commit 4271989815b5fc933c1e29bc75507c2726dc3738
Author: Julian Seward <jseward@acm.org>
Date:   Tue Nov 20 10:52:33 2018 +0100

    Add some new IROps to support improved Memcheck analysis of strlen etc.
    
    This is part of the fix for bug 386945.  It adds the following IROps, plus
    their supporting type- and printing- fragments:
    
    Iop_Reverse8sIn32_x1: 32-bit byteswap.  A fancy name, but it is consistent
    with naming for the other swapping IROps that already exist.
    
    Iop_PopCount64, Iop_PopCount32: population count
    
    Iop_ClzNat64, Iop_ClzNat32, Iop_CtzNat64, Iop_CtzNat32: counting leading and
    trailing zeroes, with "natural" (Nat) semantics for a zero input, meaning, in
    the case of zero input, return the number of bits in the word.  These
    functionally overlap with the existing Iop_Clz64, Iop_Clz32, Iop_Ctz64,
    Iop_Ctz32.  The existing operations are undefined in case of a zero input.
    Adding these new variants avoids the complexity of having to change the
    declared semantics of the existing operations.  Instead they are deprecated
    but still available for use.

diff --git a/VEX/priv/ir_defs.c b/VEX/priv/ir_defs.c
index 823b6be..3221033 100644
--- a/VEX/priv/ir_defs.c
+++ b/VEX/priv/ir_defs.c
@@ -194,6 +194,14 @@ void ppIROp ( IROp op )
       case Iop_Ctz64:    vex_printf("Ctz64"); return;
       case Iop_Ctz32:    vex_printf("Ctz32"); return;
 
+      case Iop_ClzNat64: vex_printf("ClzNat64"); return;
+      case Iop_ClzNat32: vex_printf("ClzNat32"); return;
+      case Iop_CtzNat64: vex_printf("CtzNat64"); return;
+      case Iop_CtzNat32: vex_printf("CtzNat32"); return;
+
+      case Iop_PopCount64: vex_printf("PopCount64"); return;
+      case Iop_PopCount32: vex_printf("PopCount32"); return;
+
       case Iop_CmpLT32S: vex_printf("CmpLT32S"); return;
       case Iop_CmpLE32S: vex_printf("CmpLE32S"); return;
       case Iop_CmpLT32U: vex_printf("CmpLT32U"); return;
@@ -395,6 +403,7 @@ void ppIROp ( IROp op )
 
       case Iop_CmpNEZ16x2: vex_printf("CmpNEZ16x2"); return;
       case Iop_CmpNEZ8x4:  vex_printf("CmpNEZ8x4"); return;
+      case Iop_Reverse8sIn32_x1: vex_printf("Reverse8sIn32_x1"); return;
 
       case Iop_CmpF64:    vex_printf("CmpF64"); return;
 
@@ -2719,6 +2728,7 @@ void typeOfPrimop ( IROp op,
          UNARY(Ity_I16, Ity_I16);
       case Iop_Not32:
       case Iop_CmpNEZ16x2: case Iop_CmpNEZ8x4:
+      case Iop_Reverse8sIn32_x1:
          UNARY(Ity_I32, Ity_I32);
 
       case Iop_Not64:
@@ -2782,9 +2792,13 @@ void typeOfPrimop ( IROp op,
          BINARY(Ity_I64,Ity_I64, Ity_I128);
 
       case Iop_Clz32: case Iop_Ctz32:
+      case Iop_ClzNat32: case Iop_CtzNat32:
+      case Iop_PopCount32:
          UNARY(Ity_I32, Ity_I32);
 
       case Iop_Clz64: case Iop_Ctz64:
+      case Iop_ClzNat64: case Iop_CtzNat64:
+      case Iop_PopCount64:
          UNARY(Ity_I64, Ity_I64);
 
       case Iop_DivU32: case Iop_DivS32: case Iop_DivU32E: case Iop_DivS32E:
diff --git a/VEX/pub/libvex_ir.h b/VEX/pub/libvex_ir.h
index 17bcb55..93fa5ac 100644
--- a/VEX/pub/libvex_ir.h
+++ b/VEX/pub/libvex_ir.h
@@ -452,12 +452,21 @@ typedef
       Iop_MullS8, Iop_MullS16, Iop_MullS32, Iop_MullS64,
       Iop_MullU8, Iop_MullU16, Iop_MullU32, Iop_MullU64,
 
-      /* Wierdo integer stuff */
+      /* Counting bits */
+      /* Ctz64/Ctz32/Clz64/Clz32 are UNDEFINED when given arguments of zero.
+         You must ensure they are never given a zero argument.  As of
+         2018-Nov-14 they are deprecated.  Try to use the Nat variants
+         immediately below, if you can.
+      */
       Iop_Clz64, Iop_Clz32,   /* count leading zeroes */
       Iop_Ctz64, Iop_Ctz32,   /* count trailing zeros */
-      /* Ctz64/Ctz32/Clz64/Clz32 are UNDEFINED when given arguments of
-         zero.  You must ensure they are never given a zero argument.
-      */
+      /* Count leading/trailing zeroes, with "natural" semantics for the
+         case where the input is zero: then the result is the number of bits
+         in the word. */
+      Iop_ClzNat64, Iop_ClzNat32,
+      Iop_CtzNat64, Iop_CtzNat32,
+      /* Population count -- compute the number of 1 bits in the argument. */
+      Iop_PopCount64, Iop_PopCount32,
 
       /* Standard integer comparisons */
       Iop_CmpLT32S, Iop_CmpLT64S,
@@ -831,6 +840,9 @@ typedef
       /* MISC (vector integer cmp != 0) */
       Iop_CmpNEZ16x2, Iop_CmpNEZ8x4,
 
+      /* Byte swap in a 32-bit word */
+      Iop_Reverse8sIn32_x1,
+
       /* ------------------ 64-bit SIMD FP ------------------------ */
 
       /* Convertion to/from int */
@@ -1034,8 +1046,9 @@ typedef
       Iop_Slice64,  // (I64, I64, I8) -> I64
 
       /* REVERSE the order of chunks in vector lanes.  Chunks must be
-         smaller than the vector lanes (obviously) and so may be 8-,
-         16- and 32-bit in size. */
+         smaller than the vector lanes (obviously) and so may be 8-, 16- and
+         32-bit in size.  Note that the degenerate case,
+         Iop_Reverse8sIn64_x1, is a simply a vanilla byte-swap. */
       /* Examples:
             Reverse8sIn16_x4([a,b,c,d,e,f,g,h]) = [b,a,d,c,f,e,h,g]
             Reverse8sIn32_x2([a,b,c,d,e,f,g,h]) = [d,c,b,a,h,g,f,e]