From nobody Thu Dec 18 04:51:15 2025 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id C50D5E7734F for ; Sat, 30 Sep 2023 03:31:39 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233972AbjI3Dbi (ORCPT ); Fri, 29 Sep 2023 23:31:38 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:52698 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229764AbjI3Dbe (ORCPT ); Fri, 29 Sep 2023 23:31:34 -0400 Received: from mail-yw1-x1131.google.com (mail-yw1-x1131.google.com [IPv6:2607:f8b0:4864:20::1131]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 35BB8BA for ; Fri, 29 Sep 2023 20:31:31 -0700 (PDT) Received: by mail-yw1-x1131.google.com with SMTP id 00721157ae682-5a22f9e2f40so23100777b3.1 for ; Fri, 29 Sep 2023 20:31:31 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1696044690; x=1696649490; darn=vger.kernel.org; h=mime-version:references:message-id:in-reply-to:subject:cc:to:from :date:from:to:cc:subject:date:message-id:reply-to; bh=QspsCGzxpxq3SSLDt+LJp9Jy//8oF6+tYorlwqGdsRA=; b=FVeE6k2L8wP8t7TzSUr7z3tKhbLL753juRIs1Pm1jUQpzO5VguculLP9NNS9J/izpc uwF8bc+zEkdBVEPuBqkn7M+zXM7fHFe32zxL8gf6PI+Ul75ofc/93NuSDuaCh2Vchc/G FVifpXEfHbF9pvvqRIGKOaMlpLY64Pg8h3ijY/jUuzsCPrCSt/KOwMQyKP5nNsEwFKoD epzd3p2+Gk1HSM7tl8WSygnGViXjzW8of00Doe58eQajwEr24n4s0DTYag5WTRImFvss Fxnc87PJzT1D2w9fx7jELbfruPab3T7gzvikmmHHxokA0WsCtAY07MhDjJTzz720FtMz UF9g== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1696044690; x=1696649490; h=mime-version:references:message-id:in-reply-to:subject:cc:to:from :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=QspsCGzxpxq3SSLDt+LJp9Jy//8oF6+tYorlwqGdsRA=; b=vQ3XrAP14f0x6cVv4DTRO3OZzxLaLvunGQCmITSnTn4/93MObV/iy21uxujoODK6ka Ajcb5r2sjOFG6n2haghziIi5+DjpAhR6/vPBdyEGMtJZLkmE43yB7E9JzLr+xi0l3mF2 NFPY/8rV/EFddeBO6QhfTEklr4G46WXRuF+DS23FbCIt4wX5NDWplcIJTHKKo3QIiZrl /TWuSTE/pcFZ5ouZGy8NujoyMy72tWcgZG8+T4YOegwU8MBRpCqlDAu4ccv6i4v2VnUo YGMJR2XeeKSbtEZJEly2lQCkujfF1o4ADr5J5isjn5XXFiT6g6Haq62dCjWx5RbbOxC8 73RA== X-Gm-Message-State: AOJu0YwUVx9Dw/hhZVwdoV/pqW3kB+sKGRrZ/tCeDMFpFakomuoocXJC 3f2FO1+f/FxowwmQpP2tlFoM1aoI1xttBDnLYQojTg== X-Google-Smtp-Source: AGHT+IEgXqnD7PMNlAh5dIBX1/Qy7aURmmr6eaqI4X8ytb5LLyV63aow9Zb+c+tnCOaAvJkKxO8m7Q== X-Received: by 2002:a0d:d3c5:0:b0:585:ef4e:6d93 with SMTP id v188-20020a0dd3c5000000b00585ef4e6d93mr5637010ywd.47.1696044690238; Fri, 29 Sep 2023 20:31:30 -0700 (PDT) Received: from ripple.attlocal.net (172-10-233-147.lightspeed.sntcca.sbcglobal.net. [172.10.233.147]) by smtp.gmail.com with ESMTPSA id d71-20020a814f4a000000b0059be3d854b1sm5928458ywb.109.2023.09.29.20.31.28 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 29 Sep 2023 20:31:29 -0700 (PDT) Date: Fri, 29 Sep 2023 20:31:27 -0700 (PDT) From: Hugh Dickins X-X-Sender: hugh@ripple.attlocal.net To: Andrew Morton cc: Christian Brauner , Carlos Maiolino , Chuck Lever , Jan Kara , Matthew Wilcox , Johannes Weiner , Axel Rasmussen , linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org Subject: [PATCH 6/8] shmem: move memcg charge out of shmem_add_to_page_cache() In-Reply-To: Message-ID: <4b2143c5-bf32-64f0-841-81a81158dac@google.com> References: MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Extract shmem's memcg charging out of shmem_add_to_page_cache(): it's misleading done there, because many calls are dealing with a swapcache page, whose memcg is nowadays always remembered while swapped out, then the charge re-levied when it's brought back into swapcache. Temporarily move it back up to the shmem_get_folio_gfp() level, where the memcg was charged before v5.8; but the next commit goes on to move it back down to a new home. In making this change, it becomes clear that shmem_swapin_folio() does not need to know the vma, just the fault mm (if any): call it fault_mm rather than charge_mm - let mem_cgroup_charge() decide whom to charge. Signed-off-by: Hugh Dickins Reviewed-by: Jan Kara --- mm/shmem.c | 68 +++++++++++++++++++++++------------------------------- 1 file changed, 29 insertions(+), 39 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index 63ba6037b23a..0a7f7b567b80 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -146,9 +146,8 @@ static unsigned long shmem_default_max_inodes(void) #endif =20 static int shmem_swapin_folio(struct inode *inode, pgoff_t index, - struct folio **foliop, enum sgp_type sgp, - gfp_t gfp, struct vm_area_struct *vma, - vm_fault_t *fault_type); + struct folio **foliop, enum sgp_type sgp, gfp_t gfp, + struct mm_struct *fault_mm, vm_fault_t *fault_type); =20 static inline struct shmem_sb_info *SHMEM_SB(struct super_block *sb) { @@ -760,12 +759,10 @@ static unsigned long shmem_unused_huge_shrink(struct = shmem_sb_info *sbinfo, */ static int shmem_add_to_page_cache(struct folio *folio, struct address_space *mapping, - pgoff_t index, void *expected, gfp_t gfp, - struct mm_struct *charge_mm) + pgoff_t index, void *expected, gfp_t gfp) { XA_STATE_ORDER(xas, &mapping->i_pages, index, folio_order(folio)); long nr =3D folio_nr_pages(folio); - int error; =20 VM_BUG_ON_FOLIO(index !=3D round_down(index, nr), folio); VM_BUG_ON_FOLIO(!folio_test_locked(folio), folio); @@ -776,16 +773,7 @@ static int shmem_add_to_page_cache(struct folio *folio, folio->mapping =3D mapping; folio->index =3D index; =20 - if (!folio_test_swapcache(folio)) { - error =3D mem_cgroup_charge(folio, charge_mm, gfp); - if (error) { - if (folio_test_pmd_mappable(folio)) { - count_vm_event(THP_FILE_FALLBACK); - count_vm_event(THP_FILE_FALLBACK_CHARGE); - } - goto error; - } - } + gfp &=3D GFP_RECLAIM_MASK; folio_throttle_swaprate(folio, gfp); =20 do { @@ -813,15 +801,12 @@ static int shmem_add_to_page_cache(struct folio *foli= o, } while (xas_nomem(&xas, gfp)); =20 if (xas_error(&xas)) { - error =3D xas_error(&xas); - goto error; + folio->mapping =3D NULL; + folio_ref_sub(folio, nr); + return xas_error(&xas); } =20 return 0; -error: - folio->mapping =3D NULL; - folio_ref_sub(folio, nr); - return error; } =20 /* @@ -1324,10 +1309,8 @@ static int shmem_unuse_swap_entries(struct inode *in= ode, =20 if (!xa_is_value(folio)) continue; - error =3D shmem_swapin_folio(inode, indices[i], - &folio, SGP_CACHE, - mapping_gfp_mask(mapping), - NULL, NULL); + error =3D shmem_swapin_folio(inode, indices[i], &folio, SGP_CACHE, + mapping_gfp_mask(mapping), NULL, NULL); if (error =3D=3D 0) { folio_unlock(folio); folio_put(folio); @@ -1810,12 +1793,11 @@ static void shmem_set_folio_swapin_error(struct ino= de *inode, pgoff_t index, */ static int shmem_swapin_folio(struct inode *inode, pgoff_t index, struct folio **foliop, enum sgp_type sgp, - gfp_t gfp, struct vm_area_struct *vma, + gfp_t gfp, struct mm_struct *fault_mm, vm_fault_t *fault_type) { struct address_space *mapping =3D inode->i_mapping; struct shmem_inode_info *info =3D SHMEM_I(inode); - struct mm_struct *charge_mm =3D vma ? vma->vm_mm : NULL; struct swap_info_struct *si; struct folio *folio =3D NULL; swp_entry_t swap; @@ -1843,7 +1825,7 @@ static int shmem_swapin_folio(struct inode *inode, pg= off_t index, if (fault_type) { *fault_type |=3D VM_FAULT_MAJOR; count_vm_event(PGMAJFAULT); - count_memcg_event_mm(charge_mm, PGMAJFAULT); + count_memcg_event_mm(fault_mm, PGMAJFAULT); } /* Here we actually start the io */ folio =3D shmem_swapin(swap, gfp, info, index); @@ -1880,8 +1862,7 @@ static int shmem_swapin_folio(struct inode *inode, pg= off_t index, } =20 error =3D shmem_add_to_page_cache(folio, mapping, index, - swp_to_radix_entry(swap), gfp, - charge_mm); + swp_to_radix_entry(swap), gfp); if (error) goto failed; =20 @@ -1929,7 +1910,7 @@ static int shmem_get_folio_gfp(struct inode *inode, p= goff_t index, struct address_space *mapping =3D inode->i_mapping; struct shmem_inode_info *info =3D SHMEM_I(inode); struct shmem_sb_info *sbinfo; - struct mm_struct *charge_mm; + struct mm_struct *fault_mm; struct folio *folio; pgoff_t hindex; gfp_t huge_gfp; @@ -1946,7 +1927,7 @@ static int shmem_get_folio_gfp(struct inode *inode, p= goff_t index, } =20 sbinfo =3D SHMEM_SB(inode->i_sb); - charge_mm =3D vma ? vma->vm_mm : NULL; + fault_mm =3D vma ? vma->vm_mm : NULL; =20 folio =3D filemap_get_entry(mapping, index); if (folio && vma && userfaultfd_minor(vma)) { @@ -1958,7 +1939,7 @@ static int shmem_get_folio_gfp(struct inode *inode, p= goff_t index, =20 if (xa_is_value(folio)) { error =3D shmem_swapin_folio(inode, index, &folio, - sgp, gfp, vma, fault_type); + sgp, gfp, fault_mm, fault_type); if (error =3D=3D -EEXIST) goto repeat; =20 @@ -2044,9 +2025,16 @@ static int shmem_get_folio_gfp(struct inode *inode, = pgoff_t index, if (sgp =3D=3D SGP_WRITE) __folio_set_referenced(folio); =20 - error =3D shmem_add_to_page_cache(folio, mapping, hindex, - NULL, gfp & GFP_RECLAIM_MASK, - charge_mm); + error =3D mem_cgroup_charge(folio, fault_mm, gfp); + if (error) { + if (folio_test_pmd_mappable(folio)) { + count_vm_event(THP_FILE_FALLBACK); + count_vm_event(THP_FILE_FALLBACK_CHARGE); + } + goto unacct; + } + + error =3D shmem_add_to_page_cache(folio, mapping, hindex, NULL, gfp); if (error) goto unacct; =20 @@ -2644,8 +2632,10 @@ int shmem_mfill_atomic_pte(pmd_t *dst_pmd, if (unlikely(pgoff >=3D max_off)) goto out_release; =20 - ret =3D shmem_add_to_page_cache(folio, mapping, pgoff, NULL, - gfp & GFP_RECLAIM_MASK, dst_vma->vm_mm); + ret =3D mem_cgroup_charge(folio, dst_vma->vm_mm, gfp); + if (ret) + goto out_release; + ret =3D shmem_add_to_page_cache(folio, mapping, pgoff, NULL, gfp); if (ret) goto out_release; =20 --=20 2.35.3