From a469b69ddae4665aec02fd75a51c31ad1cff043d Mon Sep 17 00:00:00 2001 From: Dale Johannesen Date: Fri, 18 May 2007 18:46:40 +0000 Subject: Document an inefficiency in tail merging. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@37235 91177308-0d34-0410-b5e6-96231b3b80d8 --- lib/CodeGen/README.txt | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) (limited to 'lib/CodeGen/README.txt') diff --git a/lib/CodeGen/README.txt b/lib/CodeGen/README.txt index 30fc27b015..419f885fee 100644 --- a/lib/CodeGen/README.txt +++ b/lib/CodeGen/README.txt @@ -142,3 +142,22 @@ load [T + 4] load [T + 7] ... load [T + 15] +//===---------------------------------------------------------------------===// +Tail merging issue: +When we're trying to merge the tails of predecessors of a block I, and there +are more than 2 predecessors, we don't do it optimally. Suppose predecessors +are A,B,C where B and C have 5 instructions in common, and A has 2 in common +with B or C. We want to get: +A: + jmp C3 +B: + jmp C2 +C: +C2: 3 common to B and C but not A +C3: 2 common to all 3 +You get this if B and C are merged first, but currently it might randomly decide +to merge A and B first, which results in not sharing the C2 instructions. We +could look at all N*(N-1) combinations of predecessors and merge the ones with +the most instructions in common first. Usually that will be fast, but it +could get slow on big graphs (e.g. large switches tend to have blocks with many +predecessors). -- cgit v1.2.3