Memory leak while compiling boost
Hi! I tried to compile the boost lib from source with $ bjam "-sTOOLS=gcc" but while compiling the process "cc1plus" tends to consume memory quite fast, finally bringing the machine to a halt. Environment: SuSE 8.2, Kernel 2.4.20, all updates gcc (GCC) 3.3 20030226 (prerelease) (SuSE Linux) boost-jam 3.1.10 (also made from source) boost 1.32.0 My fault or a boost/bjam problem? Thanks in advance! Martin
On Wed, 29 Dec 2004 12:05:00 +0100, Martin Gebert
Hi!
I tried to compile the boost lib from source with
$ bjam "-sTOOLS=gcc"
but while compiling the process "cc1plus" tends to consume memory quite fast, finally bringing the machine to a halt. Environment:
SuSE 8.2, Kernel 2.4.20, all updates gcc (GCC) 3.3 20030226 (prerelease) (SuSE Linux) boost-jam 3.1.10 (also made from source) boost 1.32.0
My fault or a boost/bjam problem? Thanks in advance!
There is a known issue on MacOS X compiling one of the files in the serialization library in release mode (xml_grammar.cpp I believe). Perhaps you're hitting the same bug on this platform. See: http://thread.gmane.org/gmane.comp.lib.boost.user/7978 -- Caleb Epstein caleb dot epstein at gmail dot com
Caleb Epstein schrieb:
There is a known issue on MacOS X compiling one of the files in the serialization library in release mode (xml_grammar.cpp I believe). Perhaps you're hitting the same bug on this platform.
See:
Sounds familiar. I'll try the workaround suggested there tomorrow. Thanks for the hint! If there are boost developers which are interested in finding the cause of this problem, just contact me if you need further information. Martin
On Wed, 29 Dec 2004 23:11:50 +0100, Martin Gebert > Sounds familiar. I'll try the workaround suggested there tomorrow. Thanks for the hint!
If there are boost developers which are interested in finding the cause of this problem, just contact me if you need further information.
It is a bug in a pre-release version of GCC that should not be used in any production systems. I doubt Robert is interested in anything other than a simple work-around. -- Caleb Epstein caleb dot epstein at gmail dot com
Martin Gebert schrieb:
See:
The workaround suggested there doesn't help; I get the same behaviour when building --with-regex only. I think I'll go with the RPM version for the time being. Martin
participants (2)
-
Caleb Epstein -
Martin Gebert