linux-perf-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: David Hildenbrand <david@redhat.com>
To: linux-kernel@vger.kernel.org
Cc: linux-doc@vger.kernel.org, dri-devel@lists.freedesktop.org,
	linux-mm@kvack.org, nouveau@lists.freedesktop.org,
	linux-trace-kernel@vger.kernel.org,
	linux-perf-users@vger.kernel.org, damon@lists.linux.dev,
	"David Hildenbrand" <david@redhat.com>,
	"Andrew Morton" <akpm@linux-foundation.org>,
	"Jérôme Glisse" <jglisse@redhat.com>,
	"Jonathan Corbet" <corbet@lwn.net>, "Alex Shi" <alexs@kernel.org>,
	"Yanteng Si" <si.yanteng@linux.dev>,
	"Karol Herbst" <kherbst@redhat.com>,
	"Lyude Paul" <lyude@redhat.com>,
	"Danilo Krummrich" <dakr@kernel.org>,
	"David Airlie" <airlied@gmail.com>,
	"Simona Vetter" <simona@ffwll.ch>,
	"Masami Hiramatsu" <mhiramat@kernel.org>,
	"Oleg Nesterov" <oleg@redhat.com>,
	"Peter Zijlstra" <peterz@infradead.org>,
	"SeongJae Park" <sj@kernel.org>,
	"Liam R. Howlett" <Liam.Howlett@oracle.com>,
	"Lorenzo Stoakes" <lorenzo.stoakes@oracle.com>,
	"Vlastimil Babka" <vbabka@suse.cz>,
	"Jann Horn" <jannh@google.com>,
	"Pasha Tatashin" <pasha.tatashin@soleen.com>,
	"Peter Xu" <peterx@redhat.com>,
	"Alistair Popple" <apopple@nvidia.com>,
	"Jason Gunthorpe" <jgg@nvidia.com>
Subject: [PATCH v2 13/17] mm/page_idle: handle device-exclusive entries correctly in page_idle_clear_pte_refs_one()
Date: Mon, 10 Feb 2025 20:37:55 +0100	[thread overview]
Message-ID: <20250210193801.781278-14-david@redhat.com> (raw)
In-Reply-To: <20250210193801.781278-1-david@redhat.com>

Ever since commit b756a3b5e7ea ("mm: device exclusive memory access")
we can return with a device-exclusive entry from page_vma_mapped_walk().

page_idle_clear_pte_refs_one() is not prepared for that, so let's
teach it what to do with these PFN swap PTEs. Note that device-private
entries are so far not applicable on that path, as page_idle_get_folio()
filters out non-lru folios.

Should we just skip PFN swap PTEs completely? Possible, but it seems
straight forward to just handle them correctly.

Note that we could currently only run into this case with
device-exclusive entries on THPs. We still adjust the mapcount on
conversion to device-exclusive; this makes the rmap walk
abort early for small folios, because we'll always have
!folio_mapped() with a single device-exclusive entry. We'll adjust the
mapcount logic once all page_vma_mapped_walk() users can properly
handle device-exclusive entries.

Fixes: b756a3b5e7ea ("mm: device exclusive memory access")
Signed-off-by: David Hildenbrand <david@redhat.com>
---
 mm/page_idle.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/mm/page_idle.c b/mm/page_idle.c
index 947c7c7a37289..408aaf29a3ea6 100644
--- a/mm/page_idle.c
+++ b/mm/page_idle.c
@@ -62,9 +62,14 @@ static bool page_idle_clear_pte_refs_one(struct folio *folio,
 			/*
 			 * For PTE-mapped THP, one sub page is referenced,
 			 * the whole THP is referenced.
+			 *
+			 * PFN swap PTEs, such as device-exclusive ones, that
+			 * actually map pages are "old" from a CPU perspective.
+			 * The MMU notifier takes care of any device aspects.
 			 */
-			if (ptep_clear_young_notify(vma, addr, pvmw.pte))
-				referenced = true;
+			if (likely(pte_present(ptep_get(pvmw.pte))))
+				referenced |= ptep_test_and_clear_young(vma, addr, pvmw.pte);
+			referenced |= mmu_notifier_clear_young(vma->vm_mm, addr, addr + PAGE_SIZE);
 		} else if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE)) {
 			if (pmdp_clear_young_notify(vma, addr, pvmw.pmd))
 				referenced = true;
-- 
2.48.1


  parent reply	other threads:[~2025-02-10 19:38 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-02-10 19:37 [PATCH v2 00/17] mm: fixes for device-exclusive entries (hmm) David Hildenbrand
2025-02-10 19:37 ` [PATCH v2 01/17] mm/gup: reject FOLL_SPLIT_PMD with hugetlb VMAs David Hildenbrand
2025-02-10 19:37 ` [PATCH v2 02/17] mm/rmap: reject hugetlb folios in folio_make_device_exclusive() David Hildenbrand
2025-02-10 19:37 ` [PATCH v2 03/17] mm/rmap: convert make_device_exclusive_range() to make_device_exclusive() David Hildenbrand
2025-02-11  5:00   ` Andrew Morton
2025-02-11  8:33     ` David Hildenbrand
2025-02-17  0:01       ` Alistair Popple
2025-02-17  9:32         ` David Hildenbrand
2025-02-10 19:37 ` [PATCH v2 04/17] mm/rmap: implement make_device_exclusive() using folio_walk instead of rmap walk David Hildenbrand
2025-02-10 19:37 ` [PATCH v2 05/17] mm/memory: detect writability in restore_exclusive_pte() through can_change_pte_writable() David Hildenbrand
2025-02-10 19:37 ` [PATCH v2 06/17] mm: use single SWP_DEVICE_EXCLUSIVE entry type David Hildenbrand
2025-02-10 19:37 ` [PATCH v2 07/17] mm/page_vma_mapped: device-exclusive entries are not migration entries David Hildenbrand
2025-02-10 19:37 ` [PATCH v2 08/17] kernel/events/uprobes: handle device-exclusive entries correctly in __replace_page() David Hildenbrand
2025-02-10 19:37 ` [PATCH v2 09/17] mm/ksm: handle device-exclusive entries correctly in write_protect_page() David Hildenbrand
2025-02-10 19:37 ` [PATCH v2 10/17] mm/rmap: handle device-exclusive entries correctly in try_to_unmap_one() David Hildenbrand
2025-02-10 19:37 ` [PATCH v2 11/17] mm/rmap: handle device-exclusive entries correctly in try_to_migrate_one() David Hildenbrand
2025-02-10 19:37 ` [PATCH v2 12/17] mm/rmap: handle device-exclusive entries correctly in page_vma_mkclean_one() David Hildenbrand
2025-02-10 19:37 ` David Hildenbrand [this message]
2025-02-11 20:48   ` [PATCH v2 13/17] mm/page_idle: handle device-exclusive entries correctly in page_idle_clear_pte_refs_one() SeongJae Park
2025-02-10 19:37 ` [PATCH v2 14/17] mm/damon: handle device-exclusive entries correctly in damon_folio_young_one() David Hildenbrand
2025-02-11  6:59   ` SeongJae Park
2025-02-10 19:37 ` [PATCH v2 15/17] mm/damon: handle device-exclusive entries correctly in damon_folio_mkold_one() David Hildenbrand
2025-02-11  7:00   ` SeongJae Park
2025-02-10 19:37 ` [PATCH v2 16/17] mm/rmap: keep mapcount untouched for device-exclusive entries David Hildenbrand
2025-02-10 19:37 ` [PATCH v2 17/17] mm/rmap: avoid -EBUSY from make_device_exclusive() David Hildenbrand
2025-02-10 23:05 ` [PATCH v2 00/17] mm: fixes for device-exclusive entries (hmm) Andrew Morton
2025-02-10 23:39   ` Barry Song
2025-02-13 11:03 ` Alistair Popple
2025-02-13 11:15   ` David Hildenbrand
2025-02-14  1:25     ` Alistair Popple
2025-02-14 10:37       ` David Hildenbrand

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=20250210193801.781278-14-david@redhat.com \
    --to=david@redhat.com \
    --cc=Liam.Howlett@oracle.com \
    --cc=airlied@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=alexs@kernel.org \
    --cc=apopple@nvidia.com \
    --cc=corbet@lwn.net \
    --cc=dakr@kernel.org \
    --cc=damon@lists.linux.dev \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=jannh@google.com \
    --cc=jgg@nvidia.com \
    --cc=jglisse@redhat.com \
    --cc=kherbst@redhat.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=linux-trace-kernel@vger.kernel.org \
    --cc=lorenzo.stoakes@oracle.com \
    --cc=lyude@redhat.com \
    --cc=mhiramat@kernel.org \
    --cc=nouveau@lists.freedesktop.org \
    --cc=oleg@redhat.com \
    --cc=pasha.tatashin@soleen.com \
    --cc=peterx@redhat.com \
    --cc=peterz@infradead.org \
    --cc=si.yanteng@linux.dev \
    --cc=simona@ffwll.ch \
    --cc=sj@kernel.org \
    --cc=vbabka@suse.cz \
    /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).