>    ./configure --with-device=ch4:ucx --with-hcoll=/opt/mellanox/hcoll --with-pmix --prefix=/home/publique/depot/mpi/mpich-ch4_ucx-3.4a3 --enable-fast=all --enable-romio --with-file-system=ufs+nfs --enable-shared --enable-sharedlibs=gcc

 

I believe the offending option is `--with-pmix`, since that defaults to `--with-pmix=yes`. You are supposed to pass in the path to pmix installation, e.g. `--with-pmix=/usr/local`.

 

I admit this is a bit not obvious.

 

-- 
Hui Zhou

 

 

From: "Audet, Martin via discuss" <discuss@mpich.org>
Reply-To: "discuss@mpich.org" <discuss@mpich.org>
Date: Tuesday, September 22, 2020 at 9:41 PM
To: "discuss@mpich.org" <discuss@mpich.org>
Cc: "Audet, Martin" <Martin.Audet@cnrc-nrc.gc.ca>
Subject: [mpich-discuss] Problem configuring mpich-3.4a3 and 3.4a2 --with-device=ch4:ucx

 

Hello MPICH_Developers,

 

I am having trouble running the configure script on two latest mpich release (3.4a2 and 3.4a3) when I use the flag --with-device=ch4:ucx (no such problems with ch3:sock or ch3:nemesis). It seems that the generated Makefile often contains two strange compilation and link flags: -Iyes/include  and –Lyes/lib.

 

To be able to compile the library I had to use the following command to remove those strange flags:

 

   find . -name Makefile -exec sed -i -e 's/-Iyes\/include\>//g' -e 's/-Lyes\/lib\>//g' {} \;

 

After this step I am able to compile and the resulting library seems ok and appears to performs well (not extensive testing).

 

For your information I call the configure script like this:

 

   ./configure --with-device=ch4:ucx --with-hcoll=/opt/mellanox/hcoll --with-pmix --prefix=/home/publique/depot/mpi/mpich-ch4_ucx-3.4a3 --enable-fast=all --enable-romio --with-file-system=ufs+nfs --enable-shared --enable-sharedlibs=gcc

 

I had experienced this problem on the following three configurations:

 

   mpich-3.4a2  CentOS 7.6  MOFED 4.7

   mpich-3.4a3  CentOS 7.6  MOFED 4.7

   mpich-3.4a3  CentOS 7.8  MOFED 4.9

 

The architecture is x86_64.

 

Could someone take a look at this problem ?

 

Thanks,

 

Martin Audet