[llvm-commits] [llvm] r94726 - in /llvm/trunk: docs/ExceptionHandling.html include/llvm/CodeGen/MachineModuleInfo.h include/llvm/Intrinsics.td lib/CodeGen/AsmPrinter/DwarfException.cpp lib/CodeGen/MachineModuleInfo.cpp lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp lib/CodeGen/SjLjEHPrepare.cpp

[Jim Grosbach]

On Jan 28, 2010, at 9:39 AM, Eric Christopher wrote:

> 
> On Jan 28, 2010, at 8:16 AM, Jim Grosbach wrote:
> 
>> That said, there's no theoretical reason it couldn't be, which you're right, is a bad thing, and is why I shied away from doing it this way the first time around. At this point, though, I think it's the lesser of two evils.
> 
> I think I'd rather put it on the invoke instruction than have another builtin like this.

I like that, too. Two things made me hesitate to go that route initially. First, it's a bitcode change that's not backward compatible, which is a bit of a pain. Second, it'd be used only for SJLJ. The DWARF EH stuff doesn't need it, and I've been trying to keep the sjlj stuff from impacting anything to do with DWARF EH as much as possible. Now, I don't know that either of those concerns are deal-breakers. If there's consensus that it's worth it, I'll gladly change to that approach.

-Jim