Skip to content

[Runtime][CoreML] Fix FFI casts in CoreML runtime #19762

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
tqchen merged 1 commit into from
Jun 14, 2026
+9 −21
Merged

[Runtime][CoreML] Fix FFI casts in CoreML runtime #19762

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
30 changes: 9 additions & 21 deletions src/runtime/extra/contrib/coreml/coreml_runtime.mm
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,7 @@
*/
#include <tvm/ffi/function.h>
#include <tvm/ffi/reflection/registry.h>
#include <tvm/runtime/logging.h>

#include "../../../../support/bytes_io.h"
#include "coreml_runtime.h"
Expand Down Expand Up @@ -136,12 +137,12 @@
return ffi::Function([this](ffi::PackedArgs args, ffi::Any* rv) { model_->Invoke(); });
} else if (name == "set_input") {
return ffi::Function([this](ffi::PackedArgs args, ffi::Any* rv) {
const auto& input_name = args[0].operator std::string();
model_->SetInput(input_name, args[1]);
model_->SetInput(args[0].cast<std::string>(), args[1].cast<DLTensor*>());
});
Comment thread
tlopex marked this conversation as resolved.
} else if (name == "get_output") {
return ffi::Function(
[this](ffi::PackedArgs args, ffi::Any* rv) { *rv = model_->GetOutput(args[0]); });
return ffi::Function([this](ffi::PackedArgs args, ffi::Any* rv) {
*rv = model_->GetOutput(args[0].cast<int>());
});
Comment thread
tlopex marked this conversation as resolved.
} else if (name == "get_num_outputs") {
return ffi::Function(
[this](ffi::PackedArgs args, ffi::Any* rv) { *rv = model_->GetNumOutputs(); });
Expand All @@ -154,32 +155,19 @@
NSDictionary* json = [NSJSONSerialization JSONObjectWithData:data
options:NSJSONReadingAllowFragments
error:nil];
NSffi::Array<NSString*>* input_names = json[@"inputs"];
NSArray<NSString*>* input_names = json[@"inputs"];

// Copy input tensors to corresponding data entries.
for (auto i = 0; i < args.size() - 1; ++i) {
TVM_FFI_ICHECK(args[i].type_code() == kTVMDLTensorHandle ||
args[i].type_code() == kTVMTensorHandle)
<< "Expect Tensor or DLTensor as inputs\n";
if (args[i].type_code() == kTVMDLTensorHandle || args[i].type_code() == kTVMTensorHandle) {
model_->SetInput([input_names[i] UTF8String], args[i]);
} else {
LOG(FATAL) << "Not implemented";
}
model_->SetInput([input_names[i] UTF8String], args[i].cast<DLTensor*>());
}
Comment thread
tlopex marked this conversation as resolved.

// Execute the subgraph.
model_->Invoke();

// TODO: Support multiple outputs.
Tensor out = model_->GetOutput(0);
if (args[args.size() - 1].type_code() == kTVMDLTensorHandle) {
DLTensor* arg = args[args.size() - 1];
out.CopyTo(arg);
} else {
Tensor arg = args[args.size() - 1];
out.CopyTo(arg);
}
out.CopyTo(args[args.size() - 1].cast<DLTensor*>());
*rv = out;
});
} else {
Expand All @@ -196,7 +184,7 @@
TVM_FFI_STATIC_INIT_BLOCK() {
namespace refl = tvm::ffi::reflection;
refl::GlobalDef().def_packed("tvm.coreml_runtime.create", [](ffi::PackedArgs args, ffi::Any* rv) {
*rv = CoreMLRuntimeCreate(args[0], args[1]);
*rv = CoreMLRuntimeCreate(args[0].cast<std::string>(), args[1].cast<std::string>());
});
Comment thread
tlopex marked this conversation as resolved.
}

Expand Down
Loading