The patch titled
Subject: mm/hugetlb: take the free hpage during the iteration directly
has been added to the -mm tree. Its filename is
mm-hugetlb-take-the-free-hpage-during-the-iteration-directly.patch
This patch should soon appear at
https://ozlabs.org/~akpm/mmots/broken-out/mm-hugetlb-take-the-free-hpage-during-the-iteration-directly.patch
and later at
https://ozlabs.org/~akpm/mmotm/broken-out/mm-hugetlb-take-the-free-hpage-during-the-iteration-directly.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Wei Yang <richard.weiyang@xxxxxxxxxxxxxxxxx>
Subject: mm/hugetlb: take the free hpage during the iteration directly
Function dequeue_huge_page_node_exact() iterates the free list and return
the first valid free hpage.
Instead of break and check the loop variant, we could return in the loop
directly. This could reduce some redundant check.
Link: https://lkml.kernel.org/r/20200831022351.20916-8-richard.weiyang@xxxxxxxxxxxxxxxxx
Signed-off-by: Wei Yang <richard.weiyang@xxxxxxxxxxxxxxxxx>
Cc: Baoquan He <bhe@xxxxxxxxxx>
Cc: Mike Kravetz <mike.kravetz@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/hugetlb.c | 20 ++++++++------------
1 file changed, 8 insertions(+), 12 deletions(-)
--- a/mm/hugetlb.c~mm-hugetlb-take-the-free-hpage-during-the-iteration-directly
+++ a/mm/hugetlb.c
@@ -1040,21 +1040,17 @@ static struct page *dequeue_huge_page_no
if (nocma && is_migrate_cma_page(page))
continue;
- if (!PageHWPoison(page))
+ if (PageHWPoison(page))
break;
+
+ list_move(&page->lru, &h->hugepage_activelist);
+ set_page_refcounted(page);
+ h->free_huge_pages--;
+ h->free_huge_pages_node[nid]--;
+ return page;
}
- /*
- * if 'non-isolated free hugepage' not found on the list,
- * the allocation fails.
- */
- if (&h->hugepage_freelists[nid] == &page->lru)
- return NULL;
- list_move(&page->lru, &h->hugepage_activelist);
- set_page_refcounted(page);
- h->free_huge_pages--;
- h->free_huge_pages_node[nid]--;
- return page;
+ return NULL;
}
static struct page *dequeue_huge_page_nodemask(struct hstate *h, gfp_t gfp_mask, int nid,
_
Patches currently in -mm which might be from richard.weiyang@xxxxxxxxxxxxxxxxx are
mm-mmap-rename-__vma_unlink_common-to-__vma_unlink.patch
mm-mmap-leverage-vma_rb_erase_ignore-to-implement-vma_rb_erase.patch
mm-mmap-leave-adjust_next-as-virtual-address-instead-of-page-frame-number.patch
mm-hugetlb-not-necessary-to-coalesce-regions-recursively.patch
mm-hugetlb-remove-vm_bug_onnrg-in-get_file_region_entry_from_cache.patch
mm-hugetlb-use-list_splice-to-merge-two-list-at-once.patch
mm-hugetlb-count-file_region-to-be-added-when-regions_needed-=-null.patch
mm-hugetlb-a-page-from-buddy-is-not-on-any-list.patch
mm-hugetlb-narrow-the-hugetlb_lock-protection-area-during-preparing-huge-page.patch
mm-hugetlb-take-the-free-hpage-during-the-iteration-directly.patch
mm-page_reporting-drop-stale-list-head-check-in-page_reporting_cycle.patch
bitops-simplify-get_count_order_long.patch
bitops-use-the-same-mechanism-for-get_count_order.patch
