This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
RE: Memory corruption due to word sharing
- From: Bernd Petrovitsch <bernd at petrovitsch dot priv dot at>
- To: "Boehm, Hans" <hans dot boehm at hp dot com>
- Cc: Linus Torvalds <torvalds at linux-foundation dot org>, Torvald Riegel <triegel at redhat dot com>, Jan Kara <jack at suse dot cz>, LKML <linux-kernel at vger dot kernel dot org>, "linux-ia64 at vger dot kernel dot org" <linux-ia64 at vger dot kernel dot org>, "dsterba at suse dot cz" <dsterba at suse dot cz>, "ptesarik at suse dot cz" <ptesarik at suse dot cz>, "rguenther at suse dot de" <rguenther at suse dot de>, "gcc at gcc dot gnu dot org" <gcc at gcc dot gnu dot org>
- Date: Thu, 02 Feb 2012 10:28:42 +0100
- Subject: RE: Memory corruption due to word sharing
- References: <20120201151918.GC16714@quack.suse.cz> <CA+55aFy55Q=+pFCZcS9cOM6SL+ZT3sNDB+c4qFvVqwwSpTqJ7g@mail.gmail.com> <1328118174.15992.6206.camel@triegel.csb> <CA+55aFw=z2SKFGoSE5e_0ZmKcAAjK5q0DBM4PaxBH2D9tuikzA@mail.gmail.com> <A3E67C2071F49C4CBC4F17E6D77CDDD232754151@G4W3299.americas.hpqcorp.net>
On Mit, 2012-02-01 at 21:04 +0000, Boehm, Hans wrote:
[...]
> The C11 memory model potentially adds overhead in only two cases:
>
> 1. When current code involves touching a field that wouldn't otherwise
> be touched. There are odd cases in which this measurably slows down
> code, but I think all agree that we need it. In addition to
> bitfields, it can affect speculatively promoting a value to a register
> in a loop, which at least older versions of gcc also do.
Just adding an -f option for this and/or activating it only for -O5 (or
whatever the highest level is) and - in case that feature is activated -
emit warnings if bitfields (and/or any other data types that might be
affected)?
Kind regards,
Bernd
--
Bernd Petrovitsch Email : bernd@petrovitsch.priv.at
LUGA : http://www.luga.at