DescriptionSubzero: Implementation of "advanced Phi lowering".
Delays Phi lowering until after register allocation. This lets the Phi assignment order take register allocation into account and avoid creating false dependencies.
All edges that lead to Phi instructions are split, and the new node gets mov instructions in the correct topological order, using available physical registers as needed.
This lowering style is controllable under -O2 using -phi-edge-split (enabled by default).
The result is faster translation time (due to fewer temporaries leading to faster liveness analysis and register allocation) as well as better code quality (due to better register allocation and fewer phi-based assignments).
BUG= none
R=jvoung@chromium.org
Committed: https://gerrit.chromium.org/gerrit/gitweb?p=native_client/pnacl-subzero.git;a=commit;h=336f6c4a90e7f5ee156bc9a339d80c7def5ddeff
Patch Set 1 #Patch Set 2 : Fix lit tests #Patch Set 3 : Clean up lowerPhiAssignments() #Patch Set 4 : Minor cleanup #
Total comments: 35
Patch Set 5 : Code review changes #
Total comments: 14
Patch Set 6 : Code review round 2 #Patch Set 7 : Refactor the mem=mem pattern during phi lowering #
Total comments: 4
Patch Set 8 : Fix vector const undef lowering for phis. #
Created: 6 years, 1 month ago
Messages
Total messages: 14 (1 generated)
|