From: Jan Kara <jack@suse.cz> To: Dan Williams <dan.j.williams@intel.com> Cc: Jan Kara <jack@suse.cz>, "Darrick J. Wong" <darrick.wong@oracle.com>, linux-nvdimm@lists.01.org, linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org Subject: Re: [PATCH 3/7] dax: Check the end of the block-device capacity with dax_direct_access() Date: Wed, 13 Feb 2019 11:22:02 +0100 [thread overview] Message-ID: <20190213102202.GA13313@quack2.suse.cz> (raw) In-Reply-To: <155000669646.348031.16690970886357498896.stgit@dwillia2-desk3.amr.corp.intel.com> On Tue 12-02-19 13:24:56, Dan Williams wrote: > The checks in __bdev_dax_supported() helped mitigate a potential data > corruption bug in the pmem driver's handling of section alignment > padding. Strengthen the checks, including checking the end of the range, > to validate the dev_pagemap, Xarray entries, and sector-to-pfn > translation established for pmem namespaces. > > Cc: Jan Kara <jack@suse.cz> > Cc: "Darrick J. Wong" <darrick.wong@oracle.com> > Signed-off-by: Dan Williams <dan.j.williams@intel.com> > --- > drivers/dax/super.c | 39 +++++++++++++++++++++++++++++---------- > 1 file changed, 29 insertions(+), 10 deletions(-) > > diff --git a/drivers/dax/super.c b/drivers/dax/super.c > index 6e928f37d084..a27395cfcec6 100644 > --- a/drivers/dax/super.c > +++ b/drivers/dax/super.c > @@ -86,12 +86,14 @@ bool __bdev_dax_supported(struct block_device *bdev, int blocksize) > { > struct dax_device *dax_dev; > bool dax_enabled = false; > + pgoff_t pgoff, pgoff_end; > struct request_queue *q; > - pgoff_t pgoff; > - int err, id; > - pfn_t pfn; > - long len; > char buf[BDEVNAME_SIZE]; > + void *kaddr, *end_kaddr; > + pfn_t pfn, end_pfn; > + sector_t last_page; > + long len, len2; > + int err, id; > > if (blocksize != PAGE_SIZE) { > pr_debug("%s: error: unsupported blocksize for dax\n", > @@ -113,6 +115,15 @@ bool __bdev_dax_supported(struct block_device *bdev, int blocksize) > return false; > } > > + last_page = ALIGN_DOWN(part_nr_sects_read(bdev->bd_part) > + - PAGE_SIZE / 512, PAGE_SIZE / 512); Why not just (i_size_read(bdev->bd_inode) - 1) >> PAGE_SHIFT? Otherwise the patch looks good to me. Honza > + err = bdev_dax_pgoff(bdev, last_page, PAGE_SIZE, &pgoff_end); > + if (err) { > + pr_debug("%s: error: unaligned partition for dax\n", > + bdevname(bdev, buf)); > + return false; > + } > + > dax_dev = dax_get_by_host(bdev->bd_disk->disk_name); > if (!dax_dev) { > pr_debug("%s: error: device does not support dax\n", > @@ -121,14 +132,15 @@ bool __bdev_dax_supported(struct block_device *bdev, int blocksize) > } > > id = dax_read_lock(); > - len = dax_direct_access(dax_dev, pgoff, 1, NULL, &pfn); > + len = dax_direct_access(dax_dev, pgoff, 1, &kaddr, &pfn); > + len2 = dax_direct_access(dax_dev, pgoff_end, 1, &end_kaddr, &end_pfn); > dax_read_unlock(id); > > put_dax(dax_dev); > > - if (len < 1) { > + if (len < 1 || len2 < 1) { > pr_debug("%s: error: dax access failed (%ld)\n", > - bdevname(bdev, buf), len); > + bdevname(bdev, buf), len < 1 ? len : len2); > return false; > } > > @@ -143,13 +155,20 @@ bool __bdev_dax_supported(struct block_device *bdev, int blocksize) > */ > WARN_ON(IS_ENABLED(CONFIG_ARCH_HAS_PMEM_API)); > dax_enabled = true; > - } else if (pfn_t_devmap(pfn)) { > - struct dev_pagemap *pgmap; > + } else if (pfn_t_devmap(pfn) && pfn_t_devmap(end_pfn)) { > + struct dev_pagemap *pgmap, *end_pgmap; > > pgmap = get_dev_pagemap(pfn_t_to_pfn(pfn), NULL); > - if (pgmap && pgmap->type == MEMORY_DEVICE_FS_DAX) > + end_pgmap = get_dev_pagemap(pfn_t_to_pfn(end_pfn), NULL); > + if (pgmap && pgmap == end_pgmap && pgmap->type == MEMORY_DEVICE_FS_DAX > + && pfn_t_to_page(pfn)->pgmap == pgmap > + && pfn_t_to_page(end_pfn)->pgmap == pgmap > + && pfn_t_to_pfn(pfn) == PHYS_PFN(__pa(kaddr)) > + && pfn_t_to_pfn(end_pfn) == PHYS_PFN(__pa(end_kaddr))) > dax_enabled = true; > put_dev_pagemap(pgmap); > + put_dev_pagemap(end_pgmap); > + > } > > if (!dax_enabled) { > -- Jan Kara <jack@suse.com> SUSE Labs, CR _______________________________________________ Linux-nvdimm mailing list Linux-nvdimm@lists.01.org https://lists.01.org/mailman/listinfo/linux-nvdimm
WARNING: multiple messages have this Message-ID (diff)
From: Jan Kara <jack@suse.cz> To: Dan Williams <dan.j.williams@intel.com> Cc: linux-nvdimm@lists.01.org, Jan Kara <jack@suse.cz>, "Darrick J. Wong" <darrick.wong@oracle.com>, linux-kernel@vger.kernel.org, vishal.l.verma@intel.com, linux-fsdevel@vger.kernel.org Subject: Re: [PATCH 3/7] dax: Check the end of the block-device capacity with dax_direct_access() Date: Wed, 13 Feb 2019 11:22:02 +0100 [thread overview] Message-ID: <20190213102202.GA13313@quack2.suse.cz> (raw) In-Reply-To: <155000669646.348031.16690970886357498896.stgit@dwillia2-desk3.amr.corp.intel.com> On Tue 12-02-19 13:24:56, Dan Williams wrote: > The checks in __bdev_dax_supported() helped mitigate a potential data > corruption bug in the pmem driver's handling of section alignment > padding. Strengthen the checks, including checking the end of the range, > to validate the dev_pagemap, Xarray entries, and sector-to-pfn > translation established for pmem namespaces. > > Cc: Jan Kara <jack@suse.cz> > Cc: "Darrick J. Wong" <darrick.wong@oracle.com> > Signed-off-by: Dan Williams <dan.j.williams@intel.com> > --- > drivers/dax/super.c | 39 +++++++++++++++++++++++++++++---------- > 1 file changed, 29 insertions(+), 10 deletions(-) > > diff --git a/drivers/dax/super.c b/drivers/dax/super.c > index 6e928f37d084..a27395cfcec6 100644 > --- a/drivers/dax/super.c > +++ b/drivers/dax/super.c > @@ -86,12 +86,14 @@ bool __bdev_dax_supported(struct block_device *bdev, int blocksize) > { > struct dax_device *dax_dev; > bool dax_enabled = false; > + pgoff_t pgoff, pgoff_end; > struct request_queue *q; > - pgoff_t pgoff; > - int err, id; > - pfn_t pfn; > - long len; > char buf[BDEVNAME_SIZE]; > + void *kaddr, *end_kaddr; > + pfn_t pfn, end_pfn; > + sector_t last_page; > + long len, len2; > + int err, id; > > if (blocksize != PAGE_SIZE) { > pr_debug("%s: error: unsupported blocksize for dax\n", > @@ -113,6 +115,15 @@ bool __bdev_dax_supported(struct block_device *bdev, int blocksize) > return false; > } > > + last_page = ALIGN_DOWN(part_nr_sects_read(bdev->bd_part) > + - PAGE_SIZE / 512, PAGE_SIZE / 512); Why not just (i_size_read(bdev->bd_inode) - 1) >> PAGE_SHIFT? Otherwise the patch looks good to me. Honza > + err = bdev_dax_pgoff(bdev, last_page, PAGE_SIZE, &pgoff_end); > + if (err) { > + pr_debug("%s: error: unaligned partition for dax\n", > + bdevname(bdev, buf)); > + return false; > + } > + > dax_dev = dax_get_by_host(bdev->bd_disk->disk_name); > if (!dax_dev) { > pr_debug("%s: error: device does not support dax\n", > @@ -121,14 +132,15 @@ bool __bdev_dax_supported(struct block_device *bdev, int blocksize) > } > > id = dax_read_lock(); > - len = dax_direct_access(dax_dev, pgoff, 1, NULL, &pfn); > + len = dax_direct_access(dax_dev, pgoff, 1, &kaddr, &pfn); > + len2 = dax_direct_access(dax_dev, pgoff_end, 1, &end_kaddr, &end_pfn); > dax_read_unlock(id); > > put_dax(dax_dev); > > - if (len < 1) { > + if (len < 1 || len2 < 1) { > pr_debug("%s: error: dax access failed (%ld)\n", > - bdevname(bdev, buf), len); > + bdevname(bdev, buf), len < 1 ? len : len2); > return false; > } > > @@ -143,13 +155,20 @@ bool __bdev_dax_supported(struct block_device *bdev, int blocksize) > */ > WARN_ON(IS_ENABLED(CONFIG_ARCH_HAS_PMEM_API)); > dax_enabled = true; > - } else if (pfn_t_devmap(pfn)) { > - struct dev_pagemap *pgmap; > + } else if (pfn_t_devmap(pfn) && pfn_t_devmap(end_pfn)) { > + struct dev_pagemap *pgmap, *end_pgmap; > > pgmap = get_dev_pagemap(pfn_t_to_pfn(pfn), NULL); > - if (pgmap && pgmap->type == MEMORY_DEVICE_FS_DAX) > + end_pgmap = get_dev_pagemap(pfn_t_to_pfn(end_pfn), NULL); > + if (pgmap && pgmap == end_pgmap && pgmap->type == MEMORY_DEVICE_FS_DAX > + && pfn_t_to_page(pfn)->pgmap == pgmap > + && pfn_t_to_page(end_pfn)->pgmap == pgmap > + && pfn_t_to_pfn(pfn) == PHYS_PFN(__pa(kaddr)) > + && pfn_t_to_pfn(end_pfn) == PHYS_PFN(__pa(end_kaddr))) > dax_enabled = true; > put_dev_pagemap(pgmap); > + put_dev_pagemap(end_pgmap); > + > } > > if (!dax_enabled) { > -- Jan Kara <jack@suse.com> SUSE Labs, CR
next prev parent reply other threads:[~2019-02-13 10:22 UTC|newest] Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top 2019-02-12 21:24 [PATCH 0/7] libnvdimm/pfn: Fix section-alignment padding Dan Williams 2019-02-12 21:24 ` Dan Williams 2019-02-12 21:24 ` [PATCH 1/7] libnvdimm/pfn: Account for PAGE_SIZE > info-block-size in nd_pfn_init() Dan Williams 2019-02-12 21:24 ` Dan Williams 2019-02-12 21:24 ` [PATCH 2/7] libnvdimm/pmem: Honor force_raw for legacy pmem regions Dan Williams 2019-02-12 21:24 ` Dan Williams [not found] ` <155000669130.348031.13764885263577554365.stgit-p8uTFz9XbKj2zm6wflaqv1nYeNYlB/vhral2JQCrhuEAvxtiuMwx3w@public.gmane.org> 2019-02-15 15:48 ` Sasha Levin 2019-02-18 21:14 ` Sasha Levin 2019-02-12 21:24 ` [PATCH 3/7] dax: Check the end of the block-device capacity with dax_direct_access() Dan Williams 2019-02-12 21:24 ` Dan Williams 2019-02-13 10:22 ` Jan Kara [this message] 2019-02-13 10:22 ` Jan Kara 2019-02-13 16:49 ` Dan Williams 2019-02-13 16:49 ` Dan Williams 2019-02-21 5:15 ` [PATCH v2] " Dan Williams 2019-02-21 5:15 ` Dan Williams 2019-02-12 21:25 ` [PATCH 4/7] libnvdimm/pfn: Introduce super-block minimum version requirements Dan Williams 2019-02-12 21:25 ` Dan Williams 2019-02-12 21:25 ` [PATCH 5/7] libnvdimm/pfn: Remove dax_label_reserve Dan Williams 2019-02-12 21:25 ` Dan Williams 2019-02-12 21:25 ` [PATCH 6/7] libnvdimm/pfn: Introduce 'struct pfn_map_info' Dan Williams 2019-02-12 21:25 ` Dan Williams 2019-02-12 21:25 ` [PATCH 7/7] libnvdimm/pfn: Fix 'start_pad' implementation Dan Williams 2019-02-12 21:25 ` Dan Williams 2019-02-21 23:47 ` Jeff Moyer 2019-02-21 23:47 ` Jeff Moyer 2019-02-22 3:58 ` Dan Williams 2019-02-22 3:58 ` Dan Williams 2019-02-22 3:58 ` Dan Williams 2019-02-22 15:42 ` Jeff Moyer 2019-02-22 15:42 ` Jeff Moyer 2019-02-22 17:12 ` Dan Williams 2019-02-22 17:12 ` Dan Williams 2019-02-22 17:12 ` Dan Williams 2019-02-22 17:21 ` Jeff Moyer 2019-02-22 17:21 ` Jeff Moyer 2019-02-22 17:21 ` Jeff Moyer 2019-02-20 17:11 ` [PATCH 0/7] libnvdimm/pfn: Fix section-alignment padding Dan Williams 2019-02-20 17:11 ` Dan Williams 2019-02-20 17:19 ` Johannes Thumshirn 2019-02-20 17:19 ` Johannes Thumshirn 2019-02-20 17:45 ` Jeff Moyer 2019-02-20 17:45 ` Jeff Moyer
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20190213102202.GA13313@quack2.suse.cz \ --to=jack@suse.cz \ --cc=dan.j.williams@intel.com \ --cc=darrick.wong@oracle.com \ --cc=linux-fsdevel@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-nvdimm@lists.01.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.