XIAO-XIA opened a new issue #7682:
URL: https://github.com/apache/tvm/issues/7682
Segmentation fault occurs when the `out_shape` of `threefry_generate` setted
to some cases.
### To Reproduce
#### environment:
Ubuntu 18.04, llvm version 8.0.1
#### script:
```python
import tvm
import tvm.relay
import tvm.testing
import tvm.topi
import numpy as np
def threefry_generate(target, ctx, gen, size):
gen_placeholder = tvm.te.placeholder(gen.shape, name="gen",
dtype="uint64")
left_placeholder, right_placeholder =
tvm.topi.random.threefry_generate(gen_placeholder, size)
s = tvm.topi.generic.schedule_extern([left_placeholder,
right_placeholder])
f = tvm.build(s, [gen_placeholder, left_placeholder, right_placeholder])
out_gen = tvm.nd.array(np.zeros(gen.shape, dtype="uint64"))
rands = tvm.nd.array(np.zeros(size, dtype="uint64"))
f(tvm.nd.array(gen), out_gen, rands)
return out_gen.asnumpy(), rands.asnumpy()
@tvm.testing.parametrize_targets
def test_threefry_generate(target, ctx):
gen = tvm.relay.random.threefry_key(0).data.asnumpy()
shape = (1, 3, 224, 224)
a, rands = threefry_generate(target, ctx, gen, shape)
if __name__ == "__main__":
test_threefry_generate(tvm.target.Target("llvm"), tvm.context("cpu"))
```
#### error outputs:
```
malloc(): memory corruption
Aborted (core dumped)
```
#### error cases:
`out_shape = (1, 3, 224, 224), (1, 3, 32, 32), (1, 3, 16, 16)`
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
