From nobody Thu Apr 9 13:31:48 2026 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id F42238634C for ; Mon, 9 Mar 2026 08:17:57 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1773044278; cv=none; b=en4y/YOZyXTweYMb5DtBXCl2SrvakwLjUM5ZmVTkFN32WKKiPJxYcn2v8dH8tSbUbgsraxZ7InYC3L68/PGzZQ7aIRMSiAWcFA1GAWsbBAHXQnXso4/J+jEm0vhLgF6Me6hHsW0e7M0VXVU43AMRSsJBxs8sGrcs+Ohk+wBV/Vg= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1773044278; c=relaxed/simple; bh=e/JIvBv1sq81twuWbFDup0095/v1157Z2wL8P2ZuYk4=; h=From:Date:Subject:MIME-Version:Content-Type:Message-Id:References: In-Reply-To:To:Cc; b=EY/KeXxdLlmA865SMR3eB5z6AvCgLYoLY4H7g1fx7Y2w/DS47oEgtbX+lkYNWUAqbgxECj5utArWJ7y+jH02xg8YNhSMzqb6HlK3vZk98Ibt3X+y9PhJDWUiSfRc/tpMhV3Yjps5YuP40eCRvXgLR3736OrzHK+Ua3kHa0fUrRo= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=fEFsOthZ; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="fEFsOthZ" Received: by smtp.kernel.org (Postfix) with ESMTPS id AF73EC2BCB0; Mon, 9 Mar 2026 08:17:57 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1773044277; bh=e/JIvBv1sq81twuWbFDup0095/v1157Z2wL8P2ZuYk4=; h=From:Date:Subject:References:In-Reply-To:To:Cc:Reply-To:From; b=fEFsOthZygR0BWYTfCXCH2TGc208NbRZ5QSmSyVWpEhuYy6oGDg3sU9yNzMhwsDSX cb2NRnRxLHEiFLhXVPTKoS3dHlHkPx4tQPO4RCiVZaLvFZNgOX/pkE6op8T9fdSjYM eqJqpbS/keXc5Sa2RwbJbW3B7EUNl1KoOTvMIgBYZIAozKHrHlZVJDsTmK+VullhF6 hrYpXiDBEfLXcTB50QN5SMBm9rBJSWqNZCmUqW/0WA8nVU+c5Ik2DyefYLK8DuZGG1 LjPDjjIJAgkZkpoXCjK2myXOvBJSCF6/RX+uzzsqlf1shz4wQSYsQrDZLFk9weJzUc NHO1/cIZfKNPw== Received: from aws-us-west-2-korg-lkml-1.web.codeaurora.org (localhost.localdomain [127.0.0.1]) by smtp.lore.kernel.org (Postfix) with ESMTP id A5F57EF3714; Mon, 9 Mar 2026 08:17:57 +0000 (UTC) From: Zhang Peng via B4 Relay Date: Mon, 09 Mar 2026 16:17:42 +0800 Subject: [PATCH 2/2] mm, vmscan: flush TLB for every 31 folios evictions Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Message-Id: <20260309-batch-tlb-flush-v1-2-eb8fed7d1a9e@icloud.com> References: <20260309-batch-tlb-flush-v1-0-eb8fed7d1a9e@icloud.com> In-Reply-To: <20260309-batch-tlb-flush-v1-0-eb8fed7d1a9e@icloud.com> To: Andrew Morton , David Hildenbrand , Lorenzo Stoakes , "Liam R. Howlett" , Vlastimil Babka , Mike Rapoport , Suren Baghdasaryan , Michal Hocko , Johannes Weiner , Qi Zheng , Shakeel Butt , Axel Rasmussen , Yuanchu Xie , Wei Xu , Michal Hocko Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, Kairui Song , Zhang Peng X-Mailer: b4 0.14.3 X-Developer-Signature: v=1; a=ed25519-sha256; t=1773044270; l=4427; i=zippermonkey@icloud.com; s=20260309; h=from:subject:message-id; bh=VtbVoUdutucQYTv61lHy5DXkSCu6zJZpJudQ30ChVHA=; b=hb84Un+QVXegPOTad/271FdH2TnK0FCOWCwyQrjIcYqmfgBJ2uz0HVL0r8lRGfBXbWq4q5Q3B MJJHWQGILuwAzFjHJquHifaJ+c6i/rVn/faJo05uEQWnci0nAx0TqPg X-Developer-Key: i=zippermonkey@icloud.com; a=ed25519; pk=tPCLpFnBfIyHsp0k7eaUTUREEa36bQNW/69X+NS8wBU= X-Endpoint-Received: by B4 Relay for zippermonkey@icloud.com/20260309 with auth_id=671 X-Original-From: Zhang Peng Reply-To: zippermonkey@icloud.com From: bruzzhang Currently we flush TLB for every dirty folio, which is a bottleneck for systems with many cores as this causes heavy IPI usage. So instead, batch the folios, and flush once for every 31 folios (one folio_batch). These folios will be held in a folio_batch releasing their lock, then when folio_batch is full, do following steps: - For each folio: lock - check still evictable - unlock - If no longer evictable, return the folio to the caller. - Flush TLB once for the batch - Pageout the folios (refcount freeze happens in the pageout path) Note we can't hold a frozen folio in folio_batch for long as it will cause filemap/swapcache lookup to livelock. Fortunately pageout usually won't take too long; sync IO is fast, and non-sync IO will be issued with the folio marked writeback. Suggested-by: Kairui Song Signed-off-by: bruzzhang --- mm/vmscan.c | 68 ++++++++++++++++++++++++++++++++++++++++++++++++++++++---= ---- 1 file changed, 61 insertions(+), 7 deletions(-) diff --git a/mm/vmscan.c b/mm/vmscan.c index a336f7fc7dae..69cdd3252ff8 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1240,6 +1240,48 @@ static void pageout_one(struct folio *folio, struct = list_head *ret_folios, VM_BUG_ON_FOLIO(folio_test_lru(folio) || folio_test_unevictable(folio), folio); } + +static void pageout_batch(struct folio_batch *fbatch, + struct list_head *ret_folios, + struct folio_batch *free_folios, + struct scan_control *sc, struct reclaim_stat *stat, + struct swap_iocb **plug, struct list_head *folio_list) +{ + int i =3D 0, count =3D folio_batch_count(fbatch); + struct folio *folio; + + folio_batch_reinit(fbatch); + do { + folio =3D fbatch->folios[i]; + if (!folio_trylock(folio)) { + list_add(&folio->lru, ret_folios); + continue; + } + + if (folio_test_writeback(folio) || folio_test_lru(folio) || + folio_mapped(folio)) + goto next; + folio_batch_add(fbatch, folio); + continue; +next: + folio_unlock(folio); + list_add(&folio->lru, ret_folios); + } while (++i !=3D count); + + i =3D 0; + count =3D folio_batch_count(fbatch); + if (!count) + return; + /* One TLB flush for the batch */ + try_to_unmap_flush_dirty(); + do { + folio =3D fbatch->folios[i]; + pageout_one(folio, ret_folios, free_folios, sc, stat, plug, + folio_list); + } while (++i !=3D count); + folio_batch_reinit(fbatch); +} + /* * Reclaimed folios are counted in stat->nr_reclaimed. */ @@ -1249,6 +1291,8 @@ static void shrink_folio_list(struct list_head *folio= _list, struct mem_cgroup *memcg) { struct folio_batch free_folios; + struct folio_batch flush_folios; + LIST_HEAD(ret_folios); LIST_HEAD(demote_folios); unsigned int nr_demoted =3D 0; @@ -1257,6 +1301,8 @@ static void shrink_folio_list(struct list_head *folio= _list, struct swap_iocb *plug =3D NULL; =20 folio_batch_init(&free_folios); + folio_batch_init(&flush_folios); + memset(stat, 0, sizeof(*stat)); cond_resched(); do_demote_pass =3D can_demote(pgdat->node_id, sc, memcg); @@ -1578,15 +1624,19 @@ static void shrink_folio_list(struct list_head *fol= io_list, goto keep_locked; if (!sc->may_writepage) goto keep_locked; - /* - * Folio is dirty. Flush the TLB if a writable entry - * potentially exists to avoid CPU writes after I/O - * starts and then write it out here. + * For anon, we should only see swap cache (anon) and + * the list pinning the page. For file page, the filemap + * and the list pins it. Combined with the page_ref_freeze + * in pageout_batch ensure nothing else touches the page + * during lock unlocked. */ - try_to_unmap_flush_dirty(); - pageout_one(folio, &ret_folios, &free_folios, sc, stat, - &plug, folio_list); + folio_unlock(folio); + if (!folio_batch_add(&flush_folios, folio)) + pageout_batch(&flush_folios, + &ret_folios, &free_folios, + sc, stat, &plug, + folio_list); goto next; } =20 @@ -1614,6 +1664,10 @@ static void shrink_folio_list(struct list_head *foli= o_list, next: continue; } + if (folio_batch_count(&flush_folios)) { + pageout_batch(&flush_folios, &ret_folios, &free_folios, sc, + stat, &plug, folio_list); + } /* 'folio_list' is always empty here */ =20 /* Migrate folios selected for demotion */ --=20 2.43.7