[llvm-commits] [llvm] r56247 - in /llvm/trunk: lib/Analysis/ScalarEvolution.cpp test/Analysis/ScalarEvolution/2008-02-11-ReversedCondition.ll test/Analysis/ScalarEvolution/avoid-smax-1.ll

Evan Cheng evan.cheng at apple.com
Tue Sep 16 12:55:08 PDT 2008


Woohoo! Thanks Dan.

Evan

On Sep 16, 2008, at 11:52 AM, Dan Gohman wrote:

> Author: djg
> Date: Tue Sep 16 13:52:57 2008
> New Revision: 56247
>
> URL: http://llvm.org/viewvc/llvm-project?rev=56247&view=rev
> Log:
> Finally re-apply r46959. This is made feasible by the combination
> of r56230, r56232, and r56246.
>
> Modified:
>    llvm/trunk/lib/Analysis/ScalarEvolution.cpp
>    llvm/trunk/test/Analysis/ScalarEvolution/2008-02-11- 
> ReversedCondition.ll
>    llvm/trunk/test/Analysis/ScalarEvolution/avoid-smax-1.ll
>
> Modified: llvm/trunk/lib/Analysis/ScalarEvolution.cpp
> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Analysis/ScalarEvolution.cpp?rev=56247&r1=56246&r2=56247&view=diff
>
> = 
> = 
> = 
> = 
> = 
> = 
> = 
> = 
> ======================================================================
> --- llvm/trunk/lib/Analysis/ScalarEvolution.cpp (original)
> +++ llvm/trunk/lib/Analysis/ScalarEvolution.cpp Tue Sep 16 13:52:57  
> 2008
> @@ -1981,8 +1981,8 @@
>
>   // At this point, we would like to compute how many iterations of  
> the
>   // loop the predicate will return true for these inputs.
> -  if (isa<SCEVConstant>(LHS) && !isa<SCEVConstant>(RHS)) {
> -    // If there is a constant, force it into the RHS.
> +  if (LHS->isLoopInvariant(L) && !RHS->isLoopInvariant(L)) {
> +    // If there is a loop-invariant, force it into the RHS.
>     std::swap(LHS, RHS);
>     Cond = ICmpInst::getSwappedPredicate(Cond);
>   }
>
> Modified: llvm/trunk/test/Analysis/ScalarEvolution/2008-02-11- 
> ReversedCondition.ll
> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Analysis/ScalarEvolution/2008-02-11-ReversedCondition.ll?rev=56247&r1=56246&r2=56247&view=diff
>
> = 
> = 
> = 
> = 
> = 
> = 
> = 
> = 
> ======================================================================
> --- llvm/trunk/test/Analysis/ScalarEvolution/2008-02-11- 
> ReversedCondition.ll (original)
> +++ llvm/trunk/test/Analysis/ScalarEvolution/2008-02-11- 
> ReversedCondition.ll Tue Sep 16 13:52:57 2008
> @@ -1,5 +1,4 @@
> ; RUN: llvm-as < %s | opt -scalar-evolution -analyze | grep {Loop  
> header: (0 smax %n) iterations!}
> -; XFAIL: *
>
> define void @foo(i32 %n) {
> entry:
>
> Modified: llvm/trunk/test/Analysis/ScalarEvolution/avoid-smax-1.ll
> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Analysis/ScalarEvolution/avoid-smax-1.ll?rev=56247&r1=56246&r2=56247&view=diff
>
> = 
> = 
> = 
> = 
> = 
> = 
> = 
> = 
> ======================================================================
> --- llvm/trunk/test/Analysis/ScalarEvolution/avoid-smax-1.ll  
> (original)
> +++ llvm/trunk/test/Analysis/ScalarEvolution/avoid-smax-1.ll Tue Sep  
> 16 13:52:57 2008
> @@ -1,11 +1,13 @@
> ; RUN: llvm-as < %s | opt -indvars | llvm-dis > %t
> -; RUN: not grep select %t
> +; RUN: grep select %t | count 2
> ; RUN: grep {icmp ne i32.\* %w	} %t
>
> ; Indvars should be able to insert a canonical induction variable
> ; for the bb6 loop without using a maximum calculation (icmp, select)
> ; because it should be able to prove that the comparison is guarded
> -; by an appropriate conditional branch.
> +; by an appropriate conditional branch. Unfortunately, indvars is
> +; not yet able to find the comparison for the other too loops in
> +; this testcase.
>
> target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32- 
> i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64- 
> f80:128:128"
> target triple = "i386-apple-darwin9"
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits




More information about the llvm-commits mailing list