diff mbox

[_eh,dawin,Version2] Allow targets to suppress epilogues in _eh frames.

Message ID 3F8A8575-5DB3-4FF1-9828-98C84A04E309@sandoe-acoustics.co.uk
State New
Headers show

Commit Message

Iain Sandoe Aug. 16, 2010, 7:45 a.m. UTC
On 16 Aug 2010, at 01:18, Jack Howarth wrote:

>   On x86_64-apple-darwin10, I am seeing the following new failures
> compard to your previous version of this patch...
>
> FAIL: g++.dg/eh/async-unwind1.C (test for excess errors)
> FAIL: g++.dg/eh/async-unwind2.C (test for excess errors)


Well, these were already discussed, (and g++.dg/eh/async-unwind2.C was  
failing with the previous version).

I was awaiting Mike's preference as to what to do in the case of a  
failure (which is no longer present).
attached update to the testsuite (just recognize the new warning).

cheers
Iain


testsuite:

	* g++.dg/eh/async-unwind1.C: Recognize warning on darwin.
	* g++.dg/eh/async-unwind2.C: Ditto.
diff mbox

Patch

Index: gcc/testsuite/g++.dg/eh/async-unwind2.C
===================================================================
--- gcc/testsuite/g++.dg/eh/async-unwind2.C	(revision 163276)
+++ gcc/testsuite/g++.dg/eh/async-unwind2.C	(working copy)
@@ -2,6 +2,7 @@ 
 // { dg-do run { target { { i?86-*-* x86_64-*-* } && ilp32 } } }
 // { dg-require-effective-target fpic }
 // { dg-options "-Os -fasynchronous-unwind-tables -fpic -fno-inline" }
+// { dg-warning "this architecture does not fully support" "" { target *-*-darwin* } 0 }
 
 #include <stdarg.h>
 
Index: gcc/testsuite/g++.dg/eh/async-unwind1.C
===================================================================
--- gcc/testsuite/g++.dg/eh/async-unwind1.C	(revision 163276)
+++ gcc/testsuite/g++.dg/eh/async-unwind1.C	(working copy)
@@ -1,6 +1,7 @@ 
 // PR rtl-optimization/36419
 // { dg-do run { target { { i?86-*-* x86_64-*-* } && ilp32 } } }
 // { dg-options "-Os -fasynchronous-unwind-tables -mpreferred-stack-boundary=4" }
+// { dg-warning "this architecture does not fully support" "" { target *-*-darwin* } 0 }
 
 extern "C" void abort ();