Hi Danny,
thanks for this background info.
I have changed Make.rules to check if 'uname -p' returns unknown, and if
so, fall back to 'uname -m'. (There is at least one platform - AIX -
where 'uname -m' doesn't return the machine architecture so we must
still use 'uname -p'.)
As you're building from SVN HEAD, you could update (r27672) to confirm
whether it solves the problem for your machine. I have also recorded it
in JIRA - http://jira.codehaus.org/browse/XTENLANG-3387.
Cheers,
Josh
On 30/04/14 05:56, Danny Munera wrote:
Hi Josh,
Thank you for the answer.
I searched about why the command "uname -p" returns unknown on this
system.
There is some information about this behaviour in this site:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=655428
It appears that the "uname" command is making improper use of the
uname(2) syscall. The Debian maintainer of coreutils (the GNU package
which includes "uname") says it's a "won't fix" bug.
I tried changing in the Make.rules file every "uname -p" to "uname -m"
and It works fine, the x10 is built correctly.
Do you think is right to use the option "-m" instead "-p" in the
command uname?
Danny Múnera
On Tue, Apr 29, 2014 at 5:30 PM, Josh Milthorpe
<josh.miltho...@anu.edu.au <mailto:josh.miltho...@anu.edu.au>> wrote:
Hi Danny,
x10.runtime/Make.rules sets a variable jarch for the architecture
using "uname -p". I assume that "uname -p" returns "unknown" on
your machine? - can you confirm?
As your platform is Linux, you could try unconditionally setting
jarch=amd64 (lines 99-101 of Make.rules). However, it's probably
better if you can find out why "uname -p" isn't working.
Cheers,
Josh
------------------------------------------
Josh Milthorpe
Visiting Fellow, Research School of Computer Science
Australian National University, Building 108
Canberra, ACT 0200
Australia
E-mail:josh.miltho...@anu.edu.au <mailto:josh.miltho...@anu.edu.au>
Web:http://cs.anu.edu.au/~Josh.Milthorpe/ <http://cs.anu.edu.au/%7EJosh.Milthorpe/>
On 29/04/14 11:19, Danny Munera wrote:
Hello everyone,
I'm trying to build the X10 runtime (with MPI) on a linux
machine. I'm getting an error when the runtime is compiled:
...
[exec] mpicxx -DNO_TRACING -DHOMOGENEOUS -O2 -finline-functions
-pthread -Iinclude -Icommon
-I"/usr/lib/jvm/java-6-oracle"/include
-I"/usr/lib/jvm/java-6-oracle"/include/linux -ansi -pedantic
-Wall -Wextra -Wno-long-long -Wno-unused-parameter
-Wno-unused-variable -Wno-unused-value -Werror=return-type
mpi/Java.cc -o mpi/X10MPIJava -ljvm
-L"/usr/lib/jvm/java-6-oracle"/jre/lib/unknown/j9vm
-L"/usr/lib/jvm/java-6-oracle"/jre/lib/*unknown/*server
-L"/usr/lib/jvm/java-6-oracle"/jre/lib/unknown/client
[exec] /usr/bin/ld: cannot find -ljvm
[exec] collect2: error: ld returned 1 exit status
[exec] make: *** [mpi/X10MPIJava] Error 1
BUILD FAILED
/home/danny/x10/x10-trunk/x10.dist/build.xml:397: The following
error occurred while executing this line:
/home/danny/x10/x10-trunk/x10.runtime/build.xml:436: exec returned: 2
I suppose the problem is because the ant script is not able to
find out the machine architecture, and the path of the ljvm
library is created with the *unknown* folder (instead amd64 which
is the right one).
Is there any way to set the architecture manually? Why the ant
script is not working on this machine?
Regards,
Danny Múneracompile
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos. Get
unparalleled scalability from the best Selenium testing platform available.
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
X10-users mailing list
X10-users@lists.sourceforge.net <mailto:X10-users@lists.sourceforge.net>
https://lists.sourceforge.net/lists/listinfo/x10-users
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos. Get
unparalleled scalability from the best Selenium testing platform
available.
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
X10-users mailing list
X10-users@lists.sourceforge.net
<mailto:X10-users@lists.sourceforge.net>
https://lists.sourceforge.net/lists/listinfo/x10-users
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos. Get
unparalleled scalability from the best Selenium testing platform available.
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
X10-users mailing list
X10-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/x10-users
------------------------------------------------------------------------------
Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
• 3 signs your SCM is hindering your productivity
• Requirements for releasing software faster
• Expert tips and advice for migrating your SCM now
http://p.sf.net/sfu/perforce
_______________________________________________
X10-users mailing list
X10-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/x10-users
