On Mon, Jul 14, 2008 at 07:37:32PM +0900, Isaku Yamahata wrote: > On Mon, Jul 14, 2008 at 07:21:43PM +1000, Simon Horman wrote: > > --- xen-unstable.hg.orig/xen/arch/ia64/linux-xen/mca_asm.S 2008-07-14 > > 17:41:29.000000000 +1000 > > +++ xen-unstable.hg/xen/arch/ia64/linux-xen/mca_asm.S 2008-07-14 > > 19:16:56.000000000 +1000 > ... > > @@ -320,7 +321,44 @@ ia64_do_tlb_purge: > > srlz.i > > ;; > > #ifdef XEN > > - // 5. VHPT > > + // 5. shared_info > > + GET_THIS_PADDR(r2, inserted_shared_info);; > > + ld8 r16=[r2] > > + mov r18=XSI_SHIFT<<2 > > + ;; > > + ptr.d r16,r18 > > + ;; > > + srlz.d > > + ;; > > + > > + // 6. mapped_regs > > + GET_THIS_PADDR(r2, inserted_mapped_regs);; > > + ld8 r16=[r2] > > + mov r18=XMAPPEDREGS_SHIFT<<2 > > + ;; > > + ptr.d r16,r18 > > + ;; > > + srlz.d > > + ;; > > + > > + // 7. VPD > > + // The VPD will not be mapped in the case where > > + // a VMX domain hasn't been started since boot > > + GET_THIS_PADDR(r2, inserted_vpd);; > > + ld8 r16=[r2] > > + mov r18=XMAPPEDREGS_SHIFT<<2 > > + ;; > > + cmp.eq p7,p0=r2,r0 > > + ;; > > +(p7) br.cond.sptk .vpd_not_mapped > > + ;; > > + ptr.d r16,r18 > > VPD is pinned down by both DTR and ITR. > so isn't ptr.i needed?
Yes, the ptr.d should be ptr.i. The ptr.d portion is handled by the puging of mapped_regs above. I will post a revised patch. _______________________________________________ Xen-ia64-devel mailing list Xen-ia64-devel@lists.xensource.com http://lists.xensource.com/xen-ia64-devel
