[llvm] r238932 - [Statepoints] Mark statepoint intrinsic with Throws attribute

Philip Reames listmail at philipreames.com
Wed Jun 3 13:30:56 PDT 2015


Igor, please add a test case for a statepoint wrapping a potential 
throwing call combined with an optimization pass like SimplifyCFG. Show 
that a invoke doesn't get folded to a call.

Sorry, should have caught this in the review.

On 06/03/2015 09:18 AM, Igor Laevsky wrote:
> Author: igor.laevsky
> Date: Wed Jun  3 11:18:58 2015
> New Revision: 238932
>
> URL: http://llvm.org/viewvc/llvm-project?rev=238932&view=rev
> Log:
> [Statepoints] Mark statepoint intrinsic with Throws attribute
>
> Differential Revision: http://reviews.llvm.org/D10215
>
>
> Modified:
>      llvm/trunk/include/llvm/IR/Intrinsics.td
>
> Modified: llvm/trunk/include/llvm/IR/Intrinsics.td
> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/include/llvm/IR/Intrinsics.td?rev=238932&r1=238931&r2=238932&view=diff
> ==============================================================================
> --- llvm/trunk/include/llvm/IR/Intrinsics.td (original)
> +++ llvm/trunk/include/llvm/IR/Intrinsics.td Wed Jun  3 11:18:58 2015
> @@ -537,7 +537,8 @@ def int_experimental_patchpoint_i64 : In
>   def int_experimental_gc_statepoint : Intrinsic<[llvm_i32_ty],
>                                  [llvm_i64_ty, llvm_i32_ty,
>                                   llvm_anyptr_ty, llvm_i32_ty,
> -                                llvm_i32_ty, llvm_vararg_ty]>;
> +                                llvm_i32_ty, llvm_vararg_ty],
> +                                [Throws]>;
>   
>   def int_experimental_gc_result   : Intrinsic<[llvm_any_ty], [llvm_i32_ty]>;
>   def int_experimental_gc_relocate : Intrinsic<[llvm_anyptr_ty],
>
>
> _______________________________________________
> 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