This is the mail archive of the gcc@gcc.gnu.org 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: Faster compilation speed


David Edelsohn wrote:

	Here's an interesting (aka depressing) data point.  My previous
cache miss statistics were for GCC -O2.  At -O0, GCC's cache miss
statistics stay the same or get up to 20% *worse*.  In comparison, the
cache statistics for IBM's compiler without optimization enabled *improve*
up to 50 for the same reload.c and insn-recog.c input files compared to
optimized.

	GCC has some sort of overhead, maybe the tree->RTL conversion as
Dan mentioned, which really hurts re-use at -O0.

David

Doesn't the garbage collector have to periodically scan all nodes,
which would mean that it trashes the cache each time it runs?

Is there some way to run these statistics while keeping the
garbage collector from running? For a test run on a reasonably
large machine and reasonably small program, you should be
able to get by without freeing the memory at all.  It would tell
you if GC was the problem with the cache, or if you should
look elsewhere.




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