- Registered on: 04/21/2016
- Last connection: 02/14/2018
- 03:38 PM GROMACS Bug #2409: PME gather CUDA kernel failing on Fermi
- One thing TO DO here would be to have a huge input system sanity test, but Szilard brought up a good point that such ...
- 11:00 AM GROMACS Bug #2409 (Resolved): PME gather CUDA kernel failing on Fermi
- Applied in changeset commit:fa92dbed34ec7b916faaa2dd6f17c1b08a8ede0a.
- 10:53 AM GROMACS Revision fa92dbed: Fix PME for large systems with Fermi GPUs
- PME spread/gather CUDA kernel scheduling did not account for
compute capability limitations. Realistically this has o...
- 08:09 PM GROMACS Revision 3005706f: Update FFTW build to 3.3.7
- Refs #2358
- 07:00 PM GROMACS Revision a4987be3: Add freeDeviceBuffer GPU utility function
- This allows to make the GPU pairlist deletion code look same
for CUDA and OpenCL. More cleanup TODOs are left.
- 03:19 PM GROMACS Revision 03cb7aba: Add the missing comma in the quote
- This was printed as:
GROMACS reminds you: "It's just B I O L O G Y, can't you see?Joe Jackson" ((null))
- 06:33 PM GROMACS Bug #2409: PME gather CUDA kernel failing on Fermi
- Thanks, reproduced on C2075 on gromacs3!
- 01:04 PM GROMACS Revision e3851eb6: Replace the remaining EXPECT_DEATH with EXPECT_DEATH_IF_SUPPORTED
- Change-Id: Ibce3d4f240bf3552893762b74206d8e54575d18d
- 01:04 PM GROMACS Revision b2c6f452: Add DeviceBuffer as a typedef for cl_mem/device ponter
- It is applied to unify the pairlist structure.
- 12:49 PM GROMACS Task #2408: device busy error with CC 2.0 and 6.1 in the same run
- Well, we can check CCs and driver version and print a message "You are likely suffering from NVIDIA disallowing combi...
Also available in: Atom