The patch titled
Subject: mm: skip invalid pages block at a time in zero_resv_unresv()
has been removed from the -mm tree. Its filename was
mm-skip-invalid-pages-block-at-a-time-in-zero_resv_unresv.patch
This patch was dropped because it was merged into mainline or a subsystem tree
------------------------------------------------------
From: Pavel Tatashin <pasha.tatashin@xxxxxxxxxx>
Subject: mm: skip invalid pages block at a time in zero_resv_unresv()
The role of zero_resv_unavail() is to make sure that every struct page
that is allocated but is not backed by memory that is accessible by kernel
is zeroed and not in some uninitialized state.
Since struct pages are allocated in blocks (2M pages in x86 case), we can
skip pageblock_nr_pages at a time, when the first one is found to be
invalid.
This optimization may help since now on x86 every hole in e820 maps is
marked as reserved in memblock, and thus will go through this function.
This function is called before sched_clock() is initialized, so I used my
x86 early boot clock patches to measure the performance improvement.
With 1T hole on i7-8700 currently we would take 0.606918s of boot time,
but with this optimization 0.001103s.
Link: http://lkml.kernel.org/r/20180615155733.1175-1-pasha.tatashin@xxxxxxxxxx
Signed-off-by: Pavel Tatashin <pasha.tatashin@xxxxxxxxxx>
Reviewed-by: Oscar Salvador <osalvador@xxxxxxx>
Reviewed-by: Naoya Horiguchi <n-horiguchi@xxxxxxxxxxxxx>
Cc: Pasha Tatashin <Pavel.Tatashin@xxxxxxxxxxxxx>
Cc: Steven Sistare <steven.sistare@xxxxxxxxxx>
Cc: Daniel Jordan <daniel.m.jordan@xxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxx>
Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Dan Williams <dan.j.williams@xxxxxxxxx>
Cc: "Huang, Ying" <ying.huang@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/page_alloc.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
--- a/mm/page_alloc.c~mm-skip-invalid-pages-block-at-a-time-in-zero_resv_unresv
+++ a/mm/page_alloc.c
@@ -6405,8 +6405,11 @@ void __paginginit zero_resv_unavail(void
pgcnt = 0;
for_each_resv_unavail_range(i, &start, &end) {
for (pfn = PFN_DOWN(start); pfn < PFN_UP(end); pfn++) {
- if (!pfn_valid(ALIGN_DOWN(pfn, pageblock_nr_pages)))
+ if (!pfn_valid(ALIGN_DOWN(pfn, pageblock_nr_pages))) {
+ pfn = ALIGN_DOWN(pfn, pageblock_nr_pages)
+ + pageblock_nr_pages - 1;
continue;
+ }
mm_zero_struct_page(pfn_to_page(pfn));
pgcnt++;
}
_
Patches currently in -mm which might be from pasha.tatashin@xxxxxxxxxx are
mm-make-memmap_init-a-proper-function.patch
mm-calculate-deferred-pages-after-skipping-mirrored-memory.patch
mm-calculate-deferred-pages-after-skipping-mirrored-memory-v2.patch
mm-move-mirrored-memory-specific-code-outside-of-memmap_init_zone.patch
mm-move-mirrored-memory-specific-code-outside-of-memmap_init_zone-v2.patch
mm-access-zone-node-via-zone_to_nid-and-zone_set_nid.patch
mm-remove-__paginginit.patch
