From nobody Tue Dec 2 01:06:25 2025 Received: from mail-pj1-f54.google.com (mail-pj1-f54.google.com [209.85.216.54]) (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 502171AA1D2 for ; Sat, 22 Nov 2025 09:03:54 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.216.54 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1763802236; cv=none; b=LL4HgK1MshHVmQkyJahYaYZGTQzfmL/J5Dec6wUwug755/kfgMQdCeUx9pC0mTbOU/BsDHQaiABodw6iElYcCNwRpLUZwwehSZeGxObUdwNh3iE01RbRqkiYl4wyDSQy3lERJTvVa2lg8eAIyx9dbjhw8/SzAPhJrz0yiEOzJVA= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1763802236; c=relaxed/simple; bh=U/xhl13Y8Bu89uCI0yP4WX3zU48aYZVZiWrbkWM6UQQ=; h=From:To:Cc:Subject:Date:Message-Id:MIME-Version:Content-Type; b=clhNchP1pSFT+QpkCKg2zYlfYY1MPOYxiVytokiPPb4NSpo9+hM8mgXvX/m1qa793SZfhE9GhDet1GyH7zyCI2J76tXHJC+GiQs5L0hRX8w3xYVAk1/g1jtxJJFE4u6JTNc7ORYQDqWZrtijhIb+sHeiArqkVL7iLqY6Ur7OsWM= 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=jitEC+Ef; arc=none smtp.client-ip=209.85.216.54 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="jitEC+Ef" Received: by mail-pj1-f54.google.com with SMTP id 98e67ed59e1d1-3436a97f092so3490774a91.3 for ; Sat, 22 Nov 2025 01:03:54 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1763802233; x=1764407033; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:message-id:date:subject:cc :to:from:from:to:cc:subject:date:message-id:reply-to; bh=gWSIc38wakQFkjPaEOd32h4AOI7qtYUllCcYDLleeHs=; b=jitEC+EfPeFm35omVyqHZ1fVO2sJxs3Spm1D50za9KMa6y6b1/Vp/ouvUAJ0gcQ/yk ss4yMXIwJWxRRQ7EnPtnbkTp9ea6MYMA/19uZ7r0NlfRyfa4jS1DHGjsW7rKRjAeUcAy f/brj0I+odTdxchK8EdZf75j2tfcRfoIJXC85+zJy+vxqVWjQ9lZKRm4wOmYERQgwmaa 8VZ58/buNWg0kVWGurTty6CcRpoSD30LDmIY5ow5aEDM1vyTd82Dmj9A0Qy9+DlwreNI LB4Q110UTDLvssz32Y4zUVVPxiWfqb39MKN1KUD+v5pys4fOwTt3jvjA2awOEFpxOgAA OciA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1763802233; x=1764407033; h=content-transfer-encoding:mime-version:message-id:date:subject:cc :to:from:x-gm-gg:x-gm-message-state:from:to:cc:subject:date :message-id:reply-to; bh=gWSIc38wakQFkjPaEOd32h4AOI7qtYUllCcYDLleeHs=; b=bBWRSa6NMAKKyVZlB20aLAykwVhPkLwh6M11GQOtomRcPTh8z4oG51xXEYgQN79RrB fdKxWGWp8Hsdo++/i5Z0wev7L9zFccei6NT9g16kwNlDyZU6yUhtGP/VJfvvEhw9jE8c lZzSKLtYNosQD3TwPF4zMdz1XAUCTEFLailnobqQ4x5RS/Ip/hXKU5zWYJA/k+/pjFqc Vk7lL4tFe2AgNwPIobYg+Z5ObbzLg0OuwozvtMcgvmyNE1mhb9WrRdLFoIVewzh7uTsR 9xG6Z7FmeBJSvyl53wlv9opO34K/8Ju1XH70ayFYQlG4ua/LAmeLtlX12L86Dk/IqxHd MOhw== X-Forwarded-Encrypted: i=1; AJvYcCVeIpZ2vUGTKRbJU4ePt81TLSZYKUAuwAmc3kuFlKfvKk30V/3XXoTgcC/qdSmzrPk/d+aHC0ypz4w1bp4=@vger.kernel.org X-Gm-Message-State: AOJu0YwaN8hXvtwLiSGtI5aqvBLDPKcj5nxf5p+mX2eBqzXDd1bo4aFj mz61IxjiJVLpIiTfuC2M+fLKA8esm14BT1Fvg0uXAj2do5FVOEVATCwR X-Gm-Gg: ASbGncu8fH5BREIBeIKk6OOyp6YwR3rDUmiRM9hx50uV2LGB0ZP0tiZBqP0vSTmvTph H+M0laVySXD/NHvUXs80nD+KIkLSClJD12SjaITz/uspommRfM8KCjUGfHxAEY2Sr2WPxZlmfBa Pbqa26iYxFn9efxVneN039bqH2qUMvByc7+da+EJoFT7umAQ4DtPp+y+J5xFLHxt6+ZqZDZZiU8 t+zFVeor4jWg4n7/ebALmaLU/ru6FxfMK0sqYPzi86aG6qYEbtD1QwYdqOn0gssu+u9qNas7ZN1 +jjmvML4pbLjVNk8pILhbG/GZhDWcuzf3TTKcHKG+ryAVo3dQV0Lh5Qh3Nx9Gk26UD70/+LSJzN 1jrGOD0hnihWFqZHTErpShVoMB2ei4XMMmzF7zkcgIJHCuMXvXjUnXcFdZBoem7ff9kawmvkrIQ SuHmALFFygXf2i21Q9JA55pY31 X-Google-Smtp-Source: AGHT+IFEXYWimQk4DcIAFQyt9lEPKtWhsqJCLh0YYLXBxKfxO64RbGZR0X6+cGmFQNdZPcXXMZdWtg== X-Received: by 2002:a17:90b:540b:b0:33f:ebc2:634 with SMTP id 98e67ed59e1d1-34733e78ea2mr5873399a91.9.1763802232568; Sat, 22 Nov 2025 01:03:52 -0800 (PST) Received: from Barrys-MBP.hub ([47.72.129.29]) by smtp.gmail.com with ESMTPSA id 98e67ed59e1d1-3472692e5c8sm7842401a91.11.2025.11.22.01.03.48 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Sat, 22 Nov 2025 01:03:52 -0800 (PST) From: Barry Song <21cnbao@gmail.com> To: akpm@linux-foundation.org, linux-mm@kvack.org Cc: linux-media@vger.kernel.org, dri-devel@lists.freedesktop.org, linaro-mm-sig@lists.linaro.org, linux-kernel@vger.kernel.org, Barry Song , Uladzislau Rezki , Sumit Semwal , John Stultz , Maxime Ripard Subject: [PATCH RFC] mm/vmap: map contiguous pages in batches whenever possible Date: Sat, 22 Nov 2025 17:03:43 +0800 Message-Id: <20251122090343.81243-1-21cnbao@gmail.com> X-Mailer: git-send-email 2.39.3 (Apple Git-146) 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 From: Barry Song In many cases, the pages passed to vmap() may include high-order pages=E2=80=94for example, the systemheap often allocates pages in descending order: order 8, then 4, then 0. Currently, vmap() iterates over every page individually=E2=80=94even the pages inside a high-order block are handled one by one. This patch detects high-order pages and maps them as a single contiguous block whenever possible. Another possibility is to implement a new API, vmap_sg(). However, that change seems to be quite large in scope. When vmapping a 128MB dma-buf using the systemheap, this RFC appears to make system_heap_do_vmap() 16=C3=97 faster: W/ patch: [ 51.363682] system_heap_do_vmap took 2474000 ns [ 53.307044] system_heap_do_vmap took 2469008 ns [ 55.061985] system_heap_do_vmap took 2519008 ns [ 56.653810] system_heap_do_vmap took 2674000 ns W/o patch: [ 8.260880] system_heap_do_vmap took 39490000 ns [ 32.513292] system_heap_do_vmap took 38784000 ns [ 82.673374] system_heap_do_vmap took 40711008 ns [ 84.579062] system_heap_do_vmap took 40236000 ns Cc: Uladzislau Rezki Cc: Sumit Semwal Cc: John Stultz Cc: Maxime Ripard Signed-off-by: Barry Song --- mm/vmalloc.c | 49 +++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 43 insertions(+), 6 deletions(-) diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 0832f944544c..af2e3e8c052a 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -642,6 +642,34 @@ static int vmap_small_pages_range_noflush(unsigned lon= g addr, unsigned long end, return err; } =20 +static inline int get_vmap_batch_order(struct page **pages, + unsigned int stride, + int max_steps, + unsigned int idx) +{ + /* + * Currently, batching is only supported in vmap_pages_range + * when page_shift =3D=3D PAGE_SHIFT. + */ + if (stride !=3D 1) + return 0; + + struct page *base =3D pages[idx]; + if (!PageHead(base)) + return 0; + + int order =3D compound_order(base); + int nr_pages =3D 1 << order; + + if (max_steps < nr_pages) + return 0; + + for (int i =3D 0; i < nr_pages; i++) + if (pages[idx + i] !=3D base + i) + return 0; + return order; +} + /* * vmap_pages_range_noflush is similar to vmap_pages_range, but does not * flush caches. @@ -655,23 +683,32 @@ int __vmap_pages_range_noflush(unsigned long addr, un= signed long end, pgprot_t prot, struct page **pages, unsigned int page_shift) { unsigned int i, nr =3D (end - addr) >> PAGE_SHIFT; + unsigned int stride; =20 WARN_ON(page_shift < PAGE_SHIFT); =20 + /* + * Some users may allocate pages from high-order down to order 0. + * We roughly check if the first page is a compound page. If so, + * there is a chance to batch multiple pages together. + */ if (!IS_ENABLED(CONFIG_HAVE_ARCH_HUGE_VMALLOC) || - page_shift =3D=3D PAGE_SHIFT) + (page_shift =3D=3D PAGE_SHIFT && !PageCompound(pages[0]))) return vmap_small_pages_range_noflush(addr, end, prot, pages); =20 - for (i =3D 0; i < nr; i +=3D 1U << (page_shift - PAGE_SHIFT)) { - int err; + stride =3D 1U << (page_shift - PAGE_SHIFT); + for (i =3D 0; i < nr; ) { + int err, order; =20 - err =3D vmap_range_noflush(addr, addr + (1UL << page_shift), + order =3D get_vmap_batch_order(pages, stride, nr - i, i); + err =3D vmap_range_noflush(addr, addr + (1UL << (page_shift + order)), page_to_phys(pages[i]), prot, - page_shift); + page_shift + order); if (err) return err; =20 - addr +=3D 1UL << page_shift; + addr +=3D 1UL << (page_shift + order); + i +=3D 1U << (order + page_shift - PAGE_SHIFT); } =20 return 0; --=20 2.39.3 (Apple Git-146)