>>> On 29.08.17 at 15:27, <boris.ostrov...@oracle.com> wrote: > On 08/29/2017 09:22 AM, Jan Beulich wrote: >>>>> On 29.08.17 at 13:56, <wei.l...@citrix.com> wrote: >>> On Mon, Aug 28, 2017 at 04:40:28PM -0400, Boris Ostrovsky wrote: >>>> There is no need to scrub pages freed during dom0 construction >>>> since heap will be scrubbed once dom0 is ready (by scrub_heap_pages()). >>>> >>>> Since boot_scrub_done will not be set if boot-time scrubbing is off we >>>> also check for domain state. >>>> >>>> Signed-off-by: Boris Ostrovsky <boris.ostrov...@oracle.com> >>>> --- >>>> xen/common/page_alloc.c | 2 +- >>>> 1 file changed, 1 insertion(+), 1 deletion(-) >>>> >>>> diff --git a/xen/common/page_alloc.c b/xen/common/page_alloc.c >>>> index 34a7992..b93dae9 100644 >>>> --- a/xen/common/page_alloc.c >>>> +++ b/xen/common/page_alloc.c >>>> @@ -2259,7 +2259,7 @@ void free_domheap_pages(struct page_info *pg, >>>> unsigned > int order) >>>> */ >>>> scrub = !!d->is_dying; >>>> #else >>>> - scrub = true; >>>> + scrub = boot_scrub_done || !!d->is_dying; >>> It seems that the debug and non-debug case should use the same >>> predicate. >> No - boot_scrub_done doesn't even exist in the other case. > > > I read Wei's message as suggesting > > scrub = !!d->is_dying; > #ifdef CONFIG_SCRUB_DEBUG > scrub |= boot_scrub_done; > #endif > > > which I could do.
Yes, but please consider the "#define boot_scrub_done 0" approach too. Jan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org https://lists.xen.org/xen-devel
