diff options
| author | Yisheng Xie <xieyisheng1@huawei.com> | 2017-02-24 23:59:42 +0100 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2017-02-25 02:46:56 +0100 |
| commit | 22c5cef16278bf83aa54ac96a7735ce82f93524e (patch) | |
| tree | 2be9e81101c4a3370addbfcf2b3694ac1f34a49b /mm/zsmalloc.c | |
| parent | mm: vmpressure: fix sending wrong events on underflow (diff) | |
| download | linux-22c5cef16278bf83aa54ac96a7735ce82f93524e.tar.xz linux-22c5cef16278bf83aa54ac96a7735ce82f93524e.zip | |
mm/zsmalloc: remove redundant SetPagePrivate2 in create_page_chain
We had used page->lru to link the component pages (except the first
page) of a zspage, and used INIT_LIST_HEAD(&page->lru) to init it.
Therefore, to get the last page's next page, which is NULL, we had to
use page flag PG_Private_2 to identify it.
But now, we use page->freelist to link all of the pages in zspage and
init the page->freelist as NULL for last page, so no need to use
PG_Private_2 anymore.
This remove redundant SetPagePrivate2 in create_page_chain and
ClearPagePrivate2 in reset_page(). Save a few cycles for migration of
zsmalloc page :)
Link: http://lkml.kernel.org/r/1487076509-49270-1-git-send-email-xieyisheng1@huawei.com
Signed-off-by: Yisheng Xie <xieyisheng1@huawei.com>
Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'mm/zsmalloc.c')
| -rw-r--r-- | mm/zsmalloc.c | 6 |
1 files changed, 1 insertions, 5 deletions
diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c index a1f24989ac23..25d0fb6443d1 100644 --- a/mm/zsmalloc.c +++ b/mm/zsmalloc.c @@ -24,7 +24,6 @@ * * Usage of struct page flags: * PG_private: identifies the first component page - * PG_private2: identifies the last component page * PG_owner_priv_1: identifies the huge component page * */ @@ -938,7 +937,6 @@ static void reset_page(struct page *page) { __ClearPageMovable(page); ClearPagePrivate(page); - ClearPagePrivate2(page); set_page_private(page, 0); page_mapcount_reset(page); ClearPageHugeObject(page); @@ -1085,7 +1083,7 @@ static void create_page_chain(struct size_class *class, struct zspage *zspage, * 2. each sub-page point to zspage using page->private * * we set PG_private to identify the first page (i.e. no other sub-page - * has this flag set) and PG_private_2 to identify the last page. + * has this flag set). */ for (i = 0; i < nr_pages; i++) { page = pages[i]; @@ -1100,8 +1098,6 @@ static void create_page_chain(struct size_class *class, struct zspage *zspage, } else { prev_page->freelist = page; } - if (i == nr_pages - 1) - SetPagePrivate2(page); prev_page = page; } } |