From patchwork Thu Jan 4 07:25:11 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vivek Kasireddy X-Patchwork-Id: 13510857 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 gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 4C5CFC4706F for ; Thu, 4 Jan 2024 07:49:01 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id E2B4110E3D0; Thu, 4 Jan 2024 07:48:53 +0000 (UTC) Received: from mgamail.intel.com (mgamail.intel.com [192.55.52.43]) by gabe.freedesktop.org (Postfix) with ESMTPS id 8F22710E36F for ; Thu, 4 Jan 2024 07:48:51 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1704354531; x=1735890531; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=U5Op4CjhDDAHfKtiDzpe6+9a1b7R8CllBxVTE/TesJE=; b=miYAvEANdccu4/PaXnby6nepDWfoqyPZE/H+20HXeQMf/GQ++Izk0hx+ 0pXbBte8bqmGgLXB3ZyUkZS2Byv3QP4fREzbK3qwl6u9LP8iI7BDMqT2x XTm8J4Y2SLM1eoDJkeISJV0JBUEBwTj1FgtzNxc9yxMrTCVxD+uAS0bHb wjZInvqftlIGgexrr3nttHCgTw+6zHIMcwZ4ByPgnBbMzzMwuuh8Iu8Gz yIx1LpRLeLxaBtpg8OGpFi52l4uREe2muaWuMV/ull3f7Bk0pQsWWvDq6 EnOnqOfVk9cJC962s34GJ4lCBorY9ZhJRMogwjeCoEMoiXshR02ap9hqr Q==; X-IronPort-AV: E=McAfee;i="6600,9927,10942"; a="483331096" X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="483331096" Received: from fmviesa002.fm.intel.com ([10.60.135.142]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:49 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="14776719" Received: from vkasired-desk2.fm.intel.com ([10.105.128.132]) by fmviesa002-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:47 -0800 From: Vivek Kasireddy To: dri-devel@lists.freedesktop.org, linux-mm@kvack.org Subject: [PATCH v10 1/8] udmabuf: Use vmf_insert_pfn and VM_PFNMAP for handling mmap Date: Wed, 3 Jan 2024 23:25:11 -0800 Message-Id: <20240104072518.807950-2-vivek.kasireddy@intel.com> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20240104072518.807950-1-vivek.kasireddy@intel.com> References: <20240104072518.807950-1-vivek.kasireddy@intel.com> MIME-Version: 1.0 X-BeenThere: dri-devel@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Direct Rendering Infrastructure - Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Dongwon Kim , David Hildenbrand , Daniel Vetter , Hugh Dickins , Vivek Kasireddy , Peter Xu , Gerd Hoffmann , Jason Gunthorpe , Junxiao Chang , Mike Kravetz Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" Add VM_PFNMAP to vm_flags in the mmap handler to ensure that the mappings would be managed without using struct page. And, in the vm_fault handler, use vmf_insert_pfn to share the page's pfn to userspace instead of directly sharing the page (via struct page *). Cc: David Hildenbrand Cc: Daniel Vetter Cc: Mike Kravetz Cc: Hugh Dickins Cc: Peter Xu Cc: Jason Gunthorpe Cc: Gerd Hoffmann Cc: Dongwon Kim Cc: Junxiao Chang Suggested-by: David Hildenbrand Acked-by: David Hildenbrand Signed-off-by: Vivek Kasireddy --- drivers/dma-buf/udmabuf.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/dma-buf/udmabuf.c b/drivers/dma-buf/udmabuf.c index c40645999648..820c993c8659 100644 --- a/drivers/dma-buf/udmabuf.c +++ b/drivers/dma-buf/udmabuf.c @@ -35,12 +35,13 @@ static vm_fault_t udmabuf_vm_fault(struct vm_fault *vmf) struct vm_area_struct *vma = vmf->vma; struct udmabuf *ubuf = vma->vm_private_data; pgoff_t pgoff = vmf->pgoff; + unsigned long pfn; if (pgoff >= ubuf->pagecount) return VM_FAULT_SIGBUS; - vmf->page = ubuf->pages[pgoff]; - get_page(vmf->page); - return 0; + + pfn = page_to_pfn(ubuf->pages[pgoff]); + return vmf_insert_pfn(vma, vmf->address, pfn); } static const struct vm_operations_struct udmabuf_vm_ops = { @@ -56,6 +57,7 @@ static int mmap_udmabuf(struct dma_buf *buf, struct vm_area_struct *vma) vma->vm_ops = &udmabuf_vm_ops; vma->vm_private_data = ubuf; + vm_flags_set(vma, VM_PFNMAP | VM_DONTEXPAND | VM_DONTDUMP); return 0; } From patchwork Thu Jan 4 07:25:12 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vivek Kasireddy X-Patchwork-Id: 13510855 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 gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id CBA14C47073 for ; Thu, 4 Jan 2024 07:48:52 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id DA6E510E3CB; Thu, 4 Jan 2024 07:48:51 +0000 (UTC) Received: from mgamail.intel.com (mgamail.intel.com [192.55.52.43]) by gabe.freedesktop.org (Postfix) with ESMTPS id 020C510E3A5 for ; Thu, 4 Jan 2024 07:48:50 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1704354531; x=1735890531; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=O1c3JIhGATN1dxcRTvUuRK9RhQSZEkJPgfU5c33zkAo=; b=iofLwCJ9zoCQtvHGReg5G2UaKufud/5YLGR48hT4t9E04+XMJd8MNT9v iQKqEVt5E8iQSyIrCDYoJvFOYXfMvOVAiu7v0k3s+QFqkEMcE+AjuqkJR dY+0vtgYPRRDdXrV2OS3TZTFhnTNBGuWibL8z/RspNWSXJiBdqlYNOBGg EqnTR9efmgu/bYKhksBm09JhO/rkHY/wwsL5C1FLpcCJtWFtU/V1kGTZO 1cvc9zhU/rmdbbUo8/q1ukd7wSUZXFGsUi5tZqyt4AOrHMYSDKr64VKMy LVTeQumPzRfx6aD3yrmnKT3kZPWqmj4qV/Q3s0cnfXhVsBLgtpZCCq13s w==; X-IronPort-AV: E=McAfee;i="6600,9927,10942"; a="483331113" X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="483331113" Received: from fmviesa002.fm.intel.com ([10.60.135.142]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:50 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="14776723" Received: from vkasired-desk2.fm.intel.com ([10.105.128.132]) by fmviesa002-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:47 -0800 From: Vivek Kasireddy To: dri-devel@lists.freedesktop.org, linux-mm@kvack.org Subject: [PATCH v10 2/8] udmabuf: Add back support for mapping hugetlb pages (v6) Date: Wed, 3 Jan 2024 23:25:12 -0800 Message-Id: <20240104072518.807950-3-vivek.kasireddy@intel.com> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20240104072518.807950-1-vivek.kasireddy@intel.com> References: <20240104072518.807950-1-vivek.kasireddy@intel.com> MIME-Version: 1.0 X-BeenThere: dri-devel@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Direct Rendering Infrastructure - Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Dongwon Kim , David Hildenbrand , Daniel Vetter , Hugh Dickins , Vivek Kasireddy , Peter Xu , Gerd Hoffmann , Jason Gunthorpe , Junxiao Chang , Mike Kravetz Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" A user or admin can configure a VMM (Qemu) Guest's memory to be backed by hugetlb pages for various reasons. However, a Guest OS would still allocate (and pin) buffers that are backed by regular 4k sized pages. In order to map these buffers and create dma-bufs for them on the Host, we first need to find the hugetlb pages where the buffer allocations are located and then determine the offsets of individual chunks (within those pages) and use this information to eventually populate a scatterlist. Testcase: default_hugepagesz=2M hugepagesz=2M hugepages=2500 options were passed to the Host kernel and Qemu was launched with these relevant options: qemu-system-x86_64 -m 4096m.... -device virtio-gpu-pci,max_outputs=1,blob=true,xres=1920,yres=1080 -display gtk,gl=on -object memory-backend-memfd,hugetlb=on,id=mem1,size=4096M -machine memory-backend=mem1 Replacing -display gtk,gl=on with -display gtk,gl=off above would exercise the mmap handler. v2: Updated get_sg_table() to manually populate the scatterlist for both huge page and non-huge-page cases. v3: s/offsets/subpgoff/g s/hpoff/mapidx/g v4: Replaced find_get_page_flags() with __filemap_get_folio() to ensure that we only obtain head pages from the mapping v5: Fix the calculation of mapidx to ensure that it is a order-n page multiple v6: - Split the processing of hugetlb or shmem pages into helpers to simplify the code in udmabuf_create() (Christoph) - Move the creation of offsets array out of hugetlb context and into common code Cc: David Hildenbrand Cc: Daniel Vetter Cc: Mike Kravetz Cc: Hugh Dickins Cc: Peter Xu Cc: Jason Gunthorpe Cc: Gerd Hoffmann Cc: Dongwon Kim Cc: Junxiao Chang Acked-by: Mike Kravetz (v2) Signed-off-by: Vivek Kasireddy --- drivers/dma-buf/udmabuf.c | 122 +++++++++++++++++++++++++++++++------- 1 file changed, 101 insertions(+), 21 deletions(-) diff --git a/drivers/dma-buf/udmabuf.c b/drivers/dma-buf/udmabuf.c index 820c993c8659..274defd3fa3e 100644 --- a/drivers/dma-buf/udmabuf.c +++ b/drivers/dma-buf/udmabuf.c @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -28,6 +29,7 @@ struct udmabuf { struct page **pages; struct sg_table *sg; struct miscdevice *device; + pgoff_t *offsets; }; static vm_fault_t udmabuf_vm_fault(struct vm_fault *vmf) @@ -41,6 +43,8 @@ static vm_fault_t udmabuf_vm_fault(struct vm_fault *vmf) return VM_FAULT_SIGBUS; pfn = page_to_pfn(ubuf->pages[pgoff]); + pfn += ubuf->offsets[pgoff] >> PAGE_SHIFT; + return vmf_insert_pfn(vma, vmf->address, pfn); } @@ -90,23 +94,29 @@ static struct sg_table *get_sg_table(struct device *dev, struct dma_buf *buf, { struct udmabuf *ubuf = buf->priv; struct sg_table *sg; + struct scatterlist *sgl; + unsigned int i = 0; int ret; sg = kzalloc(sizeof(*sg), GFP_KERNEL); if (!sg) return ERR_PTR(-ENOMEM); - ret = sg_alloc_table_from_pages(sg, ubuf->pages, ubuf->pagecount, - 0, ubuf->pagecount << PAGE_SHIFT, - GFP_KERNEL); + + ret = sg_alloc_table(sg, ubuf->pagecount, GFP_KERNEL); if (ret < 0) - goto err; + goto err_alloc; + + for_each_sg(sg->sgl, sgl, ubuf->pagecount, i) + sg_set_page(sgl, ubuf->pages[i], PAGE_SIZE, ubuf->offsets[i]); + ret = dma_map_sgtable(dev, sg, direction, 0); if (ret < 0) - goto err; + goto err_map; return sg; -err: +err_map: sg_free_table(sg); +err_alloc: kfree(sg); return ERR_PTR(ret); } @@ -143,6 +153,7 @@ static void release_udmabuf(struct dma_buf *buf) for (pg = 0; pg < ubuf->pagecount; pg++) put_page(ubuf->pages[pg]); + kfree(ubuf->offsets); kfree(ubuf->pages); kfree(ubuf); } @@ -196,17 +207,77 @@ static const struct dma_buf_ops udmabuf_ops = { #define SEALS_WANTED (F_SEAL_SHRINK) #define SEALS_DENIED (F_SEAL_WRITE) +static int handle_hugetlb_pages(struct udmabuf *ubuf, struct file *memfd, + pgoff_t offset, pgoff_t pgcnt, + pgoff_t *pgbuf) +{ + struct hstate *hpstate = hstate_file(memfd); + pgoff_t mapidx = offset >> huge_page_shift(hpstate); + pgoff_t subpgoff = (offset & ~huge_page_mask(hpstate)) >> PAGE_SHIFT; + pgoff_t maxsubpgs = huge_page_size(hpstate) >> PAGE_SHIFT; + struct page *hpage = NULL; + struct folio *folio; + pgoff_t pgidx; + + mapidx <<= huge_page_order(hpstate); + for (pgidx = 0; pgidx < pgcnt; pgidx++) { + if (!hpage) { + folio = __filemap_get_folio(memfd->f_mapping, + mapidx, + FGP_ACCESSED, 0); + if (IS_ERR(folio)) + return PTR_ERR(folio); + + hpage = &folio->page; + } + + get_page(hpage); + ubuf->pages[*pgbuf] = hpage; + ubuf->offsets[*pgbuf] = subpgoff << PAGE_SHIFT; + (*pgbuf)++; + if (++subpgoff == maxsubpgs) { + put_page(hpage); + hpage = NULL; + subpgoff = 0; + mapidx += pages_per_huge_page(hpstate); + } + } + + if (hpage) + put_page(hpage); + + return 0; +} + +static int handle_shmem_pages(struct udmabuf *ubuf, struct file *memfd, + pgoff_t offset, pgoff_t pgcnt, + pgoff_t *pgbuf) +{ + pgoff_t pgidx, pgoff = offset >> PAGE_SHIFT; + struct page *page; + + for (pgidx = 0; pgidx < pgcnt; pgidx++) { + page = shmem_read_mapping_page(memfd->f_mapping, + pgoff + pgidx); + if (IS_ERR(page)) + return PTR_ERR(page); + + ubuf->pages[*pgbuf] = page; + (*pgbuf)++; + } + + return 0; +} + static long udmabuf_create(struct miscdevice *device, struct udmabuf_create_list *head, struct udmabuf_create_item *list) { DEFINE_DMA_BUF_EXPORT_INFO(exp_info); struct file *memfd = NULL; - struct address_space *mapping = NULL; struct udmabuf *ubuf; struct dma_buf *buf; - pgoff_t pgoff, pgcnt, pgidx, pgbuf = 0, pglimit; - struct page *page; + pgoff_t pgcnt, pgbuf = 0, pglimit; int seals, ret = -EINVAL; u32 i, flags; @@ -234,6 +305,12 @@ static long udmabuf_create(struct miscdevice *device, ret = -ENOMEM; goto err; } + ubuf->offsets = kcalloc(ubuf->pagecount, sizeof(*ubuf->offsets), + GFP_KERNEL); + if (!ubuf->offsets) { + ret = -ENOMEM; + goto err; + } pgbuf = 0; for (i = 0; i < head->count; i++) { @@ -241,8 +318,7 @@ static long udmabuf_create(struct miscdevice *device, memfd = fget(list[i].memfd); if (!memfd) goto err; - mapping = memfd->f_mapping; - if (!shmem_mapping(mapping)) + if (!shmem_file(memfd) && !is_file_hugepages(memfd)) goto err; seals = memfd_fcntl(memfd, F_GET_SEALS, 0); if (seals == -EINVAL) @@ -251,16 +327,19 @@ static long udmabuf_create(struct miscdevice *device, if ((seals & SEALS_WANTED) != SEALS_WANTED || (seals & SEALS_DENIED) != 0) goto err; - pgoff = list[i].offset >> PAGE_SHIFT; - pgcnt = list[i].size >> PAGE_SHIFT; - for (pgidx = 0; pgidx < pgcnt; pgidx++) { - page = shmem_read_mapping_page(mapping, pgoff + pgidx); - if (IS_ERR(page)) { - ret = PTR_ERR(page); - goto err; - } - ubuf->pages[pgbuf++] = page; - } + + pgcnt = list[i].size >> PAGE_SHIFT; + if (is_file_hugepages(memfd)) + ret = handle_hugetlb_pages(ubuf, memfd, + list[i].offset, + pgcnt, &pgbuf); + else + ret = handle_shmem_pages(ubuf, memfd, + list[i].offset, + pgcnt, &pgbuf); + if (ret < 0) + goto err; + fput(memfd); memfd = NULL; } @@ -287,6 +366,7 @@ static long udmabuf_create(struct miscdevice *device, put_page(ubuf->pages[--pgbuf]); if (memfd) fput(memfd); + kfree(ubuf->offsets); kfree(ubuf->pages); kfree(ubuf); return ret; From patchwork Thu Jan 4 07:25:13 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vivek Kasireddy X-Patchwork-Id: 13510862 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 gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id A27ABC47255 for ; Thu, 4 Jan 2024 07:49:09 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id D959E10E3D5; Thu, 4 Jan 2024 07:48:56 +0000 (UTC) Received: from mgamail.intel.com (mgamail.intel.com [192.55.52.43]) by gabe.freedesktop.org (Postfix) with ESMTPS id CD32F10E3C4 for ; Thu, 4 Jan 2024 07:48:51 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1704354531; x=1735890531; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=ACMBAqooHk2c1KcIOvZNkowcRiJJyIDQFFF7e9jD5SI=; b=ml7BmZPABmtv4J3qfWGeFD/UQqbVpexCFJaoLIOK3Zt4pwV03Ngh28PP czy5cjaz19a8/muDGdr1HklVcUWorvZOGrXzgnHcO6imBgwT1Gi8/eGIC oHq3f4tOvOp8sO3qLvFqt7tz11TmLm62GzBIa7RpLkVoJOIcrfsPNJLY2 4Lji6zYJKyjgjzCCI/r1+W6+7AVFXw5hmo7F/5JZMfKzANig3kP+yk2hD DiAeTk5DN+P1SeZ3RoKzPVMqGqpQjYqwab1sP0t0GNpEbt4oN3CPbW3Cp B1t/CPJ85sQ3xnI5exWNKyQqmiKSgRT9bXMl73BwI/Ziw+UPibmyDO/LH g==; X-IronPort-AV: E=McAfee;i="6600,9927,10942"; a="483331152" X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="483331152" Received: from fmviesa002.fm.intel.com ([10.60.135.142]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:50 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="14776726" Received: from vkasired-desk2.fm.intel.com ([10.105.128.132]) by fmviesa002-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:47 -0800 From: Vivek Kasireddy To: dri-devel@lists.freedesktop.org, linux-mm@kvack.org Subject: [PATCH v10 3/8] mm/gup: Introduce unpin_folio/unpin_folios helpers Date: Wed, 3 Jan 2024 23:25:13 -0800 Message-Id: <20240104072518.807950-4-vivek.kasireddy@intel.com> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20240104072518.807950-1-vivek.kasireddy@intel.com> References: <20240104072518.807950-1-vivek.kasireddy@intel.com> MIME-Version: 1.0 X-BeenThere: dri-devel@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Direct Rendering Infrastructure - Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: David Hildenbrand , Vivek Kasireddy , Peter Xu , Christoph Hellwig , Matthew Wilcox , Jason Gunthorpe Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" These helpers are the folio versions of unpin_user_page/unpin_user_pages. They are currently only useful for unpinning folios pinned by memfd_pin_folios() or other associated routines. However, they could find new uses in the future, when more and more folio-only helpers are added to GUP. Cc: David Hildenbrand Cc: Matthew Wilcox Cc: Christoph Hellwig Cc: Jason Gunthorpe Cc: Peter Xu Suggested-by: David Hildenbrand Signed-off-by: Vivek Kasireddy --- include/linux/mm.h | 2 ++ mm/gup.c | 81 ++++++++++++++++++++++++++++++++++++++++------ 2 files changed, 74 insertions(+), 9 deletions(-) diff --git a/include/linux/mm.h b/include/linux/mm.h index 418d26608ece..4dbfdd277014 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1585,11 +1585,13 @@ static inline void put_page(struct page *page) #define GUP_PIN_COUNTING_BIAS (1U << 10) void unpin_user_page(struct page *page); +void unpin_folio(struct folio *folio); void unpin_user_pages_dirty_lock(struct page **pages, unsigned long npages, bool make_dirty); void unpin_user_page_range_dirty_lock(struct page *page, unsigned long npages, bool make_dirty); void unpin_user_pages(struct page **pages, unsigned long npages); +void unpin_folios(struct folio **folios, unsigned long nfolios); static inline bool is_cow_mapping(vm_flags_t flags) { diff --git a/mm/gup.c b/mm/gup.c index 231711efa390..4d7bc4453819 100644 --- a/mm/gup.c +++ b/mm/gup.c @@ -30,6 +30,23 @@ struct follow_page_context { unsigned int page_mask; }; +static inline void sanity_check_pinned_folios(struct folio **folios, + unsigned long nfolios) +{ + if (!IS_ENABLED(CONFIG_DEBUG_VM)) + return; + + for (; nfolios; nfolios--, folios++) { + struct folio *folio = *folios; + + if (is_zero_folio(folio) || + !folio_test_anon(folio)) + continue; + + VM_BUG_ON_FOLIO(!PageAnonExclusive(&folio->page), folio); + } +} + static inline void sanity_check_pinned_pages(struct page **pages, unsigned long npages) { @@ -52,15 +69,11 @@ static inline void sanity_check_pinned_pages(struct page **pages, struct page *page = *pages; struct folio *folio = page_folio(page); - if (is_zero_page(page) || - !folio_test_anon(folio)) - continue; - if (!folio_test_large(folio) || folio_test_hugetlb(folio)) - VM_BUG_ON_PAGE(!PageAnonExclusive(&folio->page), page); - else - /* Either a PTE-mapped or a PMD-mapped THP. */ - VM_BUG_ON_PAGE(!PageAnonExclusive(&folio->page) && - !PageAnonExclusive(page), page); + sanity_check_pinned_folios(&folio, 1); + + /* Either a PTE-mapped or a PMD-mapped THP. */ + if (folio_test_large(folio) && !folio_test_hugetlb(folio)) + VM_BUG_ON_PAGE(!PageAnonExclusive(page), page); } } @@ -276,6 +289,21 @@ void unpin_user_page(struct page *page) } EXPORT_SYMBOL(unpin_user_page); +/** + * unpin_folio() - release a dma-pinned folio + * @folio: pointer to folio to be released + * + * Folios that were pinned via memfd_pin_folios() or other similar routines + * must be released either using unpin_folio() or unpin_folios(). This is so + * that such folios can be separately tracked and uniquely handled. + */ +void unpin_folio(struct folio *folio) +{ + sanity_check_pinned_folios(&folio, 1); + gup_put_folio(folio, 1, FOLL_PIN); +} +EXPORT_SYMBOL(unpin_folio); + /** * folio_add_pin - Try to get an additional pin on a pinned folio * @folio: The folio to be pinned @@ -488,6 +516,41 @@ void unpin_user_pages(struct page **pages, unsigned long npages) } EXPORT_SYMBOL(unpin_user_pages); +/** + * unpin_folios() - release an array of gup-pinned folios. + * @folios: array of folios to be marked dirty and released. + * @nfolios: number of folios in the @folios array. + * + * For each folio in the @folios array, release the folio using unpin_folio(). + * + * Please see the unpin_folio() documentation for details. + */ +void unpin_folios(struct folio **folios, unsigned long nfolios) +{ + unsigned long i = 0, j; + + /* + * If this WARN_ON() fires, then the system *might* be leaking folios + * (by leaving them pinned), but probably not. More likely, gup/pup + * returned a hard -ERRNO error to the caller, who erroneously passed + * it here. + */ + if (WARN_ON(IS_ERR_VALUE(nfolios))) + return; + + sanity_check_pinned_folios(folios, nfolios); + while (i < nfolios) { + for (j = i + 1; j < nfolios; j++) + if (folios[i] != folios[j]) + break; + + if (folios[i]) + gup_put_folio(folios[i], j - i, FOLL_PIN); + i = j; + } +} +EXPORT_SYMBOL(unpin_folios); + /* * Set the MMF_HAS_PINNED if not set yet; after set it'll be there for the mm's * lifecycle. Avoid setting the bit unless necessary, or it might cause write From patchwork Thu Jan 4 07:25:14 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vivek Kasireddy X-Patchwork-Id: 13510863 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 gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id C96DCC4706F for ; Thu, 4 Jan 2024 07:49:10 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id E67E210E3D9; Thu, 4 Jan 2024 07:48:56 +0000 (UTC) Received: from mgamail.intel.com (mgamail.intel.com [192.55.52.43]) by gabe.freedesktop.org (Postfix) with ESMTPS id C7DBA10E3AF for ; Thu, 4 Jan 2024 07:48:51 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1704354531; x=1735890531; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=O22F+haVAFNBQJcPOK3PWw+AnkziG39TSsVTrN+ZPik=; b=VJRYVgF5tgPsZwa92m8fxxZhrBypP3nFkxDhNdNOoOhuI1m+GMN6R+2p IuRL0dagDzStjx/5+da3kVgEwtk7tKZ6aMBtdOmtEdUFfQzS9ltWI/D+h DOz+u6lkooX8bzIn5NHrd0r3P3wG55yJIEXoK3sIVbIy2uM3DxBOH+9MW yghqcj8BDUSl0pbl1FuNeYNJvoNKLWBxr0GjqHrYy6LFxmjkkcrRdtWp9 f17308jMezv2In52HPi0InTdzel15N6vFnKXlTc8yrb7zSgQnh0TQWbPD huq+CCn3ZNlqU2vDJ3liA+Yrc08arUFTSIfJsFu9MvBKEFUqqKca1UMtA A==; X-IronPort-AV: E=McAfee;i="6600,9927,10942"; a="483331156" X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="483331156" Received: from fmviesa002.fm.intel.com ([10.60.135.142]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:50 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="14776729" Received: from vkasired-desk2.fm.intel.com ([10.105.128.132]) by fmviesa002-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:47 -0800 From: Vivek Kasireddy To: dri-devel@lists.freedesktop.org, linux-mm@kvack.org Subject: [PATCH v10 4/8] mm/gup: Introduce check_and_migrate_movable_folios() Date: Wed, 3 Jan 2024 23:25:14 -0800 Message-Id: <20240104072518.807950-5-vivek.kasireddy@intel.com> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20240104072518.807950-1-vivek.kasireddy@intel.com> References: <20240104072518.807950-1-vivek.kasireddy@intel.com> MIME-Version: 1.0 X-BeenThere: dri-devel@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Direct Rendering Infrastructure - Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: David Hildenbrand , Vivek Kasireddy , Peter Xu , Christoph Hellwig , Matthew Wilcox , Jason Gunthorpe Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" This helper is the folio equivalent of check_and_migrate_movable_pages(). Therefore, all the rules that apply to check_and_migrate_movable_pages() also apply to this one as well. Currently, this helper is only used by memfd_pin_folios(). This patch also includes changes to rename and convert the internal functions collect_longterm_unpinnable_pages() and migrate_longterm_unpinnable_pages() to work on folios. Since they are also used by check_and_migrate_movable_pages(), a temporary array is used to collect and share the folios with these functions. Cc: David Hildenbrand Cc: Matthew Wilcox Cc: Christoph Hellwig Cc: Jason Gunthorpe Cc: Peter Xu Suggested-by: David Hildenbrand Signed-off-by: Vivek Kasireddy --- mm/gup.c | 129 +++++++++++++++++++++++++++++++++++++++---------------- 1 file changed, 92 insertions(+), 37 deletions(-) diff --git a/mm/gup.c b/mm/gup.c index 4d7bc4453819..00b24a429ba8 100644 --- a/mm/gup.c +++ b/mm/gup.c @@ -2097,20 +2097,24 @@ struct page *get_dump_page(unsigned long addr) #ifdef CONFIG_MIGRATION /* - * Returns the number of collected pages. Return value is always >= 0. + * Returns the number of collected folios. Return value is always >= 0. */ -static unsigned long collect_longterm_unpinnable_pages( - struct list_head *movable_page_list, - unsigned long nr_pages, +static unsigned long collect_longterm_unpinnable_folios( + struct list_head *movable_folio_list, + unsigned long nr_folios, + struct folio **folios, struct page **pages) { unsigned long i, collected = 0; struct folio *prev_folio = NULL; bool drain_allow = true; + struct folio *folio; - for (i = 0; i < nr_pages; i++) { - struct folio *folio = page_folio(pages[i]); + for (i = 0; i < nr_folios; i++) { + if (pages) + folios[i] = page_folio(pages[i]); + folio = folios[i]; if (folio == prev_folio) continue; prev_folio = folio; @@ -2124,7 +2128,7 @@ static unsigned long collect_longterm_unpinnable_pages( continue; if (folio_test_hugetlb(folio)) { - isolate_hugetlb(folio, movable_page_list); + isolate_hugetlb(folio, movable_folio_list); continue; } @@ -2136,7 +2140,7 @@ static unsigned long collect_longterm_unpinnable_pages( if (!folio_isolate_lru(folio)) continue; - list_add_tail(&folio->lru, movable_page_list); + list_add_tail(&folio->lru, movable_folio_list); node_stat_mod_folio(folio, NR_ISOLATED_ANON + folio_is_file_lru(folio), folio_nr_pages(folio)); @@ -2146,27 +2150,28 @@ static unsigned long collect_longterm_unpinnable_pages( } /* - * Unpins all pages and migrates device coherent pages and movable_page_list. - * Returns -EAGAIN if all pages were successfully migrated or -errno for failure - * (or partial success). + * Unpins all folios and migrates device coherent folios and movable_folio_list. + * Returns -EAGAIN if all folios were successfully migrated or -errno for + * failure (or partial success). */ -static int migrate_longterm_unpinnable_pages( - struct list_head *movable_page_list, - unsigned long nr_pages, - struct page **pages) +static int migrate_longterm_unpinnable_folios( + struct list_head *movable_folio_list, + unsigned long nr_folios, + struct folio **folios) { int ret; unsigned long i; - for (i = 0; i < nr_pages; i++) { - struct folio *folio = page_folio(pages[i]); + for (i = 0; i < nr_folios; i++) { + struct folio *folio = folios[i]; if (folio_is_device_coherent(folio)) { /* - * Migration will fail if the page is pinned, so convert - * the pin on the source page to a normal reference. + * Migration will fail if the folio is pinned, so + * convert the pin on the source folio to a normal + * reference. */ - pages[i] = NULL; + folios[i] = NULL; folio_get(folio); gup_put_folio(folio, 1, FOLL_PIN); @@ -2179,23 +2184,23 @@ static int migrate_longterm_unpinnable_pages( } /* - * We can't migrate pages with unexpected references, so drop + * We can't migrate folios with unexpected references, so drop * the reference obtained by __get_user_pages_locked(). - * Migrating pages have been added to movable_page_list after + * Migrating folios have been added to movable_folio_list after * calling folio_isolate_lru() which takes a reference so the - * page won't be freed if it's migrating. + * folio won't be freed if it's migrating. */ - unpin_user_page(pages[i]); - pages[i] = NULL; + unpin_folio(folios[i]); + folios[i] = NULL; } - if (!list_empty(movable_page_list)) { + if (!list_empty(movable_folio_list)) { struct migration_target_control mtc = { .nid = NUMA_NO_NODE, .gfp_mask = GFP_USER | __GFP_NOWARN, }; - if (migrate_pages(movable_page_list, alloc_migration_target, + if (migrate_pages(movable_folio_list, alloc_migration_target, NULL, (unsigned long)&mtc, MIGRATE_SYNC, MR_LONGTERM_PIN, NULL)) { ret = -ENOMEM; @@ -2203,15 +2208,15 @@ static int migrate_longterm_unpinnable_pages( } } - putback_movable_pages(movable_page_list); + putback_movable_pages(movable_folio_list); return -EAGAIN; err: - for (i = 0; i < nr_pages; i++) - if (pages[i]) - unpin_user_page(pages[i]); - putback_movable_pages(movable_page_list); + for (i = 0; i < nr_folios; i++) + if (folios[i]) + unpin_folio(folios[i]); + putback_movable_pages(movable_folio_list); return ret; } @@ -2235,16 +2240,60 @@ static int migrate_longterm_unpinnable_pages( static long check_and_migrate_movable_pages(unsigned long nr_pages, struct page **pages) { + unsigned long nr_folios = nr_pages; unsigned long collected; - LIST_HEAD(movable_page_list); + LIST_HEAD(movable_folio_list); + struct folio **folios; + long ret; - collected = collect_longterm_unpinnable_pages(&movable_page_list, - nr_pages, pages); + folios = kmalloc_array(nr_folios, sizeof(*folios), GFP_KERNEL); + if (!folios) + return -ENOMEM; + + collected = collect_longterm_unpinnable_folios(&movable_folio_list, + nr_folios, folios, + pages); + if (!collected) { + kfree(folios); + return 0; + } + + ret = migrate_longterm_unpinnable_folios(&movable_folio_list, + nr_folios, folios); + kfree(folios); + return ret; +} + +/* + * Check whether all folios are *allowed* to be pinned. Rather confusingly, all + * folios in the range are required to be pinned via FOLL_PIN, before calling + * this routine. + * + * If any folios in the range are not allowed to be pinned, then this routine + * will migrate those folios away, unpin all the folios in the range and return + * -EAGAIN. The caller should re-pin the entire range with FOLL_PIN and then + * call this routine again. + * + * If an error other than -EAGAIN occurs, this indicates a migration failure. + * The caller should give up, and propagate the error back up the call stack. + * + * If everything is OK and all folios in the range are allowed to be pinned, + * then this routine leaves all folios pinned and returns zero for success. + */ +static long check_and_migrate_movable_folios(unsigned long nr_folios, + struct folio **folios) +{ + unsigned long collected; + LIST_HEAD(movable_folio_list); + + collected = collect_longterm_unpinnable_folios(&movable_folio_list, + nr_folios, folios, + NULL); if (!collected) return 0; - return migrate_longterm_unpinnable_pages(&movable_page_list, nr_pages, - pages); + return migrate_longterm_unpinnable_folios(&movable_folio_list, + nr_folios, folios); } #else static long check_and_migrate_movable_pages(unsigned long nr_pages, @@ -2252,6 +2301,12 @@ static long check_and_migrate_movable_pages(unsigned long nr_pages, { return 0; } + +static long check_and_migrate_movable_folios(unsigned long nr_folios, + struct folio **folios) +{ + return 0; +} #endif /* CONFIG_MIGRATION */ /* From patchwork Thu Jan 4 07:25:15 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vivek Kasireddy X-Patchwork-Id: 13510859 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 gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 28300C47074 for ; Thu, 4 Jan 2024 07:49:05 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id CA13810E3A6; Thu, 4 Jan 2024 07:48:54 +0000 (UTC) Received: from mgamail.intel.com (mgamail.intel.com [192.55.52.43]) by gabe.freedesktop.org (Postfix) with ESMTPS id 354E910E36F for ; Thu, 4 Jan 2024 07:48:51 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1704354531; x=1735890531; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=e3c0tjXA+h9k7oaKNB8qaNJNT1gODpsdEPYdMfV6qpA=; b=JZdX8TgB738M/VngJ3wZvs/2aTkGhepNqvRa0ibNUsjKq3HBXJQhunO+ e2F/J7CZoYD4PAq+YYj9OxzkzhvuUyaJR36DVmfLsrHToPQzBV+lHfpYh oIYVz0qpOC0mR14eOOY7EvVjrBqkHi0uap3zYl3CZG+UuyxfvXhbMbtE4 lTeRyjHdFEI8HaXsmSuq6dsz/iBJdnSsNWhEeyTNPJ7sjvC4Ec+FW2Es0 Pa4oRVmYYXyErYSnsNbmXYuE8i7Sne+UgZ3sx4+seIl+RzUn+3siSRTmK VeXh5Ltd84jAlgikaDTM0zKMVXuPM5jJIoidjhIoloxCCE1W0jY68/GRY w==; X-IronPort-AV: E=McAfee;i="6600,9927,10942"; a="483331122" X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="483331122" Received: from fmviesa002.fm.intel.com ([10.60.135.142]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:50 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="14776733" Received: from vkasired-desk2.fm.intel.com ([10.105.128.132]) by fmviesa002-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:47 -0800 From: Vivek Kasireddy To: dri-devel@lists.freedesktop.org, linux-mm@kvack.org Subject: [PATCH v10 5/8] mm/gup: Introduce memfd_pin_folios() for pinning memfd folios (v10) Date: Wed, 3 Jan 2024 23:25:15 -0800 Message-Id: <20240104072518.807950-6-vivek.kasireddy@intel.com> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20240104072518.807950-1-vivek.kasireddy@intel.com> References: <20240104072518.807950-1-vivek.kasireddy@intel.com> MIME-Version: 1.0 X-BeenThere: dri-devel@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Direct Rendering Infrastructure - Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Gerd Hoffmann , Dongwon Kim , David Hildenbrand , Daniel Vetter , Hugh Dickins , Vivek Kasireddy , Matthew Wilcox , Christoph Hellwig , Peter Xu , Jason Gunthorpe , Junxiao Chang , Christoph Hellwig , Mike Kravetz Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" For drivers that would like to longterm-pin the folios associated with a memfd, the memfd_pin_folios() API provides an option to not only pin the folios via FOLL_PIN but also to check and migrate them if they reside in movable zone or CMA block. This API currently works with memfds but it should work with any files that belong to either shmemfs or hugetlbfs. Files belonging to other filesystems are rejected for now. The folios need to be located first before pinning them via FOLL_PIN. If they are found in the page cache, they can be immediately pinned. Otherwise, they need to be allocated using the filesystem specific APIs and then pinned. v2: - Drop gup_flags and improve comments and commit message (David) - Allocate a page if we cannot find in page cache for the hugetlbfs case as well (David) - Don't unpin pages if there is a migration related failure (David) - Drop the unnecessary nr_pages <= 0 check (Jason) - Have the caller of the API pass in file * instead of fd (Jason) v3: (David) - Enclose the huge page allocation code with #ifdef CONFIG_HUGETLB_PAGE (Build error reported by kernel test robot ) - Don't forget memalloc_pin_restore() on non-migration related errors - Improve the readability of the cleanup code associated with non-migration related errors - Augment the comments by describing FOLL_LONGTERM like behavior - Include the R-b tag from Jason v4: - Remove the local variable "page" and instead use 3 return statements in alloc_file_page() (David) - Add the R-b tag from David v5: (David) - For hugetlb case, ensure that we only obtain head pages from the mapping by using __filemap_get_folio() instead of find_get_page_flags() - Handle -EEXIST when two or more potential users try to simultaneously add a huge page to the mapping by forcing them to retry on failure v6: (Christoph) - Rename this API to memfd_pin_user_pages() to make it clear that it is intended for memfds - Move the memfd page allocation helper from gup.c to memfd.c - Fix indentation errors in memfd_pin_user_pages() - For contiguous ranges of folios, use a helper such as filemap_get_folios_contig() to lookup the page cache in batches v7: - Rename this API to memfd_pin_folios() and make it return folios and offsets instead of pages (David) - Don't continue processing the folios in the batch returned by filemap_get_folios_contig() if they do not have correct next_idx - Add the R-b tag from Christoph v8: (David) - Have caller pass [start, end], max_folios instead of start, nr_pages - Replace offsets array with just offset into the first page - Add comments explaning the need for next_idx - Pin (and return) the folio (via FOLL_PIN) only once v9: (Matthew) - Drop the extern while declaring memfd_alloc_folio() - Fix memfd_alloc_folio() declaration to have it return struct folio * instead of struct page * when CONFIG_MEMFD_CREATE is not defined v10: - Use unpin_folios() to unpin folios on error - Use check_and_migrate_movable_folios() instead of check_and_migrate_movable_pages() for checking and migrating folios Cc: David Hildenbrand Cc: Matthew Wilcox (Oracle) Cc: Christoph Hellwig Cc: Daniel Vetter Cc: Mike Kravetz Cc: Hugh Dickins Cc: Peter Xu Cc: Gerd Hoffmann Cc: Dongwon Kim Cc: Junxiao Chang Suggested-by: Jason Gunthorpe Reviewed-by: Jason Gunthorpe (v2) Reviewed-by: David Hildenbrand (v3) Reviewed-by: Christoph Hellwig (v6) Signed-off-by: Vivek Kasireddy --- include/linux/memfd.h | 5 ++ include/linux/mm.h | 3 + mm/gup.c | 136 ++++++++++++++++++++++++++++++++++++++++++ mm/memfd.c | 34 +++++++++++ 4 files changed, 178 insertions(+) diff --git a/include/linux/memfd.h b/include/linux/memfd.h index e7abf6fa4c52..3f2cf339ceaf 100644 --- a/include/linux/memfd.h +++ b/include/linux/memfd.h @@ -6,11 +6,16 @@ #ifdef CONFIG_MEMFD_CREATE extern long memfd_fcntl(struct file *file, unsigned int cmd, unsigned int arg); +struct folio *memfd_alloc_folio(struct file *memfd, pgoff_t idx); #else static inline long memfd_fcntl(struct file *f, unsigned int c, unsigned int a) { return -EINVAL; } +static inline struct folio *memfd_alloc_folio(struct file *memfd, pgoff_t idx) +{ + return ERR_PTR(-EINVAL); +} #endif #endif /* __LINUX_MEMFD_H */ diff --git a/include/linux/mm.h b/include/linux/mm.h index 4dbfdd277014..cbc0c6360690 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -2474,6 +2474,9 @@ long get_user_pages_unlocked(unsigned long start, unsigned long nr_pages, struct page **pages, unsigned int gup_flags); long pin_user_pages_unlocked(unsigned long start, unsigned long nr_pages, struct page **pages, unsigned int gup_flags); +long memfd_pin_folios(struct file *memfd, loff_t start, loff_t end, + struct folio **folios, unsigned int max_folios, + pgoff_t *offset); int get_user_pages_fast(unsigned long start, int nr_pages, unsigned int gup_flags, struct page **pages); diff --git a/mm/gup.c b/mm/gup.c index 00b24a429ba8..5b2b55d87998 100644 --- a/mm/gup.c +++ b/mm/gup.c @@ -5,6 +5,7 @@ #include #include +#include #include #include #include @@ -17,6 +18,7 @@ #include #include #include +#include #include #include @@ -3528,3 +3530,137 @@ long pin_user_pages_unlocked(unsigned long start, unsigned long nr_pages, &locked, gup_flags); } EXPORT_SYMBOL(pin_user_pages_unlocked); + +/** + * memfd_pin_folios() - pin folios associated with a memfd + * @memfd: the memfd whose folios are to be pinned + * @start: the first memfd offset + * @end: the last memfd offset (inclusive) + * @folios: array that receives pointers to the folios pinned + * @max_folios: maximum number of entries in @folios + * @offset: the offset into the first folio + * + * Attempt to pin folios associated with a memfd in the contiguous range + * [start, end]. Given that a memfd is either backed by shmem or hugetlb, + * the folios can either be found in the page cache or need to be allocated + * if necessary. Once the folios are located, they are all pinned via + * FOLL_PIN and @offset is populatedwith the offset into the first folio. + * And, eventually, these pinned folios must be released either using + * unpin_folios() or unpin_folio(). + * + * It must be noted that the folios may be pinned for an indefinite amount + * of time. And, in most cases, the duration of time they may stay pinned + * would be controlled by the userspace. This behavior is effectively the + * same as using FOLL_LONGTERM with other GUP APIs. + * + * Returns number of folios pinned, which could be less than @max_folios + * as it depends on the folio sizes that cover the range [start, end]. + * If no folios were pinned, it returns -errno. + */ +long memfd_pin_folios(struct file *memfd, loff_t start, loff_t end, + struct folio **folios, unsigned int max_folios, + pgoff_t *offset) +{ + unsigned int flags, nr_folios, nr_found; + unsigned int i, pgshift = PAGE_SHIFT; + pgoff_t start_idx, end_idx, next_idx; + struct folio *folio = NULL; + struct folio_batch fbatch; + struct hstate *h; + long ret; + + if (start > end || !max_folios) + return -EINVAL; + + if (!memfd) + return -EINVAL; + + if (!shmem_file(memfd) && !is_file_hugepages(memfd)) + return -EINVAL; + + if (is_file_hugepages(memfd)) { + h = hstate_file(memfd); + pgshift = huge_page_shift(h); + } + + flags = memalloc_pin_save(); + do { + nr_folios = 0; + start_idx = start >> pgshift; + end_idx = end >> pgshift; + if (is_file_hugepages(memfd)) { + start_idx <<= huge_page_order(h); + end_idx <<= huge_page_order(h); + } + + folio_batch_init(&fbatch); + while (start_idx <= end_idx && nr_folios < max_folios) { + /* + * In most cases, we should be able to find the folios + * in the page cache. If we cannot find them for some + * reason, we try to allocate them and add them to the + * page cache. + */ + nr_found = filemap_get_folios_contig(memfd->f_mapping, + &start_idx, + end_idx, + &fbatch); + if (folio) { + folio_put(folio); + folio = NULL; + } + + next_idx = 0; + for (i = 0; i < nr_found; i++) { + /* + * As there can be multiple entries for a + * given folio in the batch returned by + * filemap_get_folios_contig(), the below + * check is to ensure that we pin and return a + * unique set of folios between start and end. + */ + if (next_idx && + next_idx != folio_index(fbatch.folios[i])) + continue; + + folio = try_grab_folio(&fbatch.folios[i]->page, + 1, FOLL_PIN); + if (!folio) { + folio_batch_release(&fbatch); + goto err; + } + + if (nr_folios == 0) + *offset = offset_in_folio(folio, start); + + folios[nr_folios] = folio; + next_idx = folio_next_index(folio); + if (++nr_folios == max_folios) + break; + } + + folio = NULL; + folio_batch_release(&fbatch); + if (!nr_found) { + folio = memfd_alloc_folio(memfd, start_idx); + if (IS_ERR(folio)) { + ret = PTR_ERR(folio); + if (ret != -EEXIST) + goto err; + } + } + } + + ret = check_and_migrate_movable_folios(nr_folios, folios); + } while (ret == -EAGAIN); + + memalloc_pin_restore(flags); + return ret ? ret : nr_folios; +err: + memalloc_pin_restore(flags); + unpin_folios(folios, nr_folios); + + return ret; +} +EXPORT_SYMBOL_GPL(memfd_pin_folios); + diff --git a/mm/memfd.c b/mm/memfd.c index d3a1ba4208c9..36a75e8249f8 100644 --- a/mm/memfd.c +++ b/mm/memfd.c @@ -63,6 +63,40 @@ static void memfd_tag_pins(struct xa_state *xas) xas_unlock_irq(xas); } +/* + * This is a helper function used by memfd_pin_user_pages() in GUP (gup.c). + * It is mainly called to allocate a page in a memfd when the caller + * (memfd_pin_user_pages()) cannot find a page in the page cache at a given + * index in the mapping. + */ +struct folio *memfd_alloc_folio(struct file *memfd, pgoff_t idx) +{ +#ifdef CONFIG_HUGETLB_PAGE + struct folio *folio; + int err; + + if (is_file_hugepages(memfd)) { + folio = alloc_hugetlb_folio_nodemask(hstate_file(memfd), + NUMA_NO_NODE, + NULL, + GFP_USER); + if (folio && folio_try_get(folio)) { + err = hugetlb_add_to_page_cache(folio, + memfd->f_mapping, + idx); + if (err) { + folio_put(folio); + free_huge_folio(folio); + return ERR_PTR(err); + } + return folio; + } + return ERR_PTR(-ENOMEM); + } +#endif + return shmem_read_folio(memfd->f_mapping, idx); +} + /* * Setting SEAL_WRITE requires us to verify there's no pending writer. However, * via get_user_pages(), drivers might have some pending I/O without any active From patchwork Thu Jan 4 07:25:16 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vivek Kasireddy X-Patchwork-Id: 13510860 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 gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id D0758C4707C for ; Thu, 4 Jan 2024 07:49:06 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 27B4810E3DF; Thu, 4 Jan 2024 07:48:56 +0000 (UTC) Received: from mgamail.intel.com (mgamail.intel.com [192.55.52.43]) by gabe.freedesktop.org (Postfix) with ESMTPS id AB1E610E3A5 for ; Thu, 4 Jan 2024 07:48:51 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1704354531; x=1735890531; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=srh+pDeiF4x93Sp9cmg48dB8ma2fshhkeNl1Nqm5xC4=; b=ZWlXgZcWUv6SOFOPpMXh7VrVsQqZ9+84GAlZIwT4YC/0iTasyefYKL2l d/Yl8uKzKCHIhjJBPoDcORwyb14LViMfrLmRQVk/lWMdcUDPpJGk3cD/0 jlSdXueEBxUlQUhTcB3gKpv5DNXSNy2jGgnQguQNN3E9cOjeQezmh+2DO 5HFMZ2wnIoCKmMlMQpTabgajngTpEYkweVAbR8J5Pag6NCw5ta0q+QJN6 t1ZRpsd585hOcoYZnbfJaQ9su+sKKl93hfAYznrt162/km59fAm3obJuM WtxQmT5Pm1g2RxE821X1qwm3Bbw17JQ+1lPo3r0zv4idC9IWk++oS6deU w==; X-IronPort-AV: E=McAfee;i="6600,9927,10942"; a="483331134" X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="483331134" Received: from fmviesa002.fm.intel.com ([10.60.135.142]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:50 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="14776736" Received: from vkasired-desk2.fm.intel.com ([10.105.128.132]) by fmviesa002-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:47 -0800 From: Vivek Kasireddy To: dri-devel@lists.freedesktop.org, linux-mm@kvack.org Subject: [PATCH v10 6/8] udmabuf: Convert udmabuf driver to use folios (v2) Date: Wed, 3 Jan 2024 23:25:16 -0800 Message-Id: <20240104072518.807950-7-vivek.kasireddy@intel.com> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20240104072518.807950-1-vivek.kasireddy@intel.com> References: <20240104072518.807950-1-vivek.kasireddy@intel.com> MIME-Version: 1.0 X-BeenThere: dri-devel@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Direct Rendering Infrastructure - Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Gerd Hoffmann , Dongwon Kim , David Hildenbrand , Daniel Vetter , Hugh Dickins , Vivek Kasireddy , Matthew Wilcox , Peter Xu , Jason Gunthorpe , Junxiao Chang , Mike Kravetz Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" This is mainly a preparatory patch to use memfd_pin_folios() API for pinning folios. Using folios instead of pages makes sense as the udmabuf driver needs to handle both shmem and hugetlb cases. However, the function vmap_udmabuf() still needs a list of pages; so, we collect all the head pages into a local array in this case. Other changes in this patch include the addition of helpers for checking the memfd seals and exporting dmabuf. Moving code from udmabuf_create() into these helpers improves readability given that udmabuf_create() is a bit long. v2: (Matthew) - Use folio_pfn() on the folio instead of page_to_pfn() on head page - Don't split the arguments to shmem_read_folio() on multiple lines Cc: David Hildenbrand Cc: Matthew Wilcox Cc: Daniel Vetter Cc: Mike Kravetz Cc: Hugh Dickins Cc: Peter Xu Cc: Jason Gunthorpe Cc: Gerd Hoffmann Cc: Dongwon Kim Cc: Junxiao Chang Signed-off-by: Vivek Kasireddy --- drivers/dma-buf/udmabuf.c | 140 ++++++++++++++++++++++---------------- 1 file changed, 83 insertions(+), 57 deletions(-) diff --git a/drivers/dma-buf/udmabuf.c b/drivers/dma-buf/udmabuf.c index 274defd3fa3e..a8f3af61f7f2 100644 --- a/drivers/dma-buf/udmabuf.c +++ b/drivers/dma-buf/udmabuf.c @@ -26,7 +26,7 @@ MODULE_PARM_DESC(size_limit_mb, "Max size of a dmabuf, in megabytes. Default is struct udmabuf { pgoff_t pagecount; - struct page **pages; + struct folio **folios; struct sg_table *sg; struct miscdevice *device; pgoff_t *offsets; @@ -42,7 +42,7 @@ static vm_fault_t udmabuf_vm_fault(struct vm_fault *vmf) if (pgoff >= ubuf->pagecount) return VM_FAULT_SIGBUS; - pfn = page_to_pfn(ubuf->pages[pgoff]); + pfn = folio_pfn(ubuf->folios[pgoff]); pfn += ubuf->offsets[pgoff] >> PAGE_SHIFT; return vmf_insert_pfn(vma, vmf->address, pfn); @@ -68,11 +68,21 @@ static int mmap_udmabuf(struct dma_buf *buf, struct vm_area_struct *vma) static int vmap_udmabuf(struct dma_buf *buf, struct iosys_map *map) { struct udmabuf *ubuf = buf->priv; + struct page **pages; void *vaddr; + pgoff_t pg; dma_resv_assert_held(buf->resv); - vaddr = vm_map_ram(ubuf->pages, ubuf->pagecount, -1); + pages = kmalloc_array(ubuf->pagecount, sizeof(*pages), GFP_KERNEL); + if (!pages) + return -ENOMEM; + + for (pg = 0; pg < ubuf->pagecount; pg++) + pages[pg] = &ubuf->folios[pg]->page; + + vaddr = vm_map_ram(pages, ubuf->pagecount, -1); + kfree(pages); if (!vaddr) return -EINVAL; @@ -107,7 +117,8 @@ static struct sg_table *get_sg_table(struct device *dev, struct dma_buf *buf, goto err_alloc; for_each_sg(sg->sgl, sgl, ubuf->pagecount, i) - sg_set_page(sgl, ubuf->pages[i], PAGE_SIZE, ubuf->offsets[i]); + sg_set_folio(sgl, ubuf->folios[i], PAGE_SIZE, + ubuf->offsets[i]); ret = dma_map_sgtable(dev, sg, direction, 0); if (ret < 0) @@ -152,9 +163,9 @@ static void release_udmabuf(struct dma_buf *buf) put_sg_table(dev, ubuf->sg, DMA_BIDIRECTIONAL); for (pg = 0; pg < ubuf->pagecount; pg++) - put_page(ubuf->pages[pg]); + folio_put(ubuf->folios[pg]); kfree(ubuf->offsets); - kfree(ubuf->pages); + kfree(ubuf->folios); kfree(ubuf); } @@ -215,36 +226,33 @@ static int handle_hugetlb_pages(struct udmabuf *ubuf, struct file *memfd, pgoff_t mapidx = offset >> huge_page_shift(hpstate); pgoff_t subpgoff = (offset & ~huge_page_mask(hpstate)) >> PAGE_SHIFT; pgoff_t maxsubpgs = huge_page_size(hpstate) >> PAGE_SHIFT; - struct page *hpage = NULL; - struct folio *folio; + struct folio *folio = NULL; pgoff_t pgidx; mapidx <<= huge_page_order(hpstate); for (pgidx = 0; pgidx < pgcnt; pgidx++) { - if (!hpage) { + if (!folio) { folio = __filemap_get_folio(memfd->f_mapping, mapidx, FGP_ACCESSED, 0); if (IS_ERR(folio)) return PTR_ERR(folio); - - hpage = &folio->page; } - get_page(hpage); - ubuf->pages[*pgbuf] = hpage; + folio_get(folio); + ubuf->folios[*pgbuf] = folio; ubuf->offsets[*pgbuf] = subpgoff << PAGE_SHIFT; (*pgbuf)++; if (++subpgoff == maxsubpgs) { - put_page(hpage); - hpage = NULL; + folio_put(folio); + folio = NULL; subpgoff = 0; mapidx += pages_per_huge_page(hpstate); } } - if (hpage) - put_page(hpage); + if (folio) + folio_put(folio); return 0; } @@ -254,31 +262,69 @@ static int handle_shmem_pages(struct udmabuf *ubuf, struct file *memfd, pgoff_t *pgbuf) { pgoff_t pgidx, pgoff = offset >> PAGE_SHIFT; - struct page *page; + struct folio *folio = NULL; for (pgidx = 0; pgidx < pgcnt; pgidx++) { - page = shmem_read_mapping_page(memfd->f_mapping, - pgoff + pgidx); - if (IS_ERR(page)) - return PTR_ERR(page); + folio = shmem_read_folio(memfd->f_mapping, pgoff + pgidx); + if (IS_ERR(folio)) + return PTR_ERR(folio); - ubuf->pages[*pgbuf] = page; + ubuf->folios[*pgbuf] = folio; (*pgbuf)++; } return 0; } +static int check_memfd_seals(struct file *memfd) +{ + int seals; + + if (!memfd) + return -EBADFD; + + if (!shmem_file(memfd) && !is_file_hugepages(memfd)) + return -EBADFD; + + seals = memfd_fcntl(memfd, F_GET_SEALS, 0); + if (seals == -EINVAL) + return -EBADFD; + + if ((seals & SEALS_WANTED) != SEALS_WANTED || + (seals & SEALS_DENIED) != 0) + return -EINVAL; + + return 0; +} + +static int export_udmabuf(struct udmabuf *ubuf, + struct miscdevice *device, + u32 flags) +{ + DEFINE_DMA_BUF_EXPORT_INFO(exp_info); + struct dma_buf *buf; + + ubuf->device = device; + exp_info.ops = &udmabuf_ops; + exp_info.size = ubuf->pagecount << PAGE_SHIFT; + exp_info.priv = ubuf; + exp_info.flags = O_RDWR; + + buf = dma_buf_export(&exp_info); + if (IS_ERR(buf)) + return PTR_ERR(buf); + + return dma_buf_fd(buf, flags); +} + static long udmabuf_create(struct miscdevice *device, struct udmabuf_create_list *head, struct udmabuf_create_item *list) { - DEFINE_DMA_BUF_EXPORT_INFO(exp_info); + pgoff_t pgcnt, pgbuf = 0, pglimit; struct file *memfd = NULL; struct udmabuf *ubuf; - struct dma_buf *buf; - pgoff_t pgcnt, pgbuf = 0, pglimit; - int seals, ret = -EINVAL; + int ret = -EINVAL; u32 i, flags; ubuf = kzalloc(sizeof(*ubuf), GFP_KERNEL); @@ -299,9 +345,9 @@ static long udmabuf_create(struct miscdevice *device, if (!ubuf->pagecount) goto err; - ubuf->pages = kmalloc_array(ubuf->pagecount, sizeof(*ubuf->pages), + ubuf->folios = kmalloc_array(ubuf->pagecount, sizeof(*ubuf->folios), GFP_KERNEL); - if (!ubuf->pages) { + if (!ubuf->folios) { ret = -ENOMEM; goto err; } @@ -314,18 +360,9 @@ static long udmabuf_create(struct miscdevice *device, pgbuf = 0; for (i = 0; i < head->count; i++) { - ret = -EBADFD; memfd = fget(list[i].memfd); - if (!memfd) - goto err; - if (!shmem_file(memfd) && !is_file_hugepages(memfd)) - goto err; - seals = memfd_fcntl(memfd, F_GET_SEALS, 0); - if (seals == -EINVAL) - goto err; - ret = -EINVAL; - if ((seals & SEALS_WANTED) != SEALS_WANTED || - (seals & SEALS_DENIED) != 0) + ret = check_memfd_seals(memfd); + if (ret < 0) goto err; pgcnt = list[i].size >> PAGE_SHIFT; @@ -341,33 +378,22 @@ static long udmabuf_create(struct miscdevice *device, goto err; fput(memfd); - memfd = NULL; } - exp_info.ops = &udmabuf_ops; - exp_info.size = ubuf->pagecount << PAGE_SHIFT; - exp_info.priv = ubuf; - exp_info.flags = O_RDWR; - - ubuf->device = device; - buf = dma_buf_export(&exp_info); - if (IS_ERR(buf)) { - ret = PTR_ERR(buf); + flags = head->flags & UDMABUF_FLAGS_CLOEXEC ? O_CLOEXEC : 0; + ret = export_udmabuf(ubuf, device, flags); + if (ret < 0) goto err; - } - flags = 0; - if (head->flags & UDMABUF_FLAGS_CLOEXEC) - flags |= O_CLOEXEC; - return dma_buf_fd(buf, flags); + return ret; err: while (pgbuf > 0) - put_page(ubuf->pages[--pgbuf]); + folio_put(ubuf->folios[--pgbuf]); if (memfd) fput(memfd); kfree(ubuf->offsets); - kfree(ubuf->pages); + kfree(ubuf->folios); kfree(ubuf); return ret; } From patchwork Thu Jan 4 07:25:17 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vivek Kasireddy X-Patchwork-Id: 13510858 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 gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 76FC2C47073 for ; Thu, 4 Jan 2024 07:49:03 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id D1C1B10E3AF; Thu, 4 Jan 2024 07:48:54 +0000 (UTC) Received: from mgamail.intel.com (mgamail.intel.com [192.55.52.43]) by gabe.freedesktop.org (Postfix) with ESMTPS id 6E70210E3A5 for ; Thu, 4 Jan 2024 07:48:51 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1704354531; x=1735890531; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=HPqmDRTHlwBlDnRt25bFtTkxF656lUtiC3CZjHC0SUs=; b=CxDt48WWpDwoHlTHB06O7LP2wCrSDDvprgkC1lNVb/2Hv6LQbKb/+tH6 KAm9o9BqAE8JSSHoKNDQUizaC9eaMNvoMmuCi/kEdH6MAgnH1SRDdh+oR RFpgxVrQwt691lMw02qHJqKCMKhfB3fb8YXll+WZOvYPx0HazGMUgxbhq 9uWsk0AHnYe8XbyABpbmKu99y6n5Opy3HkXb+Dl5pdFLpB65nB2P7JqkE svCe0TA2ZcZ8FCgTJ3zX2AvGpq5wR4+BcUNAE/ht95pAm97hpcDUalO9V M/k/tE2G8cmmw/dh88liT16MW54O7ncuG54PweikpYxkmYYh0NI2oKp4F w==; X-IronPort-AV: E=McAfee;i="6600,9927,10942"; a="483331139" X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="483331139" Received: from fmviesa002.fm.intel.com ([10.60.135.142]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:50 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="14776739" Received: from vkasired-desk2.fm.intel.com ([10.105.128.132]) by fmviesa002-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:47 -0800 From: Vivek Kasireddy To: dri-devel@lists.freedesktop.org, linux-mm@kvack.org Subject: [PATCH v10 7/8] udmabuf: Pin the pages using memfd_pin_folios() API (v8) Date: Wed, 3 Jan 2024 23:25:17 -0800 Message-Id: <20240104072518.807950-8-vivek.kasireddy@intel.com> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20240104072518.807950-1-vivek.kasireddy@intel.com> References: <20240104072518.807950-1-vivek.kasireddy@intel.com> MIME-Version: 1.0 X-BeenThere: dri-devel@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Direct Rendering Infrastructure - Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Gerd Hoffmann , Dongwon Kim , David Hildenbrand , Daniel Vetter , Hugh Dickins , Vivek Kasireddy , Matthew Wilcox , Peter Xu , Jason Gunthorpe , Junxiao Chang , Mike Kravetz Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" Using memfd_pin_folios() will ensure that the pages are pinned correctly using FOLL_PIN. And, this also ensures that we don't accidentally break features such as memory hotunplug as it would not allow pinning pages in the movable zone. Using this new API also simplifies the code as we no longer have to deal with extracting individual pages from their mappings or handle shmem and hugetlb cases separately. v2: - Adjust to the change in signature of pin_user_pages_fd() by passing in file * instead of fd. v3: - Limit the changes in this patch only to those that are required for using pin_user_pages_fd() - Slightly improve the commit message v4: - Adjust to the change in name of the API (memfd_pin_user_pages) v5: - Adjust to the changes in memfd_pin_folios which now populates a list of folios and offsets v6: - Don't unnecessarily use folio_page() (Matthew) - Pass [start, end] and max_folios to memfd_pin_folios() - Create another temporary array to hold the folios returned by memfd_pin_folios() as we populate ubuf->folios. - Unpin the folios only once as memfd_pin_folios pins them once v7: - Use a list to track the folios that need to be unpinned v8: - Use unpin_folio() instead of unpin_user_page() to unpin folio Cc: David Hildenbrand Cc: Matthew Wilcox Cc: Daniel Vetter Cc: Mike Kravetz Cc: Hugh Dickins Cc: Peter Xu Cc: Jason Gunthorpe Cc: Gerd Hoffmann Cc: Dongwon Kim Cc: Junxiao Chang Signed-off-by: Vivek Kasireddy --- drivers/dma-buf/udmabuf.c | 153 +++++++++++++++++++------------------- 1 file changed, 78 insertions(+), 75 deletions(-) diff --git a/drivers/dma-buf/udmabuf.c b/drivers/dma-buf/udmabuf.c index a8f3af61f7f2..afa8bfd2a2a9 100644 --- a/drivers/dma-buf/udmabuf.c +++ b/drivers/dma-buf/udmabuf.c @@ -30,6 +30,12 @@ struct udmabuf { struct sg_table *sg; struct miscdevice *device; pgoff_t *offsets; + struct list_head unpin_list; +}; + +struct udmabuf_folio { + struct folio *folio; + struct list_head list; }; static vm_fault_t udmabuf_vm_fault(struct vm_fault *vmf) @@ -153,17 +159,43 @@ static void unmap_udmabuf(struct dma_buf_attachment *at, return put_sg_table(at->dev, sg, direction); } +static void unpin_all_folios(struct list_head *unpin_list) +{ + struct udmabuf_folio *ubuf_folio; + + while (!list_empty(unpin_list)) { + ubuf_folio = list_first_entry(unpin_list, + struct udmabuf_folio, list); + unpin_folio(ubuf_folio->folio); + + list_del(&ubuf_folio->list); + kfree(ubuf_folio); + } +} + +static int add_to_unpin_list(struct list_head *unpin_list, + struct folio *folio) +{ + struct udmabuf_folio *ubuf_folio; + + ubuf_folio = kzalloc(sizeof(*ubuf_folio), GFP_KERNEL); + if (!ubuf_folio) + return -ENOMEM; + + ubuf_folio->folio = folio; + list_add_tail(&ubuf_folio->list, unpin_list); + return 0; +} + static void release_udmabuf(struct dma_buf *buf) { struct udmabuf *ubuf = buf->priv; struct device *dev = ubuf->device->this_device; - pgoff_t pg; if (ubuf->sg) put_sg_table(dev, ubuf->sg, DMA_BIDIRECTIONAL); - for (pg = 0; pg < ubuf->pagecount; pg++) - folio_put(ubuf->folios[pg]); + unpin_all_folios(&ubuf->unpin_list); kfree(ubuf->offsets); kfree(ubuf->folios); kfree(ubuf); @@ -218,64 +250,6 @@ static const struct dma_buf_ops udmabuf_ops = { #define SEALS_WANTED (F_SEAL_SHRINK) #define SEALS_DENIED (F_SEAL_WRITE) -static int handle_hugetlb_pages(struct udmabuf *ubuf, struct file *memfd, - pgoff_t offset, pgoff_t pgcnt, - pgoff_t *pgbuf) -{ - struct hstate *hpstate = hstate_file(memfd); - pgoff_t mapidx = offset >> huge_page_shift(hpstate); - pgoff_t subpgoff = (offset & ~huge_page_mask(hpstate)) >> PAGE_SHIFT; - pgoff_t maxsubpgs = huge_page_size(hpstate) >> PAGE_SHIFT; - struct folio *folio = NULL; - pgoff_t pgidx; - - mapidx <<= huge_page_order(hpstate); - for (pgidx = 0; pgidx < pgcnt; pgidx++) { - if (!folio) { - folio = __filemap_get_folio(memfd->f_mapping, - mapidx, - FGP_ACCESSED, 0); - if (IS_ERR(folio)) - return PTR_ERR(folio); - } - - folio_get(folio); - ubuf->folios[*pgbuf] = folio; - ubuf->offsets[*pgbuf] = subpgoff << PAGE_SHIFT; - (*pgbuf)++; - if (++subpgoff == maxsubpgs) { - folio_put(folio); - folio = NULL; - subpgoff = 0; - mapidx += pages_per_huge_page(hpstate); - } - } - - if (folio) - folio_put(folio); - - return 0; -} - -static int handle_shmem_pages(struct udmabuf *ubuf, struct file *memfd, - pgoff_t offset, pgoff_t pgcnt, - pgoff_t *pgbuf) -{ - pgoff_t pgidx, pgoff = offset >> PAGE_SHIFT; - struct folio *folio = NULL; - - for (pgidx = 0; pgidx < pgcnt; pgidx++) { - folio = shmem_read_folio(memfd->f_mapping, pgoff + pgidx); - if (IS_ERR(folio)) - return PTR_ERR(folio); - - ubuf->folios[*pgbuf] = folio; - (*pgbuf)++; - } - - return 0; -} - static int check_memfd_seals(struct file *memfd) { int seals; @@ -321,16 +295,19 @@ static long udmabuf_create(struct miscdevice *device, struct udmabuf_create_list *head, struct udmabuf_create_item *list) { - pgoff_t pgcnt, pgbuf = 0, pglimit; + pgoff_t pgoff, pgcnt, pglimit, pgbuf = 0; + long nr_folios, ret = -EINVAL; struct file *memfd = NULL; + struct folio **folios; struct udmabuf *ubuf; - int ret = -EINVAL; - u32 i, flags; + u32 i, j, k, flags; + loff_t end; ubuf = kzalloc(sizeof(*ubuf), GFP_KERNEL); if (!ubuf) return -ENOMEM; + INIT_LIST_HEAD(&ubuf->unpin_list); pglimit = (size_limit_mb * 1024 * 1024) >> PAGE_SHIFT; for (i = 0; i < head->count; i++) { if (!IS_ALIGNED(list[i].offset, PAGE_SIZE)) @@ -366,17 +343,44 @@ static long udmabuf_create(struct miscdevice *device, goto err; pgcnt = list[i].size >> PAGE_SHIFT; - if (is_file_hugepages(memfd)) - ret = handle_hugetlb_pages(ubuf, memfd, - list[i].offset, - pgcnt, &pgbuf); - else - ret = handle_shmem_pages(ubuf, memfd, - list[i].offset, - pgcnt, &pgbuf); - if (ret < 0) + folios = kmalloc_array(pgcnt, sizeof(*folios), GFP_KERNEL); + if (!folios) { + ret = -ENOMEM; goto err; + } + end = list[i].offset + (pgcnt << PAGE_SHIFT) - 1; + ret = memfd_pin_folios(memfd, list[i].offset, end, + folios, pgcnt, &pgoff); + if (ret < 0) { + kfree(folios); + goto err; + } + + nr_folios = ret; + pgoff >>= PAGE_SHIFT; + for (j = 0, k = 0; j < pgcnt; j++) { + ubuf->folios[pgbuf] = folios[k]; + ubuf->offsets[pgbuf] = pgoff << PAGE_SHIFT; + + if (j == 0 || ubuf->folios[pgbuf-1] != folios[k]) { + ret = add_to_unpin_list(&ubuf->unpin_list, + folios[k]); + if (ret < 0) { + kfree(folios); + goto err; + } + } + + pgbuf++; + if (++pgoff == folio_nr_pages(folios[k])) { + pgoff = 0; + if (++k == nr_folios) + break; + } + } + + kfree(folios); fput(memfd); } @@ -388,10 +392,9 @@ static long udmabuf_create(struct miscdevice *device, return ret; err: - while (pgbuf > 0) - folio_put(ubuf->folios[--pgbuf]); if (memfd) fput(memfd); + unpin_all_folios(&ubuf->unpin_list); kfree(ubuf->offsets); kfree(ubuf->folios); kfree(ubuf); From patchwork Thu Jan 4 07:25:18 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vivek Kasireddy X-Patchwork-Id: 13510861 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 gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 56118C4707B for ; Thu, 4 Jan 2024 07:49:08 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id D60AA10E3C5; Thu, 4 Jan 2024 07:48:56 +0000 (UTC) Received: from mgamail.intel.com (mgamail.intel.com [192.55.52.43]) by gabe.freedesktop.org (Postfix) with ESMTPS id AB8D010E3A6 for ; Thu, 4 Jan 2024 07:48:51 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1704354531; x=1735890531; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=r39C9fURv/tYjjTgfRvVfOiacLkGBXJF65fG2x4l+b0=; b=nHQAaXj3gCoEecB1hdRjBdWgAK8SQXKiglTKPUFPYE5NiNwE+Z2EYUm5 vtk/YixS+Rd5Enhp1kKUMDgDXWNMQMsIwP5cvhR/44iqkJ4IWxmkWsq4x 4xSndc/26GBrsOulw+Ti+hX/NzXZWXw10dIo9I2+tMchfNliIuiIfLsOa Z0NvDDEAvW91xHuylzajkHV3eLKfaFjcIuIz6HQ6N6bF/W4s1yQRMkHvq VAVG7Mj12u5ioGryD6h7G3s98K+Cd/El+AlGUNUl7cq+DGLxaOkLz1Sj1 JvXs/mdcjD1fKlmFa0WVrE+hdjaRn8mzyDPhcqJ9iahxYa6ck5B4DWNLx A==; X-IronPort-AV: E=McAfee;i="6600,9927,10942"; a="483331149" X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="483331149" Received: from fmviesa002.fm.intel.com ([10.60.135.142]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:50 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.04,330,1695711600"; d="scan'208";a="14776743" Received: from vkasired-desk2.fm.intel.com ([10.105.128.132]) by fmviesa002-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Jan 2024 23:48:47 -0800 From: Vivek Kasireddy To: dri-devel@lists.freedesktop.org, linux-mm@kvack.org Subject: [PATCH v10 8/8] selftests/dma-buf/udmabuf: Add tests to verify data after page migration Date: Wed, 3 Jan 2024 23:25:18 -0800 Message-Id: <20240104072518.807950-9-vivek.kasireddy@intel.com> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20240104072518.807950-1-vivek.kasireddy@intel.com> References: <20240104072518.807950-1-vivek.kasireddy@intel.com> MIME-Version: 1.0 X-BeenThere: dri-devel@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Direct Rendering Infrastructure - Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Dongwon Kim , David Hildenbrand , Daniel Vetter , Hugh Dickins , Vivek Kasireddy , Peter Xu , Gerd Hoffmann , Jason Gunthorpe , Junxiao Chang , Shuah Khan , Mike Kravetz Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" Since the memfd pages associated with a udmabuf may be migrated as part of udmabuf create, we need to verify the data coherency after successful migration. The new tests added in this patch try to do just that using 4k sized pages and also 2 MB sized huge pages for the memfd. Successful completion of the tests would mean that there is no disconnect between the memfd pages and the ones associated with a udmabuf. And, these tests can also be augmented in the future to test newer udmabuf features (such as handling memfd hole punch). Cc: Shuah Khan Cc: David Hildenbrand Cc: Daniel Vetter Cc: Mike Kravetz Cc: Hugh Dickins Cc: Peter Xu Cc: Jason Gunthorpe Cc: Gerd Hoffmann Cc: Dongwon Kim Cc: Junxiao Chang Based-on-patch-by: Mike Kravetz Signed-off-by: Vivek Kasireddy --- .../selftests/drivers/dma-buf/udmabuf.c | 151 +++++++++++++++++- 1 file changed, 147 insertions(+), 4 deletions(-) diff --git a/tools/testing/selftests/drivers/dma-buf/udmabuf.c b/tools/testing/selftests/drivers/dma-buf/udmabuf.c index c812080e304e..d76c813fe652 100644 --- a/tools/testing/selftests/drivers/dma-buf/udmabuf.c +++ b/tools/testing/selftests/drivers/dma-buf/udmabuf.c @@ -9,26 +9,132 @@ #include #include #include +#include #include #include +#include #include #include #define TEST_PREFIX "drivers/dma-buf/udmabuf" #define NUM_PAGES 4 +#define NUM_ENTRIES 4 +#define MEMFD_SIZE 1024 /* in pages */ -static int memfd_create(const char *name, unsigned int flags) +static unsigned int page_size; + +static int create_memfd_with_seals(off64_t size, bool hpage) +{ + int memfd, ret; + unsigned int flags = MFD_ALLOW_SEALING; + + if (hpage) + flags |= MFD_HUGETLB; + + memfd = memfd_create("udmabuf-test", flags); + if (memfd < 0) { + printf("%s: [skip,no-memfd]\n", TEST_PREFIX); + exit(77); + } + + ret = fcntl(memfd, F_ADD_SEALS, F_SEAL_SHRINK); + if (ret < 0) { + printf("%s: [skip,fcntl-add-seals]\n", TEST_PREFIX); + exit(77); + } + + ret = ftruncate(memfd, size); + if (ret == -1) { + printf("%s: [FAIL,memfd-truncate]\n", TEST_PREFIX); + exit(1); + } + + return memfd; +} + +static int create_udmabuf_list(int devfd, int memfd, off64_t memfd_size) +{ + struct udmabuf_create_list *list; + int ubuf_fd, i; + + list = malloc(sizeof(struct udmabuf_create_list) + + sizeof(struct udmabuf_create_item) * NUM_ENTRIES); + if (!list) { + printf("%s: [FAIL, udmabuf-malloc]\n", TEST_PREFIX); + exit(1); + } + + for (i = 0; i < NUM_ENTRIES; i++) { + list->list[i].memfd = memfd; + list->list[i].offset = i * (memfd_size / NUM_ENTRIES); + list->list[i].size = getpagesize() * NUM_PAGES; + } + + list->count = NUM_ENTRIES; + list->flags = UDMABUF_FLAGS_CLOEXEC; + ubuf_fd = ioctl(devfd, UDMABUF_CREATE_LIST, list); + free(list); + if (ubuf_fd < 0) { + printf("%s: [FAIL, udmabuf-create]\n", TEST_PREFIX); + exit(1); + } + + return ubuf_fd; +} + +static void write_to_memfd(void *addr, off64_t size, char chr) +{ + int i; + + for (i = 0; i < size / page_size; i++) { + *((char *)addr + (i * page_size)) = chr; + } +} + +static void *mmap_fd(int fd, off64_t size) { - return syscall(__NR_memfd_create, name, flags); + void *addr; + + addr = mmap(NULL, size, PROT_READ|PROT_WRITE, MAP_SHARED, fd, 0); + if (addr == MAP_FAILED) { + printf("%s: ubuf_fd mmap fail\n", TEST_PREFIX); + exit(1); + } + + return addr; +} + +static int compare_chunks(void *addr1, void *addr2, off64_t memfd_size) +{ + off64_t off; + int i = 0, j, k = 0, ret = 0; + char char1, char2; + + while (i < NUM_ENTRIES) { + off = i * (memfd_size / NUM_ENTRIES); + for (j = 0; j < NUM_PAGES; j++, k++) { + char1 = *((char *)addr1 + off + (j * getpagesize())); + char2 = *((char *)addr2 + (k * getpagesize())); + if (char1 != char2) { + ret = -1; + goto err; + } + } + i++; + } +err: + munmap(addr1, memfd_size); + munmap(addr2, NUM_ENTRIES * NUM_PAGES * getpagesize()); + return ret; } int main(int argc, char *argv[]) { struct udmabuf_create create; int devfd, memfd, buf, ret; - off_t size; - void *mem; + off64_t size; + void *addr1, *addr2; devfd = open("/dev/udmabuf", O_RDWR); if (devfd < 0) { @@ -90,6 +196,9 @@ int main(int argc, char *argv[]) } /* should work */ + page_size = getpagesize(); + addr1 = mmap_fd(memfd, size); + write_to_memfd(addr1, size, 'a'); create.memfd = memfd; create.offset = 0; create.size = size; @@ -98,6 +207,40 @@ int main(int argc, char *argv[]) printf("%s: [FAIL,test-4]\n", TEST_PREFIX); exit(1); } + munmap(addr1, size); + close(buf); + close(memfd); + + /* should work (migration of 4k size pages)*/ + size = MEMFD_SIZE * page_size; + memfd = create_memfd_with_seals(size, false); + addr1 = mmap_fd(memfd, size); + write_to_memfd(addr1, size, 'a'); + buf = create_udmabuf_list(devfd, memfd, size); + addr2 = mmap_fd(buf, NUM_PAGES * NUM_ENTRIES * getpagesize()); + write_to_memfd(addr1, size, 'b'); + ret = compare_chunks(addr1, addr2, size); + if (ret < 0) { + printf("%s: [FAIL,test-5]\n", TEST_PREFIX); + exit(1); + } + close(buf); + close(memfd); + + /* should work (migration of 2MB size huge pages)*/ + page_size = getpagesize() * 512; /* 2 MB */ + size = MEMFD_SIZE * page_size; + memfd = create_memfd_with_seals(size, true); + addr1 = mmap_fd(memfd, size); + write_to_memfd(addr1, size, 'a'); + buf = create_udmabuf_list(devfd, memfd, size); + addr2 = mmap_fd(buf, NUM_PAGES * NUM_ENTRIES * getpagesize()); + write_to_memfd(addr1, size, 'b'); + ret = compare_chunks(addr1, addr2, size); + if (ret < 0) { + printf("%s: [FAIL,test-6]\n", TEST_PREFIX); + exit(1); + } fprintf(stderr, "%s: ok\n", TEST_PREFIX); close(buf);