[PATCH] D19341: Add micro-optimization for 'icmp slt (or A, B), A' to instsimplify.

[Nick Lewycky via llvm-commits]

nlewycky marked 3 inline comments as done.

================
Comment at: lib/Analysis/InstructionSimplify.cpp:2630-2639
@@ +2629,12 @@
+        bool RHSKnownNonNegative, RHSKnownNegative;
+        bool YKnownNonNegative, YKnownNegative;
+        ComputeSignBit(RHS, RHSKnownNonNegative, RHSKnownNegative, Q.DL, 0,
+                       Q.AC, Q.CxtI, Q.DT);
+        ComputeSignBit(Y, YKnownNonNegative, YKnownNegative, Q.DL, 0, Q.AC,
+                       Q.CxtI, Q.DT);
+        if (RHSKnownNonNegative && YKnownNegative)
+          return Pred == ICmpInst::ICMP_SLT ? getTrue(ITy) : getFalse(ITy);
+        if (RHSKnownNegative || YKnownNonNegative)
+          return Pred == ICmpInst::ICMP_SLT ? getFalse(ITy) : getTrue(ITy);
+      }
+    }
----------------
Counterexample "(-2|1)<-2" --> "-1<-2" --> "false".

To sanity-check, I wrote a program that enumerates A=[-128..127], B=[-128..127] and emits all the results. That's what convinced me that only the sign bit matters.


http://reviews.llvm.org/D19341