This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
RE: [patch 4/4] -fstrict-volatile-bitfields cleanup v3: remove from defaults on all targets
- From: Bernd Edlinger <bernd dot edlinger at hotmail dot de>
- To: DJ Delorie <dj at redhat dot com>
- Cc: "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>
- Date: Wed, 4 Sep 2013 15:45:48 +0200
- Subject: RE: [patch 4/4] -fstrict-volatile-bitfields cleanup v3: remove from defaults on all targets
- Authentication-results: sourceware.org; auth=none
- References: <DUB122-W35943644A5D83EC2CAFDB7E4370 at phx dot gbl>,<52245D01 dot 1030702 at arm dot com>,<Pine dot LNX dot 4 dot 64 dot 1309021600390 dot 17654 at digraph dot polyomino dot org dot uk>,<CAFiYyc0he723Zn7jhpXxZCuhL+=KkmDU5FWHvjDpC4juJcsvvg at mail dot gmail dot com>,<DUB122-W24F7F5549EDAC48D8889B0E4310 at phx dot gbl>,<Pine dot LNX dot 4 dot 64 dot 1309032118460 dot 27960 at digraph dot polyomino dot org dot uk> <DUB122-W47E9398116D13B0436C57EE4320 at phx dot gbl>,<201309041329 dot r84DT2gZ023416 at greed dot delorie dot com>
On Wed, 4 Sep 2013 09:29:02, DJ Delorie wrote:
>
>> I fully agree with you, the current default state of
>> -fstrict-volatile-bitfields should be disabled for all targets.
>
> Please don't do that until gcc produces code that does the same
> things. Most of my targets rely on gcc not doing the old behavior, to
> generate correct code.
>
>> For portability of application code, the default should be always
>> off, unless specifically requested.
>
> The vendors of my targets specificially requested it be the default.
>
>> Even driver code rarely uses bit-fields for register access,
>
> People keep saying this, and people are wrong. For the targets I
> support, they *all* use bitfields for *all* the peripherals, because
> that's what's in the headers the vendor ships with each chip.
I really do respect your customer's requirements.
I do not even think about changing the default state of
-fstrict-volatile-bitfields on gcc 4.7.x or gcc 4.8.x, although
I hope that the other parts of this patch will make their way
to these old versions too.
How about this for a compromise: Let's make the
default of -fstrict-volatile-bitfields an optional
configure argument for gcc 4.9, that can be used for every
target, even for X86_64 of you want ?
Regards
Bernd.