This is the mail archive of the 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] PR39077

On Sun, Feb 21, 2010 at 12:51 PM, Richard Guenther
<> wrote:
> Isn't the default of 20 incoming edges unnecessarily high? ?Or did you
> just want to avoid extreme cases like in the PR?

We just want to avoid the extreme cases. This situation is highly
unusual anyway.

> I suppose doing the check at insertion time isn't possible? (Well, I see
> it doesn't fit easily into the existing code that way).

Sadly no.In LCM you compute the insert points in advance, and you
can't decide *not* to insert afterwards. Imagine an incoming critical
edge, and you decide not to insert. If you have to insert on another
successor edge of the predecessor, you still end up inserting new
computations, but you don't eliminate the redundancy. Or worse: you
don't insert but you have to insert to make a computation redundant on
another path.

> The patch is ok with me if Jeff is fine with it.

Thanks. I think Jeff is fine with it (he said he'd OK a patch like
this in the bug audit trail) but I'll wait a bit to give him the
chance to have a fresh look.


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