Add HDP flush for gfx940 (#1434)
* Fix collective trace * Use nontemporal for st_global * Fix previous commit * Add HDP flush to data receive path * Fix previous commit * Control flushing by NCCL_NET_FORCE_FLUSH and RCCL_NET_HDP_FLUSH * Introduce RCCL_NET_HDP_FLUSH and RCCL_NET_GDR_FLUSH Both are on by default. Turn both off will skip all flush will likely result in data error. * Enable GDR copy by default * Remove GDR flush env var because it is disabled by GDC flush * Output kernel collective trace at comm destroy by default * Limit kernel timeout messages to 100 * Use system relaxed atomic for loadInt * Refine timeout messages and use atomic for setting offset from CPU * Add kernel trace for barrier timeout * Add backup barrier to avoid race in atomicAdd * Use different counters for different warps * Rework barrier implementation * Fix for other GFX * Use __hip_atomic_store and __hip_atomic_load * Fix bug in previous commit * Don't reset barrier values in running kernel * Update trace format * Fix typo * Switch back to hip_atomic_fetch_add * Use same barrier implementation for all GFX * Remove extra threadfence * Turn off HDP flush by default Please use RCCL_NET_HDP_FLUSH=1 to switch on HDP flush * Remove unnecessary changes from alterative barrier implementation * Added back __threadfence_block * Revert back to threadfence for gfx other than gfx94x
This commit is contained in:
@@ -23,7 +23,7 @@ inline __device__ int min(int a, ssize_t b) { return (a < b) ? a : b; }
|
||||
|
||||
inline __device__ int loadInt(int* ptr) {
|
||||
int v;
|
||||
v = atomicAdd((unsigned long long *)ptr, 0);
|
||||
v = __atomic_load_n(ptr, __ATOMIC_RELAXED);
|
||||
return v;
|
||||
}
|
||||
|
||||
|
||||
Reference in New Issue
Block a user