[mvapich-discuss] Source file too large to compile at this optimization level (src/mpi/coll/bcast_tuning.c)

Subramoni, Hari subramoni.1 at osu.edu
Fri Aug 30 12:12:38 EDT 2019 

Dear, Chris.

The other way to work around this would be to add –disable-fast at configure time. This will avoid some of the optimizations that is leading to the large file size.

Best,
Hari.

From: Christopher Harrop - NOAA Affiliate <christopher.w.harrop at noaa.gov>
Sent: Friday, August 30, 2019 12:04 PM
To: Subramoni, Hari <subramoni.1 at osu.edu>
Cc: David Norton <dnorton at nvidia.com>; mvapich-discuss at cse.ohio-state.edu <mvapich-discuss at mailman.cse.ohio-state.edu>
Subject: Re: [mvapich-discuss] Source file too large to compile at this optimization level (src/mpi/coll/bcast_tuning.c)

Dear Hari,

Thank you for the update.  We’ll try an older version of PGI and/or mvapich2 for now while we wait for the 19.9 release.

Chris


Hi, All.

We had an offline discussion on this. PGI engineering has addressed this in both the LLVM and the noLLVM backends and should be available in the 19.9 release.

Best,
Hari.

From: David Norton <dnorton at nvidia.com<mailto:dnorton at nvidia.com>>
Sent: Tuesday, August 20, 2019 12:09 PM
To: Christopher W. Harrop <christopher.w.harrop at noaa.gov<mailto:christopher.w.harrop at noaa.gov>>; Subramoni, Hari <subramoni.1 at osu.edu<mailto:subramoni.1 at osu.edu>>
Cc: mvapich-discuss at cse.ohio-state.edu<mailto:mvapich-discuss at cse.ohio-state.edu> <mvapich-discuss at mailman.cse.ohio-state.edu<mailto:mvapich-discuss at mailman.cse.ohio-state.edu>>
Subject: RE: [mvapich-discuss] Source file too large to compile at this optimization level (src/mpi/coll/bcast_tuning.c)

Hi Chris

The file exceeds one of the compiler’s internal limits.  If I hear of a way around this from engineering, I’ll let you know.  Otherwise it may be that we have to wait for the next compiler release for increased limits.

-dave

From: Christopher W. Harrop <christopher.w.harrop at noaa.gov<mailto:christopher.w.harrop at noaa.gov>>
Sent: Tuesday, August 20, 2019 8:56 AM
To: Subramoni, Hari <subramoni.1 at osu.edu<mailto:subramoni.1 at osu.edu>>
Cc: David Norton <dnorton at nvidia.com<mailto:dnorton at nvidia.com>>; mvapich-discuss at cse.ohio-state.edu<mailto:mvapich-discuss at cse.ohio-state.edu> <mvapich-discuss at mailman.cse.ohio-state.edu<mailto:mvapich-discuss at mailman.cse.ohio-state.edu>>
Subject: Re: [mvapich-discuss] Source file too large to compile at this optimization level (src/mpi/coll/bcast_tuning.c)

Dear Hari,

Thank you for your and PGI’s work on this.  I’ll watch this space for an update.

Chris

On Aug 19, 2019, at 3:45 PM, Subramoni, Hari <subramoni.1 at osu.edu<mailto:subramoni.1 at osu.edu>> wrote:

Dear, Chris.

Thanks for reporting the issue. We're working on this with the PGI folks and hope to have a solution soon.
Best,
Hari.

Sent from my phone

On Aug 19, 2019 4:27 PM, "Christopher W. Harrop" <christopher.w.harrop at noaa.gov<mailto:christopher.w.harrop at noaa.gov>> wrote:
Hi,

I downloaded the mvapich2 2.3.2 source tarball and configure and compiled with PGI 19.4.  The build stops with an error:

PGC-F-0007-Source file too large to compile at this optimization level  (src/mpi/coll/bcast_tuning.c)
I’ve ensured there is plenty of memory available, but that doesn’t seem to matter.  Is this a known issue?

Here are my build details:

CentOS Linux release 7.6.1810 (Core)

pgcc 19.4-0 LLVM 64-bit target on x86-64 Linux -tp skylake

./configure --prefix=/path/to/mvapich2/2.3.2/pgi/19.4 --enable-fortran=yes --enable-cxx --with-pmi=pmi2 --with-pm=slurm --with-slurm=/path/to/slurm/default --with-device=ch3:mrail --with-rdma=gen2 --enable-romio

Mellanox HDR 100

Here is the error in more context:

/bin/sh ./libtool  --tag=CC   --mode=compile pgcc -DHAVE_CONFIG_H -I. -I./src/include  -I./src/include -I./src/mpi/datatype -I./src/mpi/spawn -I./src/util/dbg -Isrc/util/logging/common -I./src/util/mem -I./src/util/thread -I./src/u\
til/type -I./src/util/wrappers -I./src/util/wrappers -I./src/util/assert -I./src/util/timers -I./src/util/pointer -I./src/binding/cxx -I/path/to/mvapich2/2.3.2/pgi/19.4/src/binding/fortran/mpif_h -Isrc/binding/fortran/use_mpi  -I./\
src/mpid/ch3/include -I./src/mpid/ch3/include -I/path/to/mvapich2/2.3.2/pgi/19.4/src/mpid/ch3/channels/common/include      -I./src/mpid/ch3/channels/mrail/include -I./src/mpid/ch3/channels/mrail/include -I./src/mpid/ch3/channels/co\
mmon/include -I./src/mpid/ch3/channels/common/include -I./src/mpid/common/locks -I./src/mpid/common/locks -I./src/util/wrappers -I./src/util/wrappers -I./src/mpid/ch3/channels/mrail/src/gen2 -I./src/mpid/ch3/channels/mrail/src/gen2\
 -D_GNU_SOURCE -I./src/mpid/ch3/channels/mrail/src/gen2 -I./src/mpi/coll -I./src/mpi/romio/adio/include   -I./src/mpid/ch3/channels/common/include -I./src/mpid/ch3/channels/common/src/ft     -I./src/mpid/common/datatype -I./src/mpi\
d/common/datatype -I./src/mpid/common/sched   -I./src/mpid/common/thread -I./src/mpid/common/timers     -I./src/pmi/simple -Icontrib/hwloc/include/ -I./contrib/hwloc/include -I/path/to/mvapich2/2.3.2/pgi/19.4/src/mpl/include -I/pat\
h/to/mvapich2/2.3.2/pgi/19.4/src/mpl/include -I/path/to/mvapich2/2.3.2/pgi/19.4/src/openpa/src -I/path/to/mvapich2/2.3.2/pgi/19.4/src/openpa/src -D_REENTRANT -I/path/to/mvapich2/2.3.2/pgi/19.4/src/mpi/romio/include -I/path/to/slurm\
/default/include  -DNDEBUG -DNVALGRIND -O2 -c -o src/mpi/coll/lib_libmpi_la-bcast_tuning.lo `test -f 'src/mpi/coll/bcast_tuning.c' || echo './'`src/mpi/coll/bcast_tuning.c
libtool: compile:  pgcc -DHAVE_CONFIG_H -I. -I./src/include -I./src/include -I./src/mpi/datatype -I./src/mpi/spawn -I./src/util/dbg -Isrc/util/logging/common -I./src/util/mem -I./src/util/thread -I./src/util/type -I./src/util/wrapp\
ers -I./src/util/wrappers -I./src/util/assert -I./src/util/timers -I./src/util/pointer -I./src/binding/cxx -I/path/to/mvapich2/2.3.2/pgi/19.4/src/binding/fortran/mpif_h -Isrc/binding/fortran/use_mpi -I./src/mpid/ch3/include -I./src\
/mpid/ch3/include -I/path/to/mvapich2/2.3.2/pgi/19.4/src/mpid/ch3/channels/common/include -I./src/mpid/ch3/channels/mrail/include -I./src/mpid/ch3/channels/mrail/include -I./src/mpid/ch3/channels/common/include -I./src/mpid/ch3/cha\
nnels/common/include -I./src/mpid/common/locks -I./src/mpid/common/locks -I./src/util/wrappers -I./src/util/wrappers -I./src/mpid/ch3/channels/mrail/src/gen2 -I./src/mpid/ch3/channels/mrail/src/gen2 -D_GNU_SOURCE -I./src/mpid/ch3/c\
hannels/mrail/src/gen2 -I./src/mpi/coll -I./src/mpi/romio/adio/include -I./src/mpid/ch3/channels/common/include -I./src/mpid/ch3/channels/common/src/ft -I./src/mpid/common/datatype -I./src/mpid/common/datatype -I./src/mpid/common/s\
ched -I./src/mpid/common/thread -I./src/mpid/common/timers -I./src/pmi/simple -Icontrib/hwloc/include/ -I./contrib/hwloc/include -I/path/to/mvapich2/2.3.2/pgi/19.4/src/mpl/include -I/path/to/mvapich2/2.3.2/pgi/19.4/src/mpl/include \
-I/path/to/mvapich2/2.3.2/pgi/19.4/src/openpa/src -I/path/to/mvapich2/2.3.2/pgi/19.4/src/openpa/src -D_REENTRANT -I/path/to/mvapich2/2.3.2/pgi/19.4/src/mpi/romio/include -I/path/to/slurm/default/include -DNDEBUG -DNVALGRIND -O2 -c \
src/mpi/coll/bcast_tuning.c -MD  -fpic -DPIC -o src/mpi/coll/.libs/lib_libmpi_la-bcast_tuning.o
PGC-F-0007-Source file too large to compile at this optimization level  (src/mpi/coll/bcast_tuning.c)
PGC/x86-64 Linux 19.4-0: compilation aborted
make[2]: *** [src/mpi/coll/lib_libmpi_la-bcast_tuning.lo] Error 1
make[2]: Leaving directory `/path/to/mvapich2/2.3.2/pgi/19.4'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/path/to/mvapich2/2.3.2/pgi/19.4'
make: *** [all] Error 2

Chris
-----------------------------------------------------------------------------------------------------------
Christopher W. Harrop                               email: christopher.w.harrop at noaa.gov<mailto:christopher.w.harrop at noaa.gov>
Global Systems Division                                                      voice: (303) 497-6808<tel:%28303%29%20497-6808>
NOAA Earth System Research Laboratory                             fax: (303) 497-7259<tel:%28303%29%20497-7259>
325 Broadway R/GSD6
Boulder, CO 80303






________________________________
This email message is for the sole use of the intended recipient(s) and may contain confidential information.  Any unauthorized review, use, disclosure or distribution is prohibited.  If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cse.ohio-state.edu/pipermail/mvapich-discuss/attachments/20190830/25900a97/attachment-0001.html>

More information about the mvapich-discuss mailing list