Skip to content
Permalink
Browse files

Document the precomputation algorithm's purpose

  • Loading branch information...
oli-obk committed Mar 12, 2019
1 parent 1ae1312 commit 2a1eb1cef1a36c6f0a9d2e347529561c1293044e
Showing with 7 additions and 0 deletions.
  1. +7 −0 src/librustc_mir/interpret/memory.rs
@@ -791,6 +791,13 @@ impl<'a, 'mir, 'tcx, M: Machine<'a, 'mir, 'tcx>> Memory<'a, 'mir, 'tcx, M> {

let undef_mask = &self.get(src.alloc_id)?.undef_mask;

// Since we are copying `size` bytes from `src` to `dest + i * size` (`for i in 0..repeat`),
// a naive undef mask copying algorithm would repeatedly have to read the undef mask from
// the source and write it to the destination. Even if we optimized the memory accesses,
// we'd be doing all of this `repeat` times.
// Therefor we precompute a compressed version of the undef mask of the source value and
// then write it back `repeat` times without computing any more information from the source.

// a precomputed cache for ranges of defined/undefined bits
// 0000010010001110 will become
// [5, 1, 2, 1, 3, 3, 1]

0 comments on commit 2a1eb1c

Please sign in to comment.
You can’t perform that action at this time.