tqchen commented on a change in pull request #7785:
URL: https://github.com/apache/tvm/pull/7785#discussion_r611874827
##########
File path: src/target/source/codegen_c_host.cc
##########
@@ -40,13 +40,16 @@ namespace codegen {
CodeGenCHost::CodeGenCHost() { module_name_ =
GetUniqueName("__tvm_module_ctx"); }
-void CodeGenCHost::Init(bool output_ssa, bool emit_asserts, std::string
target_str) {
+void CodeGenCHost::Init(bool output_ssa, bool emit_asserts, bool
is_aot_executor,
+ std::string target_str) {
Review comment:
Thanks @giuseros ! Yes CRT is a more generic term than aot, so it might
strike a right balance
cc @areusch as well. We should explicit document the assumption being made
in CRT.
For example, if we really go with the direct function invocation
route(instead of TVMFuncCall into a global symbol), then it might be harder for
us to handle device functions that does not directly corresponds to a
symbol(which is fine, but just need documenting the assumption.
##########
File path: src/target/source/codegen_c_host.cc
##########
@@ -40,13 +40,16 @@ namespace codegen {
CodeGenCHost::CodeGenCHost() { module_name_ =
GetUniqueName("__tvm_module_ctx"); }
-void CodeGenCHost::Init(bool output_ssa, bool emit_asserts, std::string
target_str) {
+void CodeGenCHost::Init(bool output_ssa, bool emit_asserts, bool
is_aot_executor,
+ std::string target_str) {
Review comment:
Thanks @giuseros ! Yes CRT is a more generic term than aot, so it might
strike a right balance.
cc @areusch as well. We should explicit document the assumption being made
in CRT.
For example, if we really go with the direct function invocation
route(instead of TVMFuncCall into a global symbol), then it might be harder for
us to handle device functions that does not directly corresponds to a
symbol(which is fine, but just need documenting the assumption.
--
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]
