This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

RE: [patch] do not always push DImode constants to memory before loading them to registers on ARM


> > So I suspect that on balance, the optimum is likely to be around 3
> > instructions regardless of the optimization level, except 
> when there are
> > no load delay slots (arm7 or earlier) when it should be 4.
> 
> OK, what about changing the test to:
> 
>   return (  arm_gen_constant (SET, SImode, NULL_RTX, parts[0],
>                               NULL_RTX, NULL_RTX, 0, 0)
>           + arm_gen_constant (SET, SImode, NULL_RTX, parts[1],
>                               NULL_RTX, NULL_RTX, 0, 0)
>           > ((optimize_size || (tune_flags & FL_LDSCHED)) ? 3 : 4));
> 
> then?
> 

That would be great.  Thanks.

Consider the patch pre-approved.

R.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]