This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [RFC/CFT] Hookize TARGET_UNWIND_INFO and related macros

From: Michael Eager <eager at eagerm dot com>
To: Richard Henderson <rth at redhat dot com>
Cc: Ulrich Weigand <uweigand at de dot ibm dot com>, GCC Patches <gcc-patches at gcc dot gnu dot org>, joseph at codesourcery dot com, sje at cup dot hp dot com, richard dot earnshaw at arm dot com, ktietz70 at googlemail dot com
Date: Tue, 19 Oct 2010 10:08:50 -0700
Subject: Re: [RFC/CFT] Hookize TARGET_UNWIND_INFO and related macros
References: <201009300820.o8U8KUKX006458@d12av02.megacenter.de.ibm.com> <4CA4AE7F.8060106@redhat.com>

Richard Henderson wrote:

On 09/30/2010 01:20 AM, Ulrich Weigand wrote:

The problem is that we define INCOMING_RETURN_ADDR_RTX to get correct
.debug_frame DWARF2 debugging info.  However, we do not support run-time
unwinding using DWARF2 CFI, and do not define EH_RETURN_DATA_REGNO and
related macros.

This caused MUST_USE_SJLJ_EXCEPTIONS to be set in the old logic...

Interesting. I missed that side effect.

	* config/spu/spu.c (TARGET_EXCEPT_UNWIND_INFO): Set to
	sjlj_except_unwind_info.


Yes, this is exactly what I intended.  Eventually I'd like all
targets to explicitly use dwarf2, sjlj, or their own target hook
function.  The messy logic of default should go away.

Similarly:


	* config/microblaze/microblaze.c (TARGET_EXCEPT_UNWIND_INFO):
	Use sjlj unwind info for exceptions.

Checked in, revision 165701


--
Michael Eager	 eager@eagercon.com
1960 Park Blvd., Palo Alto, CA 94306  650-325-8077

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]