Date: Thu, 12 Feb 2009 11:00:12 GMT From: Mikolaj Golub <to.my.trociny@gmail.com> To: freebsd-bugs@FreeBSD.org Subject: Re: misc/131597: c++ exceptions very slow on FreeBSD 7.1/amd64 Message-ID: <200902121100.n1CB0Cps074819@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
The following reply was made to PR misc/131597; it has been noted by GNATS.
From: Mikolaj Golub <to.my.trociny@gmail.com>
To: bug-followup@FreeBSD.org
Cc: Guillaume Morin <guillaume@morinfr.org>
Subject: Re: misc/131597: c++ exceptions very slow on FreeBSD 7.1/amd64
Date: Thu, 12 Feb 2009 12:25:18 +0200
It looks like 6.x is not affected by this problem (or at least the situation
is much better here). I have checked on several 6.x hosts and the output looks
like this:
0.55 real 0.55 user 0.00 sys
1000 maximum resident set size
4 average shared memory size
33 average unshared data size
133 average unshared stack size
98 page reclaims
0 page faults
0 swaps
0 block input operations
0 block output operations
0 messages sent
0 messages received
0 signals received
1 voluntary context switches
48 involuntary context switches
In all cases sys time was zero. While on my 7.1-STABLE i386 (Core(TM)2 CPU 6400 @ 2.13GHz) box :
1.93 real 0.58 user 1.34 sys
1252 maximum resident set size
4 average shared memory size
732 average unshared data size
128 average unshared stack size
115 page reclaims
0 page faults
0 swaps
0 block input operations
0 block output operations
0 messages sent
0 messages received
0 signals received
1 voluntary context switches
32 involuntary context switches
or 7.1-RELEASE-p1 amd64 (Xeon(R) CPU E5450 @ 3.00GHz):
3.93 real 3.80 user 0.12 sys
1736 maximum resident set size
4 average shared memory size
1024 average unshared data size
128 average unshared stack size
135 page reclaims
0 page faults
0 swaps
0 block input operations
0 block output operations
0 messages sent
0 messages received
0 signals received
1 voluntary context switches
41 involuntary context switches
I have tried g++ both of v3.4 and v4.2 and hasn't noticed significant
difference, in my tests 3.4 was a bit faster but it might be influence of
external factors (host load).
--
Mikolaj Golub
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200902121100.n1CB0Cps074819>