Re: [PATCH] fixed the memory leak in record.c

[Hui Zhu <teawater@gmail.com>]

On Fri, Aug 28, 2009 at 09:45, Michael Snyder<msnyder@vmware.com> wrote:
> Hui Zhu wrote:
>>
>> On Thu, Aug 27, 2009 at 13:35, Hui Zhu<teawater@gmail.com> wrote:
>>>
>>> On Thu, Aug 27, 2009 at 11:11, Jiang Jilin<freephp@gmail.com> wrote:
>>>>
>>>>       * record.c (record_list_release_next) : fixed memory leak when
>>>> record type is record_reg
>>>>
>>>> Signed-off-by: Jiang Jilin <freephp@gmail.com>
>>>> ---
>>>>  gdb/record.c |    7 ++++---
>>>>  1 files changed, 4 insertions(+), 3 deletions(-)
>>>>
>>>> diff --git a/gdb/record.c b/gdb/record.c
>>>> index 8afca6b..07e9e80 100644
>>>> --- a/gdb/record.c
>>>> +++ b/gdb/record.c
>>>> @@ -158,9 +158,10 @@ record_list_release_next (void)
>>>>    {
>>>>      rec = tmp->next;
>>>>      if (tmp->type == record_reg)
>>>> -       record_insn_num--;
>>>> -      else if (tmp->type == record_reg)
>>>> -       xfree (tmp->u.reg.val);
>>>> +       {
>>>> +         record_insn_num--;
>>>> +         xfree (tmp->u.reg.val);
>>>> +       }
>>>>      else if (tmp->type == record_mem)
>>>>       xfree (tmp->u.mem.val);
>>>>      xfree (tmp);
>>>> --
>>>> 1.5.4.3
>>>>
>>>>
>>
>> Hi Michael,
>>
>> Could you please help me review this patch?
>
> Yes, I think it looks fine.  Please commit.

Checked in.  Thank you, Michael and Jilin.

Hui


>
>> Following is my patch:
>>
>>> Cool.  Thanks for find it out.  :)
>>> But this patch is not OK.  This record_reg should be record_end.
>>>
>>> I make a patch for it.
>>>
>>> Thanks,
>>> Hui
>>>
>>> 2009-08-27  Hui Zhu  <teawater@gmail.com>
>>>
>>>       * record.c (record_list_release_next): Change the first
>>>       record_reg to record_end.
>>>
>> ---
>>  record.c |    2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> --- a/record.c
>> +++ b/record.c
>> @@ -157,7 +157,7 @@ record_list_release_next (void)
>>   while (tmp)
>>     {
>>       rec = tmp->next;
>> -      if (tmp->type == record_reg)
>> +      if (tmp->type == record_end)
>>        record_insn_num--;
>>       else if (tmp->type == record_reg)
>>        xfree (tmp->u.reg.val);
>
>