old-cross-binutils/gdb/testsuite/gdb.base/return.exp
Stephane Carrez d6f5fea1d2 * gdb.base/return2.exp: return of long long and double fails for
68HC11; don't execute these tests on that platform.
	* gdb.base/return.exp: Return of a double fails for 68hc11.
2001-07-16 18:49:43 +00:00

129 lines
4.5 KiB
Text

# Copyright (C) 1992, 1997, 2001 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 2 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, write to the Free Software
# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
# Please email any bugs, comments, and/or additions to this file to:
# bug-gdb@prep.ai.mit.edu
# This file was written by Jeff Law. (law@cs.utah.edu)
if $tracelevel then {
strace $tracelevel
}
set prms_id 0
set bug_id 0
set testfile "return"
set srcfile ${testfile}.c
set binfile ${objdir}/${subdir}/${testfile}
if { [gdb_compile "${srcdir}/${subdir}/${srcfile}" "${binfile}" executable {debug}] != "" } {
gdb_suppress_entire_file "Testcase compile failed, so all tests in this file will automatically fail."
}
proc return_tests { } {
global gdb_prompt
if { ! [ runto func1 ] } then { return 0 }
send_gdb "return\n"
gdb_expect {
-re "Make .* return now.*y or n. $" {
send_gdb "y\n"
exp_continue
}
-re "func1 ..;.*$gdb_prompt $" {
send_gdb "step\n"
exp_continue
}
-re ".*in main after func1.*$gdb_prompt $" { pass "simple return" }
-re "$gdb_prompt $" { fail "simple return" }
timeout { fail "(timeout) simple return" }
}
# Set breakpoints in other interesting functions.
gdb_test "break func2" "" "break func2"
gdb_test "break func3" "" "break func3"
gdb_test "continue" "return -5;" "continue to return of -5"
send_gdb "return 5\n"
gdb_expect {
-re "Make .* return now.*y or n. $" {
send_gdb "y\n"
exp_continue
}
-re ".*tmp2 = func2.*$gdb_prompt $" { }
-re "$gdb_prompt $" { fail "did not return (integer test)" }
timeout { fail "(timeout) did not return (integer test)" }
}
gdb_test "next" "tmp3 = func3.*" "next over call to func2"
gdb_test "p tmp2" ".* = 5" "correct value returned (integer test)"
gdb_test "continue" "return -5.0;" "continue to return of -5.0"
# Return of a double does not work for 68hc11 (need struct return
# in memory).
setup_xfail "m6811-*-*"
send_gdb "return 5.0\n"
gdb_expect {
-re "Make .* return now.*y or n. $" {
send_gdb "y\n"
exp_continue
}
-re ".*tmp3 = func3.*$gdb_prompt $" { }
-re "$gdb_prompt $" { fail "did not return (double test)" }
timeout { fail "(timeout) did not return (double test)" }
}
setup_xfail "m6811-*-*"
gdb_test "next" "printf.*" "next over call to func3"
# This test is going to fail on all i*86 systems using an i*87.
# When returning a floating point value from a function, all known
# compilers do this via a `fldl' instruction, which pushes the floating
# value on the i387 stack. This causes two problems:
# a) Most i*86 targets do not store (or cannot store, see comment in
# in i386v-nat.c:i386_register_u_addr) the floating point registers
# to the target.
# b) gdb would have to figure out if the `fldl' instruction (or variants
# of it) has already been executed. If not, it would have to simulate
# a push instruction, as it is not enough to write the register,
# the floating point `stack pointer' has to be updated too.
# Do not expect this to get fixed anytime soon.
# This test also fails for sparc Solaris 2.3 & 2.4, but passes under 2.5
# At the time the `next' is issued, the floating point unit for the
# process is not yet initialized, and the storing of the floating
# point value to the floating point return register is ignored.
# Xfail it for current versions that are known to fail. Presumably
# if some future version does initialize the floating point unit at
# process start, making this test pass, it will be for a version that
# is not xfailed.
setup_xfail "i*86-*-*" "sparc-*-solaris2.3*" "sparc-*-solaris2.4*" "m6811-*-*"
gdb_test "p tmp3" ".* = 5.*" "correct value returned double test (known problem with i*86 and sparc solaris"
}
# Start with a fresh gdb.
gdb_exit
gdb_start
gdb_reinitialize_dir $srcdir/$subdir
gdb_load ${binfile}
set timeout 30
return_tests