[llvm-commits] [llvm] r107565 - /llvm/trunk/lib/Target/README.txt

[Eli Friedman]

Author: efriedma
Date: Sat Jul  3 02:38:12 2010
New Revision: 107565

URL: http://llvm.org/viewvc/llvm-project?rev=107565&view=rev
Log:
Note switch-lowering inefficiency.


Modified:
    llvm/trunk/lib/Target/README.txt

Modified: llvm/trunk/lib/Target/README.txt
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/README.txt?rev=107565&r1=107564&r2=107565&view=diff
==============================================================================
--- llvm/trunk/lib/Target/README.txt (original)
+++ llvm/trunk/lib/Target/README.txt Sat Jul  3 02:38:12 2010
@@ -1888,3 +1888,39 @@
 (That said, we are doing a lot better than gcc on this testcase. :) )
 
 //===---------------------------------------------------------------------===//
+
+Switch lowering generates less than ideal code for the following switch:
+define void @a(i32 %x) nounwind {
+entry:
+  switch i32 %x, label %if.end [
+    i32 0, label %if.then
+    i32 1, label %if.then
+    i32 2, label %if.then
+    i32 3, label %if.then
+    i32 5, label %if.then
+  ]
+if.then:
+  tail call void @foo() nounwind
+  ret void
+if.end:
+  ret void
+}
+declare void @foo()
+
+Generated code on x86-64 (other platforms give similar results):
+a:
+	cmpl	$5, %edi
+	ja	.LBB0_2
+	movl	%edi, %eax
+	movl	$47, %ecx
+	btq	%rax, %rcx
+	jb	.LBB0_3
+.LBB0_2:
+	ret
+.LBB0_3:
+	xorb	%al, %al
+	jmp	foo at PLT  # TAILCALL
+
+The movl+movl+btq+jb could be simplified to a cmpl+jne.
+
+//===---------------------------------------------------------------------===//