Use monotonic clock for interval measurement (#6211)
Signed-off-by: Lifu Huang <lifu.hlf@gmail.com>
This commit is contained in:
@@ -228,7 +228,7 @@ class EAGLEWorker(TpModelWorker):
|
||||
return
|
||||
|
||||
# Capture draft
|
||||
tic = time.time()
|
||||
tic = time.perf_counter()
|
||||
before_mem = get_available_gpu_memory(self.device, self.gpu_id)
|
||||
logger.info(
|
||||
f"Capture draft cuda graph begin. This can take up to several minutes. avail mem={before_mem:.2f} GB"
|
||||
@@ -236,7 +236,7 @@ class EAGLEWorker(TpModelWorker):
|
||||
self.cuda_graph_runner = EAGLEDraftCudaGraphRunner(self)
|
||||
after_mem = get_available_gpu_memory(self.device, self.gpu_id)
|
||||
logger.info(
|
||||
f"Capture draft cuda graph end. Time elapsed: {time.time() - tic:.2f} s. avail mem={after_mem:.2f} GB. mem usage={(before_mem - after_mem):.2f} GB."
|
||||
f"Capture draft cuda graph end. Time elapsed: {time.perf_counter() - tic:.2f} s. avail mem={after_mem:.2f} GB. mem usage={(before_mem - after_mem):.2f} GB."
|
||||
)
|
||||
|
||||
# Capture extend
|
||||
|
||||
Reference in New Issue
Block a user