--hpx:print-bind outputs incorrect information

[eschnett]

I am using --hpx:print-bind on Stampede.

I am starting my application with

ibrun ./bin/block_matrix --hpx:1:pu-offset=8 --hpx:threads=8 --hpx:pu-step=1 --hpx:print-bind

which uses 2 MPI processes and 8 threads each, i.e. one full node.

--hpx:print-bind outputs

   0   0: PU L#0(P#0), Core L#0(P#0), Socket L#0(P#0), Node L#0(P#0)
   1: PU L#1(P#1), Core L#1(P#1), Socket L#0(P#0), Node L#0(P#0)
   2: PU L#2(P#2), Core L#2(P#2), Socket L#0(P#0), Node L#0(P#0)
   3: PU L#3(P#3), Core L#3(P#3), Socket L#0(P#0), Node L#0(P#0)
   4: PU L#4(P#4), Core L#4(P#4), Socket L#0(P#0), Node L#0(P#0)
   5: PU L#5(P#5), Core L#5(P#5), Socket L#0(P#0), Node L#0(P#0)
   6: PU L#6(P#6), Core L#6(P#6), Socket L#0(P#0), Node L#0(P#0)
   7: PU L#7(P#7), Core L#7(P#7), Socket L#0(P#0), Node L#0(P#0)
: PU L#0(P#0), Core L#0(P#0), Socket L#0(P#0), Node L#0(P#0)
   1: PU L#1(P#1), Core L#1(P#1), Socket L#0(P#0), Node L#0(P#0)
   2: PU L#2(P#2), Core L#2(P#2), Socket L#0(P#0), Node L#0(P#0)
   3: PU L#3(P#3), Core L#3(P#3), Socket L#0(P#0), Node L#0(P#0)
   4: PU L#4(P#4), Core L#4(P#4), Socket L#0(P#0), Node L#0(P#0)
   5: PU L#5(P#5), Core L#5(P#5), Socket L#0(P#0), Node L#0(P#0)
   6: PU L#6(P#6), Core L#6(P#6), Socket L#0(P#0), Node L#0(P#0)
   7: PU L#7(P#7), Core L#7(P#7), Socket L#0(P#0), Node L#0(P#0)

which (apart from the mis-formatting) is wrong: This output indicates that both localities are running on the first 8 processes. I used hwloc to check, and the second locality actually runs on PUs 8 to 15, not 0 to 7.

[sithhell]

This has been fixed by fixing #421