From 9c6a0dcf0e5ccc99e55ad9c1f70eede532004775 Mon Sep 17 00:00:00 2001 From: Chris Lattner Date: Thu, 26 Nov 2009 01:51:18 +0000 Subject: [PATCH] update some notes slightly git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89913 91177308-0d34-0410-b5e6-96231b3b80d8 --- lib/Target/README.txt | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/lib/Target/README.txt b/lib/Target/README.txt index aad621f440a..6d90f05a46e 100644 --- a/lib/Target/README.txt +++ b/lib/Target/README.txt @@ -220,7 +220,7 @@ so cool to turn it into something like: ... which would only do one 32-bit XOR per loop iteration instead of two. It would also be nice to recognize the reg->size doesn't alias reg->node[i], but -alas. +this requires TBAA. //===---------------------------------------------------------------------===// @@ -280,6 +280,9 @@ unsigned int popcount(unsigned int input) { return count; } +This is a form of idiom recognition for loops, the same thing that could be +useful for recognizing memset/memcpy. + //===---------------------------------------------------------------------===// These should turn into single 16-bit (unaligned?) loads on little/big endian @@ -343,7 +346,7 @@ PHI Slicing could be extended to do this. //===---------------------------------------------------------------------===// -LSR should know what GPR types a target has. This code: +LSR should know what GPR types a target has from TargetData. This code: volatile short X, Y; // globals @@ -369,7 +372,6 @@ LBB1_2: LSR should reuse the "+" IV for the exit test. - //===---------------------------------------------------------------------===// Tail call elim should be more aggressive, checking to see if the call is