[llvm-commits] [llvm] r92382 - /llvm/trunk/lib/Target/README.txt

[Chris Lattner]

Author: lattner
Date: Thu Dec 31 19:29:26 2009
New Revision: 92382

URL: http://llvm.org/viewvc/llvm-project?rev=92382&view=rev
Log:
update this.  To take the next step, llvm.powi should be generalized to work
on integers as well and codegen should lower them to branch trees.

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=92382&r1=92381&r2=92382&view=diff

==============================================================================
--- llvm/trunk/lib/Target/README.txt (original)
+++ llvm/trunk/lib/Target/README.txt Thu Dec 31 19:29:26 2009
@@ -106,7 +106,17 @@
 
 //===---------------------------------------------------------------------===//
 
-Reassociate should turn: X*X*X*X -> t=(X*X) (t*t) to eliminate a multiply.
+Reassociate should turn things like:
+
+int factorial(int X) {
+ return X*X*X*X*X*X*X*X;
+}
+
+into llvm.powi calls, allowing the code generator to produce balanced
+multiplication trees.
+
+First, the intrinsic needs to be extended to support integers, and second the
+code generator needs to be enhanced to lower these to multiplication trees.
 
 //===---------------------------------------------------------------------===//
 
@@ -119,7 +129,32 @@
   return bar(z, n) + bar(2*z, 2*n);
 }
 
-Reassociate should handle the example in GCC PR16157.
+This is blocked on not handling X*X*X -> powi(X, 3) (see note above).  The issue
+is that we end up getting t = 2*X  s = t*t   and don't turn this into 4*X*X,
+which is the same number of multiplies and is canonical, because the 2*X has
+multiple uses.  Here's a simple example:
+
+define i32 @test15(i32 %X1) {
+  %B = mul i32 %X1, 47   ; X1*47
+  %C = mul i32 %B, %B
+  ret i32 %C
+}
+
+
+//===---------------------------------------------------------------------===//
+
+Reassociate should handle the example in GCC PR16157:
+
+extern int a0, a1, a2, a3, a4; extern int b0, b1, b2, b3, b4; 
+void f () {  /* this can be optimized to four additions... */ 
+        b4 = a4 + a3 + a2 + a1 + a0; 
+        b3 = a3 + a2 + a1 + a0; 
+        b2 = a2 + a1 + a0; 
+        b1 = a1 + a0; 
+} 
+
+This requires reassociating to forms of expressions that are already available,
+something that reassoc doesn't think about yet.
 
 //===---------------------------------------------------------------------===//
 
@@ -721,17 +756,6 @@
 
 //===---------------------------------------------------------------------===//
 
-Reassociate should turn things like:
-
-int factorial(int X) {
- return X*X*X*X*X*X*X*X;
-}
-
-into llvm.powi calls, allowing the code generator to produce balanced
-multiplication trees.
-
-//===---------------------------------------------------------------------===//
-
 We generate a horrible  libcall for llvm.powi.  For example, we compile:
 
 #include <cmath>