This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] PR optimization/12324
- From: "Joseph S. Myers" <jsm at polyomino dot org dot uk>
- To: Bernardo Innocenti <bernie at develer dot com>
- Cc: Jan Hubicka <hubicka at ucw dot cz>, Jan Hubicka <jh at suse dot cz>, gcc-patches at gcc dot gnu dot org, rth at redhat dot com
- Date: Fri, 3 Oct 2003 21:58:16 +0100 (BST)
- Subject: Re: [PATCH] PR optimization/12324
- References: <20030927135507.GG23075@kam.mff.cuni.cz><Pine.LNX.4.58.0309271654050.21283@kern.srcf.societies.cam.ac.uk><20030927183054.GA25269@atrey.karlin.mff.cuni.cz> <3F7D2FEC.5040200@develer.com>
On Fri, 3 Oct 2003, Bernardo Innocenti wrote:
> (const phy_cmd_t []) { /* config */
So here the initializer is a pointer to a compound literal - and that
compound literal would naturally get written out exactly once (in the
absence of being deferred). The problem case which I think is why Jakub
made them deferred is if the initializer is a compound literal of
structure - not array - type (permitted as an extension in gnu89 mode),
where it gets included in the larger initializer and shouldn't get emitted
on its own as well.
--
Joseph S. Myers
jsm@polyomino.org.uk