[llvm] r274753 - Code size optimisation: don't rewrite fputs to fwrite when optimising for size

[Benjamin Kramer via llvm-commits]

On Thu, Jul 7, 2016 at 3:56 PM, Sjoerd Meijer via llvm-commits
<llvm-commits at lists.llvm.org> wrote:
> Author: sjoerdmeijer
> Date: Thu Jul  7 08:56:23 2016
> New Revision: 274753
>
> URL: http://llvm.org/viewvc/llvm-project?rev=274753&view=rev
> Log:
> Code size optimisation: don't rewrite fputs to fwrite when optimising for size
> because fwrite requires more arguments and thus extra MOVs are required.
>
> Added:
>     llvm/trunk/test/Transforms/InstCombine/fputs-opt-size.ll
> Modified:
>     llvm/trunk/lib/Transforms/Utils/SimplifyLibCalls.cpp
>
> Modified: llvm/trunk/lib/Transforms/Utils/SimplifyLibCalls.cpp
> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/Utils/SimplifyLibCalls.cpp?rev=274753&r1=274752&r2=274753&view=diff
> ==============================================================================
> --- llvm/trunk/lib/Transforms/Utils/SimplifyLibCalls.cpp (original)
> +++ llvm/trunk/lib/Transforms/Utils/SimplifyLibCalls.cpp Thu Jul  7 08:56:23 2016
> @@ -2056,6 +2056,8 @@ Value *LibCallSimplifier::optimizeCall(C
>      case LibFunc::fwrite:
>        return optimizeFWrite(CI, Builder);
>      case LibFunc::fputs:
> +      if (CI->getParent()->getParent()->optForSize())
> +        return nullptr;

Why not do this check inside of optimizeFPuts? There should also be a
comment on why we don't want to optimize it when targeting minimum
size, i.e. replicate your commit message in the source code.

>        return optimizeFPuts(CI, Builder);
>      case LibFunc::log:
>      case LibFunc::log10:
>
> Added: llvm/trunk/test/Transforms/InstCombine/fputs-opt-size.ll
> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/InstCombine/fputs-opt-size.ll?rev=274753&view=auto
> ==============================================================================
> --- llvm/trunk/test/Transforms/InstCombine/fputs-opt-size.ll (added)
> +++ llvm/trunk/test/Transforms/InstCombine/fputs-opt-size.ll Thu Jul  7 08:56:23 2016
> @@ -0,0 +1,28 @@
> +; When optimising for size, we don't want to rewrite fputs to fwrite
> +; because it requires more arguments and thus extra MOVs are required.
> +;
> +; RUN: opt < %s -instcombine -S | FileCheck %s
> +
> +%struct._IO_FILE = type { i32, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, %struct._IO_marker*, %struct._IO_FILE*, i32, i32, i32, i16, i8, [1 x i8], i8*, i64, i8*, i8*, i8*, i8*, i32, i32, [40 x i8] }
> +%struct._IO_marker = type { %struct._IO_marker*, %struct._IO_FILE*, i32 }
> +
> + at .str = private unnamed_addr constant [10 x i8] c"mylog.txt\00", align 1
> + at .str.1 = private unnamed_addr constant [2 x i8] c"a\00", align 1
> + at .str.2 = private unnamed_addr constant [27 x i8] c"Hello world this is a test\00", align 1
> +
> +define i32 @main() local_unnamed_addr #0 {
> +entry:
> +; CHECK-LABEL: @main(
> +; CHECK-NOT: call i64 @fwrite
> +; CHECK: call i32 @fputs
> +
> +  %call = tail call %struct._IO_FILE* @fopen(i8* getelementptr inbounds ([10 x i8], [10 x i8]* @.str, i32 0, i32 0), i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.str.1, i32 0, i32 0)) #2
> +  %call1 = tail call i32 @fputs(i8* getelementptr inbounds ([27 x i8], [27 x i8]* @.str.2, i32 0, i32 0), %struct._IO_FILE* %call) #2
> +  ret i32 0
> +}
> +
> +declare noalias %struct._IO_FILE* @fopen(i8* nocapture readonly, i8* nocapture readonly) local_unnamed_addr #1
> +declare i32 @fputs(i8* nocapture readonly, %struct._IO_FILE* nocapture) local_unnamed_addr #1
> +
> +attributes #0 = { nounwind optsize }
> +attributes #1 = { nounwind optsize  }
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits