Re: [PATCH 2/2] Add options to skip unavailable locals

[Pedro Alves <palves at redhat dot com>]

On 08/01/2013 02:59 PM, Yao Qi wrote:
> On 08/01/2013 08:33 PM, Pedro Alves wrote:

>> I don't think this has been updated right for entry values.  With
>> entry values, we now have_two_  values to account for.  I think
>> we need to do this once for each of the regular arg and the
>> entry arg?
>>
> 
> OK, will update the patch to handle both regular arg and entry value.
> 
> Is it possible that one is available and the other is not?  

Sure.  The debug info just tells us where to fetch the entry
value from (registers, memory, etc.) just like the regular arg.
So it'll depend on which chunks of memory/registers have
been collected.

> If yes, 
> when option '--skip-unavailable' is used, some thing we have to think about,
> 
> 1. if regular arg is available but entry is unavailable, like this,
> 
>  
> {name="j",type="int",value="4"},{name="j@entry",type="int",value="<unavailable>"}
> 
> what is the expected output?  how about displaying regular arg only as 
> it is available? like:
> 
>    {name="j",type="int",value="4"}
> 
> 2. if entry is available but regular arg is not, like this,
> 
>  
> {name="j",type="int",value="<unavailable>"},{name="j@entry",type="int",value="4"}
> 
> what is the expected output? how about displaying entry value only as it 
> is available? like:
> 
>    {name="j@entry",type="int",value="4"}
> 

Yes, that's what I'd expect.

-- 
Pedro Alves