mainline "exploding"

Gerald Pfeifer gerald@pfeifer.com
Wed Oct 15 00:31:00 GMT 2003


On Tue, 14 Oct 2003, Joe Buck wrote:
> Certainly if we have two inlining algorithms we have to keep maintaining,
> tuning, and bug-fixing two inlining algorithms.  Making the poorer one go
> away seems advisable.

Definitely, from an engineering point of view.

> I might have been confused, but the main argument I heard against using
> unit-at-a-time all the time was memory, so if we actually need more
> memory to use the old algorithm in many cases, that kills that.

The case where we (currently) do not use unit-at-a-time has regressed
extremely, that's why I found it more expensive.  Until recently, this
case was much lighter, but still perhaps unit-at-a-time plus some lower
threshold is a good idea.

Gerald
-- 
Gerald Pfeifer (Jerry)   gerald@pfeifer.com   http://www.pfeifer.com/gerald/



More information about the Gcc mailing list