binutils-gdb/gdb/testsuite/gdb.base/foll-vfork.exp

489 lines
15 KiB
Plaintext
Raw Normal View History

# Copyright 1997-2016 Free Software Foundation, Inc.
1999-06-29 01:04:32 +02:00
# 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
1999-06-29 01:04:32 +02:00
# (at your option) any later version.
#
1999-06-29 01:04:32 +02:00
# 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.
#
1999-06-29 01:04:32 +02:00
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
1999-06-29 01:04:32 +02:00
# Various tests of gdb's ability to follow the parent or child of a
# Unix vfork system call. A vfork parent is blocked until the child
# either execs or exits --- since those events take somewhat different
# code paths in GDB, both variants are exercised.
# Until "set follow-fork-mode" and "catch vfork" are implemented on
# other targets...
#
Remove HP-UX references fom testsuite This patch removes all special cases for HP-UX, for which support has been removed earlier, that I found in the testsuite. Note that the hppa architecture != HP-UX, since other OSes can run on hppa, so I tried to leave everything that is not HP-UX specific. Two complete tests were completely HP-UX specific, so I removed them. I ran the testsuite on Linux x86-64, native and native-gdbserver, and noticed no regressions. gdb/testsuite/ChangeLog: * gdb.asm/asm-source.exp: Remove HP-UX references. * gdb.base/annota1.exp: Likewise. * gdb.base/annota3.exp: Likewise. * gdb.base/attach.exp: Likewise. * gdb.base/bigcore.exp: Likewise. * gdb.base/break.exp: Likewise. * gdb.base/call-ar-st.exp: Likewise. * gdb.base/callfuncs.exp: Likewise. * gdb.base/catch-fork-static.exp: Likewise. * gdb.base/display.exp: Likewise. * gdb.base/foll-exec-mode.exp: Likewise. * gdb.base/foll-exec.exp: Likewise. * gdb.base/foll-fork.exp: Likewise. * gdb.base/foll-vfork.exp: Likewise. * gdb.base/funcargs.exp: Likewise. * gdb.base/hbreak2.exp: Likewise. * gdb.base/inferior-died.exp: Likewise. * gdb.base/interrupt.exp: Likewise. * gdb.base/multi-forks.exp: Likewise. * gdb.base/nodebug.exp: Likewise. * gdb.base/sepdebug.exp: Likewise. * gdb.base/solib1.c: Likewise. * gdb.base/step-test.exp: Likewise. * gdb.mi/non-stop.c: Likewise. * gdb.mi/pthreads.c: Likewise. * gdb.multi/bkpt-multi-exec.ex: Likewise. * gdb.threads/pthreads.c: Likewise. * gdb.threads/staticthreads.exp: Likewise. * lib/future.exp: Likewise. * lib/gdb.exp: Likewise. * gdb.base/so-indr-cl.c: Remove. * gdb.base/so-indr-cl.exp: Likewise. * gdb.base/solib.c: Likewise. * gdb.base/solib.exp: Likewise. * gdb.base/solib2.c: Likewise.
2015-12-21 18:51:54 +01:00
if {![istarget "*-linux*"]} then {
continue
}
standard_testfile
1999-06-29 01:04:32 +02:00
set compile_options debug
set dirname [relative_filename [pwd] [file dirname $binfile]]
lappend compile_options "additional_flags=-DBASEDIR=\"$dirname\""
if {[build_executable $testfile.exp $testfile $srcfile $compile_options] == -1} {
untested "failed to compile $testfile"
return -1
1999-06-29 01:04:32 +02:00
}
set testfile2 "vforked-prog"
set srcfile2 ${testfile2}.c
if {[build_executable $testfile.exp $testfile2 $srcfile2 $compile_options] == -1} {
untested "failed to compile $testfile2"
return -1
1999-06-29 01:04:32 +02:00
}
# A few of these tests require a little more time than the standard
# timeout allows.
set oldtimeout $timeout
set timeout [expr "$timeout + 10"]
# Start with a fresh GDB, with verbosity enabled, and run to main. On
# error, behave as "return", so we don't try to continue testing with
# a borked session.
proc setup_gdb {} {
global testfile
clean_restart $testfile
# The "Detaching..." and "Attaching..." messages may be hidden by
# default.
gdb_test_no_output "set verbose"
if ![runto_main] {
return -code return
}
}
proc check_vfork_catchpoints {} {
global gdb_prompt
global has_vfork_catchpoints
setup_gdb
# Verify that the system supports "catch vfork".
gdb_test "catch vfork" "Catchpoint \[0-9\]* \\(vfork\\)" "insert first vfork catchpoint"
set has_vfork_catchpoints 0
gdb_test_multiple "continue" "continue to first vfork catchpoint" {
2010-01-11 Thiago Jung Bauermann <bauerman@br.ibm.com> Convert hardware watchpoints to use breakpoint_ops. gdb/ * breakpoint.h (breakpoint_ops) <insert>: Rename to... <insert_location>: ... this. Return int instead of void. Accept pointer to struct bp_location instead of pointer to struct breakpoint. Adapt all implementations. (breakpoint_ops) <remove>: Rename to... <remove_location>: ... this. Accept pointer to struct bp_location instead of pointer to struct breakpoint. Adapt all implementations. * breakpoint.c (insert_catchpoint): Delete function. (insert_bp_location): Call the watchpoint or catchpoint's breakpoint_ops.insert method. (remove_breakpoint_1): Call the watchpoint or catchpoint's breakpoint_ops.remove method. (insert_watchpoint, remove_watchpoint): New functions. (watchpoint_breakpoint_ops): New structure. (watch_command_1): Initialize the OPS field. * inf-child.c (inf_child_insert_fork_catchpoint) (inf_child_remove_fork_catchpoint, inf_child_insert_vfork_catchpoint) (inf_child_remove_vfork_catchpoint, inf_child_insert_exec_catchpoint) (inf_child_remove_exec_catchpoint, inf_child_set_syscall_catchpoint): Delete functions. (inf_child_target): Remove initialization of to_insert_fork_catchpoint, to_remove_fork_catchpoint, to_insert_vfork_catchpoint, to_remove_vfork_catchpoint, to_insert_exec_catchpoint, to_remove_exec_catchpoint and to_set_syscall_catchpoint. * target.c (update_current_target): Change default implementation of to_insert_fork_catchpoint, to_remove_fork_catchpoint, to_insert_vfork_catchpoint, to_remove_vfork_catchpoint, to_insert_exec_catchpoint, to_remove_exec_catchpoint and to_set_syscall_catchpoint to return_one. (debug_to_insert_fork_catchpoint, debug_to_insert_vfork_catchpoint) (debug_to_insert_exec_catchpoint): Report return value. * target.h (to_insert_fork_catchpoint, to_insert_vfork_catchpoint) (to_insert_exec_catchpoint): Change declaration to return int instead of void. gdb/testsuite/ * gdb.base/foll-exec.exp: Adapt to new error string when the catchpoint type is not supported. * gdb.base/foll-fork.exp: Likewise. * gdb.base/foll-vfork.exp: Likewise.
2011-01-11 20:16:23 +01:00
-re ".*Your system does not support this type\r\nof catchpoint.*$gdb_prompt $" {
unsupported "continue to first vfork catchpoint"
}
-re ".*Catchpoint.*$gdb_prompt $" {
set has_vfork_catchpoints 1
pass "continue to first vfork catchpoint"
}
}
if {$has_vfork_catchpoints == 0} {
unsupported "vfork catchpoints"
return -code return
}
}
1999-06-29 01:04:32 +02:00
proc vfork_parent_follow_through_step {} {
with_test_prefix "vfork parent follow, through step" {
1999-06-29 01:04:32 +02:00
global gdb_prompt
setup_gdb
gdb_test_no_output "set follow-fork parent"
set test "step"
gdb_test_multiple "next" $test {
Follow-fork message printing improvements This commit modifies the code that prints attach and detach messages related to following fork and vfork. The changes include using target_terminal_ours_for_output instead of target_terminal_ours, printing "vfork" instead of "fork" for all vfork-related messages, and using _() for the format strings of all of the messages. We also add a "detach" message for when a fork parent is detached. Previously in this case the only message was notification of attaching to the child. We still do not print any messages when following the parent and detaching the child (the default). The rationale for this is that from the user's perspective the new child was never attached. Note that all of these messages are only printed when 'verbose' is set or when debugging is turned on. The tests gdb.base/foll-fork.exp and gdb.base/foll-vfork.exp were modified to check for the new message. Tested on x64 Ubuntu Lucid, native only. gdb/ChangeLog: * infrun.c (follow_fork_inferior): Update fork message printing to use target_terminal_ours_for_output instead of target_terminal_ours, to use _() for all format strings, to print "vfork" instead of "fork" for vforks, and to add a detach message. (handle_vfork_child_exec_or_exit): Update message printing to use target_terminal_ours_for_output instead of target_terminal_ours, to use _() for all format strings, and to fix some formatting. gdb/testsuite/ChangeLog: * gdb.base/foll-fork.exp (test_follow_fork, catch_fork_child_follow): Check for updated fork messages emitted from infrun.c. * gdb.base/foll-vfork.exp (vfork_parent_follow_through_step, vfork_parent_follow_to_bp, vfork_and_exec_child_follow_to_main_bp, vfork_and_exec_child_follow_through_step): Check for updated vfork messages emitted from infrun.c.
2014-10-24 20:36:06 +02:00
-re "Detaching after vfork from.*if \\(pid == 0\\).*$gdb_prompt " {
pass $test
}
1999-06-29 01:04:32 +02:00
}
# The child has been detached; allow time for any output it might
# generate to arrive, so that output doesn't get confused with
# any gdb_expected debugger output from a subsequent testpoint.
#
exec sleep 1
}}
1999-06-29 01:04:32 +02:00
proc vfork_parent_follow_to_bp {} {
with_test_prefix "vfork parent follow, to bp" {
1999-06-29 01:04:32 +02:00
global gdb_prompt
global srcfile
1999-06-29 01:04:32 +02:00
setup_gdb
gdb_test_no_output "set follow-fork parent"
set bp_location [gdb_get_line_number "printf (\"I'm the proud parent of child"]
gdb_test "break ${srcfile}:${bp_location}" ".*" "break, vfork to bp"
set test "continue to bp"
gdb_test_multiple "continue" $test {
Follow-fork message printing improvements This commit modifies the code that prints attach and detach messages related to following fork and vfork. The changes include using target_terminal_ours_for_output instead of target_terminal_ours, printing "vfork" instead of "fork" for all vfork-related messages, and using _() for the format strings of all of the messages. We also add a "detach" message for when a fork parent is detached. Previously in this case the only message was notification of attaching to the child. We still do not print any messages when following the parent and detaching the child (the default). The rationale for this is that from the user's perspective the new child was never attached. Note that all of these messages are only printed when 'verbose' is set or when debugging is turned on. The tests gdb.base/foll-fork.exp and gdb.base/foll-vfork.exp were modified to check for the new message. Tested on x64 Ubuntu Lucid, native only. gdb/ChangeLog: * infrun.c (follow_fork_inferior): Update fork message printing to use target_terminal_ours_for_output instead of target_terminal_ours, to use _() for all format strings, to print "vfork" instead of "fork" for vforks, and to add a detach message. (handle_vfork_child_exec_or_exit): Update message printing to use target_terminal_ours_for_output instead of target_terminal_ours, to use _() for all format strings, and to fix some formatting. gdb/testsuite/ChangeLog: * gdb.base/foll-fork.exp (test_follow_fork, catch_fork_child_follow): Check for updated fork messages emitted from infrun.c. * gdb.base/foll-vfork.exp (vfork_parent_follow_through_step, vfork_parent_follow_to_bp, vfork_and_exec_child_follow_to_main_bp, vfork_and_exec_child_follow_through_step): Check for updated vfork messages emitted from infrun.c.
2014-10-24 20:36:06 +02:00
-re ".*Detaching after vfork from child process.*Breakpoint.*${bp_location}.*$gdb_prompt " {
pass $test
}
1999-06-29 01:04:32 +02:00
}
# The child has been detached; allow time for any output it might
# generate to arrive, so that output doesn't get confused with
# any expected debugger output from a subsequent testpoint.
#
exec sleep 1
}}
1999-06-29 01:04:32 +02:00
proc vfork_child_follow_to_exit {} {
with_test_prefix "vfork child follow, to exit" {
global gdb_prompt
setup_gdb
gdb_test_no_output "set follow-fork child"
set test "continue to child exit"
gdb_test_multiple "continue" $test {
-re "Couldn't get registers.*$gdb_prompt " {
# PR gdb/14766
fail "$test"
}
-re "Attaching after.* vfork to.*Detaching vfork parent .* after child exit.*$gdb_prompt " {
pass $test
}
}
# The parent has been detached; allow time for any output it might
# generate to arrive, so that output doesn't get confused with
# any gdb_expected debugger output from a subsequent testpoint.
#
exec sleep 1
}}
1999-06-29 01:04:32 +02:00
proc vfork_and_exec_child_follow_to_main_bp {} {
with_test_prefix "vfork and exec child follow, to main bp" {
1999-06-29 01:04:32 +02:00
global gdb_prompt
global srcfile2
1999-06-29 01:04:32 +02:00
setup_gdb
gdb_test_no_output "set follow-fork child"
set linenum [gdb_get_line_number "printf(\"Hello from vforked-prog" ${srcfile2}]
set test "continue to bp"
gdb_test_multiple "continue" $test {
Follow-fork message printing improvements This commit modifies the code that prints attach and detach messages related to following fork and vfork. The changes include using target_terminal_ours_for_output instead of target_terminal_ours, printing "vfork" instead of "fork" for all vfork-related messages, and using _() for the format strings of all of the messages. We also add a "detach" message for when a fork parent is detached. Previously in this case the only message was notification of attaching to the child. We still do not print any messages when following the parent and detaching the child (the default). The rationale for this is that from the user's perspective the new child was never attached. Note that all of these messages are only printed when 'verbose' is set or when debugging is turned on. The tests gdb.base/foll-fork.exp and gdb.base/foll-vfork.exp were modified to check for the new message. Tested on x64 Ubuntu Lucid, native only. gdb/ChangeLog: * infrun.c (follow_fork_inferior): Update fork message printing to use target_terminal_ours_for_output instead of target_terminal_ours, to use _() for all format strings, to print "vfork" instead of "fork" for vforks, and to add a detach message. (handle_vfork_child_exec_or_exit): Update message printing to use target_terminal_ours_for_output instead of target_terminal_ours, to use _() for all format strings, and to fix some formatting. gdb/testsuite/ChangeLog: * gdb.base/foll-fork.exp (test_follow_fork, catch_fork_child_follow): Check for updated fork messages emitted from infrun.c. * gdb.base/foll-vfork.exp (vfork_parent_follow_through_step, vfork_parent_follow_to_bp, vfork_and_exec_child_follow_to_main_bp, vfork_and_exec_child_follow_through_step): Check for updated vfork messages emitted from infrun.c.
2014-10-24 20:36:06 +02:00
-re "Attaching after.* vfork to.*Detaching vfork parent.*xecuting new program.*Breakpoint.*vforked-prog.c:${linenum}.*$gdb_prompt " {
pass $test
}
1999-06-29 01:04:32 +02:00
}
# The parent has been detached; allow time for any output it might
# generate to arrive, so that output doesn't get confused with
# any gdb_expected debugger output from a subsequent testpoint.
#
exec sleep 1
}}
1999-06-29 01:04:32 +02:00
proc vfork_and_exec_child_follow_through_step {} {
with_test_prefix "vfork and exec child follow, through step" {
1999-06-29 01:04:32 +02:00
global gdb_prompt
global srcfile2
1999-06-29 01:04:32 +02:00
setup_gdb
gdb_test_no_output "set follow-fork child"
set test "step over vfork"
Remove HP-UX references fom testsuite This patch removes all special cases for HP-UX, for which support has been removed earlier, that I found in the testsuite. Note that the hppa architecture != HP-UX, since other OSes can run on hppa, so I tried to leave everything that is not HP-UX specific. Two complete tests were completely HP-UX specific, so I removed them. I ran the testsuite on Linux x86-64, native and native-gdbserver, and noticed no regressions. gdb/testsuite/ChangeLog: * gdb.asm/asm-source.exp: Remove HP-UX references. * gdb.base/annota1.exp: Likewise. * gdb.base/annota3.exp: Likewise. * gdb.base/attach.exp: Likewise. * gdb.base/bigcore.exp: Likewise. * gdb.base/break.exp: Likewise. * gdb.base/call-ar-st.exp: Likewise. * gdb.base/callfuncs.exp: Likewise. * gdb.base/catch-fork-static.exp: Likewise. * gdb.base/display.exp: Likewise. * gdb.base/foll-exec-mode.exp: Likewise. * gdb.base/foll-exec.exp: Likewise. * gdb.base/foll-fork.exp: Likewise. * gdb.base/foll-vfork.exp: Likewise. * gdb.base/funcargs.exp: Likewise. * gdb.base/hbreak2.exp: Likewise. * gdb.base/inferior-died.exp: Likewise. * gdb.base/interrupt.exp: Likewise. * gdb.base/multi-forks.exp: Likewise. * gdb.base/nodebug.exp: Likewise. * gdb.base/sepdebug.exp: Likewise. * gdb.base/solib1.c: Likewise. * gdb.base/step-test.exp: Likewise. * gdb.mi/non-stop.c: Likewise. * gdb.mi/pthreads.c: Likewise. * gdb.multi/bkpt-multi-exec.ex: Likewise. * gdb.threads/pthreads.c: Likewise. * gdb.threads/staticthreads.exp: Likewise. * lib/future.exp: Likewise. * lib/gdb.exp: Likewise. * gdb.base/so-indr-cl.c: Remove. * gdb.base/so-indr-cl.exp: Likewise. * gdb.base/solib.c: Likewise. * gdb.base/solib.exp: Likewise. * gdb.base/solib2.c: Likewise.
2015-12-21 18:51:54 +01:00
# The ideal support is to be able to debug the child even
# before it execs. Thus, "next" lands on the next line after
# the vfork.
gdb_test_multiple "next" $test {
-re "Attaching after .* vfork to child.*if \\(pid == 0\\).*$gdb_prompt " {
pass "$test"
}
1999-06-29 01:04:32 +02:00
}
# The parent has been detached; allow time for any output it might
# generate to arrive, so that output doesn't get confused with
# any expected debugger output from a subsequent testpoint.
#
exec sleep 1
}}
1999-06-29 01:04:32 +02:00
proc tcatch_vfork_then_parent_follow {} {
with_test_prefix "vfork parent follow, finish after tcatch vfork" {
1999-06-29 01:04:32 +02:00
global gdb_prompt
global srcfile
setup_gdb
gdb_test_no_output "set follow-fork parent"
gdb_test "tcatch vfork" "Catchpoint .*(vfork).*"
gdb_test "continue" "0x\[0-9a-fA-F\]*.*(vfork|__kernel_v?syscall).*" \
"continue to vfork"
set linenum [gdb_get_line_number "pid = vfork ();"]
set test "finish"
gdb_test_multiple "finish" $test {
-re "Run till exit from.*vfork.*0x\[0-9a-fA-F\]* in main .* at .*${srcfile}:${linenum}.*$gdb_prompt " {
pass $test
}
-re "Run till exit from.*__kernel_v?syscall.*0x\[0-9a-fA-F\]* in vfork .*$gdb_prompt " {
send_gdb "finish\n"
exp_continue
}
1999-06-29 01:04:32 +02:00
}
# The child has been detached; allow time for any output it might
# generate to arrive, so that output doesn't get confused with
# any expected debugger output from a subsequent testpoint.
#
exec sleep 1
}}
1999-06-29 01:04:32 +02:00
proc tcatch_vfork_then_child_follow_exec {} {
with_test_prefix "vfork child follow, finish after tcatch vfork" {
1999-06-29 01:04:32 +02:00
global gdb_prompt
global srcfile
1999-06-29 01:04:32 +02:00
global srcfile2
setup_gdb
gdb_test_no_output "set follow-fork child"
gdb_test "tcatch vfork" "Catchpoint .*(vfork).*"
gdb_test "continue" "0x\[0-9a-fA-F\]*.*(vfork|__kernel_v?syscall).*" \
"continue to vfork"
set linenum1 [gdb_get_line_number "pid = vfork ();"]
set linenum2 [gdb_get_line_number "printf(\"Hello from vforked-prog" ${srcfile2}]
set test "finish"
gdb_test_multiple "finish" $test {
-re "Run till exit from.*vfork.*${srcfile}:${linenum1}.*$gdb_prompt " {
pass $test
}
-re "Run till exit from.*__kernel_v?syscall.*0x\[0-9a-fA-F\]* in vfork .*$gdb_prompt " {
send_gdb "finish\n"
exp_continue
}
-re "Run till exit from.*vfork.*${srcfile2}:${linenum2}.*$gdb_prompt " {
pass "$test (followed exec)"
}
1999-06-29 01:04:32 +02:00
}
# The parent has been detached; allow time for any output it might
# generate to arrive, so that output doesn't get confused with
# any expected debugger output from a subsequent testpoint.
#
exec sleep 1
}}
1999-06-29 01:04:32 +02:00
proc tcatch_vfork_then_child_follow_exit {} {
with_test_prefix "vfork child follow, finish after tcatch vfork" {
1999-06-29 01:04:32 +02:00
global gdb_prompt
global srcfile
setup_gdb
gdb_test_no_output "set follow-fork child"
gdb_test "tcatch vfork" "Catchpoint .*(vfork).*"
gdb_test "continue" "0x\[0-9a-fA-F\]*.*(vfork|__kernel_v?syscall).*" \
"continue to vfork"
set test "finish"
gdb_test_multiple "finish" $test {
-re "Run till exit from.*vfork.*exited normally.*$gdb_prompt " {
setup_kfail "gdb/14762" *-*-*
fail $test
}
-re "Run till exit from.*vfork.*pid = vfork \\(\\).*$gdb_prompt " {
pass $test
}
-re "Run till exit from.*__kernel_v?syscall.*0x\[0-9a-fA-F\]* in vfork .*$gdb_prompt " {
send_gdb "finish\n"
exp_continue
}
}
# The parent has been detached; allow time for any output it might
# generate to arrive, so that output doesn't get confused with
# any expected debugger output from a subsequent testpoint.
#
exec sleep 1
}}
1999-06-29 01:04:32 +02:00
proc vfork_relations_in_info_inferiors { variant } {
with_test_prefix "vfork relations in info inferiors" {
global gdb_prompt
setup_gdb
gdb_test_no_output "set follow-fork child"
set test "step over vfork"
gdb_test_multiple "next" $test {
-re "Attaching after .* vfork to child.*if \\(pid == 0\\).*$gdb_prompt " {
pass "$test"
}
}
gdb_test "info inferiors" \
".*is vfork child of inferior 1.*is vfork parent of inferior 2" \
"info inferiors shows vfork parent/child relation"
if { $variant == "exec" } {
global srcfile2
set linenum [gdb_get_line_number "printf(\"Hello from vforked-prog" ${srcfile2}]
set test "continue to bp"
gdb_test_multiple "continue" $test {
-re ".*xecuting new program.*Breakpoint.*vforked-prog.c:${linenum}.*$gdb_prompt " {
pass $test
}
}
} else {
set test "continue to child exit"
gdb_test_multiple "continue" $test {
-re "exited normally.*$gdb_prompt " {
pass $test
}
}
}
set test "vfork relation no longer appears in info inferiors"
gdb_test_multiple "info inferiors" $test {
-re "is vfork child of inferior 1.*$gdb_prompt $" {
fail $test
}
-re "is vfork parent of inferior 2.*$gdb_prompt $" {
fail $test
}
-re "$gdb_prompt $" {
pass $test
}
}
}}
proc do_vfork_and_follow_parent_tests {} {
global gdb_prompt
1999-06-29 01:04:32 +02:00
# Try following the parent process by stepping through a call to
# vfork. Do this without catchpoints.
vfork_parent_follow_through_step
1999-06-29 01:04:32 +02:00
# Try following the parent process by setting a breakpoint on the
# other side of a vfork, and running to that point. Do this
# without catchpoints.
vfork_parent_follow_to_bp
1999-06-29 01:04:32 +02:00
# Try catching a vfork, and stepping out to the parent.
#
tcatch_vfork_then_parent_follow
}
proc do_vfork_and_follow_child_tests_exec {} {
1999-06-29 01:04:32 +02:00
# Try following the child process by just continuing through the
# vfork, and letting the parent's breakpoint on "main" be auto-
# magically reset in the child.
#
vfork_and_exec_child_follow_to_main_bp
1999-06-29 01:04:32 +02:00
# Try following the child process by stepping through a call to
# vfork. The child also executes an exec. Since the child cannot
# be debugged until after it has exec'd, and since there's a bp on
# "main" in the parent, and since the bp's for the parent are
# recomputed in the exec'd child, the step through a vfork should
# land us in the "main" for the exec'd child, too.
#
vfork_and_exec_child_follow_through_step
1999-06-29 01:04:32 +02:00
# Try catching a vfork, and stepping out to the child.
#
tcatch_vfork_then_child_follow_exec
1999-06-29 01:04:32 +02:00
# Test the ability to follow both child and parent of a vfork. Do
# this without catchpoints.
# ??rehrauer: NYI. Will add testpoints here when implemented.
#
# Test the ability to have the debugger ask the user at vfork-time
# whether to follow the parent, child or both. Do this without
# catchpoints.
# ??rehrauer: NYI. Will add testpoints here when implemented.
#
# Step over a vfork in the child, do "info inferiors" and check the
# parent/child relation is displayed. Run the child over the exec,
# and confirm the relation is no longer displayed in "info
# inferiors".
#
vfork_relations_in_info_inferiors "exec"
1999-06-29 01:04:32 +02:00
}
proc do_vfork_and_follow_child_tests_exit {} {
# Try following the child process by just continuing through the
# vfork, and letting the child exit.
#
vfork_child_follow_to_exit
# Try catching a vfork, and stepping out to the child.
#
tcatch_vfork_then_child_follow_exit
# Step over a vfork in the child, do "info inferiors" and check the
# parent/child relation is displayed. Run the child to completion,
# and confirm the relation is no longer displayed in "info
# inferiors".
#
vfork_relations_in_info_inferiors "exit"
}
with_test_prefix "check vfork support" {
# Check that vfork catchpoints are supported, as an indicator for
# whether vfork-following is supported.
check_vfork_catchpoints
}
Extended-remote exec test This patch updates several exec-related tests and some of the library functions in order to get them running with extended-remote. There were three changes that were required, as follows: In gdb.base/foll-exec.exp, use 'clean_start' in place of proc 'zap_session' to reset the state of the debugger between tests. This sets 'remote exec-file' to execute the correct binary file in each subsequent test. In gdb.base/pie-execl.exp, there is an expect statement with an expression that is used to match output from both gdb and the program under debug. For the remote target, this had to be split into two expressions, using $inferior_spawn_id to match the output from the program. Because I had encountered problems with extended-remote exec events in non-stop mode in my manual testing, I added non-stop testing to the non-ldr-exc-[1234].exp tests. In order to set non-stop mode for remote targets, it is necessary to 'set non-stop on' after gdb has started, but before it connects to gdbserver. This is done using 'save_vars' to set non-stop mode in GDBFLAGS, so GDB sets non-stop mode on startup. gdb/testsuite/ChangeLog: * gdb.base/foll-exec.c: Add copyright header. Fix formatting issues. * gdb.base/foll-exec.exp (zap_session): Delete proc. (do_exec_tests): Use clean_restart in place of zap_session, and for test initialization. Fix formatting issues. Use fail in place of perror. * gdb.base/pie-execl.exp (main): Use 'inferior_spawn_id' in an expect statement to match an expression with output from the program under debug. * gdb.threads/non-ldr-exc-1.exp (do_test, main): Add non-stop tests and pass stop mode argument to clean_restart. Use save_vars to enable non-stop in GDBFLAGS. * gdb.threads/non-ldr-exc-2.exp: Likewise. * gdb.threads/non-ldr-exc-3.exp: Likewise. * gdb.threads/non-ldr-exc-4.exp: Likewise.
2015-09-11 20:06:02 +02:00
# Follow parent and follow child vfork tests with a child that execs.
with_test_prefix "exec" {
# These are tests of gdb's ability to follow the parent of a Unix
# vfork system call. The child will subsequently call a variant
# of the Unix exec system call.
do_vfork_and_follow_parent_tests
# These are tests of gdb's ability to follow the child of a Unix
# vfork system call. The child will subsequently call a variant
# of a Unix exec system call.
#
do_vfork_and_follow_child_tests_exec
}
# Switch to test the case of the child exiting. We can't use
# standard_testfile here because we don't want to overwrite the binary
# of the previous tests.
set testfile "foll-vfork-exit"
set srcfile ${testfile}.c
set binfile [standard_output_file ${testfile}]
if {[build_executable $testfile.exp $testfile $srcfile] == -1} {
untested "failed to build $testfile"
return
}
# Follow parent and follow child vfork tests with a child that exits.
with_test_prefix "exit" {
# These are tests of gdb's ability to follow the parent of a Unix
# vfork system call. The child will subsequently exit.
do_vfork_and_follow_parent_tests
# These are tests of gdb's ability to follow the child of a Unix
# vfork system call. The child will subsequently exit.
#
do_vfork_and_follow_child_tests_exit
}
1999-06-29 01:04:32 +02:00
set timeout $oldtimeout
return 0