Re: [PATCH v2] mm: Optimise put_pages_list()
From: Anthony Yznaga
Date: Sat Oct 23 2021 - 02:34:35 EST
On 10/22/21 6:13 PM, Matthew Wilcox wrote:
On Fri, Oct 22, 2021 at 04:26:59PM -0700, Anthony Yznaga wrote:
On 10/7/21 12:21 PM, Matthew Wilcox (Oracle) wrote:
Instead of calling put_page() one page at a time, pop pages off
the list if their refcount was too high and pass the remainder to
put_unref_page_list(). This should be a speed improvement, but I have
no measurements to support that. Current callers do not care about
performance, but I hope to add some which do.
Signed-off-by: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx>
---
v2:
- Handle compound pages (Mel)
- Comment why we don't need to handle PageLRU
- Added call to __ClearPageWaiters(), matching that in release_pages()
mm/swap.c | 23 ++++++++++++++++-------
1 file changed, 16 insertions(+), 7 deletions(-)
diff --git a/mm/swap.c b/mm/swap.c
index af3cad4e5378..9f334d503fd2 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -134,18 +134,27 @@ EXPORT_SYMBOL(__put_page);
* put_pages_list() - release a list of pages
* @pages: list of pages threaded on page->lru
*
- * Release a list of pages which are strung together on page.lru. Currently
- * used by read_cache_pages() and related error recovery code.
+ * Release a list of pages which are strung together on page.lru.
*/
void put_pages_list(struct list_head *pages)
{
- while (!list_empty(pages)) {
- struct page *victim;
+ struct page *page, *next;
- victim = lru_to_page(pages);
- list_del(&victim->lru);
- put_page(victim);
+ list_for_each_entry_safe(page, next, pages, lru) {
+ if (!put_page_testzero(page)) {
+ list_del(&page->lru);
+ continue;
+ }
I know that compound pages are not currently passed to put_pages_list(),
but I assume the put_page_testzero() should only be done on the head
page similar to release_pages()?
Fun fact about pages: You can't put a tail page on an LRU list. Why?
struct page {
...
union {
struct { /* Page cache and anonymous pages */
struct list_head lru;
...
struct { /* Tail pages of compound page */
unsigned long compound_head; /* Bit zero is set */
so if you try to put a tail page on the LRU list, it becomes no longer
a tail page.
Ah, makes sense. Thanks!
In that case here's my r-b:
Reviewed-by: Anthony Yznaga <anthony.yznaga@xxxxxxxxxx>
Anthony