This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: gcc 3.3 garbage collector defaults
- From: Matt Austern <austern at apple dot com>
- To: Zack Weinberg <zack at codesourcery dot com>
- Cc: Ziemowit Laski <zlaski at apple dot com>, Neil Booth <neil at daikokuya dot co dot uk>, Benjamin Kosnik <bkoz at redhat dot com>, Andi Kleen <ak at suse dot de>, gcc at gcc dot gnu dot org, libstdc++ at gcc dot gnu dot org
- Date: Mon, 27 Jan 2003 16:48:58 -0800
- Subject: Re: gcc 3.3 garbage collector defaults
On Monday, January 27, 2003, at 04:37 PM, Zack Weinberg wrote:
cvs annotate says that the #defines in 3.2.x have not changed since
version 1.1 of ggc-page.c. It was my intent, when I replaced the
#defines with parameters, not to change the values.
What I suspect was either that these settings were chosen without much
testing, or that they were good at the time they were chosen, but are
not so good for newer machines and/or newer compiler's allocation
thresholds.
That's quite possible, but I hope we can focus on the things
that *have* changed between 3.2 and 3.3.
When I compare 3.2 and 3.3 on the same file and the same
machine, 3.3 is much slower than 3.2 and a good part of the
extra time, according to -ftime-report, is from the gc. If
those measurements are right (and I think they probably are,
since several other people report similar results), then
there aren't a whole lot of possibilities for what caused
the regression. We now know it wasn't the gc parameters.
What possibilities haven't we ruled out yet?
--Matt