Re: [reverse RFA] no singlestep-over-BP in reverse

[Michael Snyder <msnyder@vmware.com>]

[-- Attachment #1: Type: text/plain, Size: 969 bytes --]

Daniel Jacobowitz wrote:
> On Mon, Sep 15, 2008 at 11:31:33AM -0700, Michael Snyder wrote:
>> When we're stopped at a breakpoint and we want to
>> continue in reverse, we're not actually going to
>> execute the instruction at the breakpoint -- we're
>> going to de-execute the previous instruction.
>>
>> Therefore there's no need to singlestep before
>> inserting breakpoints.  In fact it would be a bad
>> idea to do so, because if there is a breakpoint at
>> the previous instruction, we WANT to hit it.
>>
>> Note that this patch is to be applied to the reverse branch.
> 
> If there is a breakpoint on the previous instruction, will you hit it
> before or after de-executing that instruction?  It seems like this
> logic should be somehow still necessary... but I can't put my finger
> on when.

The attached test does just that -- and it passes on three
targets that can go in reverse (teawater's, gdb-freeplay,
and the as yet unreleased VMware implementation).


[-- Attachment #2: consecutive-reverse.exp --]
[-- Type: text/plain, Size: 3284 bytes --]

#   Copyright 2001, 2007, 2008 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/>.

# Please email any bugs, comments, and/or additions to this file to:
# bug-gdb@prep.ai.mit.edu

# Based on a file written by Michael Snyder. (msnyder@redhat.com)

#
# Test breakpoints at consecutive instruction addresses.
#

set testfile "consecutive-reverse"
set srcfile ${testfile}.c
set binfile ${objdir}/${subdir}/${testfile}

if { [prepare_for_testing ${testfile}.exp $testfile $srcfile] } {
    untested "${testfile}.exp"
    return -1
}

runto main

# FIXME 'record' command should say something so we know it's working.
gdb_test "record" "" "start recording"

gdb_breakpoint foo
gdb_test "continue" "Breakpoint $decimal, foo .*" \
	"continue to breakpoint in foo"

set foo1_addr 0
set foo2_addr 0
set stop_addr 0

send_gdb "x /2i \$pc\n"
gdb_expect {
    global hex
    global foo1_addr
    global foo2_addr
    global gdb_prompt

    -re "($hex).*\[\r\n\]+($hex).*$gdb_prompt $" {
	set foo1_addr $expect_out(1,string)
	set foo2_addr $expect_out(2,string)
	pass "get breakpoint address for foo"
    }
    -re ".*$gdb_prompt $" {
	fail "get breakpoint address for foo"
	return 0;
    }
    timeout {
	fail "get breakpoint address for foo (timeout)"
	return 0;
    }
}

gdb_test "break \*$foo2_addr" "Breakpoint $decimal at $foo2_addr: file .*" \
	"set bp, 2nd instr"

send_gdb "step\n"
gdb_expect {
    -re "Breakpoint $decimal, ($hex) in foo.*$gdb_prompt $" {
	set stop_addr $expect_out(1,string)
	if [eval expr "$foo2_addr == $stop_addr"] then {
	    pass "stopped at bp, 2nd instr"
	} else {
	    fail "stopped at bp, 2nd instr (wrong address)"
	}
    }
    -re ".*$gdb_prompt $" {
	fail "stopped at bp, 2nd instr"
    }
    timeout {
	fail "stopped at bp, 2nd instr (timeout)"

    }
}

###
###
###

# Set reverse execution direction
# FIXME: command needs to acknowledge, so we can test if it succeeded.

gdb_test "set exec-dir reverse" "" "set reverse execution"

# Now step backward and hope to hit the first breakpoint.

set test_msg "stopped at bp in reverse, 1st instr"
gdb_test_multiple "step" "$test_msg" {
    -re "Breakpoint $decimal, ($hex) in foo.*$gdb_prompt $" {
	set stop_addr $expect_out(1,string)
	if [eval expr "$foo1_addr == $stop_addr"] then {
	    pass "$test_msg"
	} else {
	    fail "$test_msg (wrong address)"
	}
    }
    -re "Breakpoint $decimal, foo.*$gdb_prompt $" {
	send_gdb "print \$pc == $foo1_addr\n"
	gdb_expect {
	    -re "$decimal = 1\[\r\n\]+$gdb_prompt $" {
		pass "$test_msg"
	    }
	    -re "$decimal = 0\[\r\n\]+$gdb_prompt $" {
		fail "$test_msg (wrong address)"
	    }
	}
    }
    -re ".*$gdb_prompt $" {
	fail "$test_msg"
    }
}