You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@tvm.apache.org by gi...@apache.org on 2024/03/05 05:03:02 UTC

(tvm) branch nightly updated (ad1da4ee57 -> 46aaf61119)

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 ad1da4ee57 [Runtime][Builtin] Using float32 accumulation in attention kernel (#16667)
     add ae2ab58ad6 [KVCache] Fix the reference counter in sequence fork (#16666)
     add 31bb4b58fe [TVMScript] Infer T.reads() for DeclBuffer nodes (#16663)
     add 880af308ee Simplify Windows CMake Command (#16656)
     add 46aaf61119 [BugFix] add the default value for DFT in ONNX frontend (#16659)

No new revisions were added by this update.

Summary of changes:
 cmake/config.cmake                                 |   6 +
 docs/install/from_source.rst                       |   2 +-
 python/tvm/relay/frontend/onnx.py                  |   6 +-
 src/runtime/relax_vm/paged_kv_cache.cc             |   1 +
 src/tir/ir/script/script_complete.cc               |  11 +
 tests/python/frontend/onnx/test_forward.py         |   2 +-
 ...runtime_builtin_paged_attention_kv_cache_tir.py | 290 +++++++++++++--------
 tests/python/tvmscript/test_tvmscript_complete.py  |  36 ++-
 8 files changed, 227 insertions(+), 127 deletions(-)