Lunderberg commented on a change in pull request #10558:
URL: https://github.com/apache/tvm/pull/10558#discussion_r825964159
##########
File path: src/runtime/hexagon/hexagon/hexagon_device_api_v2.cc
##########
@@ -163,6 +177,63 @@
TVM_REGISTER_GLOBAL("device_api.hexagon.mem_copy").set_body([](TVMArgs args, TVM
*rv = static_cast<int32_t>(0);
});
+std::map<void*, HexagonBuffer*> vtcmallocs;
+
+TVM_REGISTER_GLOBAL("device_api.hexagon.AllocNd").set_body([](TVMArgs args,
TVMRetValue* rv) {
+ int32_t device_type = args[0];
+ int32_t device_id = args[1];
+ int32_t dtype_code_hint = args[2];
+ int32_t dtype_bits_hint = args[3];
+ std::string scope = args[4];
+ CHECK(scope.find("vtcm") != std::string::npos);
+ int64_t ndim = args[5];
+ // Forcing contiguous allocation, for now
+ // TODO(Straw): Enable discontiguous allocation
+ CHECK_EQ(ndim, 1);
+ std::vector<int64_t> shape;
+ for (int i = 0; i < ndim; ++i) {
+ shape.push_back(args[6 + i]);
+ }
Review comment:
I like that idea, and it matches some of the discussion @adstraw and I
had on Friday on making a cleaner interface without needing the variadic `Call`
signature. For my understanding, what does "codegen a `builtin::kArrShape
struct`" mean in this context? My understanding was that the
`builtin::kArrShape` was used to specify access of the `int64_t*
DLTensor::shape` field. I can see how a `const int64_t* shape` could be passed
in using the same `NDArrayToLLVMArray` that is used by `CodeGenLLVM`'s
`AllocateConst` visitor, but I'm not quite seeing where `builtin::kArrShape`
would fit into it.
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
