tqchen commented on a change in pull request #7382:
URL: https://github.com/apache/tvm/pull/7382#discussion_r568044848
##########
File path: src/target/source/codegen_c.cc
##########
@@ -648,6 +648,21 @@ void CodeGenC::VisitExpr_(const CallNode* op,
std::ostream& os) { // NOLINT(*)
os << " != ";
this->PrintExpr(op->args[0], os);
os << ")";
+ } else if (op->op.as<OpNode>()->name == "tir.round") {
+ CHECK_EQ(op->args.size(), 1U);
Review comment:
We should ideally handle these intriniscs lowering stage before the
codegen. The main reason is we have not yet registered the dispatch rule for C
backend, or default
- LLVM intrinsic dispatch rule:
https://github.com/apache/tvm/blob/main/src/target/llvm/intrin_rule_llvm.cc#L84
- https://github.com/apache/tvm/blob/main/src/target/intrin_rule.cc#L54
----------------------------------------------------------------
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]
