This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [ping] Re: PATCH: Disable sched1 when optimizing for size
Dave Korn wrote:
> Richard Earnshaw wrote:
>
>>> Nevertheless, for both ARM and Thumb-2 there's a significant penalty to
>>> running this pass in terms of code size (0.3% and 1.3% respectively);
>
> What about all the other platforms? Unless I'm misreading your patch it
> disables the pass for them too; can you be sure it's not a win, possibly even
> a vitally necessary one, on other targets?
I'm getting a real sense of deja vu. Haven't we had this discussion several
times before, over the past decade or two? My recollection is that early
scheduling is always a loss on register starved processors, and that this
was well-known. However, I can's find the discussion in the archives. I
guess I may have imagined it all.
Andrew.