[DOC] Clean up MLA internal type usage and docs - #460
Merged
Conversation
Allow the fused RoPE cache launcher to reuse one compiled artifact across token counts by making only token-dependent tensors use dynamic layout. Made-with: Cursor
Keep the public fused RoPE launcher focused on marking token-dependent tensor layouts dynamic before delegating to the internal JIT launcher. Made-with: Cursor
Normalize the TopK gating softmax register memref construction at the call site so it matches the generated binding signature without changing the shared primitive API. Made-with: Cursor
Prefer FlyDSL internal numeric/vector APIs in MLA and documentation while preserving the verified MLA performance profile. Made-with: Cursor
coderfeli
changed the title
omuhamma
pushed a commit
that referenced
this pull request
May 8, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
fx.*,ArithValue,Vector, andrange(..., init=...)patterns.const_expr(...)and local@flyc.jitdispatch around runtime helper branches.Test plan
python -m py_compile kernels/mla_fwd_decode_m16x8_fp8_fp8.pygit diff --checkFLYDSL_DUMP_IR=1 FLYDSL_RUNTIME_ENABLE_CACHE=0 PYTHONPATH=python:.MLA small-shape compile; verified ASM hashc97855f56b17874e1ae389cebba408aea055d06e03a909824f5a6048ec26b47dPYTHONPATH=python:.MLA perf/correctness shapes:b=1,c=128,b=4,c=2048,b=32,c=8192Made with Cursor