From nobody Sat Feb 7 15:15:36 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 E2851360741; Wed, 28 Jan 2026 13:56:05 +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=1769608566; cv=none; b=f6IJRItDk73trSmazBzPiSEzIBDZJXdI1PO0s+T+nZMQSlqZOZN1z4Dw4U1GXiq/s9zlzFcqNcoX0TQJ1118B/jIhvlgfL6nGeC+dQ+Qf5QWF+iddh2YYFolnUewLbjcKOHYSe8kj0/diiO7H0WAWXkm2xbYuGlLVp5vZ9w4TTg= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769608566; c=relaxed/simple; bh=tEvq02HqTGhVbwvuqPMQOCqNdmF5urkjY+a3mR1Nbe0=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=MgWwMomE6aOI7vC4/Om4f4mt1i2j0VJvavdxl4S2goEJpJM6Fbydpbxjw/oIBSljmA2fKQSQ50fHvXv9c2GOBqrB+gM+ZFHTvIi8KD5l0V2Pq7bznhhAPP0rrddK6UY6R4wiPluH/ckHKWCJNu9A0h4SZOGYc6F0cp9AZhEP9iM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=aeTbquYU; 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="aeTbquYU" Received: by smtp.kernel.org (Postfix) with ESMTPSA id B7D37C4AF09; Wed, 28 Jan 2026 13:56:04 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1769608565; bh=tEvq02HqTGhVbwvuqPMQOCqNdmF5urkjY+a3mR1Nbe0=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=aeTbquYUpbBvXgcn+nCLRd30N7kj0HLzpvLxDW5A9YB06CXqxDquhlGU/DUwBnCJ4 9BqYsssvsTQ1X/JVYtwpoAzofFrTCz5DwalTfDvtZs0zxsxT1PtF7J1opmJPuAZr0M P7laqQA4a++A18xQFluFhXX88iUR5oCf8gY7inOMsZRXUCfT03PXpTWPVoHFmayesC B9seaV9C1YfjsS+bEuVrnnOm7WFmrwzm4Oiz1z1YfLc2LQ4lj7f1szS1sTCgWqCUu+ MaFeOtb81DCtEObzJjhskbv7P7qdAaQRCK+adXNSe6C8qxcMRpHn41p3zcb9cueiRh tNYMQzNwRf3mQ== Received: from phl-compute-02.internal (phl-compute-02.internal [10.202.2.42]) by mailfauth.phl.internal (Postfix) with ESMTP id DFA50F4006D; Wed, 28 Jan 2026 08:56:03 -0500 (EST) Received: from phl-frontend-04 ([10.202.2.163]) by phl-compute-02.internal (MEProxy); Wed, 28 Jan 2026 08:56:03 -0500 X-ME-Sender: X-ME-Received: X-ME-Proxy-Cause: gggruggvucftvghtrhhoucdtuddrgeefgedrtddtgdduieefheegucetufdoteggodetrf dotffvucfrrhhofhhilhgvmecuhfgrshhtofgrihhlpdfurfetoffkrfgpnffqhgenuceu rghilhhouhhtmecufedttdenucesvcftvggtihhpihgvnhhtshculddquddttddmnecujf gurhephffvvefufffkofgjfhgggfestdekredtredttdenucfhrhhomhepmfhirhihlhcu ufhhuhhtshgvmhgruhcuoehkrghssehkvghrnhgvlhdrohhrgheqnecuggftrfgrthhtvg hrnhephfdufeejhefhkedtuedvfeevjeffvdfhvedtudfgudffjeefieekleehvdetvdev necuvehluhhsthgvrhfuihiivgeptdenucfrrghrrghmpehmrghilhhfrhhomhepkhhirh hilhhlodhmvghsmhhtphgruhhthhhpvghrshhonhgrlhhithihqdduieduudeivdeiheeh qddvkeeggeegjedvkedqkhgrsheppehkvghrnhgvlhdrohhrghesshhhuhhtvghmohhvrd hnrghmvgdpnhgspghrtghpthhtohepvdekpdhmohguvgepshhmthhpohhuthdprhgtphht thhopegrkhhpmheslhhinhhugidqfhhouhhnuggrthhiohhnrdhorhhgpdhrtghpthhtoh epmhhutghhuhhnrdhsohhngheslhhinhhugidruggvvhdprhgtphhtthhopegurghvihgu sehrvgguhhgrthdrtghomhdprhgtphhtthhopeifihhllhihsehinhhfrhgruggvrggurd horhhgpdhrtghpthhtohepuhhsrghmrggrrhhifheigedvsehgmhgrihhlrdgtohhmpdhr tghpthhtohepfhhvughlsehgohhoghhlvgdrtghomhdprhgtphhtthhopehoshgrlhhvrg guohhrsehsuhhsvgdruggvpdhrtghpthhtoheprhhpphhtsehkvghrnhgvlhdrohhrghdp rhgtphhtthhopehvsggrsghkrgesshhushgvrdgtii X-ME-Proxy: Feedback-ID: i10464835:Fastmail Received: by mail.messagingengine.com (Postfix) with ESMTPA; Wed, 28 Jan 2026 08:56:02 -0500 (EST) From: Kiryl Shutsemau To: Andrew Morton , Muchun Song , David Hildenbrand , Matthew Wilcox , Usama Arif , Frank van der Linden Cc: Oscar Salvador , Mike Rapoport , Vlastimil Babka , Lorenzo Stoakes , Zi Yan , Baoquan He , Michal Hocko , Johannes Weiner , Jonathan Corbet , Huacai Chen , WANG Xuerui , Palmer Dabbelt , Paul Walmsley , Albert Ou , Alexandre Ghiti , kernel-team@meta.com, linux-mm@kvack.org, linux-kernel@vger.kernel.org, linux-doc@vger.kernel.org, loongarch@lists.linux.dev, linux-riscv@lists.infradead.org, Kiryl Shutsemau Subject: [PATCHv5 07/17] mm: Rework compound_head() for power-of-2 sizeof(struct page) Date: Wed, 28 Jan 2026 13:54:48 +0000 Message-ID: <20260128135500.22121-8-kas@kernel.org> X-Mailer: git-send-email 2.51.2 In-Reply-To: <20260128135500.22121-1-kas@kernel.org> References: <20260128135500.22121-1-kas@kernel.org> 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" For tail pages, the kernel uses the 'compound_info' field to get to the head page. The bit 0 of the field indicates whether the page is a tail page, and if set, the remaining bits represent a pointer to the head page. For cases when size of struct page is power-of-2, change the encoding of compound_info to store a mask that can be applied to the virtual address of the tail page in order to access the head page. It is possible because struct page of the head page is naturally aligned with regards to order of the page. The significant impact of this modification is that all tail pages of the same order will now have identical 'compound_info', regardless of the compound page they are associated with. This paves the way for eliminating fake heads. The HugeTLB Vmemmap Optimization (HVO) creates fake heads and it is only applied when the sizeof(struct page) is power-of-2. Having identical tail pages allows the same page to be mapped into the vmemmap of all pages, maintaining memory savings without fake heads. If sizeof(struct page) is not power-of-2, there is no functional changes. Limit mask usage to HugeTLB vmemmap optimization (HVO) where it makes a difference. The approach with mask would work in the wider set of conditions, but it requires validating that struct pages are naturally aligned for all orders up to the MAX_FOLIO_ORDER, which can be tricky. Signed-off-by: Kiryl Shutsemau Reviewed-by: Muchun Song Reviewed-by: Zi Yan --- include/linux/page-flags.h | 81 ++++++++++++++++++++++++++++++++++---- mm/slab.h | 16 ++++++-- mm/util.c | 16 ++++++-- 3 files changed, 97 insertions(+), 16 deletions(-) diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index d14a17ffb55b..8f2c7fbc739b 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -198,6 +198,29 @@ enum pageflags { =20 #ifndef __GENERATING_BOUNDS_H =20 +/* + * For tail pages, if the size of struct page is power-of-2 ->compound_info + * encodes the mask that converts the address of the tail page address to + * the head page address. + * + * Otherwise, ->compound_info has direct pointer to head pages. + */ +static __always_inline bool compound_info_has_mask(void) +{ + /* + * Limit mask usage to HugeTLB vmemmap optimization (HVO) where it + * makes a difference. + * + * The approach with mask would work in the wider set of conditions, + * but it requires validating that struct pages are naturally aligned + * for all orders up to the MAX_FOLIO_ORDER, which can be tricky. + */ + if (!IS_ENABLED(CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP)) + return false; + + return is_power_of_2(sizeof(struct page)); +} + #ifdef CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP DECLARE_STATIC_KEY_FALSE(hugetlb_optimize_vmemmap_key); =20 @@ -210,6 +233,10 @@ static __always_inline const struct page *page_fixed_f= ake_head(const struct page if (!static_branch_unlikely(&hugetlb_optimize_vmemmap_key)) return page; =20 + /* Fake heads only exists if compound_info_has_mask() is true */ + if (!compound_info_has_mask()) + return page; + /* * Only addresses aligned with PAGE_SIZE of struct page may be fake head * struct page. The alignment check aims to avoid access the fields ( @@ -223,10 +250,14 @@ static __always_inline const struct page *page_fixed_= fake_head(const struct page * because the @page is a compound page composed with at least * two contiguous pages. */ - unsigned long head =3D READ_ONCE(page[1].compound_info); + unsigned long info =3D READ_ONCE(page[1].compound_info); =20 - if (likely(head & 1)) - return (const struct page *)(head - 1); + /* See set_compound_head() */ + if (likely(info & 1)) { + unsigned long p =3D (unsigned long)page; + + return (const struct page *)(p & info); + } } return page; } @@ -281,11 +312,26 @@ static __always_inline int page_is_fake_head(const st= ruct page *page) =20 static __always_inline unsigned long _compound_head(const struct page *pag= e) { - unsigned long head =3D READ_ONCE(page->compound_info); + unsigned long info =3D READ_ONCE(page->compound_info); =20 - if (unlikely(head & 1)) - return head - 1; - return (unsigned long)page_fixed_fake_head(page); + /* Bit 0 encodes PageTail() */ + if (!(info & 1)) + return (unsigned long)page_fixed_fake_head(page); + + /* + * If compound_info_has_mask() is false, the rest of compound_info is + * the pointer to the head page. + */ + if (!compound_info_has_mask()) + return info - 1; + + /* + * If compoun_info_has_mask() is true the rest of the info encodes + * the mask that converts the address of the tail page to the head page. + * + * No need to clear bit 0 in the mask as 'page' always has it clear. + */ + return (unsigned long)page & info; } =20 #define compound_head(page) ((typeof(page))_compound_head(page)) @@ -294,7 +340,26 @@ static __always_inline void set_compound_head(struct p= age *page, const struct page *head, unsigned int order) { - WRITE_ONCE(page->compound_info, (unsigned long)head + 1); + unsigned int shift; + unsigned long mask; + + if (!compound_info_has_mask()) { + WRITE_ONCE(page->compound_info, (unsigned long)head | 1); + return; + } + + /* + * If the size of struct page is power-of-2, bits [shift:0] of the + * virtual address of compound head are zero. + * + * Calculate mask that can be applied to the virtual address of + * the tail page to get address of the head page. + */ + shift =3D order + order_base_2(sizeof(struct page)); + mask =3D GENMASK(BITS_PER_LONG - 1, shift); + + /* Bit 0 encodes PageTail() */ + WRITE_ONCE(page->compound_info, mask | 1); } =20 static __always_inline void clear_compound_head(struct page *page) diff --git a/mm/slab.h b/mm/slab.h index 8a2a9c6c697b..f68c3ac8126f 100644 --- a/mm/slab.h +++ b/mm/slab.h @@ -137,11 +137,19 @@ static_assert(IS_ALIGNED(offsetof(struct slab, freeli= st), sizeof(struct freelist */ static inline struct slab *page_slab(const struct page *page) { - unsigned long head; + unsigned long info; + + info =3D READ_ONCE(page->compound_info); + if (info & 1) { + /* See compound_head() */ + if (compound_info_has_mask()) { + unsigned long p =3D (unsigned long)page; + page =3D (struct page *)(p & info); + } else { + page =3D (struct page *)(info - 1); + } + } =20 - head =3D READ_ONCE(page->compound_head); - if (head & 1) - page =3D (struct page *)(head - 1); if (data_race(page->page_type >> 24) !=3D PGTY_slab) page =3D NULL; =20 diff --git a/mm/util.c b/mm/util.c index 3ebcb9e6035c..20dccf2881d7 100644 --- a/mm/util.c +++ b/mm/util.c @@ -1237,7 +1237,7 @@ static void set_ps_flags(struct page_snapshot *ps, co= nst struct folio *folio, */ void snapshot_page(struct page_snapshot *ps, const struct page *page) { - unsigned long head, nr_pages =3D 1; + unsigned long info, nr_pages =3D 1; struct folio *foliop; int loops =3D 5; =20 @@ -1247,8 +1247,8 @@ void snapshot_page(struct page_snapshot *ps, const st= ruct page *page) again: memset(&ps->folio_snapshot, 0, sizeof(struct folio)); memcpy(&ps->page_snapshot, page, sizeof(*page)); - head =3D ps->page_snapshot.compound_info; - if ((head & 1) =3D=3D 0) { + info =3D ps->page_snapshot.compound_info; + if (!(info & 1)) { ps->idx =3D 0; foliop =3D (struct folio *)&ps->page_snapshot; if (!folio_test_large(foliop)) { @@ -1259,7 +1259,15 @@ void snapshot_page(struct page_snapshot *ps, const s= truct page *page) } foliop =3D (struct folio *)page; } else { - foliop =3D (struct folio *)(head - 1); + /* See compound_head() */ + if (compound_info_has_mask()) { + unsigned long p =3D (unsigned long)page; + + foliop =3D (struct folio *)(p & info); + } else { + foliop =3D (struct folio *)(info - 1); + } + ps->idx =3D folio_page_idx(foliop, page); } =20 --=20 2.51.2