Files
binutils-gdb/gdb/testsuite/gdb.threads/signal-command-multiple-signals-pending.exp
Andrew Burgess f302f9e26e gdb/testsuite: squash duplicate test names in gdb.threads/*.exp
Resolve all of the duplicate test names in the gdb.threads/*.exp set
of tests (that I see).  Nothing very exciting here, mostly either
giving tests explicit testnames, or adding with_test_prefix.

The only interesting one is gdb.threads/execl.exp, I believe the
duplicate test name was caused by an actual duplicate test.  I've
remove the simpler form of the test.  I don't believe we've lost any
test coverage with this change.

gdb/testsuite/ChangeLog:

	* gdb.threads/execl.exp: Remove duplicate 'info threads' test.
	Make use of $gdb_test_name instead of creating a separate $test
	variable.
	* gdb.threads/print-threads.exp: Add a with_test_prefix instead of
	adding a '($name)' at the end of each test.  This also catches the
	one place where '($name)' was missing, and so caused a duplicate
	test name.
	* gdb.threads/queue-signal.exp: Give tests unique names to avoid
	duplicate test names based on the command being tested.
	* gdb.threads/signal-command-multiple-signals-pending.exp:
	Likewise.
	* lib/gdb.exp (gdb_compile_shlib_pthreads): Tweak test name to
	avoid duplicate testnames when a test script uses this proc and
	also gdb_compile_pthreads.
	* lib/prelink-support.exp (build_executable_own_libs): Use
	with_test_prefix to avoid duplicate test names when we call
	build_executable twice.
2021-03-16 09:31:06 +00:00

176 lines
5.8 KiB
Plaintext

# Copyright (C) 2014-2021 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 that "signal FOO" behaves correctly when we have multiple
# threads that have stopped for a signal.
standard_testfile
if [target_info exists gdb,nosignals] {
verbose "Skipping ${testfile}.exp because of nosignals."
return -1
}
if {[gdb_compile_pthreads "${srcdir}/${subdir}/${srcfile}" "${binfile}" \
executable { debug }] != "" } {
return -1
}
# Run the test proper. SCHEDLOCK indicates which variant (around
# scheduler-locking) of the test to perform.
proc test { schedlock } {
global srcfile binfile
with_test_prefix "schedlock $schedlock" {
clean_restart ${binfile}
if ![runto_main] then {
fail "can't run to main"
return 0
}
gdb_test "handle SIGUSR1 stop print pass"
gdb_test "handle SIGUSR2 stop print pass"
gdb_test "break all_threads_started" "Breakpoint .* at .*$srcfile.*"
# Create threads one at a time, to insure stable thread
# numbers between runs and targets.
gdb_test "break thread_function" "Breakpoint .* at .*$srcfile.*"
gdb_test "continue" "thread_function.*" "thread 2 created"
gdb_test "continue" "thread_function.*" "thread 3 created"
gdb_test "continue" "all_threads_started.*" \
"continue to all_threads_started"
# Using schedlock, let the main thread queue a signal for each
# non-main thread.
gdb_test_no_output "set scheduler-locking on"
gdb_test "break all_threads_signalled" "Breakpoint .* at .*$srcfile.*"
gdb_test "continue" "all_threads_signalled.*" \
"continue to all_threads signalled"
gdb_test "info threads" "\\\* 1\[ \t\]+Thread.*" "thread 1 selected"
# With schedlock still enabled, let each thread report its
# signal.
gdb_test "thread 3" "Switching to thread 3.*"
gdb_test "continue" "Thread 3 .*received signal SIGUSR2.*" "stop with SIGUSR2"
gdb_test "thread 2" "Switching to thread 2.*"
gdb_test "continue" "Thread 2 .*received signal SIGUSR1.*" "stop with SIGUSR1"
gdb_test "break handler_sigusr1" "Breakpoint .* at .*$srcfile.*"
gdb_test "break handler_sigusr2" "Breakpoint .* at .*$srcfile.*"
set handler_re "Breakpoint .*, handler_sigusr. \\(sig=.*\\) at .*"
# Now test the "signal" command with either scheduler locking
# enabled or disabled.
if { $schedlock == "off" } {
# With scheduler locking off, switch to the main thread
# and issue "signal 0". "signal 0" should then warn that
# two threads have signals that will be delivered. When
# we let the command proceed, a signal should be
# delivered, and thus the corresponding breakpoint in the
# signal handler should trigger.
gdb_test_no_output "set scheduler-locking off"
gdb_test "thread 1" "Switching to thread 1.*"
set queried 0
set test "signal command queries"
gdb_test_multiple "signal 0" $test {
-re "stopped with.*stopped with.*stopped with.*Continue anyway.*y or n. $" {
fail "$test (too many threads noted)"
set queried 1
}
-re "stopped with signal SIGUSR.*\r\nContinuing .*still deliver .*Continue anyway.*y or n. $" {
pass $test
set queried 1
}
-re "Continue anyway.*y or n. $" {
fail "$test (no threads noted)"
set queried 1
}
}
# Continuing should stop in one of the signal handlers.
# Which thread runs first is not determinate.
if {$queried} {
gdb_test "y" "$handler_re" "one signal delivered"
}
# Continuing a second time should stop in the other
# handler.
with_test_prefix "second signal" {
gdb_test "continue" "$handler_re" "signal delivered"
}
} else {
# With scheduler locking on, stay with thread 2 selected,
# and try to deliver its signal explicitly. The "signal"
# command should then warn that one other thread has a
# signal that will be delivered. When we let the command
# proceed, the current thread's signal should be
# delivered, and thus the corresponding breakpoint in the
# signal handler should trigger.
gdb_test "signal SIGUSR1" \
"Breakpoint .*, handler_sigusr1 \\(sig=.*\\) at .*" \
"signal command does not query, signal delivered"
with_test_prefix "second signal" {
# The other thread had stopped for a signal too, and
# it wasn't resumed yet. Disabling schedlock and
# trying "signal 0" from the main thread should warn
# again.
gdb_test_no_output "set scheduler-locking off"
set queried 0
set test "signal command queries"
gdb_test_multiple "signal 0" $test {
-re "stopped with.*stopped with.*Continue anyway.*y or n. $" {
fail "$test (too many threads noted)"
set queried 1
}
-re "stopped with signal SIGUSR.*\r\nContinuing .*still deliver .*Continue anyway.*y or n. $" {
pass $test
set queried 1
}
-re "Continue anyway.*y or n. $" {
fail "$test (no threads noted)"
set queried 1
}
}
if {$queried} {
gdb_test "y" "Breakpoint .*, handler_sigusr2 \\(sig=.*\\) at .*" "signal delivered"
}
}
}
# Both threads got their signal. Continuing again should
# neither intercept nor deliver any other signal.
gdb_test "b end" "Breakpoint .* at .*$srcfile.*"
gdb_test "continue" "end .*" "no more signals"
}
}
foreach schedlock {"off" "on"} {
test $schedlock
}