Skip to content

update flight recorder usage #2292

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

Merged
merged 1 commit into from
Nov 12, 2025
Merged

update flight recorder usage #2292

Show file tree
Hide file tree
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
24 changes: 17 additions & 7 deletions src/xccl/ProcessGroupXCCL.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -508,7 +508,7 @@ c10::intrusive_ptr<ProcessGroupXCCL::WorkXCCL> ProcessGroupXCCL::initWork(
xpuEventCacheEnabled_.load());

if (record) {
r->trace_id_ = FlightRecorderXCCL::get()->record(
auto traceId = FlightRecorderXCCL::get()->recordWithResetEnabled(
local_id_,
std::make_tuple(pg_uid_, pg_desc_), // PG name tuple
seqCollective_,
Expand All @@ -522,6 +522,9 @@ c10::intrusive_ptr<ProcessGroupXCCL::WorkXCCL> ProcessGroupXCCL::initWork(
options_->timeout,
pgStatus_,
isP2P);

r->trace_id_ = traceId.id;
r->trace_reset_epoch_ = traceId.reset_epoch;
}
return r;
}
Expand Down Expand Up @@ -847,9 +850,11 @@ c10::intrusive_ptr<Work> ProcessGroupXCCL::collective(
c10::ListType::create(c10::TensorType::get()), devices);
work->future_->markCompleted(at::IValue(*work->outputs_));
auto id = work->trace_id_;
auto reset_epoch = work->trace_reset_epoch_;
work->future_->addCallback(
[id](at::ivalue::Future&) {
FlightRecorderXCCL::get()->retire_id(id, /*compute_duration*/ false);
[id, reset_epoch](at::ivalue::Future&) {
FlightRecorderXCCL::get()->retire_id(
id, reset_epoch, /*compute_duration*/ false);
},
/*use_future*/ false);
work->blockingWait_ = blockingWait_;
Expand Down Expand Up @@ -929,7 +934,7 @@ c10::intrusive_ptr<Work> ProcessGroupXCCL::pointToPoint(

c10::intrusive_ptr<ProcessGroupXCCL::WorkXCCL> work;
if (coalescing_state_) {
FlightRecorderXCCL::get()->record(
FlightRecorderXCCL::get()->recordWithResetEnabled(
local_id_,
std::make_tuple(pg_uid_, pg_desc_), // PG name tuple
seqCollective_,
Expand All @@ -948,7 +953,7 @@ c10::intrusive_ptr<Work> ProcessGroupXCCL::pointToPoint(
work->outputs_ = std::make_shared<std::vector<at::Tensor>>();
work->outputs_->push_back(tensor);

work->trace_id_ = FlightRecorderXCCL::get()->record(
auto traceId = FlightRecorderXCCL::get()->recordWithResetEnabled(
local_id_,
std::make_tuple(pg_uid_, pg_desc_), // PG name tuple
seqCollective_,
Expand All @@ -962,6 +967,9 @@ c10::intrusive_ptr<Work> ProcessGroupXCCL::pointToPoint(
options_->timeout,
pgStatus_,
true);

work->trace_id_ = traceId.id;
work->trace_reset_epoch_ = traceId.reset_epoch;
}

if (enableNanCheck_ && opType == OpType::SEND) {
Expand Down Expand Up @@ -999,9 +1007,11 @@ c10::intrusive_ptr<Work> ProcessGroupXCCL::pointToPoint(
}

auto id = work->trace_id_;
auto reset_epoch = work->trace_reset_epoch_;
work->future_->addCallback(
[id](at::ivalue::Future&) {
FlightRecorderXCCL::get()->retire_id(id, /*compute_duration*/ false);
[id, reset_epoch](at::ivalue::Future&) {
FlightRecorderXCCL::get()->retire_id(
id, reset_epoch, /*compute_duration*/ false);
},
/*use_future*/ false);
setEnqueuedPgStatus(work);
Expand Down
3 changes: 2 additions & 1 deletion src/xccl/ProcessGroupXCCL.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -121,7 +121,8 @@ class TORCH_API ProcessGroupXCCL : public Backend {
std::chrono::time_point<std::chrono::steady_clock> workStartTime_;
uint64_t seq_;
bool isP2P_;
std::optional<uint64_t> trace_id_;
std::optional<size_t> trace_id_;
std::optional<size_t> trace_reset_epoch_;
size_t numelIn_ = -1;
size_t numelOut_ = -1;

Expand Down