zhuwenxi commented on issue #7596:
URL: https://github.com/apache/tvm/issues/7596#issuecomment-793783685


   @leeexyz It works when I comment out the ICHECK line, but only for 
"tvm.build()". It still crashes when trying to run the module:
   
![image](https://user-images.githubusercontent.com/4969797/110467406-efa61c00-8111-11eb-880a-83d45b2c7665.png)
   
   Below is my code to reproduce the crash:
   <pre>
   import tvm
   from tvm import te
   
   n = 1024
   l = 128
   m = 235
   bias = te.var("bias", dtype="float32")`
   A = te.placeholder((n, l), name="A")`
   B = te.placeholder((l, m), name="B")`
   C = te.extern(`
       (n, m),
       [A, B],
       lambda ins, outs: tvm.tir.call_packed(`
           "tvm.contrib.cblas.matmul", ins[0], ins[1], outs[0], False, False`
       ),`
       name="C",`
   )
   D = te.compute(C.shape, lambda i, j: C[i, j] + bias, name="D")`
   s = te.create_schedule(D.op)`
   
   ctx = tvm.cpu(0)`
   f = tvm.build(s, [A, B, D, bias], "c")
   a = tvm.nd.array(np.random.uniform(size=(n, l)).astype(A.dtype), ctx)
   b = tvm.nd.array(np.random.uniform(size=(l, m)).astype(B.dtype), ctx)
   d = tvm.nd.array(np.zeros((n, m), dtype=D.dtype), ctx)
   bb = 10.0
   f(a, b, d, bb)
   </pre>


----------------------------------------------------------------
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]


Reply via email to