search for: shadow_fault

Just found this patch in house which I thought sent out already. :-) Still one thing remaining is to special case for UP guest which suffers overhead from this change... ---- Remove lock on first guest table walk Existing shadow fault path grabs big lock before walking guest tables, to ensure consistency with shadow content lest concurrent change from other vcpu in a bad OS. But this lock

...Is it reasonable to implement the back link only for the GPT > pages? Of course, this will increase the complexity of the code a little. > 3) Can you show us the statistics between the current shadow pgt and XI > pgt for some critical operations, such as shadow_resync_all, gva_to_gpa, > shadow_fault and so on. I''m really curious about it. > > I have to say I''m not very familiar with the current shadow pgt > implementation so I could miss some important considerations when I post > these questions. Please point it out. > Thanks for sharing your idea and code wit...

...gt;domain_id, v->vcpu_id, va, regs->error_code, regs->rip); +#elif defined(__i386__) + SHADOW_PRINTK("d:v=%u:%u va=%#lx err=%u\n", + v->domain->domain_id, v->vcpu_id, va, regs->error_code); +#endif /* __i386__ */ perfc_incr(shadow_fault); -- TAMURA, Yoshiaki NTT Cyber Space Labs OSS Computing Project Kernel Group E-mail: tamura.yoshiaki@lab.ntt.co.jp TEL: +81-46-859-2771 FAX: +81-46-855-1152 Address: 1-1 Hikarinooka, Yokosuka Kanagawa 239-0847 JAPAN _______________________________________________ Xen-devel mailing list Xen-d...

I would like to know approximately the overall execution flow when the guest O/S page fault - how is the page fault transfer to a page fault in the host O/S? Thanks. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

...please feel free help polish a better name. :-) > - In shadow_check_gwalk, maybe return 1 at the top of the function if > the version number hasn''t changed, rather than putting most of the > function inside an if()? Agree. > - You''ve added a second "not a shadow_fault" printout without removing > the first. The first one is branched with lock held, and some audit functions also required lock held. To differentiate, I put a new label. Now I just removed the new label since only few places use it. > - We can remove the comment at the top of multi.c...

...gs(gpte) & _PAGE_PRESENT) ) return 0; gpa = l1e_get_paddr(gpte) + (va & ~PAGE_MASK); /* Use 1:1 page table to identify MMIO address space */ - if (mmio_space(gpa)) + if (mmio_space(gpa)){ handle_mmio(va, gpa); + return 1; + } result = shadow_fault(va, regs); @@ -451,10 +454,9 @@ p->port_mm = 0; /* Check if the packet needs to be intercepted */ - if (vmx_io_intercept(p)) { + if (vmx_portio_intercept(p)) /* no blocking & no evtchn notification */ return; - } set_bit(ARCH_VMX_IO_WAIT, &d->...

...mp;spte); - if (__put_user(spte, (unsigned long *) - &shadow_linear_pg_table[va >> PAGE_SHIFT])) { + if (__copy_to_user(&shadow_linear_pg_table[va >> PAGE_SHIFT], + &spte, sizeof(spte))) { return; } } @@ -715,17 +719,18 @@ int shadow_fault(unsigned long va, struc void shadow_l1_normal_pt_update( - unsigned long pa, unsigned long gpte, + unsigned long pa, l1_pgentry_t gpte, unsigned long *prev_smfn_ptr, l1_pgentry_t **prev_spl1e_ptr) { - unsigned long smfn, spte, prev_smfn = *prev_smfn_ptr; + l1_pgentry...

Hi, Robert, I found out another confusing code snippet: in void xi_invl_mfn(struct domain *d, unsigned long mfn) if (ext && pfn < ext->large_page_aligned_size) According to the code, it should be if (ext && (pfn>>SPT_ENTRIES_ORDER) < ext->large_page_aligned_size) If I made any mistake, please point it out.