This is now done.
authorDan Gohman <gohman@apple.com>
Mon, 1 Mar 2010 17:43:57 +0000 (17:43 +0000)
committerDan Gohman <gohman@apple.com>
Mon, 1 Mar 2010 17:43:57 +0000 (17:43 +0000)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97450 91177308-0d34-0410-b5e6-96231b3b80d8

lib/Target/X86/README.txt

index 3c6138bb1e6a64089439a068c328c0e4834b5cc3..d4545a6fcfd37150820e69ae6bb268a45c2d7e4d 100644 (file)
@@ -227,11 +227,6 @@ lambda, siod, optimizer-eval, ackermann, hash2, nestedloop, strcat, and Treesor.
 
 //===---------------------------------------------------------------------===//
 
-Teach the coalescer to coalesce vregs of different register classes. e.g. FR32 /
-FR64 to VR128.
-
-//===---------------------------------------------------------------------===//
-
 Adding to the list of cmp / test poor codegen issues:
 
 int test(__m128 *A, __m128 *B) {