This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Work in progress: "Super Sib Calls"; opinions sought
- From: Richard Henderson <rth at redhat dot com>
- To: Jeff Law <law at redhat dot com>
- Cc: Andreas Bauer <baueran at in dot tum dot de>, gcc at gcc dot gnu dot org, pizka at in dot tum dot de, jason dot ozolins at anu dot edu dot au
- Date: Tue, 10 Sep 2002 11:03:55 -0700
- Subject: Re: Work in progress: "Super Sib Calls"; opinions sought
- References: <20020910063439.GA18565@redhat.com> <200209101715.g8AHFCC10633@porcupine.slc.redhat.com>
On Tue, Sep 10, 2002 at 11:15:12AM -0600, Jeff Law wrote:
> BTW, we are making the assumption that the register allocator will
> realize that call-clobbered registers are available at the sibcall
> site. It should probably already do that and failure to get a reg
> would indicate a serious problem in the allocator ...
Nope. Since we've not yet emitted the sibcall_epilogue
(and how could we before register allocation), there's
nothing in the viscinity of the sibcall pattern that would
indicate that a call-saved register wasn't usable.
r~