- Email: firstname.lastname@example.org
- Registered on: 08/19/2015
- Last connection: 06/16/2017
- 10:42 PM Charm++ Bug #1614 (In Progress): SDAG Rdma example /examples/rdma/stencil3D failing on nightly build beca...
- Running on 2 processors: ./stencil3d 64 32 +balancer RefineLB +ppn 2
charmrun> /usr/bin/setarch x86_64 -R mpirun ...
- 12:04 AM Charm++ Bug #1613 (Implemented): Fix autobuild failure in mpi-win-x86_64 caused by examples/rdma/stencil3d
- The received sdag message (whose pointers were buffered and used later) was getting freed by the runtime system. Sinc...
- 12:04 AM Charm++ 3.00 hours (Bug #1613 (Merged): Fix autobuild failure in mpi-win-x86_64 caused by examples/rdma/s...
- 04:34 PM Charm++ Bug #1613 (Merged): Fix autobuild failure in mpi-win-x86_64 caused by examples/rdma/stencil3d
- make: Entering directory '/home/nikhil/autobuild/mpi-win-x86_64/charm/mpi-win-x86_64/examples/charm++/rdma/stencil...
- 12:58 PM Charm++ Bug #1610 (Implemented): Fix broken builds due to minor errors in examples/rdma/simpleRdma
- Fix: https://charm.cs.illinois.edu/gerrit/#/c/2700/
- 12:39 PM Charm++ Documentation #1611 (New): Document network dependent rdma thresholds, above which benefits of th...
- It's unclear to the user as to when the Zero Copy API should be used. So, it'll be good to document the buffer sizes ...
- 12:15 PM Charm++ Bug #1610 (Merged): Fix broken builds due to minor errors in examples/rdma/simpleRdma
- After merging the SDAG change related to RDMA (https://charm.cs.illinois.edu/gerrit/#/c/2572/), the nightly build ran...
- 03:03 PM Charm++ Bug #1539: Failure in migration when using RDMA sends in AMPI
- The AMPI RDMA bug (examples/ampi/Cjacobi3D) occurs intermittently with the following steps:
1. Migration occurs an...
- 06:17 PM Charm++ Bug #1539: Failure in migration when using RDMA sends in AMPI
- Looking at the core dump on failure for multicore-linux-x86_64, it looks like the segfault doesn't happen through the...
- 01:44 PM Charm++ Bug #738: prevent direct calls to sdag entry functions
- Changed the prefix for sdag entry methods to "_sdag_fnc_" as "_sdag_init" is an already generated xi function.
Also available in: Atom