Bug 217831
| Summary: | [patch] science/avogadro: fix for pkg-plist | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | Ports & Packages | Reporter: | Kenneth Salerno <kennethsalerno> | ||||||
| Component: | Individual Port(s) | Assignee: | Dmitry Marakasov <amdmi3> | ||||||
| Status: | Closed Not A Bug | ||||||||
| Severity: | Affects Only Me | CC: | me, tcberner | ||||||
| Priority: | --- | Keywords: | patch | ||||||
| Version: | Latest | ||||||||
| Hardware: | i386 | ||||||||
| OS: | Any | ||||||||
| Attachments: |
|
||||||||
|
Description
Kenneth Salerno
2017-03-16 10:25:58 UTC
Cannot repodiduce. It builds cleanly both with PYTHON enabled and disabled. (In reply to Dmitry Marakasov from comment #1) Can you try building with poudriere? I used poudriere to create the package and consistently reproduce this problem. Created attachment 183671 [details]
build log on 11.0-RELEASE/amd64
For 1 month, I now see the same error on 11.0-RELEASE/amd64. Lacking a poudriere setup, I installed only the build dependencies into a chroot and started the build:
pkg-static add pkg-1.10.1.txz
pkg add openbabel-2.4.1.txz boost-python-libs-1.64.0.txz libGLU-9.0.0_3.txz qt4-gui-4.8.7_3.txz qt4-network-4.8.7_3.txz qt4-opengl-4.8.7_3.txz eigen2-2.0.17,1.txz py27-numpy-1.11.2_3,1.txz py27-sip-4.19.2,1.txz cmake-3.8.2.txz desktop-file-utils-0.23.txz pkgconf-1.3.7,1.txz python27-2.7.13_4.txz python2-2_3.txz mesa-libs-17.1.2.txz glproto-1.4.17.txz dri2proto-2.8.txz dri3proto-1.0.txz glproto-1.4.17.txz qt4-linguisttools-4.8.7_2.txz qt4-moc-4.8.7_2.txz qt4-qmake-4.8.7_2.txz qt4-rcc-4.8.7_1.txz qt4-uic-4.8.7_1.txz
make -C /usr/ports/science/avogadro/ install 2>&1 | tee avogadro-build_install-20170621-1
-- [4/5] Numpy Module
Traceback (most recent call last):
File "<string>", line 1, in <module>
File "/usr/local/lib/python2.7/site-packages/numpy/__init__.py", line 142, in <module>
from . import add_newdocs
File "/usr/local/lib/python2.7/site-packages/numpy/add_newdocs.py", line 13, in <module>
from numpy.lib import add_newdoc
File "/usr/local/lib/python2.7/site-packages/numpy/lib/__init__.py", line 8, in <module>
from .type_check import *
File "/usr/local/lib/python2.7/site-packages/numpy/lib/type_check.py", line 11, in <module>
import numpy.core.numeric as _nx
File "/usr/local/lib/python2.7/site-packages/numpy/core/__init__.py", line 14, in <module>
from . import multiarray
ImportError: /lib/libgcc_s.so.1: version GCC_4.6.0 required by /usr/local/lib/gcc5/libgfortran.so.3 not found
-- Numpy NOT found - Python support disabled.
CMake Warning at CMakeLists.txt:281 (message):
Not all python dependencies are found - Python support diabled
This part in your log looks suspicious.
(In reply to Tobias C. Berner from comment #4) Yeah, that was pretty obvious. I was thinking in a wrong direction and it took me some rebuilding of packages to find the problem: My libblas.so still had an RPATH of /usr/local/lib/gcc48. Rebuilding blas solved the problem. Moreover, I found 3 libraries from avidemux-qt4 still having in RPATH containing gcc49 looking for similar problems with: for L in /usr/local/lib/*.so* ; do readelf -d "$L" 2>/dev/null | grep -q 'RPATH.*gcc4' && pkg which "$L" && readelf -d "$L" | grep gcc4 ; done I guess when the default gcc version is changed, more ports need a PORTREVISION bump -- all that might have gcc in RPATH somewhere. Thanks, Jan Henrik So may this PR be closed then? (In reply to Dmitry Marakasov from comment #6) From my perspective, the problem is solved. On my system, it was not in avogadro, but in blas. By now, I guess I had upgraded gcc improperly. Without a further comment from Kenneth or his build logs, it is not completely clear, whether or not he had the same problem. Anyhow, the pkg-plist seems to be correct, if all dependencies are freshly build. Oki, I'll close this then. Kenneth, if the problem reoccurs, feel free to reopen it. woops, I had my wires crossed, and thought this was a kde@ port |
