From: Andi Kleen <[EMAIL PROTECTED]> Date: Tue, 7 Mar 2006 19:50:52 +0100
> On Wednesday 08 March 2006 02:29, Benjamin LaHaise wrote: > > On Tue, Mar 07, 2006 at 05:27:37PM +0100, Andi Kleen wrote: > > > On Wednesday 08 March 2006 00:26, Benjamin LaHaise wrote: > > > > Hi Andi, > > > > > > > > On x86-64 one inefficiency that shows up on profiles is the handling of > > > > struct page conversion to/from idx and addresses. This is mostly due > > > > to > > > > the fact that struct page is currently 56 bytes on x86-64, so gcc has > > > > to > > > > emit a slow division or multiplication to convert. > > > > > > Huh? > > > > You used an unsigned long, but ptrdiff_t is signed. gcc cannot use any > > shifting tricks because they round incorrectly in the signed case. > > > My vmlinux has Right, it uses the imul because gcc has determined that the shift/add tricks expand to more cycles than the multiply by reciprocol would take. It's very cpu and gcc dependant what code sequence you get here. I still think using a power-of-2 for the page struct is still probably a good thing to do, however. I've always been tempted to do this on sparc64. - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
