Re: [PRecord] Memory gets set even if users responds 'n' to query

[Hui Zhu <teawater at gmail dot com>]

On Tue, Jul 21, 2009 at 10:15, Marc Khouzam<marc.khouzam@ericsson.com> wrote:
> Hi,
>
> while looking into the query stuff we have been discussing I ran into
> the weird behavior where PRecord sets memory even if I answer 'n' to the
> query. ?I've done a little troubleshooting and it seems that the query
> code is not the problem. ?So, I thought that maybe even though PRecord
> returns
> from its method right away when I say 'n' to the query, the memory
> has already been changed. ?Could that be? ?I didn't investigate further.
>
> Also, I didn't try it with changing registers.
>
> I'm pretty sure it used to work properly.
>
> Thanks
>
> Marc
>
> Here is the session:
>
> GNU gdb (GDB) 6.8.50.20090720-cvs
> Copyright (C) 2009 Free Software Foundation, Inc.
> License GPLv3+: GNU GPL version 3 or later
> <http://gnu.org/licenses/gpl.html>
> This is free software: you are free to change and redistribute it.
> There is NO WARRANTY, to the extent permitted by law. ?Type "show
> copying"
> and "show warranty" for details.
> This GDB was configured as "i686-pc-linux-gnu".
> For bug reporting instructions, please see:
> <http://www.gnu.org/software/gdb/bugs/>...
> (gdb) l
> 1 ? ? ? int main() {
> 2 ? ? ? ? ? int a = 1;
> 3 ? ? ? ? ? int b = 10;
> 4
> 5 ? ? ? ? ? a++;
> 6 ? ? ? ? ? b++;
> 7
> 8 ? ? ? ? ? return a;
> 9 ? ? ? }
> 10
> (gdb) start
> Temporary breakpoint 1 at 0x80483f5: file b.cc, line 2.
> Starting program: /home/marc/testing/a.out
> re
> Temporary breakpoint 1, main () at b.cc:2
> 2 ? ? ? ? ? int a = 1;
> (gdb) record
> (gdb) n
> 3 ? ? ? ? ? int b = 10;
> (gdb) n
> 5 ? ? ? ? ? a++;
> (gdb) n
> 6 ? ? ? ? ? b++;
> (gdb) n
> 8 ? ? ? ? ? return a;
> (gdb) rn
> 6 ? ? ? ? ? b++;
> (gdb) p a
> $1 = 2
> (gdb) set var a = 8
> Because GDB is in replay mode, writing to memory will make the execution
> log unusable from this point onward. ?Write memory at address
> 0xbffff6b0?(y or [n]) n
> (gdb) p a
> $2 = 8
> (gdb)
>
>

Thanks Marc.

This issue is because when user answer n, function record_xfer_partial
return -1 will not really cancel the memory change operation.
I make a patch to change it to error.

And I have check the code about register, it works OK.  Because it
already use error.

Please help me review it.

Thanks,
Hui

2009-07-21  Hui Zhu  <teawater@gmail.com>
	* record.c (record_xfer_partial): Call error When nquery
	return "n".
---
 record.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/record.c
+++ b/record.c
@@ -997,7 +997,7 @@ record_xfer_partial (struct target_ops *
 		         "will make the execution log unusable from this "
 		         "point onward.  Write memory at address %s?"),
 		       paddress (target_gdbarch, offset)))
-	    return -1;
+	    error (_("Process record canceled the operation."));

 	  /* Destroy the record from here forward.  */
 	  record_list_release_next ();

Attachment: reverse-fix-change-memory.txt
Description: Text document