search for: xenpage_list

...ni patches for Linux we submitted to lkml depend on shadow mode. > > -- Keir > So this is _not_ for 3.0.2? I expect some hiccups with that if domUs start using other than log-dirty mode (e.g. SHM_translate). For this particular bug, I think the P2M table is missing the mapping for the xenpage_list. We''ll take a look at that. For PAE or x86-64, although xenlinux uses subset of the shadow sub-modes, I guess we have more bugs with domUs simply because we haven''t tested (we have higher confidence with log-dirty mode for domU live migration). HVM guests have been tested well, a...

Hi, I am trying to release domU pages from page_list and xenpage_list after domU shutdown while retaining the rest of the domain information. To achieve this in __domain_finalise_shutdown i call domain_relinquish_resources. This is failing to release pages from page_list for type PGT_l2_page_tables and crashing dom0. To be specific, while testing on mini-os i saw...

...c Tue Aug 28 13:06:41 2007 +0900 @@ -313,13 +313,13 @@ static void relinquish_memory(struct dom spin_unlock_recursive(&d->page_alloc_lock); } -void domain_relinquish_resources(struct domain *d) +int domain_relinquish_resources(struct domain *d) { relinquish_memory(d, &d->xenpage_list); relinquish_memory(d, &d->page_list); xfree(d->arch.foreign_mfns); xfree(d->arch.p2m); - return; + return 0; } void arch_dump_domain_info(struct domain *d) diff -r 58d131f1fb35 -r 2c9db26f1d0e xen/arch/x86/domain.c --- a/xen/arch/x86/domain.c Fri Aug 24 16:32:...

...e_info *alloc_domheap_pages( diff --git a/xen/include/xen/sched.h b/xen/include/xen/sched.h index 6c55039..480ef39 100644 --- a/xen/include/xen/sched.h +++ b/xen/include/xen/sched.h @@ -242,6 +242,7 @@ struct domain struct page_list_head page_list; /* linked list */ struct page_list_head xenpage_list; /* linked list (size xenheap_pages) */ unsigned int tot_pages; /* number of pages currently possesed */ + unsigned int unclaimed_pages; /* pages claimed but not possessed */ unsigned int max_pages; /* maximum value for tot_pages */ atomic_t...