GaryYuyjl commented on a change in pull request #6121:
URL: https://github.com/apache/incubator-tvm/pull/6121#discussion_r470510402
##########
File path: python/tvm/relay/op/strategy/cuda.py
##########
@@ -172,6 +172,21 @@ def conv2d_strategy_cuda(attrs, inputs, out_type, target):
wrap_topi_schedule(topi.cuda.schedule_conv2d_nhwc_tensorcore),
name="conv2d_nhwc_tensorcore.cuda",
plevel=20)
+ elif layout == "HWNC":
+ assert kernel_layout in ["HWOI", "HWOI16o16i", "HWOI8o32i",
"HWOI32o16i"]
+ _, _, N, in_channels = get_const_tuple(data.shape)
+ pre_computed = len(kernel.shape) == 6
+ if pre_computed:
+ _, _, oc_chunk, _, oc_block_factor, _ =
get_const_tuple(kernel.shape)
+ out_channels = oc_chunk * oc_block_factor
+ else:
+ _, _, out_channels, _ = get_const_tuple(kernel.shape)
+ if topi.cuda.is_shape_tensorcore_direct_qualified(batch=N,
in_channels=in_channels, num_filter=out_channels, in_dtype=data.dtype):
+ strategy.add_implementation(
+ wrap_compute_conv2d(topi.cuda.conv2d_hwnc_tensorcore),
+
wrap_topi_schedule(topi.cuda.schedule_conv2d_hwnc_tensorcore),
+ name="conv2d_hwnc_tensorcore_direct.cuda",
+ plevel=20)
Review comment:
So far the HWNC layout doesn't support non tensor core versions.
----------------------------------------------------------------
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]
