Re: PATCH][1/8] 2.6.15 mlock: make_pages_wired/unwired

[Stone Wang]

2006/3/21, Nick Piggin <nickpiggin@xxxxxxxxxxxx>:
> Stone Wang wrote:
> > We dont account HugeTLB pages for:
> >
> > 1. HugeTLB pages themselves are not reclaimable.
> >
> > 2. If we count HugeTLB pages in "Wired",then we would have no mind
> >    how many of the "Wired" are HugeTLB pages, and how many are
> > normal-size pages.
> >    Thus, hard to get a clear map of physical memory use,for example:
> >      how many pages are reclaimable?
> >    If we must count HugeTLB pages,more fields should be added to
> > "/proc/meminfo",
> >    for exmaple: "Wired HugeTLB:", "Wired Normal:".
> >
>
> Then why do you wire them at all? Your unwire function does not appear
> to be able to unwire them.

We didnt wire them.

Check get_user_pages():

        /* We dont account wired HugeTLB pages */
        if (is_vm_hugetlb_page(vma)) {
            i = follow_hugetlb_page(mm, vma, pages, vmas,
                        &start, &len, i);
            continue;
        }


Shaoping Wang

>
> --
> SUSE Labs, Novell Inc.
> Send instant messages to your online friends http://au.messenger.yahoo.com
>
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/