I wrote: > I'll try some variations like that when I can get complete nodes on the > chassis.
It turns out that adding --mca mtl ^mxm to the 1.8 case gives results in line with 1.6 as best as I can estimate the variation (error bars -- we've heard of them). It makes no difference to 1.6 whether I enable or disable mxm explicitly, and ompi_info doesn't show anything like a verbosity setting for it. Any idea why it has that effect?
