This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Kudos to the gcc developers
- To: gcc at gcc dot gnu dot org
- Subject: Kudos to the gcc developers
- From: Brad Lucier <lucier at math dot purdue dot edu>
- Date: Thu, 30 Nov 2000 11:45:25 -0500 (EST)
- Cc: lucier at math dot purdue dot edu, feeley at iro dot umontreal dot ca, staff at math dot purdue dot edu, wilker at math dot purdue dot edu, matzmich at cs dot tu-berlin dot de
Two recent incidents really emphasized to me how much improved the
current CVS sources for gcc are over past releases:
1. After Michael Matz's dominators calculation was integrated into
gcc, I ran a profiled version to see what the next significant "hot
spot" was in the compiler. I couldn't find one. Right now, -O2 is
just about as fast as -O1 and does not take very much more memory.
(Which, unfortunately, begs the question whether -O1 should be
significantly faster and take significantly less memory, but let's
leave that one for now ;-).
2. After getting nowhere trying to diagnose the gcse problem that's
biting me, I decided to try to track down the other code generation
bugs that I was hitting a few weeks ago. They're gone; recent code
generation changes have fixed them. Right now, the only problem I have
with the CVS sources is with gcse. (BTW, perhaps someone can answer
the questions in
http://gcc.gnu.org/ml/gcc-prs/2000-q4/msg00590.html
"Yes" or "no" should suffice for most of them.)
So, for the parts of the compiler that I exercise (mainly straight C
code generation with various levels of optimization on really large
routines), things look very, very good. And that doesn't even take
into account the improvements in C++, etc.!
Brad