[PyKDE] PyKDE building and the -j flag (again)
Torsten Marek
shlomme at gmx.net
Fri Apr 14 22:54:09 BST 2006
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Danny Pansters wrote:
> Hi,
>
> I'm maintaining the FreeBSD ports for sip/pyqt/pykde. I'm once again having
> problems with the -j flag, but this time not across archs but for i386. On my
> box it works fine, on the FreeBSD build cluster it doesn't. For clarity: I'm
> dependent on the package build runs on the cluster to get feedback on archs I
> dont have, that's everything non-i386 ATM.
>
> Now, I understand what the -j flag does and obviously prefer j=1 where
> possible. In the past I've seen breakage on sparc64 and settled for j=3
> there, and j=2 for other non-i386. This seemed to work, because for a while I
> didn't get any reports from "pointyhat", the ports build cluster. But now I
> get a report that I can't reproduce on my own i386 (P4) box:
>
> c++ -c -Wno-deprecated-declarations -pipe -fPIC -O -pipe -Wall -W
> -DQT_NO_DEBUG -DQT_THREAD_SUPPORT -I. -I../extra/kde350 -I/usr/local/include
> -I/usr/local/include/python2.4 -I/usr/X11R6/include -o sipkdecorepart0.o
> sipkdecorepart0.cpp
> /usr/local/include/ktimezones.h: In function `PyObject*
> convertFrom_ZoneMap(void*)':
> /usr/local/include/ktimezones.h:191: error: `KTimezone::KTimezone(const
> KTimezone&)' is private
> sip/kdecore/ktimezones.sip:209: error: within this context
> /usr/local/include/ktimezones.h: In function `void*
> init_KTimezones(sipWrapper*, PyObject*, sipWrapper**)':
> /usr/local/include/ktimezones.h:340: error: `KTimezones::KTimezones(const
> KTimezones&)' is private
> sipkdecorepart0.cpp:9497: error: within this context
> /usr/local/include/ktimezones.h: In function `void*
> init_KTimezone(sipWrapper*, PyObject*, sipWrapper**)':
> /usr/local/include/ktimezones.h:191: error: `KTimezone::KTimezone(const
> KTimezone&)' is private
> sipkdecorepart0.cpp:10450: error: within this context
> sip/kdecore/kmountpoint.sip: In function `PyObject*
> convertFrom_KMountPoint_List(void*)':
> sip/kdecore/kmountpoint.sip:151: warning: taking address of temporary
> sipkdecorepart0.cpp: At global scope:
> sipkdecorepart0.cpp:34440: warning: unused parameter 'sipPy'
> sipkdecorepart0.cpp:34440: warning: unused parameter 'sipIsErr'
> sipkdecorepart0.cpp: In function `PyObject* convertFrom_Display(void*)':
> sipkdecorepart0.cpp:34478: warning: unused variable 'sipCpp'
> sip/kdecore/kconfigbase.sip: In function `PyObject*
> convertFrom_ulonglong(void*)':
> sip/kdecore/kconfigbase.sip:307: warning: unused variable 'LongLong'
> sip/kdecore/kwinmodule.sip: In function `PyObject* convertFrom_QValueList_2100
> (void*)':
> sip/kdecore/kwinmodule.sip:111: warning: unused variable 'inst'
> *** Error code 1
>
> On my box it builds fine (updated to latest OS code before, this is FreeBSD6,
> but got same report for 5). In other (earlier on other archs) reports I'd see
> the same thing, not in the same place, indicating some system limit got hit.
> I realize that the warnings have nothing to do with the actual build error.
>
> So if the number of "chunks" (-j) matters that much, I figure that certain
> system limits are dependent of system resources like RAM.
>
> Is this correct? Any suggestions on how I can prevent these problems (maybe
> just set j sufficiently high for all archs and FreeBSD versions)? How do
> other (source) packagers deal with this?
>
> Thanks very much for any guidance. The fbsd ports people either don't know or
> don't care, I've asked about this there several times. I need to fix this in
> some sustainable way.
>
Hi,
just as a quick guess, are the versions of sip the same?
As far as system resources are concerned, g++ 4.0 (the C++ compiler I assume you
are using) takes up huge amounts of memory (several hundreds of megabytes) when
compiling anything sip-related (at least on my system...), and I use -j4 for all
Debian builds by default right now.
best regards
Torsten
- --
Torsten Marek <shlomme at gmx.net>
ID: A244C858 -- FP: 1902 0002 5DFC 856B F146 894C 7CC5 451E A244 C858
Keyserver: subkeys.pgp.net
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFEQBoBfMVFHqJEyFgRArz+AJ4xz2oQXXl2NDz/rP3Ch9KCmMurIACfWYLP
9NgmAECTvUoY7Ugf3DzOgDU=
=2a9a
-----END PGP SIGNATURE-----
More information about the PyQt
mailing list