GCC 3.3 compile speed regression - AN ANSWER

[Michael S. Zick]

On Tuesday 11 February 2003 12:59 pm, Robert Dewar wrote:
> > I read your reference to: $$$ as a suggestion of re-introducing the
> > concept of "direct remuneration for effort" into the solution.
>
> No that's not quite what I was saying. I was saying that a lot of the
> people who work on gcc are paid to do so by their companies. They will
> naturally tend to work on those things of greatest interest to their
> companies.
>
Understood.  Thanks.

Lets declare that aspect in the context of a defined solution...

"Pinging" the users on the speed subject gets a lot of direct response.

"Pinging" the list of active developers on the speed subject gets a lot of
both direct and implicit response.

My conclusion; A separate branch would not die from lack of interest
from either of those groups.

Here is where I see this fitting in with the objective of companies...

1) Give the problem/solution the recognition it deserves in the form of
a separate branch, solely for this specific purpose.

1.a) Now those companies have a project name (tag) to include as a
line item in their budget.

1.b) Now the people working for those companies have a project name
to include on their time-sheets (or whatever they happen to use
to track their time spent with their employer).

Now the time, work and effort spent on addressing this one problem
gets recognized and acknowledged rather than remaining an
un-acknowledged (and un-budgeted) sub-component of some
other portion of the GCC project.

Then we can learn if companies fail to support this effort due to
the apparent lack of interest you have noticed;
Or if that is just a current artifact of the problem not having a 
recognizable place in the organizational structure of the GCC project.

Mike