Skip to content

Commit 10c8f05

Browse files
author
Tristan Konolige
committed
forgot fill
1 parent 445d328 commit 10c8f05

File tree

1 file changed

+1
-0
lines changed

1 file changed

+1
-0
lines changed

python/tvm/utils/roofline.py

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -175,6 +175,7 @@ def estimate_peak_fma_flops(
175175
assert random_fill, "Please make sure USE_RANDOM is ON in config.cmake"
176176

177177
a = nd.empty((nthreads, num_vector_registers, vec_width), dtype="float32", device=dev)
178+
random_fill(a)
178179
times = f.time_evaluator(f.entry_name, dev, repeat=100, number=1)(a)
179180
flops = 2 * vec_width * num_vector_registers * nthreads * iters # fma is two flops
180181
flop_s = flops / times.min

0 commit comments

Comments
 (0)