> On Thu, 2004-12-09 at 16:53, Miquel van Smoorenburg wrote: >> In article <[EMAIL PROTECTED]>, >> michael <[EMAIL PROTECTED]> wrote: >> >However, for one simple test (v.low memory, v.low comms, high compute) >> >case which is Fortran90 compiled using mpich, I actually found turning >> >HT off helped -- when it was on then for -np 2 there was no speed up >> >(top implied they ran on same physical processor). I presume this was >> >because the CPU was fully busy anyhow so the 2nd thread had to wait >> (you >> >seem to imply the opposite but I don't follow why!).
I've just heard back from the mpich guys that the current mpich isn't thread safe. Thus turning off HT does indeed seem the sane thing to do. mpich2, apparently, will be thread safe "soon" so I'll look into that and report back once it hits the distribution sites Michael -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
