Re: PCH, and more generally C++ parser performance

[Mark Mitchell <mark at codesourcery dot com>]

>>>>> "Zack" == Zack Weinberg <zack@wolery.cumb.org> writes:

    >> - Sort the pointers when marking to improve locality.

    Zack> I tried to do this, misunderstood the way ggc_mark_trees
    Zack> works, and wound up calling qsort on every iteration.
    Zack> Obviously that was worse.

I tried it too, using radix-sort (which has better asymptotic
complexity that qsort, although not necessarily better wall-clock
time), but I couldn't get a win, even on a case that was causing heavy
paging.  I think I just blew it somehow -- it should help.

    >>  The language really doesn't let you do that.  You have to
    >> parse everything.  Example:
    >> 
    >> template <class T> struct S { static int i; };
    >> 
    >> template <class T> int S<T>::i = f();
    >> 
    >> inline void g () { S<int>.i = 7; }
    >> 
    >> This program requires you to instantiate `S<int>::i', even if
    >> you don't need `g'.

    Zack> I don't understand C++ well enough to know why.  In fact, I
    Zack> can't parse

    Zack> template <class T> int S<T>::i = f();

Oh.  That's called a "static data member".  Think of it as a global
variable, but inside a class scope.  There's only one such thing
(that's why it's global) -- as opposed to a field, or ordinary data
member, of which there is one per object.

    Zack> at all.  Note that in this example the only text I would
    Zack> defer processing would be the body of g -

    Zack> { S<int>.i = 7; }

The point is that the instantiation -- in the body of G -- causes this
static data member to come in to existence, and I believe the standard
requires that we call `f'.  (Which of course might have side-effects.)
Skip parsing `g', and you skip instantiating `S<int>::i', and you
therefore skip the side-effects in `f' ...

--
Mark Mitchell                   mark@codesourcery.com
CodeSourcery, LLC               http://www.codesourcery.com