The quilt patch titled
Subject: mm/filemap: don't decrease mmap_miss when folio has workingset flag
has been removed from the -mm tree. Its filename was
mm-readahead-dont-decrease-mmap_miss-when-folio-has-workingset-flags.patch
This patch was dropped because it was merged into the mm-stable branch
of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
------------------------------------------------------
From: Liu Shixin <liushixin2@xxxxxxxxxx>
Subject: mm/filemap: don't decrease mmap_miss when folio has workingset flag
Date: Fri, 22 Mar 2024 17:35:55 +0800
If there are too many folios that are recently evicted in a file, then
they will probably continue to be evicted. In such situation, there is no
positive effect to read-ahead this file since it is only a waste of IO.
The mmap_miss is increased in do_sync_mmap_readahead() and decreased in
both do_async_mmap_readahead() and filemap_map_pages(). In order to skip
read-ahead in above scenario, the mmap_miss have to increased exceed
MMAP_LOTSAMISS. This can be done by stop decreased mmap_miss when folio
has workingset flag. The async path is not to care because in above
scenario, it's hard to run into the async path.
[liushixin2@xxxxxxxxxx: add comments]
Link: https://lkml.kernel.org/r/20240326065026.1910584-1-liushixin2@xxxxxxxxxx
Link: https://lkml.kernel.org/r/20240322093555.226789-3-liushixin2@xxxxxxxxxx
Signed-off-by: Liu Shixin <liushixin2@xxxxxxxxxx>
Reviewed-by: Jan Kara <jack@xxxxxxx>
Cc: Al Viro <viro@xxxxxxxxxxxxxxxxxx>
Cc: Christian Brauner <brauner@xxxxxxxxxx>
Cc: Jinjiang Tu <tujinjiang@xxxxxxxxxx>
Cc: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/filemap.c | 14 ++++++++++++--
1 file changed, 12 insertions(+), 2 deletions(-)
--- a/mm/filemap.c~mm-readahead-dont-decrease-mmap_miss-when-folio-has-workingset-flags
+++ a/mm/filemap.c
@@ -3492,7 +3492,15 @@ static vm_fault_t filemap_map_folio_rang
if (PageHWPoison(page + count))
goto skip;
- (*mmap_miss)++;
+ /*
+ * If there are too many folios that are recently evicted
+ * in a file, they will probably continue to be evicted.
+ * In such situation, read-ahead is only a waste of IO.
+ * Don't decrease mmap_miss in this scenario to make sure
+ * we can stop read-ahead.
+ */
+ if (!folio_test_workingset(folio))
+ (*mmap_miss)++;
/*
* NOTE: If there're PTE markers, we'll leave them to be
@@ -3541,7 +3549,9 @@ static vm_fault_t filemap_map_order0_fol
if (PageHWPoison(page))
return ret;
- (*mmap_miss)++;
+ /* See comment of filemap_map_folio_range() */
+ if (!folio_test_workingset(folio))
+ (*mmap_miss)++;
/*
* NOTE: If there're PTE markers, we'll leave them to be
_
Patches currently in -mm which might be from liushixin2@xxxxxxxxxx are
