diff options
author | Patrick Mours <pmours@nvidia.com> | 2020-06-08 18:16:10 +0300 |
---|---|---|
committer | Patrick Mours <pmours@nvidia.com> | 2020-06-08 18:55:49 +0300 |
commit | 9f7d84b656fbb56966620ecc249ce5bc7089a1d1 (patch) | |
tree | d0a022feae43f6db2166cf5214b56cce99b96a60 /intern/cycles/device/device.cpp | |
parent | 0a907657d4d525d320e0c8518f583b7210736214 (diff) |
Cycles: Add support for P2P memory distribution (e.g. via NVLink)
This change modifies the multi-device implementation to support memory distribution
across devices, to reduce the overall memory footprint of large scenes and allow scenes to
fit entirely into combined GPU memory that previously had to fall back to host memory.
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D7426
Diffstat (limited to 'intern/cycles/device/device.cpp')
-rw-r--r-- | intern/cycles/device/device.cpp | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/intern/cycles/device/device.cpp b/intern/cycles/device/device.cpp index bad156d40bf..41dd7894d93 100644 --- a/intern/cycles/device/device.cpp +++ b/intern/cycles/device/device.cpp @@ -602,6 +602,7 @@ DeviceInfo Device::get_multi_device(const vector<DeviceInfo> &subdevices, info.has_adaptive_stop_per_sample = true; info.has_osl = true; info.has_profiling = true; + info.has_peer_memory = false; foreach (const DeviceInfo &device, subdevices) { /* Ensure CPU device does not slow down GPU. */ @@ -645,6 +646,7 @@ DeviceInfo Device::get_multi_device(const vector<DeviceInfo> &subdevices, info.has_adaptive_stop_per_sample &= device.has_adaptive_stop_per_sample; info.has_osl &= device.has_osl; info.has_profiling &= device.has_profiling; + info.has_peer_memory |= device.has_peer_memory; } return info; |