From nobody Sat Feb 7 11:39:07 2026 Received: from mail-pf1-f172.google.com (mail-pf1-f172.google.com [209.85.210.172]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id E3BF333E367 for ; Fri, 23 Jan 2026 08:23:04 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769156586; cv=none; b=Ka5XFd+yQDi8W0vRumEf1gIDFuxxKMCCcr/hb27OdcF+Sn2plvKaAAvPSedmP1Dau7ljXdzJQpIEir/H9K4bgo/3HeCHHucWp3LXd1JpHE1hOdRZ1qLo0sBr+LxRzjxaJvGXZknz4uaFRVWMON3UQV8pFN2OBT0tA7HbZRijy+I= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769156586; c=relaxed/simple; bh=IkHfdbJmzL0qf+WIlOJlSWUnMkFUcaiF6Vhs50DkoNA=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=mT0I1OorV1Z8iHLr0woIThs+L5K+zRL9KDbPXuSpuL4P3mVEdSw1y6NkCcLqY6ZBgUfgZCWNJJy4sjF5koSARo6XwN8t9jWpS5shdCJm/swZZbOij2+nColT7QhiW9BaurZw1QNJrZiHvh0PjlhMpl74DakgoTIfA85F1hv3AJc= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=kF8HVcXG; arc=none smtp.client-ip=209.85.210.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="kF8HVcXG" Received: by mail-pf1-f172.google.com with SMTP id d2e1a72fcca58-7f89d0b37f0so1526069b3a.0 for ; Fri, 23 Jan 2026 00:23:04 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1769156584; x=1769761384; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=gNFjSEtLjmo3zavrqqxSndVZ5Mb9BZ08qMwOVvVVYnk=; b=kF8HVcXGhPaI665OGA7qDNTc537IXj/kjhaidOtq12Xe5nBKskkEYCun9XB4Fn7vRZ Tn6tKm5cTdnWURbHYJ0Kk1Ibh06Cp0lrmz3UOTm/AsRt+/IrzFqn78S6WAy61aaSwI2T DfTYEyJI2j9mXzCVnsGT/rSIDj0+gREglrY8l25zGwsL5s398tsHZ1LtFbcdzGZEZEnu Jpy6LuprB6+ipY9woDYXM6iKXtJPUiD/QLqbjzOA0+YijnnYXF3vVwJyzoY6xlF3yDBU 9S1sULN/XSgTQ0dFiv9DBQPA71JpEVpC+vHkeev5ICC81zFHhFqubUPtDCa+cWPnny6g aPJA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1769156584; x=1769761384; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-gg:x-gm-message-state:from :to:cc:subject:date:message-id:reply-to; bh=gNFjSEtLjmo3zavrqqxSndVZ5Mb9BZ08qMwOVvVVYnk=; b=hqXRGQwOED4IPrc6DiR8TWO+nL+HxEpdahrV9R/KuYsHhVLc3azjt0Jour3qayEj0P ynwdVr7Fj/AfFvppafs7F+S+oN+ThhZm9Z+eKibhWFPEXU3ssG5ymJ/McsYgVRx9GA6Q BBC0biu3g1YUnsQakrQE9qgFeYzpSO6ovg1ZEWqX2sYDzfNmzhXoE0oHc3jmsTJIi4Vt reaV5j1GlsMfms0jOIxFKE4TNQe0lhCAfpA9HPGk8D96Pmp7Er2d0aXMvqohusebpTTs jqJvBzH2l3ANu+6TEYqOsrN8rnhcnL86QoLTVMcnrZ+db1n+pink5basn4lXKaDgakHW xMHQ== X-Forwarded-Encrypted: i=1; AJvYcCWax2HhBBUXPm7YeBtsjSuHS6tK39lhsl/vBQbmKFzWqUTiq16iOwYeMh5q/jx3Wxln3xGxEMaTlpCoD2s=@vger.kernel.org X-Gm-Message-State: AOJu0YxWfRZiXofKiFhQYR77LVVKaqg37Um4k/6ynrLQlqWkuU05CRHm 85NRzgcxM1sfA5zKMV4f4vkc+QsXKdf0+e8l2I0MdluuQPeZZ/jABmW1 X-Gm-Gg: AZuq6aJTFThCPNBnyVl/wcGtLxFveCbFhUaBicWfeboI4EgM1pDFoS4KTr5Pty7QD1t b00TziHZLAplDf7uzCsR87kcrxAnJGGuj67aEKqj9pHzIxDQuwZzd4TSFOgqwLqK8imxICfEcL+ T6d9+wvYYwGkabEwYug8mkDhkFb/maU8NCHN6AozdPhcjUh0Tm5OaZWrvbqqsni1Hxqkdj34R7o fue6qIBy+fZiOwEoiBtl9VkSnnhiZ+Fr5ZZlLwTZksRmADslPV5BKGBSbx7iIROcl0r95UJmxgX bG7QbXsdU6NdoFQJi+fFM6Cf2mpZ6fX9QS5JQ1fOm12KROCBze2Iehpl7OWIuD28lczjUEOHw1S BZQGgasaUuZzzHaU3IJnqBql/EBccDagHP1ajZEUR3NnaVUXIcfUlDbDwNK+xqEy/tzYIzJCykX HinvIMV7FgfY8Byjn9dPMqVDBT X-Received: by 2002:a05:6a21:116:b0:35f:84c7:4012 with SMTP id adf61e73a8af0-38e7008e9e2mr1886524637.29.1769156583989; Fri, 23 Jan 2026 00:23:03 -0800 (PST) Received: from localhost.localdomain ([221.227.246.159]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-c635a43f11csm1348363a12.35.2026.01.23.00.23.01 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 23 Jan 2026 00:23:03 -0800 (PST) From: Vernon Yang To: akpm@linux-foundation.org, david@kernel.org Cc: lorenzo.stoakes@oracle.com, ziy@nvidia.com, dev.jain@arm.com, baohua@kernel.org, lance.yang@linux.dev, linux-mm@kvack.org, linux-kernel@vger.kernel.org, Vernon Yang Subject: [PATCH mm-new v5 2/5] mm: khugepaged: refine scan progress number Date: Fri, 23 Jan 2026 16:22:29 +0800 Message-ID: <20260123082232.16413-3-vernon2gm@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20260123082232.16413-1-vernon2gm@gmail.com> References: <20260123082232.16413-1-vernon2gm@gmail.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Vernon Yang Currently, each scan always increases "progress" by HPAGE_PMD_NR, even if only scanning a single PTE/PMD entry. - When only scanning a sigle PTE entry, let me provide a detailed example: static int hpage_collapse_scan_pmd() { for (addr =3D start_addr, _pte =3D pte; _pte < pte + HPAGE_PMD_NR; _pte++, addr +=3D PAGE_SIZE) { pte_t pteval =3D ptep_get(_pte); ... if (pte_uffd_wp(pteval)) { <-- first scan hit result =3D SCAN_PTE_UFFD_WP; goto out_unmap; } } } During the first scan, if pte_uffd_wp(pteval) is true, the loop exits directly. In practice, only one PTE is scanned before termination. Here, "progress +=3D 1" reflects the actual number of PTEs scanned, but previously "progress +=3D HPAGE_PMD_NR" always. - When the memory has been collapsed to PMD, let me provide a detailed example: The following data is traced by bpftrace on a desktop system. After the system has been left idle for 10 minutes upon booting, a lot of SCAN_PMD_MAPPED or SCAN_NO_PTE_TABLE are observed during a full scan by khugepaged. @scan_pmd_status[1]: 1 ## SCAN_SUCCEED @scan_pmd_status[6]: 2 ## SCAN_EXCEED_SHARED_PTE @scan_pmd_status[3]: 142 ## SCAN_PMD_MAPPED @scan_pmd_status[2]: 178 ## SCAN_NO_PTE_TABLE total progress size: 674 MB Total time : 419 seconds ## include khugepaged_scan_sleep_millisecs The khugepaged_scan list save all task that support collapse into hugepage, as long as the task is not destroyed, khugepaged will not remove it from the khugepaged_scan list. This exist a phenomenon where task has already collapsed all memory regions into hugepage, but khugepaged continues to scan it, which wastes CPU time and invalid, and due to khugepaged_scan_sleep_millisecs (default 10s) causes a long wait for scanning a large number of invalid task, so scanning really valid task is later. After applying this patch, when the memory is either SCAN_PMD_MAPPED or SCAN_NO_PTE_TABLE, just skip it, as follow: @scan_pmd_status[6]: 2 @scan_pmd_status[3]: 147 @scan_pmd_status[2]: 173 total progress size: 45 MB Total time : 20 seconds Signed-off-by: Vernon Yang --- include/linux/xarray.h | 9 ++++++++ mm/khugepaged.c | 47 ++++++++++++++++++++++++++++++++++-------- 2 files changed, 47 insertions(+), 9 deletions(-) diff --git a/include/linux/xarray.h b/include/linux/xarray.h index be850174e802..f77d97d7b957 100644 --- a/include/linux/xarray.h +++ b/include/linux/xarray.h @@ -1646,6 +1646,15 @@ static inline void xas_set(struct xa_state *xas, uns= igned long index) xas->xa_node =3D XAS_RESTART; } =20 +/** + * xas_get_index() - Get XArray operation state for a different index. + * @xas: XArray operation state. + */ +static inline unsigned long xas_get_index(struct xa_state *xas) +{ + return xas->xa_index; +} + /** * xas_advance() - Skip over sibling entries. * @xas: XArray operation state. diff --git a/mm/khugepaged.c b/mm/khugepaged.c index 6f0f05148765..de95029e3763 100644 --- a/mm/khugepaged.c +++ b/mm/khugepaged.c @@ -68,7 +68,10 @@ enum scan_result { static struct task_struct *khugepaged_thread __read_mostly; static DEFINE_MUTEX(khugepaged_mutex); =20 -/* default scan 8*HPAGE_PMD_NR ptes (or vmas) every 10 second */ +/* + * default scan 8*HPAGE_PMD_NR ptes, pmd_mapped, no_pte_table or vmas + * every 10 second. + */ static unsigned int khugepaged_pages_to_scan __read_mostly; static unsigned int khugepaged_pages_collapsed; static unsigned int khugepaged_full_scans; @@ -1240,7 +1243,8 @@ static enum scan_result collapse_huge_page(struct mm_= struct *mm, unsigned long a } =20 static enum scan_result hpage_collapse_scan_pmd(struct mm_struct *mm, - struct vm_area_struct *vma, unsigned long start_addr, bool *mmap_locked, + struct vm_area_struct *vma, unsigned long start_addr, + bool *mmap_locked, unsigned int *cur_progress, struct collapse_control *cc) { pmd_t *pmd; @@ -1255,6 +1259,9 @@ static enum scan_result hpage_collapse_scan_pmd(struc= t mm_struct *mm, =20 VM_BUG_ON(start_addr & ~HPAGE_PMD_MASK); =20 + if (cur_progress) + *cur_progress +=3D 1; + result =3D find_pmd_or_thp_or_none(mm, start_addr, &pmd); if (result !=3D SCAN_SUCCEED) goto out; @@ -1396,6 +1403,12 @@ static enum scan_result hpage_collapse_scan_pmd(stru= ct mm_struct *mm, result =3D SCAN_SUCCEED; } out_unmap: + if (cur_progress) { + if (_pte >=3D pte + HPAGE_PMD_NR) + *cur_progress +=3D HPAGE_PMD_NR - 1; + else + *cur_progress +=3D _pte - pte; + } pte_unmap_unlock(pte, ptl); if (result =3D=3D SCAN_SUCCEED) { result =3D collapse_huge_page(mm, start_addr, referenced, @@ -2286,8 +2299,9 @@ static enum scan_result collapse_file(struct mm_struc= t *mm, unsigned long addr, return result; } =20 -static enum scan_result hpage_collapse_scan_file(struct mm_struct *mm, uns= igned long addr, - struct file *file, pgoff_t start, struct collapse_control *cc) +static enum scan_result hpage_collapse_scan_file(struct mm_struct *mm, + unsigned long addr, struct file *file, pgoff_t start, + unsigned int *cur_progress, struct collapse_control *cc) { struct folio *folio =3D NULL; struct address_space *mapping =3D file->f_mapping; @@ -2376,6 +2390,18 @@ static enum scan_result hpage_collapse_scan_file(str= uct mm_struct *mm, unsigned cond_resched_rcu(); } } + if (cur_progress) { + unsigned long idx =3D xas_get_index(&xas) - start; + + if (folio =3D=3D NULL) + *cur_progress +=3D HPAGE_PMD_NR; + else if (xa_is_value(folio)) + *cur_progress +=3D idx + (1 << xas_get_order(&xas)); + else if (folio_order(folio) =3D=3D HPAGE_PMD_ORDER) + *cur_progress +=3D idx + 1; + else + *cur_progress +=3D idx + folio_nr_pages(folio); + } rcu_read_unlock(); =20 if (result =3D=3D SCAN_SUCCEED) { @@ -2456,6 +2482,7 @@ static unsigned int khugepaged_scan_mm_slot(unsigned = int pages, enum scan_result =20 while (khugepaged_scan.address < hend) { bool mmap_locked =3D true; + unsigned int cur_progress =3D 0; =20 cond_resched(); if (unlikely(hpage_collapse_test_exit_or_disable(mm))) @@ -2472,7 +2499,8 @@ static unsigned int khugepaged_scan_mm_slot(unsigned = int pages, enum scan_result mmap_read_unlock(mm); mmap_locked =3D false; *result =3D hpage_collapse_scan_file(mm, - khugepaged_scan.address, file, pgoff, cc); + khugepaged_scan.address, file, pgoff, + &cur_progress, cc); fput(file); if (*result =3D=3D SCAN_PTE_MAPPED_HUGEPAGE) { mmap_read_lock(mm); @@ -2486,7 +2514,8 @@ static unsigned int khugepaged_scan_mm_slot(unsigned = int pages, enum scan_result } } else { *result =3D hpage_collapse_scan_pmd(mm, vma, - khugepaged_scan.address, &mmap_locked, cc); + khugepaged_scan.address, &mmap_locked, + &cur_progress, cc); } =20 if (*result =3D=3D SCAN_SUCCEED) @@ -2494,7 +2523,7 @@ static unsigned int khugepaged_scan_mm_slot(unsigned = int pages, enum scan_result =20 /* move to next address */ khugepaged_scan.address +=3D HPAGE_PMD_SIZE; - progress +=3D HPAGE_PMD_NR; + progress +=3D cur_progress; if (!mmap_locked) /* * We released mmap_lock so break loop. Note @@ -2817,7 +2846,7 @@ int madvise_collapse(struct vm_area_struct *vma, unsi= gned long start, mmap_locked =3D false; *lock_dropped =3D true; result =3D hpage_collapse_scan_file(mm, addr, file, pgoff, - cc); + NULL, cc); =20 if (result =3D=3D SCAN_PAGE_DIRTY_OR_WRITEBACK && !triggered_wb && mapping_can_writeback(file->f_mapping)) { @@ -2832,7 +2861,7 @@ int madvise_collapse(struct vm_area_struct *vma, unsi= gned long start, fput(file); } else { result =3D hpage_collapse_scan_pmd(mm, vma, addr, - &mmap_locked, cc); + &mmap_locked, NULL, cc); } if (!mmap_locked) *lock_dropped =3D true; --=20 2.51.0