[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Xen-devel] [PATCH v4] memory_hotplug: Free pages as higher order
- To: Michal Hocko <mhocko@xxxxxxxxxx>
- From: Arun KS <arunks@xxxxxxxxxxxxxx>
- Date: Fri, 05 Oct 2018 12:54:45 +0530
- Cc: vatsa@xxxxxxxxxxxxxx, linux-mm@xxxxxxxxx, getarunks@xxxxxxxxx, kys@xxxxxxxxxxxxx, boris.ostrovsky@xxxxxxxxxx, sthemmin@xxxxxxxxxxxxx, xen-devel@xxxxxxxxxxxxxxxxxxxx, haiyangz@xxxxxxxxxxxxx, yasu.isimatu@xxxxxxxxx, dan.j.williams@xxxxxxxxx, akpm@xxxxxxxxxxxxxxxxxxxx, vbabka@xxxxxxx, osalvador@xxxxxxx, jgross@xxxxxxxx, aaron.lu@xxxxxxxxx, malat@xxxxxxxxxx, gregkh@xxxxxxxxxxxxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, vinmenon@xxxxxxxxxxxxxx, jrdr.linux@xxxxxxxxx, devel@xxxxxxxxxxxxxxxxxxxxxx, iamjoonsoo.kim@xxxxxxx, mgorman@xxxxxxxxxxxxxxxxxxx, kirill.shutemov@xxxxxxxxxxxxxxx
- Delivery-date: Fri, 05 Oct 2018 07:24:59 +0000
- List-id: Xen developer discussion <xen-devel.lists.xenproject.org>
On 2018-10-04 20:21, Michal Hocko wrote:
On Wed 03-10-18 19:09:39, Arun KS wrote:
[...]
+static int online_pages_blocks(unsigned long start, unsigned long
nr_pages)
+{
+ unsigned long end = start + nr_pages;
+ int order, ret, onlined_pages = 0;
+
+ while (start < end) {
+ order = min(MAX_ORDER - 1UL, __ffs(start));
+
+ while (start + (1UL << order) > end)
+ order--;
this really made me scratch my head. Wouldn't it be much simpler to do
the following?
order = min(MAX_ORDER - 1, get_order(end - start))?
Yes. Much better. Will change to,
order = min(MAX_ORDER - 1,
get_order(PFN_PHYS(end) - PFN_PHYS(start)));
+
+ ret = (*online_page_callback)(pfn_to_page(start), order);
+ if (!ret)
+ onlined_pages += (1UL << order);
+ else if (ret > 0)
+ onlined_pages += ret;
+
+ start += (1UL << order);
+ }
+ return onlined_pages;
}
[...]
-static void __init __free_pages_boot_core(struct page *page, unsigned
int order)
+void __free_pages_core(struct page *page, unsigned int order)
{
unsigned int nr_pages = 1 << order;
struct page *p = page;
unsigned int loop;
- prefetchw(p);
- for (loop = 0; loop < (nr_pages - 1); loop++, p++) {
- prefetchw(p + 1);
+ for (loop = 0; loop < nr_pages; loop++, p++) {
__ClearPageReserved(p);
set_page_count(p, 0);
}
- __ClearPageReserved(p);
- set_page_count(p, 0);
page_zone(page)->managed_pages += nr_pages;
set_page_refcounted(page);
I think this is wort a separate patch as it is unrelated to the patch.
Sure. Will split the patch.
Regards,
Arun
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel
|
