Re: [Qemu-devel] [PATCH 04/13] tcg: sync output arguments on liveness request

[Aurelien Jarno]

On Thu, Sep 27, 2012 at 11:39:06AM -0700, Richard Henderson wrote:
> On 09/27/2012 10:15 AM, Aurelien Jarno wrote:
> > Synchronize an output argument when requested by the liveness analysis.
> > This is needed so that the temp can be declared dead later.
> > 
> > For that, add a new op_sync_args table in which each bit tells if the
> > corresponding output argument needs to be synchronized with the memory.
> > Pass it to the tcg_reg_alloc_* functions, and honor this bit. We need to
> > synchronize the argument before marking it as dead, and we have to make
> > sure all the infos about the temp are correctly filled.
> > 
> > At the same time change some types from unsigned int to uint16_t when
> > passing op_dead_args.
> > 
> > Signed-off-by: Aurelien Jarno <address@hidden>
> 
> If I understand the patch correctly, this patch alone is no change,
> since no sync_args bits are ever set.

They are set later in this series, this is a preparatory work.

> Which means that
> 
> > +        if (NEED_SYNC_ARG(i)) {
> > +            tcg_reg_sync(s, reg);
> > +        }
> > +        if (IS_DEAD_ARG(i)) {
> > +            temp_dead(s, args[i]);
> > +        }
> 
> Might ought to be better written as
> 
>   if (IS_DEAD_ARG(i)) {
>      ...
>   } else if (NEED_SYNC_ARG(i)) {
>      ...
>   }
> 
> as temp_dead implies sync from patches 2 and 3.  This pattern is also

No temp_dead doesn't imply sync. temp_dead is only marking the temp as
dead, it doesn' save it. 

> replicated 3-4 times.  Subroutine?
> 

Replicated in this patch, but only 2 are left at the end.

-- 
Aurelien Jarno                          GPG: 1024D/F1BCDB73
address@hidden                 http://www.aurel32.net