From: Zi Yan <zi.yan@sent.com>
To: "Pankaj Raghav (Samsung)" <kernel@pankajraghav.com>, linux-mm@kvack.org
Cc: "Zi Yan" <ziy@nvidia.com>,
"Matthew Wilcox (Oracle)" <willy@infradead.org>,
"David Hildenbrand" <david@redhat.com>,
"Yang Shi" <shy828301@gmail.com>, "Yu Zhao" <yuzhao@google.com>,
"Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>,
"Ryan Roberts" <ryan.roberts@arm.com>,
"Michal Koutný" <mkoutny@suse.com>,
"Roman Gushchin" <roman.gushchin@linux.dev>,
"Zach O'Keefe" <zokeefe@google.com>,
"Hugh Dickins" <hughd@google.com>,
"Luis Chamberlain" <mcgrof@kernel.org>,
"Andrew Morton" <akpm@linux-foundation.org>,
linux-kernel@vger.kernel.org, cgroups@vger.kernel.org,
linux-fsdevel@vger.kernel.org, linux-kselftest@vger.kernel.org
Subject: [PATCH v5 2/8] mm: Support order-1 folios in the page cache
Date: Mon, 26 Feb 2024 15:55:28 -0500 [thread overview]
Message-ID: <20240226205534.1603748-3-zi.yan@sent.com> (raw)
In-Reply-To: <20240226205534.1603748-1-zi.yan@sent.com>
From: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Folios of order 1 have no space to store the deferred list. This is
not a problem for the page cache as file-backed folios are never
placed on the deferred list. All we need to do is prevent the core
MM from touching the deferred list for order 1 folios and remove the
code which prevented us from allocating order 1 folios.
Link: https://lore.kernel.org/linux-mm/90344ea7-4eec-47ee-5996-0c22f42d6a6a@google.com/
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Zi Yan <ziy@nvidia.com>
---
mm/filemap.c | 2 --
mm/huge_memory.c | 19 +++++++++++++++----
mm/internal.h | 3 +--
mm/readahead.c | 3 ---
4 files changed, 16 insertions(+), 11 deletions(-)
diff --git a/mm/filemap.c b/mm/filemap.c
index b7a21551fbc7..b4858d89f1b1 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1912,8 +1912,6 @@ struct folio *__filemap_get_folio(struct address_space *mapping, pgoff_t index,
gfp_t alloc_gfp = gfp;
err = -ENOMEM;
- if (order == 1)
- order = 0;
if (order > 0)
alloc_gfp |= __GFP_NORETRY | __GFP_NOWARN;
folio = filemap_alloc_folio(alloc_gfp, order);
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index b20e535e874c..9840f312c08f 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -790,8 +790,10 @@ struct deferred_split *get_deferred_split_queue(struct folio *folio)
void folio_prep_large_rmappable(struct folio *folio)
{
- VM_BUG_ON_FOLIO(folio_order(folio) < 2, folio);
- INIT_LIST_HEAD(&folio->_deferred_list);
+ if (!folio || !folio_test_large(folio))
+ return;
+ if (folio_order(folio) > 1)
+ INIT_LIST_HEAD(&folio->_deferred_list);
folio_set_large_rmappable(folio);
}
@@ -3114,7 +3116,8 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
/* Prevent deferred_split_scan() touching ->_refcount */
spin_lock(&ds_queue->split_queue_lock);
if (folio_ref_freeze(folio, 1 + extra_pins)) {
- if (!list_empty(&folio->_deferred_list)) {
+ if (folio_order(folio) > 1 &&
+ !list_empty(&folio->_deferred_list)) {
ds_queue->split_queue_len--;
list_del(&folio->_deferred_list);
}
@@ -3165,6 +3168,9 @@ void folio_undo_large_rmappable(struct folio *folio)
struct deferred_split *ds_queue;
unsigned long flags;
+ if (folio_order(folio) <= 1)
+ return;
+
/*
* At this point, there is no one trying to add the folio to
* deferred_list. If folio is not in deferred_list, it's safe
@@ -3190,7 +3196,12 @@ void deferred_split_folio(struct folio *folio)
#endif
unsigned long flags;
- VM_BUG_ON_FOLIO(folio_order(folio) < 2, folio);
+ /*
+ * Order 1 folios have no space for a deferred list, but we also
+ * won't waste much memory by not adding them to the deferred list.
+ */
+ if (folio_order(folio) <= 1)
+ return;
/*
* The try_to_unmap() in page reclaim path might reach here too,
diff --git a/mm/internal.h b/mm/internal.h
index 2b7efffbe4d7..c4853ebfa030 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -420,8 +420,7 @@ static inline struct folio *page_rmappable_folio(struct page *page)
{
struct folio *folio = (struct folio *)page;
- if (folio && folio_order(folio) > 1)
- folio_prep_large_rmappable(folio);
+ folio_prep_large_rmappable(folio);
return folio;
}
diff --git a/mm/readahead.c b/mm/readahead.c
index 1e74455f908e..130c0e7df99f 100644
--- a/mm/readahead.c
+++ b/mm/readahead.c
@@ -514,9 +514,6 @@ void page_cache_ra_order(struct readahead_control *ractl,
/* Don't allocate pages past EOF */
while (index + (1UL << order) - 1 > limit)
order--;
- /* THP machinery does not support order-1 */
- if (order == 1)
- order = 0;
err = ra_alloc_folio(ractl, index, mark, order, gfp);
if (err)
break;
--
2.43.0
next prev parent reply other threads:[~2024-02-26 20:56 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-26 20:55 [PATCH v5 0/8] Split a folio to any lower order folios Zi Yan
2024-02-26 20:55 ` [PATCH v5 1/8] mm/huge_memory: only split PMD mapping when necessary in unmap_folio() Zi Yan
2024-02-28 10:30 ` David Hildenbrand
2024-02-26 20:55 ` Zi Yan [this message]
2024-02-26 20:55 ` [PATCH v5 3/8] mm/memcg: use order instead of nr in split_page_memcg() Zi Yan
2024-02-26 20:55 ` [PATCH v5 4/8] mm/page_owner: use order instead of nr in split_page_owner() Zi Yan
2024-02-26 20:55 ` [PATCH v5 5/8] mm: memcg: make memcg huge page split support any order split Zi Yan
2024-02-26 20:55 ` [PATCH v5 6/8] mm: page_owner: add support for splitting to any order in split page_owner Zi Yan
2024-02-28 10:31 ` David Hildenbrand
2024-02-26 20:55 ` [PATCH v5 7/8] mm: thp: split huge page to any lower order pages Zi Yan
2024-02-28 8:23 ` Ryan Roberts
2024-02-28 15:42 ` Zi Yan
2024-02-28 15:44 ` Ryan Roberts
2024-02-28 15:52 ` Zi Yan
2024-03-07 14:58 ` Zi Yan
2024-02-26 20:55 ` [PATCH v5 8/8] mm: huge_memory: enable debugfs to split huge pages to any order Zi Yan
2024-03-01 9:51 ` Aishwarya TCV
2024-03-01 10:33 ` Ryan Roberts
2024-03-01 12:11 ` Mark Brown
2024-03-01 12:56 ` Zi Yan
2024-03-01 14:14 ` Mark Brown
2024-03-01 12:52 ` Zi Yan
2024-03-01 13:09 ` Ryan Roberts
2024-03-01 13:53 ` Zi Yan
2024-03-01 14:18 ` Ryan Roberts
2024-03-01 14:27 ` Mark Brown
2024-03-01 15:21 ` Zi Yan
2024-03-01 19:41 ` Zi Yan
2024-03-01 14:24 ` Mark Brown
2024-03-01 14:00 ` Zi Yan
2024-03-01 14:23 ` Ryan Roberts
2024-03-01 14:33 ` Zi Yan
2024-03-01 19:37 ` Zi Yan
2024-03-01 20:02 ` Zi Yan
2024-03-01 21:10 ` Zi Yan
2024-03-04 9:50 ` Aishwarya TCV
2024-03-04 14:58 ` Zi Yan
2024-03-04 15:44 ` Aishwarya TCV
2024-03-04 15:57 ` Zi Yan
2024-03-04 18:25 ` Aishwarya TCV
2024-03-04 18:26 ` Zi Yan
2024-03-04 18:31 ` Zi Yan
2024-03-07 15:06 ` Zi Yan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240226205534.1603748-3-zi.yan@sent.com \
--to=zi.yan@sent.com \
--cc=akpm@linux-foundation.org \
--cc=cgroups@vger.kernel.org \
--cc=david@redhat.com \
--cc=hughd@google.com \
--cc=kernel@pankajraghav.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mcgrof@kernel.org \
--cc=mkoutny@suse.com \
--cc=roman.gushchin@linux.dev \
--cc=ryan.roberts@arm.com \
--cc=shy828301@gmail.com \
--cc=willy@infradead.org \
--cc=yuzhao@google.com \
--cc=ziy@nvidia.com \
--cc=zokeefe@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).