[llvm-commits] [llvm] r49006 - in /llvm/trunk/lib: CodeGen/DwarfWriter.cpp CodeGen/LLVMTargetMachine.cpp CodeGen/SelectionDAG/SelectionDAGISel.cpp Target/PowerPC/PPCAsmPrinter.cpp Target/PowerPC/PPCRegisterInfo.cpp Target/X86/X86AsmPrinter.cpp Target/X86/X86RegisterInfo.cpp Transforms/Utils/LowerInvoke.cpp

Chris Lattner clattner at apple.com
Tue Apr 1 16:51:05 PDT 2008


On Apr 1, 2008, at 2:44 PM, Dale Johannesen wrote:

>
>
> On Apr 1, 2008, at 4:49 AM, Duncan Sands wrote:
>
>> the dwarf info has two parts related to
>> exception handling: the eh table, and frame moves info.  If
>> a function is nounwind then there is no need to generate the
>> frame moves info.  If a function has no invokes then there
>> is no need to generate the eh table.
>
> So in the testcase CodeGen/X86/2007-09-17-ObjcFrameEH.ll,
> there is no need to generate the .eh symbol the test is looking
> for.  Is that right?

Yep, from the commit message:

--
Objective-C was generating EH frame info like this:

"_-[NSString(local) isNullOrNil]".eh = 0
         .no_dead_strip  "_-[NSString(local) isNullOrNil]".eh

The ".eh" should be inside the quotes.
---

To make this test test the thing that we care about, I'd add a call to  
the body of the function.

-Chris



More information about the llvm-commits mailing list