[llvm-dev] GEP index canonicalization

[Manuel Jacob via llvm-dev]

Hi,

InstCombine canonicalizes index operands (unless they are into struct 
types) to pointer size.  The comment says: "If we are using a wider 
index than needed for this platform, shrink it to what we need.  If 
narrower, sign-extend it to what we need.  This explicit cast can make 
subsequent optimizations more obvious.".

For our architecture, the canonicalization is a bit problematic.  For 
example, our load operation can take any width and will implicitly 
sign-extend or truncate it.  The extra explicit cast however will show 
up as an extra operation in the machine code.  It is of course easy to 
eliminate the cast in a peephole optimization in the backend.

More interesting is the effect of this canonicalization on subsequent 
transformations.  Which optimizations are more obvious now, as the 
comment says?  I found examples where it enables IndVarSimplify to 
promote an index variable to pointer size.  However that introduces 
truncations, which can't be optimized away by simple peephole 
optimizations in the backend anymore.

Does it make sense to add a target hook for this?

-Manuel


Example:

define void @foo(i32 %n, i32* %a) {
entry:
   %cmp1 = icmp slt i32 0, %n
   br i1 %cmp1, label %for.body, label %for.end

for.body:                                         ; preds = %for.body, 
%entry
   %i = phi i32 [ %inc, %for.body ], [ 0, %entry ]
   %ptr = getelementptr inbounds i32, i32* %a, i32 %i
   store i32 %i, i32* %ptr, align 4
   %inc = add nsw i32 %i, 1
   %cmp = icmp slt i32 %inc, %n
   br i1 %cmp, label %for.body, label %for.end

for.end:                                          ; preds = %for.body, 
%entry
   ret void
}


InstCombine introduces a sext instruction:

define void @foo(i32 %n, i32* %a) {
entry:
   %cmp1 = icmp sgt i32 %n, 0
   br i1 %cmp1, label %for.body, label %for.end

for.body:                                         ; preds = %for.body, 
%entry
   %i = phi i32 [ %inc, %for.body ], [ 0, %entry ]
   %0 = sext i32 %i to i64
   %ptr = getelementptr inbounds i32, i32* %a, i64 %0
   store i32 %i, i32* %ptr, align 4
   %inc = add nsw i32 %i, 1
   %cmp = icmp slt i32 %inc, %n
   br i1 %cmp, label %for.body, label %for.end

for.end:                                          ; preds = %for.body, 
%entry
   ret void
}


IndVarSimplify promotes %i to i64, requiring two additional truncs:

define void @foo(i32 %n, i32* %a) {
entry:
   %cmp1 = icmp sgt i32 %n, 0
   br i1 %cmp1, label %for.body.preheader, label %for.end

for.body.preheader:                               ; preds = %entry
   br label %for.body

for.body:                                         ; preds = 
%for.body.preheader, %for.body
   %indvars.iv = phi i64 [ 0, %for.body.preheader ], [ %indvars.iv.next, 
%for.body ]
   %ptr = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
   %0 = trunc i64 %indvars.iv to i32
   store i32 %0, i32* %ptr, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
   %exitcond = icmp ne i32 %lftr.wideiv, %n
   br i1 %exitcond, label %for.body, label %for.end.loopexit

for.end.loopexit:                                 ; preds = %for.body
   br label %for.end

for.end:                                          ; preds = 
%for.end.loopexit, %entry
   ret void
}