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]: MD_FALLBACK_FRAME_STATE_FOR macro for darwin PPC


Hi Dale,

Dale Johannesen wrote:

Saved regs are described in rs6000.h of course.

From the cfa?


(FS)->regs.reg[i_].loc.offset					\
	    = (long)&(mc_->ss.r0) + (i_*4) - new_cfa_;

How is this offset organized? (Note, did not look at the source at the moment.)
There are 146 regs, afair.


Floats are F14-F31 and I think you may need CR also, some of its subregs are caller-saved.
And then there's Altivec, V20-V31 and VRSAVE. They don't exist on all machines though!

Are these the registers which are called 'nonvolatile' in the Mach-O runtime convention? Chapter 'register preservation'?


If so, then I don't need to restore a few gprs. Right?

Thx for the feedback.

Regards,
Andreas



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