We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
1 parent c4c0a49 commit 5b561a6Copy full SHA for 5b561a6
1 file changed
tests/python/unittest/test_roofline.py
@@ -42,7 +42,7 @@ def test_estimate_peak_flops_cpu(dtype):
42
target = tvm.target.Target("llvm -mattr=+fma,+avx2")
43
dev = remote.device(str(target))
44
# This test uses vectorized instructions so we need a target that supports them
45
- flops = tvm.utils.roofline.x86.estimate_peak_fma_vector_flops(target, dev, remote, "float32")
+ flops = tvm.utils.roofline.x86.estimate_peak_fma_vector_flops(target, dev, remote, dtype)
46
# Assume we can achieve 1 GFLOP/s per thread, which is 1 FLOP per cycle on a 1GHz cpu.
47
assert (
48
flops > 10**9 and flops < 10**14
0 commit comments