Oh, I missed that.
Then, yes, I agree. We can't support that.
That sounds like a bug PGI will need to fix, in general.
On Mar 11, 2011, at 4:34 PM, Ralph Castain wrote:
> That's not good enough - you have to use PGI's hacked version of if.h instead.
>
> Hence my comment that we can't fix thi
That's not good enough - you have to use PGI's hacked version of if.h instead.
Hence my comment that we can't fix this for the general case.
On Mar 11, 2011, at 1:43 PM, Jeff Squyres wrote:
> David --
>
> If you can come up with a configure-esque test for detecting that we're on a
> system tha
David --
If you can come up with a configure-esque test for detecting that we're on a
system that has this bad condition, we can add an #if around the #include.
For example, if you can write a short test that verifies that is
there, but it doesn't work, and further, if we're using the broken v
I have found this thread form two years ago. I am some kind of lost on
configuring an infiniband cluster for openmpi. What is best: use iboip
or use native infiniband ibverbs interface ? For now I am using native
infiniband withou ipoib. But I have lots of problems specially with
latency in the clu
On Thursday, March 10, 2011 08:30:19 pm Thierry LAMOUREUX wrote:
> Hello,
>
> We add recently enhanced our network with Infiniband modules on a six node
> cluster.
>
> We have install all OFED drivers related to our hardware
>
> We have set network IP like following :
> - eth : 192.168.1.0 / 255
On Mar 9, 2011, at 3:15 PM, Randall Svancara wrote:
> I am experiencing an error with an application called MPIBLAST. I am
> trying to understand more about what this error represents in terms of
> this application:
>
> mpiblast_writer.cppStreamliner::CalculateMessageSizes - miscalculate
> messa
