how == "run"} {
> +	set quit_anyway_re [make_re "killed"]
> +    } else {
> +	set quit_anyway_re [make_re "detached"]
> +    }
> +
> +    if {$quit_how == "quit"} {
> +	set test "quit with \"quit\""
> +	gdb_test_multiple "quit" $test {
> +	    -re $quit_anyway_re {
> +		send_gdb "y\n"
> +		gdb_test_multiple "" $test {
> +		    eof {
> +			pass $test
> +		    }
> +		}
> +	    }
> +	}
> +    } elseif {$quit_how == "sighup"} {
> +	test_quit_with_sig HUP
> +    } elseif {$quit_how == "sigterm"} {
> +	test_quit_with_sig TERM
> +    } else {
> +	error "unhandled '\$quit_how': $quit_how"
> +    }
> +
> +    if {$test_spawn_id != ""} {
> +	kill_wait_spawned_process $test_spawn_id
> +    }
> +}
> +
> +with_test_prefix "quit with live inferior" {

I think this prefix is not very useful, since it contains all the tests, 
although I'm not against it either.

> +    foreach_with_prefix appear_how {"run" "attach" "attach-nofile"} {
> +	if {$appear_how != "run" && ![can_spawn_for_attach]} {
> +	    continue
> +	}
> +
> +	foreach_with_prefix extra_inferior {0 1} {
> +	    foreach_with_prefix quit_how {"quit" "sigterm" "sighup"} {
> +		quit_with_live_inferior $appear_how $extra_inferior $quit_how
> +	    }
> +	}
> +    }
> +}

Thanks,

Simon

  reply	other threads:[~2017-10-12 22:11 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-10-12 10:53 Pedro Alves
2017-10-12 22:11 ` Simon Marchi [this message]
2017-10-17 10:36   ` Pedro Alves
2017-10-17 13:37     ` Simon Marchi
2017-10-17 16:28       ` Pedro Alves

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=5ecf8183d97c76fc4ca009d6f3e10432@polymtl.ca \
    --to=simon.marchi@polymtl.ca \
    --cc=gdb-patches@sourceware.org \
    --cc=palves@redhat.com \
    /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).