A status report as of 13:30 US Pacific time.
Note that all the bugs I reported last night appeared in BOTH 1.3.1 and
1.4 (despite only 1.3.1 in the initial subject lines of each thread).
With the tarball which Jeff Squyres provided off-list I've verified that
the bugs in the following threads appear FIXED.
I have no clue which branches the fixes have been applied to.
+ hwloc-1.3.1 build failure w/ Solaris Studio compilers
http://www.open-mpi.org/community/lists/hwloc-devel/2012/01/2728.php
Brice added a required "-lm" in the proper place
+ hwloc-1.3.1 install failure w/ older GNU sed
http://www.open-mpi.org/community/lists/hwloc-devel/2012/01/2730.php
Brice rewrote the install hook not to use sed's "-i" option
+ hwloc-1.3.1 install failures on Solaris-10
http://www.open-mpi.org/community/lists/hwloc-devel/2012/01/2731.php
Brice and Jeff rewrote docs/Makefile.am to avoid over-length command lines.
The following two remain, as far as I know, UNRESOLVED:
+ hwloc-1.3.1 "gmake check" failure on Solaris-10/SPARC/gccfss
http://www.open-mpi.org/community/lists/hwloc-devel/2012/01/2729.php
For this compiler the ffs() implementation is missing from libgcc
(https://forums.oracle.com/forums/thread.jspa?threadID=1997328)
Hwloc could either:
1) Detect this compiler at configure time and modify misc.h to pretend
ffs() is missing
2) Document this compiler as broken/unsupported
+ hwloc-1.3.1 assertion failures on Linux/POWER7
http://www.open-mpi.org/community/lists/hwloc-devel/2012/01/2733.php
Brice has requested more info off-list, which I have provided.
There are actually 2 issues in that thread:
1) XLC appears to be doing bad things w/ the set_affinity code (also
seen by Chris Samuel).
2) Assertion failures, possibly related to IBM's virtualization, and
present w/ both gcc and xlc.
The following two I am NOT expecting fixes for:
+ hwloc-1.3.1 libtool broken on Solaris10-x86
http://www.open-mpi.org/community/lists/hwloc-devel/2012/01/2732.php
Turned out to be a false-alarm.
libtool does the right thing when the error occurs, but doesn't suppress
the error message
+ hwloc-1.3.1 and 1.4 failures on MIPS64/qemu
http://www.open-mpi.org/community/lists/hwloc-devel/2012/01/2747.php
Looks very likely to be a compiler bug, or possibly a qemu bug.
I'll provide more info IF I can, but this is very low on my list of
priorities.
-Paul
--
Paul H. Hargrove phhargr...@lbl.gov
Future Technologies Group
HPC Research Department Tel: +1-510-495-2352
Lawrence Berkeley National Laboratory Fax: +1-510-486-6900
