feat: support single-node multi-NPUs offline inference. #267
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
yq33victor
force-pushed
the
dev/support_offline_multi_xpu3
branch
from
5d8db2a to
2c0bc1e
Compare
yq33victor
force-pushed
the
dev/support_offline_multi_xpu3
branch
2 times, most recently
from
febeb33 to
2f78aa0
Compare
XuZhang99
reviewed
Oct 22, 2025
xllm/core/distributed_runtime/spawn_worker_server/spawn_worker_server.cpp
Show resolved
Hide resolved
xllm/core/distributed_runtime/spawn_worker_server/spawn_worker_server.cpp
Outdated
Show resolved
Hide resolved
Signed-off-by: pengtao.156 <pengtao.156@jd.com>
yq33victor
force-pushed
the
dev/support_offline_multi_xpu3
branch
from
2f78aa0 to
0ac5cf7
Compare
XuZhang99
reviewed
Oct 23, 2025
|
|
||
| std::atomic<bool> done(false); | ||
| #if defined(USE_NPU) | ||
| xllm::Device device("npu:" + std::to_string(device_idx)); |
Collaborator
There was a problem hiding this comment.
maybe we can add another construction function for class Device:
Device::Device(int32_t device_id){
std::string device_name = type() + ":" + std::to_string(device_id);
device_ = torch::Device(device_name);
}and move #if defined(USE_NPU) before FLAGS_enable_atb_comm_multiprocess = true;
JimHsiung
approved these changes
Oct 24, 2025
liutongxuan
approved these changes
Oct 24, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.