Skip to content

Refactor: unify AICPU spin-wait macros to SPIN_WAIT_HINT #219

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
ChaoWao merged 1 commit into from
Mar 9, 2026
Merged

Refactor: unify AICPU spin-wait macros to SPIN_WAIT_HINT #219

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
1 change: 0 additions & 1 deletion src/a2a3/runtime/tensormap_and_ringbuffer/aicpu/aicpu_executor.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -792,7 +792,6 @@ int AicpuExecutor::resolve_and_dispatch_pto2(Runtime* runtime, int thread_idx,
} else {
SPIN_WAIT_HINT();
}
PTO2_SPIN_PAUSE_LIGHT();
CYCLE_COUNT_LAP(sched_idle_cycle);
#if PTO2_PROFILING
if (profiling_enabled) {
Expand Down
2 changes: 1 addition & 1 deletion src/a2a3/runtime/tensormap_and_ringbuffer/runtime/pto_orchestrator.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -453,7 +453,7 @@ void pto2_orchestrator_wait_all(PTO2OrchestratorState* orch) {

// Spin-wait until scheduler reports all tasks done
while (!orch->scheduler->is_done()) {
PTO2_SPIN_PAUSE();
SPIN_WAIT_HINT();
}
}

Expand Down
4 changes: 2 additions & 2 deletions src/a2a3/runtime/tensormap_and_ringbuffer/runtime/pto_ring_buffer.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -144,7 +144,7 @@ struct PTO2HeapRing {
exit(1);
}

PTO2_SPIN_PAUSE();
SPIN_WAIT_HINT();
}
}

Expand Down Expand Up @@ -343,7 +343,7 @@ struct PTO2TaskRing {
exit(1);
}

PTO2_SPIN_PAUSE();
SPIN_WAIT_HINT();
}
}

Expand Down
25 changes: 10 additions & 15 deletions src/a2a3/runtime/tensormap_and_ringbuffer/runtime/pto_runtime2_types.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -344,20 +344,15 @@ typedef void (*PTO2InCoreFunc)(void** args, int32_t num_args);
#define PTO2_MEMORY_BARRIER() __sync_synchronize()
#endif

/**
* Pause instruction for spin-wait loops
* Include sched_yield() to prevent CPU starvation of other threads
*/
#include <sched.h>
#if defined(__aarch64__)
#define PTO2_SPIN_PAUSE() do { __asm__ __volatile__("yield" ::: "memory"); sched_yield(); } while(0)
#define PTO2_SPIN_PAUSE_LIGHT() __asm__ __volatile__("yield" ::: "memory")
#elif defined(__x86_64__)
#define PTO2_SPIN_PAUSE() do { __builtin_ia32_pause(); sched_yield(); } while(0)
#define PTO2_SPIN_PAUSE_LIGHT() __builtin_ia32_pause()
// Spin-wait hint for AICPU threads. On real hardware the AICPU has dedicated
// ARM A55 cores — no OS yield is needed, so the hint is a no-op. In simulation
// all threads share host CPU cores, so we yield to prevent starvation.
// This header is also compiled into the Host .so (for struct definitions only),
// where the hint is never called — the fallback no-op keeps Host builds clean.
#if __has_include("spin_hint.h")
#include "spin_hint.h"
#else
#define PTO2_SPIN_PAUSE() sched_yield()
#define PTO2_SPIN_PAUSE_LIGHT() ((void)0)
#define SPIN_WAIT_HINT() ((void)0)
#endif
Comment thread
ChaoWao marked this conversation as resolved.

// =============================================================================
Expand Down Expand Up @@ -387,7 +382,7 @@ static inline void pto2_fanout_lock(PTO2TaskDescriptor* task,
while (task->fanout_lock.load(std::memory_order_acquire) != 0) {
contended = true;
atomic_ops++; // each load = 1 atomic
PTO2_SPIN_PAUSE_LIGHT();
SPIN_WAIT_HINT();
}
int32_t expected = 0;
if (task->fanout_lock.compare_exchange_weak(expected, 1,
Expand All @@ -408,7 +403,7 @@ static inline void pto2_fanout_lock(PTO2TaskDescriptor* task,
static inline void pto2_fanout_lock(PTO2TaskDescriptor* task) {
for (;;) {
while (task->fanout_lock.load(std::memory_order_acquire) != 0) {
PTO2_SPIN_PAUSE_LIGHT();
SPIN_WAIT_HINT();
}
int32_t expected = 0;
if (task->fanout_lock.compare_exchange_weak(expected, 1,
Expand Down
Loading