Laurawly commented on a change in pull request #4644: [WIP] Relay op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r372704546
##########
File path: topi/python/topi/intel_graphics/conv2d.py
##########
@@ -443,8 +354,32 @@ def _schedule_cl_spatialpack_NCHWc(cfg, s, op):
s[output].compute_inline()
conv = s.outputs[0]
SCHEDULE_OUTPUT = False
+ else: # conv2d_NCHWc_unpack
+ conv = op.input_tensors[0]
+ temp = s[conv].op.input_tensors[0]
+ kernel = s[conv].op.input_tensors[1]
+ temp_W = s.cache_read(temp, "warp", [conv])
+ conv_L = s.cache_write(conv, "local")
+ SCHEDULE_OUTPUT = True
kernel_L = s.cache_read(kernel, "local", [conv_L])
+ if temp.name == "pad_temp":
+ data = temp.op.input_tensors[0]
+ # TODO(@Laurawly): Do we need to schedule pad op here?
Review comment:
Yeah we need to schedule temp.
----------------------------------------------------------------
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:
us...@infra.apache.org
With regards,
Apache Git Services
