Re: [RFC] Don't allow setting register in non-innermost frame

[Yao Qi <yao@codesourcery.com>]

On 08/21/2012 04:19 AM, Doug Evans wrote:
> fwiw, I've been able to work around corrupt, or otherwise not
> completely useful, core files by setting registers in non-innermost
> frames.
> Granted, I knew what was happening underneath the covers, so to speak,
> and I could have done things differently, but I like this capability.
>

Yeah, it is convenient, but sometimes it is misleading.  People who are 
clear on what is happening underneath can modify the frame/memory 
directly where the registers are saved.

> If gdb had started out disallowing changing registers in non-innermost
> frames, we mightn't be thinking of restricting it now.
> "It's easier to relax restrictions than it is to impose them after the fact."

Even if the fact is not clear or misleading sometimes?

> I'd like to hear more justification for this change.

Unfortunately, I don't have extra justification here.  I am writing a 
new MI notification when register is modified by 'set $reg=FOO' in 
console, similar to the patches on 'command parameter change' 
notification.  When the register is modified in innermost frame, a MI 
notification (with frame info) can be sent to MI frontend, and frontend 
can update its register contents, such as "register view".  However, if 
register is modified in non-innermost frame, the notification is not 
useful to frontend.  Then, I started to think of the meaning of 'setting 
register in non-innermost frame', and proposed this change finally.

-- 
Yao