Skip to content

feat(speculative): let the MTP gate buy back exactness by dropping qmv_wide #1199

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
inureyes merged 2 commits into from
Aug 17, 2026
Merged

feat(speculative): let the MTP gate buy back exactness by dropping qmv_wide #1199

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6adf9c5
perf(speculative): add an MLX overlay switch that restores block-vs-c…
inureyes Aug 17, 2026
e161fdf
feat(speculative): let the MTP gate buy back exactness by dropping qm…
inureyes Aug 17, 2026
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
7 changes: 6 additions & 1 deletion src/lib/mlx-cpp/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,9 +29,14 @@ function(mlx_apply_source_overlays mlx_source_dir)
# The steel/gemm/mma.h overlay (issue #217: upstream safe-load fix
# cherry-pick, PRs #3560/#3565) was retired when the pin moved to
# 2026-06-11 upstream main, which carries the fix natively (issue #222).
# quantized.cpp carries one delta, the MLXCEL_QMV_WIDE off-switch on
# `use_qmv_wide` (issue #1187). Upstream has no equivalent knob and the
# predicate is unchanged as of pin 9a795735, so it has to be overlaid here.
# Refresh the file wholesale on a bump and re-apply that one hunk.
set(_metal_patch_files
"mlx/backend/metal/compiled.cpp"
"mlx/backend/metal/kernels/utils.h")
"mlx/backend/metal/kernels/utils.h"
"mlx/backend/metal/quantized.cpp")
foreach(_patch_file ${_metal_patch_files})
if(EXISTS "${CMAKE_CURRENT_SOURCE_DIR}/patches/${_patch_file}")
configure_file(
Expand Down
Loading