From nobody Thu Dec 18 15:24:28 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 E9649EE49A6 for ; Mon, 21 Aug 2023 16:10:07 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236015AbjHUQKH (ORCPT ); Mon, 21 Aug 2023 12:10:07 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:56312 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235944AbjHUQKC (ORCPT ); Mon, 21 Aug 2023 12:10:02 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 308A5103 for ; Mon, 21 Aug 2023 09:09:08 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1692634142; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=JSFFh/KLUGTq+zDQ/y6zlFb70l3lmZmKgiSBR+NyjZY=; b=MT2mNmXgc+vI20AhORo3Z/U/+lZIlvFiXxKuk/k+VoMSqkn0Le0qYfUZgp0TC0CHdEjVEn kcx2H6ct4qBwKv2cvbxn16WmQCwjX1ccdwQKL3S+FUjPcFRJvBb+BOFkW7Iz9Za2F8pbgm Mgw/yyPshAs6g+jq5HzH8YEdLMblF7c= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-520-GEYiNV70OW-yfFc5WKiRKg-1; Mon, 21 Aug 2023 12:08:59 -0400 X-MC-Unique: GEYiNV70OW-yfFc5WKiRKg-1 Received: from smtp.corp.redhat.com (int-mx09.intmail.prod.int.rdu2.redhat.com [10.11.54.9]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id E42CB187573F; Mon, 21 Aug 2023 16:08:55 +0000 (UTC) Received: from t14s.fritz.box (unknown [10.39.192.184]) by smtp.corp.redhat.com (Postfix) with ESMTP id 56E2C492C14; Mon, 21 Aug 2023 16:08:53 +0000 (UTC) From: David Hildenbrand To: linux-kernel@vger.kernel.org Cc: linux-mm@kvack.org, linux-arm-kernel@lists.infradead.org, David Hildenbrand , Andrew Morton , Matthew Wilcox , Peter Xu , Catalin Marinas , Will Deacon , Hugh Dickins , Seth Jennings , Dan Streetman , Vitaly Wool Subject: [PATCH mm-unstable v1 1/4] mm/swap: stop using page->private on tail pages for THP_SWAP Date: Mon, 21 Aug 2023 18:08:46 +0200 Message-ID: <20230821160849.531668-2-david@redhat.com> In-Reply-To: <20230821160849.531668-1-david@redhat.com> References: <20230821160849.531668-1-david@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.9 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Let's stop using page->private on tail pages, making it possible to just unconditionally reuse that field in the tail pages of large folios. The remaining usage of the private field for THP_SWAP is in the THP splitting code (mm/huge_memory.c), that we'll handle separately later. Update the THP_SWAP documentation and sanity checks in mm_types.h and __split_huge_page_tail(). Signed-off-by: David Hildenbrand Acked-by: Catalin Marinas Reviewed-by: Yosry Ahmed --- arch/arm64/mm/mteswap.c | 5 +++-- include/linux/mm_types.h | 12 +----------- include/linux/swap.h | 9 +++++++++ mm/huge_memory.c | 15 ++++++--------- mm/memory.c | 2 +- mm/rmap.c | 2 +- mm/swap_state.c | 5 +++-- mm/swapfile.c | 4 ++-- 8 files changed, 26 insertions(+), 28 deletions(-) diff --git a/arch/arm64/mm/mteswap.c b/arch/arm64/mm/mteswap.c index cd508ba80ab1..a31833e3ddc5 100644 --- a/arch/arm64/mm/mteswap.c +++ b/arch/arm64/mm/mteswap.c @@ -33,8 +33,9 @@ int mte_save_tags(struct page *page) =20 mte_save_page_tags(page_address(page), tag_storage); =20 - /* page_private contains the swap entry.val set in do_swap_page */ - ret =3D xa_store(&mte_pages, page_private(page), tag_storage, GFP_KERNEL); + /* lookup the swap entry.val from the page */ + ret =3D xa_store(&mte_pages, page_swap_entry(page).val, tag_storage, + GFP_KERNEL); if (WARN(xa_is_err(ret), "Failed to store MTE tags")) { mte_free_tag_storage(tag_storage); return xa_err(ret); diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index b9b6c88875b9..61361f1750c3 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -333,11 +333,8 @@ struct folio { atomic_t _pincount; #ifdef CONFIG_64BIT unsigned int _folio_nr_pages; - /* 4 byte gap here */ - /* private: the union with struct page is transitional */ - /* Fix THP_SWAP to not use tail->private */ - unsigned long _private_1; #endif + /* private: the union with struct page is transitional */ }; struct page __page_1; }; @@ -358,9 +355,6 @@ struct folio { /* public: */ struct list_head _deferred_list; /* private: the union with struct page is transitional */ - unsigned long _avail_2a; - /* Fix THP_SWAP to not use tail->private */ - unsigned long _private_2a; }; struct page __page_2; }; @@ -385,9 +379,6 @@ FOLIO_MATCH(memcg_data, memcg_data); offsetof(struct page, pg) + sizeof(struct page)) FOLIO_MATCH(flags, _flags_1); FOLIO_MATCH(compound_head, _head_1); -#ifdef CONFIG_64BIT -FOLIO_MATCH(private, _private_1); -#endif #undef FOLIO_MATCH #define FOLIO_MATCH(pg, fl) \ static_assert(offsetof(struct folio, fl) =3D=3D \ @@ -396,7 +387,6 @@ FOLIO_MATCH(flags, _flags_2); FOLIO_MATCH(compound_head, _head_2); FOLIO_MATCH(flags, _flags_2a); FOLIO_MATCH(compound_head, _head_2a); -FOLIO_MATCH(private, _private_2a); #undef FOLIO_MATCH =20 /** diff --git a/include/linux/swap.h b/include/linux/swap.h index bb5adc604144..84fe0e94f5cd 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -339,6 +339,15 @@ static inline swp_entry_t folio_swap_entry(struct foli= o *folio) return entry; } =20 +static inline swp_entry_t page_swap_entry(struct page *page) +{ + struct folio *folio =3D page_folio(page); + swp_entry_t entry =3D folio_swap_entry(folio); + + entry.val +=3D page - &folio->page; + return entry; +} + static inline void folio_set_swap_entry(struct folio *folio, swp_entry_t e= ntry) { folio->private =3D (void *)entry.val; diff --git a/mm/huge_memory.c b/mm/huge_memory.c index cc2f65f8cc62..c04702ae71d2 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -2446,18 +2446,15 @@ static void __split_huge_page_tail(struct page *hea= d, int tail, page_tail->index =3D head->index + tail; =20 /* - * page->private should not be set in tail pages with the exception - * of swap cache pages that store the swp_entry_t in tail pages. - * Fix up and warn once if private is unexpectedly set. - * - * What of 32-bit systems, on which folio->_pincount overlays - * head[1].private? No problem: THP_SWAP is not enabled on 32-bit, and - * pincount must be 0 for folio_ref_freeze() to have succeeded. + * page->private should not be set in tail pages. Fix up and warn once + * if private is unexpectedly set. */ - if (!folio_test_swapcache(page_folio(head))) { - VM_WARN_ON_ONCE_PAGE(page_tail->private !=3D 0, page_tail); + if (unlikely(page_tail->private)) { + VM_WARN_ON_ONCE_PAGE(true, page_tail); page_tail->private =3D 0; } + if (PageSwapCache(head)) + set_page_private(page_tail, (unsigned long)head->private + tail); =20 /* Page flags must be visible before we make the page non-compound. */ smp_wmb(); diff --git a/mm/memory.c b/mm/memory.c index d003076b218d..ff13242c1589 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -3882,7 +3882,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) * changed. */ if (unlikely(!folio_test_swapcache(folio) || - page_private(page) !=3D entry.val)) + page_swap_entry(page).val !=3D entry.val)) goto out_page; =20 /* diff --git a/mm/rmap.c b/mm/rmap.c index 1f04debdc87a..ec7f8e6c9e48 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -1647,7 +1647,7 @@ static bool try_to_unmap_one(struct folio *folio, str= uct vm_area_struct *vma, */ dec_mm_counter(mm, mm_counter(&folio->page)); } else if (folio_test_anon(folio)) { - swp_entry_t entry =3D { .val =3D page_private(subpage) }; + swp_entry_t entry =3D page_swap_entry(subpage); pte_t swp_pte; /* * Store the swap location in the pte. diff --git a/mm/swap_state.c b/mm/swap_state.c index 01f15139b7d9..2f2417810052 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -100,6 +100,7 @@ int add_to_swap_cache(struct folio *folio, swp_entry_t = entry, =20 folio_ref_add(folio, nr); folio_set_swapcache(folio); + folio_set_swap_entry(folio, entry); =20 do { xas_lock_irq(&xas); @@ -113,7 +114,6 @@ int add_to_swap_cache(struct folio *folio, swp_entry_t = entry, if (shadowp) *shadowp =3D old; } - set_page_private(folio_page(folio, i), entry.val + i); xas_store(&xas, folio); xas_next(&xas); } @@ -154,9 +154,10 @@ void __delete_from_swap_cache(struct folio *folio, for (i =3D 0; i < nr; i++) { void *entry =3D xas_store(&xas, shadow); VM_BUG_ON_PAGE(entry !=3D folio, entry); - set_page_private(folio_page(folio, i), 0); xas_next(&xas); } + entry.val =3D 0; + folio_set_swap_entry(folio, entry); folio_clear_swapcache(folio); address_space->nrpages -=3D nr; __node_stat_mod_folio(folio, NR_FILE_PAGES, -nr); diff --git a/mm/swapfile.c b/mm/swapfile.c index d46933adf789..bd9d904671b9 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -3369,7 +3369,7 @@ struct swap_info_struct *swp_swap_info(swp_entry_t en= try) =20 struct swap_info_struct *page_swap_info(struct page *page) { - swp_entry_t entry =3D { .val =3D page_private(page) }; + swp_entry_t entry =3D page_swap_entry(page); return swp_swap_info(entry); } =20 @@ -3384,7 +3384,7 @@ EXPORT_SYMBOL_GPL(swapcache_mapping); =20 pgoff_t __page_file_index(struct page *page) { - swp_entry_t swap =3D { .val =3D page_private(page) }; + swp_entry_t swap =3D page_swap_entry(page); return swp_offset(swap); } EXPORT_SYMBOL_GPL(__page_file_index); --=20 2.41.0 From nobody Thu Dec 18 15:24:28 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 58D7BEE49AA for ; Mon, 21 Aug 2023 16:10:09 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236120AbjHUQKJ (ORCPT ); Mon, 21 Aug 2023 12:10:09 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:56338 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235967AbjHUQKC (ORCPT ); Mon, 21 Aug 2023 12:10:02 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 9F69410F for ; Mon, 21 Aug 2023 09:09:11 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1692634146; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=oXgxSUiHlACYEZ8/tewWc2FJCrUF3dBlQ9Skg2Bh23k=; b=Pq/BL3S7WZzvJ9buNcwNKPZofA9JOOlKM8lT9Kk80y4FN1T0HdqTTkEk+gDv6OCFfFsDWE IrXQmQ2oYY1dT6xyZ+bmz9pKstgaJ6be3OOYwtI1ChqRyfd6ZgirrG0P2WvJNGpn2DZ5wN IVwvsbAEaBOabyXvFbuC3umJznEJjx0= Received: from mimecast-mx02.redhat.com (66.187.233.73 [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-312-gUGvI0rJOSu3AbVA6NUL9A-1; Mon, 21 Aug 2023 12:09:01 -0400 X-MC-Unique: gUGvI0rJOSu3AbVA6NUL9A-1 Received: from smtp.corp.redhat.com (int-mx09.intmail.prod.int.rdu2.redhat.com [10.11.54.9]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 807BE3C11A04; Mon, 21 Aug 2023 16:08:59 +0000 (UTC) Received: from t14s.fritz.box (unknown [10.39.192.184]) by smtp.corp.redhat.com (Postfix) with ESMTP id 5EBE9492C13; Mon, 21 Aug 2023 16:08:56 +0000 (UTC) From: David Hildenbrand To: linux-kernel@vger.kernel.org Cc: linux-mm@kvack.org, linux-arm-kernel@lists.infradead.org, David Hildenbrand , Andrew Morton , Matthew Wilcox , Peter Xu , Catalin Marinas , Will Deacon , Hugh Dickins , Seth Jennings , Dan Streetman , Vitaly Wool Subject: [PATCH mm-unstable v1 2/4] mm/swap: use dedicated entry for swap in folio Date: Mon, 21 Aug 2023 18:08:47 +0200 Message-ID: <20230821160849.531668-3-david@redhat.com> In-Reply-To: <20230821160849.531668-1-david@redhat.com> References: <20230821160849.531668-1-david@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.9 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" From: Matthew Wilcox Let's stop working on the private field and use an explicit swap field. We have to move the swp_entry_t typedef. Signed-off-by: Matthew Wilcox Signed-off-by: David Hildenbrand Reviewed-by: Chris Li --- include/linux/mm_types.h | 23 +++++++++++++---------- include/linux/swap.h | 5 ++--- 2 files changed, 15 insertions(+), 13 deletions(-) diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 61361f1750c3..438a07854f8c 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -259,6 +259,14 @@ static inline struct page *encoded_page_ptr(struct enc= oded_page *page) */ #define TAIL_MAPPING_REUSED_MAX (2) =20 +/* + * A swap entry has to fit into a "unsigned long", as the entry is hidden + * in the "index" field of the swapper address space. + */ +typedef struct { + unsigned long val; +} swp_entry_t; + /** * struct folio - Represents a contiguous set of bytes. * @flags: Identical to the page flags. @@ -269,7 +277,7 @@ static inline struct page *encoded_page_ptr(struct enco= ded_page *page) * @index: Offset within the file, in units of pages. For anonymous memor= y, * this is the index from the beginning of the mmap. * @private: Filesystem per-folio data (see folio_attach_private()). - * Used for swp_entry_t if folio_test_swapcache(). + * @swap: Used for swp_entry_t if folio_test_swapcache(). * @_mapcount: Do not access this member directly. Use folio_mapcount() to * find out how many times this folio is mapped by userspace. * @_refcount: Do not access this member directly. Use folio_ref_count() @@ -312,7 +320,10 @@ struct folio { }; struct address_space *mapping; pgoff_t index; - void *private; + union { + void *private; + swp_entry_t swap; + }; atomic_t _mapcount; atomic_t _refcount; #ifdef CONFIG_MEMCG @@ -1220,14 +1231,6 @@ enum tlb_flush_reason { NR_TLB_FLUSH_REASONS, }; =20 - /* - * A swap entry has to fit into a "unsigned long", as the entry is hidden - * in the "index" field of the swapper address space. - */ -typedef struct { - unsigned long val; -} swp_entry_t; - /** * enum fault_flag - Fault flag definitions. * @FAULT_FLAG_WRITE: Fault was a write fault. diff --git a/include/linux/swap.h b/include/linux/swap.h index 84fe0e94f5cd..82859a1944f5 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -335,8 +335,7 @@ struct swap_info_struct { =20 static inline swp_entry_t folio_swap_entry(struct folio *folio) { - swp_entry_t entry =3D { .val =3D page_private(&folio->page) }; - return entry; + return folio->swap; } =20 static inline swp_entry_t page_swap_entry(struct page *page) @@ -350,7 +349,7 @@ static inline swp_entry_t page_swap_entry(struct page *= page) =20 static inline void folio_set_swap_entry(struct folio *folio, swp_entry_t e= ntry) { - folio->private =3D (void *)entry.val; + folio->swap =3D entry; } =20 /* linux/mm/workingset.c */ --=20 2.41.0 From nobody Thu Dec 18 15:24:28 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 BC2E6EE49A6 for ; Mon, 21 Aug 2023 16:10:12 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236262AbjHUQKM (ORCPT ); Mon, 21 Aug 2023 12:10:12 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:56334 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235981AbjHUQKF (ORCPT ); Mon, 21 Aug 2023 12:10:05 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id BDDDD11F for ; Mon, 21 Aug 2023 09:09:11 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1692634146; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=GdnFUGqNF8uC/Mhnd72wov++fKDyb2cH73FrJqKy4cQ=; b=e3JNvKtoQghdCsspwNPKLI95sboueYhn/r7/VOHRgcn7kQ3IaXGyiwAiPm9gM8Mj/WNs5U IzUuNNKvwEVrBKnnG+U8HXSRh40bvwzZS/TLmd5XmHRF/Hr+tSNg94Zg4XKnAREQHg4ZxV lxIS09ET667vXTzTTeNvM9yGYCjNixE= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-343-L421dwfjPouKmahmihEKAA-1; Mon, 21 Aug 2023 12:09:02 -0400 X-MC-Unique: L421dwfjPouKmahmihEKAA-1 Received: from smtp.corp.redhat.com (int-mx09.intmail.prod.int.rdu2.redhat.com [10.11.54.9]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id DBFE01875842; Mon, 21 Aug 2023 16:09:01 +0000 (UTC) Received: from t14s.fritz.box (unknown [10.39.192.184]) by smtp.corp.redhat.com (Postfix) with ESMTP id B9873492C13; Mon, 21 Aug 2023 16:08:59 +0000 (UTC) From: David Hildenbrand To: linux-kernel@vger.kernel.org Cc: linux-mm@kvack.org, linux-arm-kernel@lists.infradead.org, David Hildenbrand , Andrew Morton , Matthew Wilcox , Peter Xu , Catalin Marinas , Will Deacon , Hugh Dickins , Seth Jennings , Dan Streetman , Vitaly Wool Subject: [PATCH mm-unstable v1 3/4] mm/swap: inline folio_set_swap_entry() and folio_swap_entry() Date: Mon, 21 Aug 2023 18:08:48 +0200 Message-ID: <20230821160849.531668-4-david@redhat.com> In-Reply-To: <20230821160849.531668-1-david@redhat.com> References: <20230821160849.531668-1-david@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.9 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Let's simply work on the folio directly and remove the helpers. Suggested-by: Matthew Wilcox Signed-off-by: David Hildenbrand Reviewed-by: Chris Li --- include/linux/swap.h | 12 +----------- mm/memory.c | 2 +- mm/shmem.c | 6 +++--- mm/swap_state.c | 7 +++---- mm/swapfile.c | 2 +- mm/util.c | 2 +- mm/vmscan.c | 2 +- mm/zswap.c | 4 ++-- 8 files changed, 13 insertions(+), 24 deletions(-) diff --git a/include/linux/swap.h b/include/linux/swap.h index 82859a1944f5..603acf813873 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -333,25 +333,15 @@ struct swap_info_struct { */ }; =20 -static inline swp_entry_t folio_swap_entry(struct folio *folio) -{ - return folio->swap; -} - static inline swp_entry_t page_swap_entry(struct page *page) { struct folio *folio =3D page_folio(page); - swp_entry_t entry =3D folio_swap_entry(folio); + swp_entry_t entry =3D folio->swap; =20 entry.val +=3D page - &folio->page; return entry; } =20 -static inline void folio_set_swap_entry(struct folio *folio, swp_entry_t e= ntry) -{ - folio->swap =3D entry; -} - /* linux/mm/workingset.c */ bool workingset_test_recent(void *shadow, bool file, bool *workingset); void workingset_age_nonresident(struct lruvec *lruvec, unsigned long nr_pa= ges); diff --git a/mm/memory.c b/mm/memory.c index ff13242c1589..c51800dbfa9b 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -3831,7 +3831,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) folio_add_lru(folio); =20 /* To provide entry to swap_readpage() */ - folio_set_swap_entry(folio, entry); + folio->swap =3D entry; swap_readpage(page, true, NULL); folio->private =3D NULL; } diff --git a/mm/shmem.c b/mm/shmem.c index 7a0c1e19d9f8..fc1afe9dfcfe 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1657,7 +1657,7 @@ static int shmem_replace_folio(struct folio **foliop,= gfp_t gfp, int error; =20 old =3D *foliop; - entry =3D folio_swap_entry(old); + entry =3D old->swap; swap_index =3D swp_offset(entry); swap_mapping =3D swap_address_space(entry); =20 @@ -1678,7 +1678,7 @@ static int shmem_replace_folio(struct folio **foliop,= gfp_t gfp, __folio_set_locked(new); __folio_set_swapbacked(new); folio_mark_uptodate(new); - folio_set_swap_entry(new, entry); + new->swap =3D entry; folio_set_swapcache(new); =20 /* @@ -1800,7 +1800,7 @@ static int shmem_swapin_folio(struct inode *inode, pg= off_t index, /* We have to do this with folio locked to prevent races */ folio_lock(folio); if (!folio_test_swapcache(folio) || - folio_swap_entry(folio).val !=3D swap.val || + folio->swap.val !=3D swap.val || !shmem_confirm_swap(mapping, index, swap)) { error =3D -EEXIST; goto unlock; diff --git a/mm/swap_state.c b/mm/swap_state.c index 2f2417810052..b3b14bd0dd64 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -100,7 +100,7 @@ int add_to_swap_cache(struct folio *folio, swp_entry_t = entry, =20 folio_ref_add(folio, nr); folio_set_swapcache(folio); - folio_set_swap_entry(folio, entry); + folio->swap =3D entry; =20 do { xas_lock_irq(&xas); @@ -156,8 +156,7 @@ void __delete_from_swap_cache(struct folio *folio, VM_BUG_ON_PAGE(entry !=3D folio, entry); xas_next(&xas); } - entry.val =3D 0; - folio_set_swap_entry(folio, entry); + folio->swap.val =3D 0; folio_clear_swapcache(folio); address_space->nrpages -=3D nr; __node_stat_mod_folio(folio, NR_FILE_PAGES, -nr); @@ -233,7 +232,7 @@ bool add_to_swap(struct folio *folio) */ void delete_from_swap_cache(struct folio *folio) { - swp_entry_t entry =3D folio_swap_entry(folio); + swp_entry_t entry =3D folio->swap; struct address_space *address_space =3D swap_address_space(entry); =20 xa_lock_irq(&address_space->i_pages); diff --git a/mm/swapfile.c b/mm/swapfile.c index bd9d904671b9..e52f486834eb 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -1536,7 +1536,7 @@ static bool swap_page_trans_huge_swapped(struct swap_= info_struct *si, =20 static bool folio_swapped(struct folio *folio) { - swp_entry_t entry =3D folio_swap_entry(folio); + swp_entry_t entry =3D folio->swap; struct swap_info_struct *si =3D _swap_info_get(entry); =20 if (!si) diff --git a/mm/util.c b/mm/util.c index cde229b05eb3..f31e2ca62cfa 100644 --- a/mm/util.c +++ b/mm/util.c @@ -764,7 +764,7 @@ struct address_space *folio_mapping(struct folio *folio) return NULL; =20 if (unlikely(folio_test_swapcache(folio))) - return swap_address_space(folio_swap_entry(folio)); + return swap_address_space(folio->swap); =20 mapping =3D folio->mapping; if ((unsigned long)mapping & PAGE_MAPPING_FLAGS) diff --git a/mm/vmscan.c b/mm/vmscan.c index c7c149cb8d66..6f13394b112e 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1423,7 +1423,7 @@ static int __remove_mapping(struct address_space *map= ping, struct folio *folio, } =20 if (folio_test_swapcache(folio)) { - swp_entry_t swap =3D folio_swap_entry(folio); + swp_entry_t swap =3D folio->swap; =20 if (reclaimed && !mapping_exiting(mapping)) shadow =3D workingset_eviction(folio, target_memcg); diff --git a/mm/zswap.c b/mm/zswap.c index 7300b98d4a03..412b1409a0d7 100644 --- a/mm/zswap.c +++ b/mm/zswap.c @@ -1190,7 +1190,7 @@ static void zswap_fill_page(void *ptr, unsigned long = value) =20 bool zswap_store(struct folio *folio) { - swp_entry_t swp =3D folio_swap_entry(folio); + swp_entry_t swp =3D folio->swap; int type =3D swp_type(swp); pgoff_t offset =3D swp_offset(swp); struct page *page =3D &folio->page; @@ -1370,7 +1370,7 @@ bool zswap_store(struct folio *folio) =20 bool zswap_load(struct folio *folio) { - swp_entry_t swp =3D folio_swap_entry(folio); + swp_entry_t swp =3D folio->swap; int type =3D swp_type(swp); pgoff_t offset =3D swp_offset(swp); struct page *page =3D &folio->page; --=20 2.41.0 From nobody Thu Dec 18 15:24:28 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 B4607EE49AB for ; Mon, 21 Aug 2023 16:10:10 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236201AbjHUQKK (ORCPT ); Mon, 21 Aug 2023 12:10:10 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:56322 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235986AbjHUQKG (ORCPT ); Mon, 21 Aug 2023 12:10:06 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 62580FE for ; Mon, 21 Aug 2023 09:09:13 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1692634152; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=J/PlBgSOglEnVLTJqu2zbMYsJDvOn/SB34qHQ+lZyNk=; b=gP6Ft9YJ0vVj1169p7RdODEkACqQ6fb1lUmjqbzc5Bf6Uv+/gWA/2UU/NVwy/13yHenr2N to8ei6q5zAv9gItA8atSHTKNkrMYggDmY/OEtcj1UFY8jetnhJxoZQIkbz4leAR4B+KFaI emg2B5tFos25WPJFOJ3Dr/5JANFzz7U= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-211-Afa7N3q2MaOpDmVf2OHBqA-1; Mon, 21 Aug 2023 12:09:09 -0400 X-MC-Unique: Afa7N3q2MaOpDmVf2OHBqA-1 Received: from smtp.corp.redhat.com (int-mx09.intmail.prod.int.rdu2.redhat.com [10.11.54.9]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 67DB9858EED; Mon, 21 Aug 2023 16:09:08 +0000 (UTC) Received: from t14s.fritz.box (unknown [10.39.192.184]) by smtp.corp.redhat.com (Postfix) with ESMTP id 219E8492C13; Mon, 21 Aug 2023 16:09:02 +0000 (UTC) From: David Hildenbrand To: linux-kernel@vger.kernel.org Cc: linux-mm@kvack.org, linux-arm-kernel@lists.infradead.org, David Hildenbrand , Andrew Morton , Matthew Wilcox , Peter Xu , Catalin Marinas , Will Deacon , Hugh Dickins , Seth Jennings , Dan Streetman , Vitaly Wool Subject: [PATCH mm-unstable v1 4/4] mm/huge_memory: work on folio->swap instead of page->private when splitting folio Date: Mon, 21 Aug 2023 18:08:49 +0200 Message-ID: <20230821160849.531668-5-david@redhat.com> In-Reply-To: <20230821160849.531668-1-david@redhat.com> References: <20230821160849.531668-1-david@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.9 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Let's work on folio->swap instead. While at it, use folio_test_anon() and folio_test_swapcache() -- the original folio remains valid even after splitting (but is then an order-0 folio). We can probably convert a lot more to folios in that code, let's focus on folio->swap handling only for now. Signed-off-by: David Hildenbrand Reviewed-by: Chris Li --- mm/huge_memory.c | 29 +++++++++++++++-------------- 1 file changed, 15 insertions(+), 14 deletions(-) diff --git a/mm/huge_memory.c b/mm/huge_memory.c index c04702ae71d2..4465915711c3 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -2401,10 +2401,16 @@ static void lru_add_page_tail(struct page *head, st= ruct page *tail, } } =20 -static void __split_huge_page_tail(struct page *head, int tail, +static void __split_huge_page_tail(struct folio *folio, int tail, struct lruvec *lruvec, struct list_head *list) { + struct page *head =3D &folio->page; struct page *page_tail =3D head + tail; + /* + * Careful: new_folio is not a "real" folio before we cleared PageTail. + * Don't pass it around before clear_compound_head(). + */ + struct folio *new_folio =3D (struct folio *)page_tail; =20 VM_BUG_ON_PAGE(atomic_read(&page_tail->_mapcount) !=3D -1, page_tail); =20 @@ -2453,8 +2459,8 @@ static void __split_huge_page_tail(struct page *head,= int tail, VM_WARN_ON_ONCE_PAGE(true, page_tail); page_tail->private =3D 0; } - if (PageSwapCache(head)) - set_page_private(page_tail, (unsigned long)head->private + tail); + if (folio_test_swapcache(folio)) + new_folio->swap.val =3D folio->swap.val + tail; =20 /* Page flags must be visible before we make the page non-compound. */ smp_wmb(); @@ -2500,11 +2506,9 @@ static void __split_huge_page(struct page *page, str= uct list_head *list, /* complete memcg works before add pages to LRU */ split_page_memcg(head, nr); =20 - if (PageAnon(head) && PageSwapCache(head)) { - swp_entry_t entry =3D { .val =3D page_private(head) }; - - offset =3D swp_offset(entry); - swap_cache =3D swap_address_space(entry); + if (folio_test_anon(folio) && folio_test_swapcache(folio)) { + offset =3D swp_offset(folio->swap); + swap_cache =3D swap_address_space(folio->swap); xa_lock(&swap_cache->i_pages); } =20 @@ -2514,7 +2518,7 @@ static void __split_huge_page(struct page *page, stru= ct list_head *list, ClearPageHasHWPoisoned(head); =20 for (i =3D nr - 1; i >=3D 1; i--) { - __split_huge_page_tail(head, i, lruvec, list); + __split_huge_page_tail(folio, i, lruvec, list); /* Some pages can be beyond EOF: drop them from page cache */ if (head[i].index >=3D end) { struct folio *tail =3D page_folio(head + i); @@ -2559,11 +2563,8 @@ static void __split_huge_page(struct page *page, str= uct list_head *list, =20 remap_page(folio, nr); =20 - if (PageSwapCache(head)) { - swp_entry_t entry =3D { .val =3D page_private(head) }; - - split_swap_cluster(entry); - } + if (folio_test_swapcache(folio)) + split_swap_cluster(folio->swap); =20 for (i =3D 0; i < nr; i++) { struct page *subpage =3D head + i; --=20 2.41.0