This is an automated email from the ASF dual-hosted git repository.
github-bot pushed a change to branch nightly
in repository https://gitbox.apache.org/repos/asf/tvm.git
from 49e519ecef Bump cutlass_fpA_intB_gemm to latest commit (#18193)
add 5bcc7f5b6e [REFACTOR] Update data type rewriter to enable recursive
rewrite in Any (#18197)
add d6c3deaf36 [Relax][ONNX] Parse ONNX Upsample to Relax resize2d (#18180)
add 532db3392a [TIR] Fix host/device function check for build (#18199)
add 0ab1351c9d [FFI][REFACTOR] Move Downcast out of ffi for now (#18198)
No new revisions were added by this update.
Summary of changes:
ffi/include/tvm/ffi/any.h | 1 -
ffi/include/tvm/ffi/cast.h | 94 +-----------------------
ffi/tests/cpp/test_string.cc | 3 +-
{ffi/include/tvm/ffi => include/tvm/node}/cast.h | 79 ++++----------------
include/tvm/node/node.h | 3 +-
include/tvm/runtime/disco/session.h | 15 +++-
include/tvm/runtime/object.h | 1 -
include/tvm/runtime/vm/vm.h | 8 +-
python/tvm/relax/frontend/onnx/onnx_frontend.py | 31 +++++++-
python/tvm/tir/build.py | 16 ++--
src/node/container_printing.cc | 3 +-
src/node/repr_printer.cc | 1 +
src/runtime/profiling.cc | 10 +--
src/runtime/vm/attn_backend.cc | 28 +++----
src/runtime/vm/cuda/cuda_graph_builtin.cc | 6 +-
src/runtime/vm/vm.cc | 6 +-
src/target/source/codegen_metal.cc | 2 +-
src/tir/ir/data_type_rewriter.cc | 11 ++-
src/tir/transforms/memhammer_coalesce.cc | 4 +-
tests/python/codegen/test_target_codegen_cuda.py | 10 ++-
20 files changed, 119 insertions(+), 213 deletions(-)
copy {ffi/include/tvm/ffi => include/tvm/node}/cast.h (54%)
