From patchwork Mon Sep 8 16:26:55 2014 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Pedro Alves X-Patchwork-Id: 2692 Received: (qmail 6448 invoked by alias); 8 Sep 2014 16:27:06 -0000 Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Unsubscribe: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org Delivered-To: mailing list gdb-patches@sourceware.org Received: (qmail 6430 invoked by uid 89); 8 Sep 2014 16:27:03 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-3.2 required=5.0 tests=AWL, BAYES_00, RP_MATCHES_RCVD, SPF_HELO_PASS, SPF_PASS autolearn=ham version=3.3.2 X-HELO: mx1.redhat.com Received: from mx1.redhat.com (HELO mx1.redhat.com) (209.132.183.28) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES256-GCM-SHA384 encrypted) ESMTPS; Mon, 08 Sep 2014 16:27:02 +0000 Received: from int-mx14.intmail.prod.int.phx2.redhat.com (int-mx14.intmail.prod.int.phx2.redhat.com [10.5.11.27]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id s88GQwtj024917 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=FAIL); Mon, 8 Sep 2014 12:26:58 -0400 Received: from [127.0.0.1] (ovpn01.gateway.prod.ext.ams2.redhat.com [10.39.146.11]) by int-mx14.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id s88GQts1027213; Mon, 8 Sep 2014 12:26:56 -0400 Message-ID: <540DD8CF.5050200@redhat.com> Date: Mon, 08 Sep 2014 17:26:55 +0100 From: Pedro Alves User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.7.0 MIME-Version: 1.0 To: Jan Kratochvil CC: Doug Evans , Mark Wielaard , gdb-patches@sourceware.org Subject: Re: Regression: GDB stopped on run with attached process (PR 17347) [Re: [pushed+7.8] Re: [PATCH] Fix "attach" command vs user input race References: <1404291574.3766.35.camel@bordewijk.wildebeest.org> <53B3CDCC.9050502@redhat.com> <53B57911.10304@redhat.com> <53B6B0B8.2050702@redhat.com> <21434.52532.737427.778289@ruffy.mtv.corp.google.com> <53BC0D0B.7040001@redhat.com> <21437.28600.751354.629884@ruffy.mtv.corp.google.com> <53BD7749.5000800@redhat.com> <20140903075858.GA23492@host2.jankratochvil.net> <540775D7.7040003@redhat.com> <20140907192818.GA17035@host2.jankratochvil.net> In-Reply-To: <20140907192818.GA17035@host2.jankratochvil.net> On 09/07/2014 08:28 PM, Jan Kratochvil wrote: > Temporary breakpoint 1 at 0x8048481: file gdb/testsuite/gdb.base/attach.c, line 13.^M > Starting program: /unsafe/home/jkratoch/hammock/20140907fedorarel21-f21/fedora-2---Type to continue, or q to quit---ERROR: Window too small. > UNRESOLVED: gdb.base/attach.exp: cmdline attach run: run to main Good catch. I see that if one runs the test with a short enough terminal, do_command_attach_tests fails as well. I'm leaving that for a separate patch. >> +proc test_command_line_attach_run {} { >> + global gdb_prompt >> + global binfile > >> + global verbose >> + global GDB >> + global INTERNAL_GDBFLAGS > > These 3 lines are unused. Thanks, removed. >> + set test "run to prompt" >> + gdb_exit >> + set res [gdb_spawn_with_cmdline_opts "--pid=$testpid -ex \"start\""] > > Here see the attachment. Thanks for this. > Missing: > /* See top.h. */ > > Unless that rule from me has been abandoned. Added. Here's the updated patch. WDYT? ---------- 8< ---------- Subject: [PATCH 2/2] gdb/17347 - Regression: GDB stopped on run with attached process Doing: gdb --pid=PID -ex run Results in GDB getting a SIGTTIN, and thus ending stopped. That's usually indicative of a missing target_terminal_ours call. E.g., from the PR: $ sleep 1h & p=$!; sleep 0.1; gdb -batch sleep $p -ex run [1] 28263 [1] Killed sleep 1h [2]+ Stopped gdb -batch sleep $p -ex run The workaround is doing: gdb -ex "attach $PID" -ex "run" instead of gdb [-p] $PID -ex "run" With the former, gdb waits for the attach command to complete before moving on to the "run" command, because the interpreter is in sync mode at this point, within execute_command. But for the latter, attach_command is called directly from captured_main, and thus misses that waiting. IOW, "run" is running before the attach continuation has run, before the program stops and attach completes. The broken terminal settings are just one symptom of that. Any command that queries or requires input results in the same. The fix is to wait in catch_command_errors (which is specific to main.c nowadays), just like we wait in execute_command. gdb/ChangeLog: 2014-09-08 Pedro Alves PR gdb/17347 * main.c: Include "infrun.h". (catch_command_errors, catch_command_errors_const): Wait for the foreground command to complete. * top.c (maybe_wait_sync_command_done): New function, factored out from ... (maybe_wait_sync_command_done): ... here. * top.h (maybe_wait_sync_command_done): New declaration. gdb/testsuite/ChangeLog: 2014-09-08 Pedro Alves PR gdb/17347 * lib/gdb.exp (gdb_spawn_with_cmdline_opts): New procedure. * gdb.base/attach.exp (test_command_line_attach_run): New procedure. (top level): Call it. --- gdb/main.c | 9 ++++++++ gdb/testsuite/gdb.base/attach.exp | 45 +++++++++++++++++++++++++++++++++++++++ gdb/testsuite/lib/gdb.exp | 16 ++++++++++++++ gdb/top.c | 28 +++++++++++++++--------- gdb/top.h | 8 +++++++ 5 files changed, 96 insertions(+), 10 deletions(-) diff --git a/gdb/main.c b/gdb/main.c index 12f0146..756dd4f 100644 --- a/gdb/main.c +++ b/gdb/main.c @@ -45,6 +45,7 @@ #include "filestuff.h" #include #include "event-top.h" +#include "infrun.h" /* The selected interpreter. This will be used as a set command variable, so it should always be malloc'ed - since @@ -369,7 +370,11 @@ catch_command_errors (catch_command_errors_ftype *command, TRY_CATCH (e, mask) { + int was_sync = sync_execution; + command (arg, from_tty); + + maybe_wait_sync_command_done (was_sync); } return handle_command_errors (e); } @@ -388,7 +393,11 @@ catch_command_errors_const (catch_command_errors_const_ftype *command, TRY_CATCH (e, mask) { + int was_sync = sync_execution; + command (arg, from_tty); + + maybe_wait_sync_command_done (was_sync); } return handle_command_errors (e); } diff --git a/gdb/testsuite/gdb.base/attach.exp b/gdb/testsuite/gdb.base/attach.exp index a20c51a..6340496 100644 --- a/gdb/testsuite/gdb.base/attach.exp +++ b/gdb/testsuite/gdb.base/attach.exp @@ -396,6 +396,49 @@ proc do_command_attach_tests {} { remote_exec build "kill -9 ${testpid}" } +# Test ' gdb --pid PID -ex "run" '. GDB used to have a bug where +# "run" would run before the attach finished - PR17347. + +proc test_command_line_attach_run {} { + global gdb_prompt + global binfile + + if ![isnative] then { + unsupported "commandline attach run test" + return 0 + } + + with_test_prefix "cmdline attach run" { + set testpid [spawn_wait_for_attach $binfile] + + set test "run to prompt" + gdb_exit + + set res [gdb_spawn_with_cmdline_opts \ + "-iex \"set height 0\" -iex \"set width 0\" --pid=$testpid -ex \"start\""] + if { $res != 0} { + fail $test + return $res + } + gdb_test_multiple "" $test { + -re {Attaching to.*Start it from the beginning\? \(y or n\) } { + pass $test + } + } + + send_gdb "y\n" + + set test "run to main" + gdb_test_multiple "" $test { + -re "Temporary breakpoint .* main .*$gdb_prompt $" { + pass $test + } + } + + # Get rid of the process + remote_exec build "kill -9 ${testpid}" + } +} # Start with a fresh gdb @@ -420,4 +463,6 @@ do_call_attach_tests do_command_attach_tests +test_command_line_attach_run + return 0 diff --git a/gdb/testsuite/lib/gdb.exp b/gdb/testsuite/lib/gdb.exp index ecf942e..c54ba4b 100644 --- a/gdb/testsuite/lib/gdb.exp +++ b/gdb/testsuite/lib/gdb.exp @@ -3334,6 +3334,22 @@ proc gdb_spawn { } { default_gdb_spawn } +# Spawn GDB with CMDLINE_FLAGS appended to the GDBFLAGS global. + +proc gdb_spawn_with_cmdline_opts { cmdline_flags } { + global GDBFLAGS + + set saved_gdbflags $GDBFLAGS + + append GDBFLAGS $cmdline_flags + + set res [gdb_spawn] + + set GDBFLAGS $saved_gdbflags + + return $res +} + # Start gdb running, wait for prompt, and disable the pagers. # Overridable function -- you can override this function in your diff --git a/gdb/top.c b/gdb/top.c index fc2b84d..d6696cd 100644 --- a/gdb/top.c +++ b/gdb/top.c @@ -373,6 +373,23 @@ check_frame_language_change (void) } } +/* See top.h. */ + +void +maybe_wait_sync_command_done (int was_sync) +{ + /* If the interpreter is in sync mode (we're running a user + command's list, running command hooks or similars), and we + just ran a synchronous command that started the target, wait + for that command to end. */ + if (!interpreter_async && !was_sync && sync_execution) + { + while (gdb_do_one_event () >= 0) + if (!sync_execution) + break; + } +} + /* Execute the line P as a command, in the current user context. Pass FROM_TTY as second argument to the defining function. */ @@ -459,16 +476,7 @@ execute_command (char *p, int from_tty) else cmd_func (c, arg, from_tty); - /* If the interpreter is in sync mode (we're running a user - command's list, running command hooks or similars), and we - just ran a synchronous command that started the target, wait - for that command to end. */ - if (!interpreter_async && !was_sync && sync_execution) - { - while (gdb_do_one_event () >= 0) - if (!sync_execution) - break; - } + maybe_wait_sync_command_done (was_sync); /* If this command has been post-hooked, run the hook last. */ execute_cmd_post_hook (c); diff --git a/gdb/top.h b/gdb/top.h index c322c33..94f6c48 100644 --- a/gdb/top.h +++ b/gdb/top.h @@ -42,6 +42,14 @@ extern void quit_command (char *, int); extern void quit_cover (void); extern void execute_command (char *, int); +/* If the interpreter is in sync mode (we're running a user command's + list, running command hooks or similars), and we just ran a + synchronous command that started the target, wait for that command + to end. WAS_SYNC indicates whether sync_execution was set before + the command was run. */ + +extern void maybe_wait_sync_command_done (int was_sync); + extern void check_frame_language_change (void); /* Prepare for execution of a command.