Skip to content

Make HCCL OOM injection persistent after trigger count #9

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
yjyang62 wants to merge 1 commit into
base: cursor/pta-oom-2.9.1-fbca
Choose a base branch
from
+1 −1
Draft

Make HCCL OOM injection persistent after trigger count #9

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion torch_npu/csrc/distributed/ProcessGroupHCCL.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -113,7 +113,7 @@ void maybeThrowHcclOom(c10d::OpType opType, c10_npu::CaptureStatus capture_statu
}

const int64_t current_count = ++g_hccl_oom_call_count;
if (current_count > trigger_count && current_count < trigger_count + 2) {
if (current_count > trigger_count) {
auto retmsg = std::string("HCCL function error: Failed to allocate memory. "
"Injected HCCL OOM after ") + std::to_string(current_count) +
" HCCL operations, op type is " + opTypeToString(opType) +
Expand Down