OK, I finally figured out what's going on. I recently upgraded to Mac OS X Yosemite, and I did not install the Xcode command line tools. My /usr/include directory was missing, however I had /usr/bin/clang (not sure wether it is there because of the default Xcode install, or it was a leftover of my previous OS X system). I also had pkg-config installed trough Homebrew, and "pkg-config --cflags libxml-2.0" prints "-I/usr/include/libxml2", however remember I did not have "/usr/include". Do this pkg-config install is assuming /usr/include do exist. The thing is that MPICH is able to build fine (and in fact, many other software packages) as long as you pass --disable-libxml2 to workaround the hwloc issue. But after spending some time looking at hwloc build system, I realized the dependence on pkg-config to look for libxml2, and then there is no easy workaround. Finally, I've installed the Xcode command line tools (with "xcode-select --install") and MPICH + embed hwloc built just fine. Now I had the cmd line tools in my system, but it makes sense anyway, after all I use it everyday for development and the lack of it will likely cause me headaches in the near future with some other software. Some final comment for hwloc folks: * Relying in pkg-config for building is totally fine, but please note this tool is non-standard in OS X. * Looking in config/hwloc_pkg.m4, I noticed you somehow do check for broken stuff by attempting to link with package libraries. Well, if would be nice to extend this macro to also check you are able to compile by #include'ing package headers. IMHO, such sanity checks save time in the long run for both developer and users. Sorry I do not offer a patch, I have very limited knowledge of autotools and M4. On 7 March 2015 at 00:24, Brice Goglin <[email protected]> wrote:

Hello,

Sorry but we don't add configure checks without knowing what's causing the problem. So far it only looks like a broken libxml install.

We cannot check for all broken installs in the world. Otherwise one day somebody will remove printf from his libc and request a new configure check for printf() as well. And another check in case he modified printf() to return different values. And another check in case printf() was renamed to pruntf(). Endless.

Please try to understand what caused this broken/partial libxml install (do you have logs of the install?).

By the way, let's make this new hwloc feature official: hwloc can now detect broken libxml installs by failing to build :)

Brice

Le 06/03/2015 21:14, Balaji, Pavan a écrit :

...

Hi,

This is a problem with hwloc, which I had reported in the past. I believe this is not fixed in hwloc yet.

The suggestion given at that point was to remove and install libxml again, which I did, and things started working correctly again. But, I agree, hwloc should detect this in configure and abort if there's an issue.

I've cc'ed hwloc-devel.

-- Pavan

...

On Mar 6, 2015, at 2:10 AM, Lisandro Dalcin <[email protected]> wrote:

This is with the 3.1.4 tarball. Not sure if this is actually a problem in MPICH or in my system.

I got the following build failure. Shouldn't configure catch that beforehand?

Making all in tools/topo/hwloc/hwloc Making all in src CC topology.lo CC traversal.lo CC distances.lo CC components.lo CC bind.lo CC bitmap.lo CC pci-common.lo CC diff.lo CC misc.lo CC base64.lo CC topology-noos.lo CC topology-synthetic.lo CC topology-custom.lo CC topology-xml.lo CC topology-xml-nolibxml.lo CC topology-xml-libxml.lo CC topology-darwin.lo CC topology-x86.lo topology-xml-libxml.c:17:10: fatal error: 'libxml/parser.h' file not found #include <libxml/parser.h> ^ 1 error generated. make[4]: *** [topology-xml-libxml.lo] Error 1 make[4]: *** Waiting for unfinished jobs.... make[3]: *** [all-recursive] Error 1 make[2]: *** [all-recursive] Error 1 make[1]: *** [all-recursive] Error 1 make: *** [all] Error 2

-- Lisandro Dalcin ============ Research Scientist Computer, Electrical and Mathematical Sciences & Engineering (CEMSE) Numerical Porous Media Center (NumPor) King Abdullah University of Science and Technology (KAUST) http://numpor.kaust.edu.sa/

4700 King Abdullah University of Science and Technology al-Khawarizmi Bldg (Bldg 1), Office # 4332 Thuwal 23955-6900, Kingdom of Saudi Arabia http://www.kaust.edu.sa

Office Phone: +966 12 808-0459 _______________________________________________ To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/devel -- Pavan Balaji ✉️ http://www.mcs.anl.gov/~balaji

_______________________________________________ hwloc-devel mailing list [email protected] Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-devel Link to this post: http://www.open-mpi.org/community/lists/hwloc-devel/2015/03/4400.php

-- Lisandro Dalcin ============ Research Scientist Computer, Electrical and Mathematical Sciences & Engineering (CEMSE) Numerical Porous Media Center (NumPor) King Abdullah University of Science and Technology (KAUST) http://numpor.kaust.edu.sa/ 4700 King Abdullah University of Science and Technology al-Khawarizmi Bldg (Bldg 1), Office # 4332 Thuwal 23955-6900, Kingdom of Saudi Arabia http://www.kaust.edu.sa Office Phone: +966 12 808-0459