From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 67227 invoked by alias); 8 Apr 2016 16:52:41 -0000 Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org Received: (qmail 66360 invoked by uid 89); 8 Apr 2016 16:52:40 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-2.9 required=5.0 tests=BAYES_00,RP_MATCHES_RCVD,SPF_HELO_PASS autolearn=ham version=3.3.2 spammy=1172, act 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; Fri, 08 Apr 2016 16:52:30 +0000 Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 47E217DCF0; Fri, 8 Apr 2016 16:52:29 +0000 (UTC) Received: from [127.0.0.1] (ovpn01.gateway.prod.ext.phx2.redhat.com [10.5.9.1]) by int-mx09.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id u38GqSaK029521; Fri, 8 Apr 2016 12:52:28 -0400 Subject: Re: [PATCH 1/7] New test case gdb.trace/signal.exp To: Yao Qi , gdb-patches@sourceware.org References: <1458749384-19793-1-git-send-email-yao.qi@linaro.org> <1458749384-19793-2-git-send-email-yao.qi@linaro.org> From: Pedro Alves Message-ID: <5707E1CC.2010609@redhat.com> Date: Fri, 08 Apr 2016 16:52:00 -0000 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.7.1 MIME-Version: 1.0 In-Reply-To: <1458749384-19793-2-git-send-email-yao.qi@linaro.org> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit X-SW-Source: 2016-04/txt/msg00183.txt.bz2 On 03/23/2016 04:09 PM, Yao Qi wrote: > This is to test whether GDBserver deliver signal to the inferior while > doing the step over. Nowadays, GDBserver doesn't deliver signal, so > there won't be spurious collection, however, if GDBserver does deliver > signal, there might be spurious collections. It'd be good to have an intro comment like this in the signal.exp file itself. Actually, it'd be good to add some general comments in the .exp file explaining what the test is doing: both an overview, and on the individual test parts. From the quick skim I did, I couldn't really figure out what is the test doing. > > gdb/testsuite: > > 2016-03-23 Yao Qi > > * gdb.trace/signal.c: New file. > * gdb.trace/signal.exp: New file. > --- > gdb/testsuite/gdb.trace/signal.c | 68 +++++++++++++++ > gdb/testsuite/gdb.trace/signal.exp | 172 +++++++++++++++++++++++++++++++++++++ > 2 files changed, 240 insertions(+) > create mode 100644 gdb/testsuite/gdb.trace/signal.c > create mode 100644 gdb/testsuite/gdb.trace/signal.exp > > diff --git a/gdb/testsuite/gdb.trace/signal.c b/gdb/testsuite/gdb.trace/signal.c > new file mode 100644 > index 0000000..b2b976d > --- /dev/null > +++ b/gdb/testsuite/gdb.trace/signal.c > @@ -0,0 +1,68 @@ > +/* This testcase is part of GDB, the GNU debugger. > + > + Copyright 2016 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 . */ > + > +#include > +#include > +#include > + > +static int counter = 0; > + > +static void > +handler (int sig) > +{ > + counter++; > +} > + > +static int iterations = 3; > + > +static void > +start (int pid) > +{ > + int i; > + > + for (i = 0; i < iterations; i++) > + { > + kill (pid, SIGABRT); > + } > +} > + > +static void > +end (void) > +{} > + > +int > +main (void) > +{ > + struct sigaction act; > + int i; > + > + memset (&act, 0, sizeof act); > + act.sa_handler = handler; > + act.sa_flags = SA_NODEFER; > + sigaction (SIGABRT, &act, NULL); > + > + for (i = 0; i < 3; i++) > + { > + kill (getpid (), SIGABRT); > + } > + > + counter = 0; > + start (getpid ()); > + > + end (); > + return 0; > +} > diff --git a/gdb/testsuite/gdb.trace/signal.exp b/gdb/testsuite/gdb.trace/signal.exp > new file mode 100644 > index 0000000..85898c3 > --- /dev/null > +++ b/gdb/testsuite/gdb.trace/signal.exp > @@ -0,0 +1,172 @@ > +# Copyright 2016 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 . > + > +set syscall_insn "" > + > +# Define the syscall instruction for each target. > + > +if { [istarget "i\[34567\]86-*-linux*"] || [istarget "x86_64-*-linux*"] } { > + set syscall_insn "\[ \t\](int|syscall|sysenter)\[ \t\]" > +} elseif { [istarget "aarch64*-*-linux*"] || [istarget "arm*-*-linux*"] } { > + set syscall_insn "\[ \t\](swi|svc)\[ \t\]" > +} else { > + return -1 Call "unsupported" before returning, perhaps? > +} > + > +load_lib "trace-support.exp" > + > +standard_testfile > + > +if {[prepare_for_testing $testfile.exp $testfile $srcfile debug]} { > + untested $testfile.exp > + return -1 > +} > + > +if ![runto_main] { > + fail "Can't run to main to check for trace support" > + return -1 > +} > + > +if ![gdb_target_supports_trace] { > + unsupported "target does not support trace" > + return -1 > +} > + > +# Start with a fresh gdb. > +clean_restart ${testfile} > +if ![runto_main] { > + fail "Can't run to main" > + return -1 > +} > + > +gdb_test "break kill" "Breakpoint \[0-9\]* at .*" > +gdb_test "handle SIGABRT nostop noprint pass" ".*" "pass SIGABRT" > + > +gdb_test "continue" "Continuing\\..*Breakpoint \[0-9\]+, (.* in |__libc_|)kill \\(\\).*" \ > + "continue to kill (1st time)" Avoid using ()s at end of test message to differentiate tests: https://sourceware.org/gdb/wiki/GDBTestcaseCookbook#Do_not_use_.22tail_parentheses.22_on_test_messages Suggest, e.g.: "continue to kill, 1st time" "continue to kill, 2nd time" > +# Hit the breakpoint on $syscall for the first time. In this time, > +# we will let PLT resolution done, and the number single steps we will > +# do later will be reduced. I found this comment quite confusing, until I realized you were talking about the previous line. > + > +gdb_test "continue" "Continuing\\..*Breakpoint \[0-9\]+, (.* in |__libc_|)kill \\(\\).*" \ > + "continue to kill (2nd time)" > +# Hit the breakpoint on $syscall for the second time. In this time, > +# the address of syscall insn and next insn of syscall are recorded. And this too. How about just moving the comments to just before each of the corresponding gdb_test calls? > + > +gdb_test "display/i \$pc" ".*" > + > +# Single step until we see a syscall insn or we reach the > +# upper bound of loop iterations. > +set msg "find syscall insn in kill" > +set steps 0 > +set max_steps 1000 > +gdb_test_multiple "stepi" $msg { > + -re ".*$syscall_insn.*$gdb_prompt $" { > + pass $msg > + } > + -re "x/i .*=>.*\r\n$gdb_prompt $" { > + incr steps > + if {$steps == $max_steps} { > + fail $msg > + } else { > + send_gdb "stepi\n" > + exp_continue > + } > + } > +} > + > +if {$steps == $max_steps} { > + return > +} > + > +# Remove the display > +gdb_test_no_output "delete display 1" > + > +set syscall_insn_addr [get_hexadecimal_valueof "\$pc" "0"] > +set syscall_insn_next 0 > +set test "x/2i \$pc" > +gdb_test_multiple $test $test { > + -re "$hex .*:\[^\r\n\]+\r\n\[ \]+($hex).*\.\r\n$gdb_prompt $" { > + set syscall_insn_next $expect_out(1,string) > + } > +} > + > +delete_breakpoints > +gdb_test "break start" "Breakpoint \[0-9\] at .*" > +gdb_continue_to_breakpoint "continue to start" > + > +gdb_assert { 0 == [get_integer_valueof "counter" "1"] } Might be nice to give this gdb_assert a more friendly test message. > + > +delete_breakpoints > + > +# Set tracepoints on syscall instruction and the next one. It is > +# more likely to get signal on these two places when GDBserver is > +# doing step-over. > +gdb_test "trace *$syscall_insn_addr" "Tracepoint \[0-9\] at .*" > +gdb_test "trace *$syscall_insn_next" "Tracepoint \[0-9\] at .*" Looks like this is putting addresses in the test messages, which is not stable. > + > +gdb_test "break end" "Breakpoint \[0-9\] at .*" > + > +gdb_test_no_output "tstart" > +gdb_test "continue" ".*Breakpoint.* end .*at.*$srcfile.*" \ > + "continue to end" > +gdb_test_no_output "tstop" > + > +set iterations [get_integer_valueof "iterations" "0"] > + > +gdb_assert { $iterations == [get_integer_valueof "counter" "0"] } Like above. > + > +# Record the hit times of each tracepoint in this array. > +array set tracepoint_hits { } > + > +set test "tfind 0" > +gdb_test_multiple $test $test { > + -re "Found trace frame 0, tracepoint ($decimal).*\r\n$gdb_prompt $" { > + > + set idx [expr $expect_out(1,string)] > + > + if {[info exists tracepoint_hits($idx)]} { > + incr tracepoint_hits($idx) > + } else { > + set tracepoint_hits($idx) 1 > + } > + } > +} > + > +set loop 1 > +while { $loop } { > + set test "tfind" > + gdb_test_multiple $test $test { > + -re "Found trace frame $decimal, tracepoint ($decimal).*\r\n$gdb_prompt $" { > + set idx [expr $expect_out(1,string)] > + > + if {[info exists tracepoint_hits($idx)]} { > + incr tracepoint_hits($idx) > + } else { > + set tracepoint_hits($idx) 1 > + } > + } > + -re "Target failed to find requested trace frame\..*\r\n$gdb_prompt $" { > + set loop 0 > + } > + } > +} > + > +for { set i 0 } { $i < [expr 20] } { incr i } { Why "20" here? > + if {[info exists tracepoint_hits($i)]} { > + gdb_assert { $tracepoint_hits($i) == $iterations } \ > + "tracepoint $i hit $iterations times" > + } > +} > I noticed several places could use $decimal. Thanks, Pedro Alves