[PATCH] D52286: [Intrinsic] Signed Saturation Intirnsic

[Bevin Hansson via Phabricator via llvm-commits]

ebevhan added a comment.

Here is a module with the function right after expansion. This is generated for x86 and not our target so both this IR and the result will probably be slightly different than what I posted.

  target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
  target triple = "x86_64-unknown-linux-gnu"
  
  ; Function Attrs: nounwind uwtable
  define signext i16 @ac(i24 signext %a, i32 %n) #0 {
  entry:
    %a.addr = alloca i24, align 4
    %n.addr = alloca i32, align 4
    %s = alloca i24, align 4
    %i = alloca i32, align 4
    store i24 %a, i24* %a.addr, align 4
    store i32 %n, i32* %n.addr, align 4
    %0 = load i24, i24* %a.addr, align 4
    %cmp = icmp slt i24 %0, 0
    br i1 %cmp, label %cond.true, label %cond.false
  
  cond.true:                                        ; preds = %entry
    br label %cond.end
  
  cond.false:                                       ; preds = %entry
    %1 = load i24, i24* %a.addr, align 4
    br label %cond.end
  
  cond.end:                                         ; preds = %cond.false, %cond.true
    %cond = phi i24 [ 0, %cond.true ], [ %1, %cond.false ]
    store i24 %cond, i24* %a.addr, align 4
    store i24 0, i24* %s, align 4
    store i32 0, i32* %i, align 4
    br label %for.cond
  
  for.cond:                                         ; preds = %for.inc, %cond.end
    %2 = load i32, i32* %i, align 4
    %3 = load i32, i32* %n.addr, align 4
    %cmp1 = icmp ult i32 %2, %3
    br i1 %cmp1, label %for.body, label %for.end
  
  for.body:                                         ; preds = %for.cond
    %4 = load i24, i24* %a.addr, align 4
    %too_large = icmp sgt i24 %4, 32767
    %too_small = icmp slt i24 %4, -32768
    %5 = select i1 %too_small, i24 -32768, i24 %4
    %6 = select i1 %too_large, i24 32767, i24 %5
    %resize = trunc i24 %6 to i16
    %resize2 = sext i16 %resize to i24
    %7 = load i24, i24* %s, align 4
    %add = add i24 %7, %resize2
    store i24 %add, i24* %s, align 4
    br label %for.inc
  
  for.inc:                                          ; preds = %for.body
    %8 = load i32, i32* %i, align 4
    %inc = add i32 %8, 1
    store i32 %inc, i32* %i, align 4
    br label %for.cond
  
  for.end:                                          ; preds = %for.cond
    %9 = load i24, i24* %s, align 4
    %too_large4 = icmp sgt i24 %9, 32767
    %too_small5 = icmp slt i24 %9, -32768
    %10 = select i1 %too_small5, i24 -32768, i24 %9
    %11 = select i1 %too_large4, i24 32767, i24 %10
    %resize3 = trunc i24 %11 to i16
    ret i16 %resize3
  }
  
  attributes #0 = { nounwind uwtable "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+fxsr,+mmx,+sse,+sse2,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" }


Repository:
  rL LLVM

https://reviews.llvm.org/D52286