linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Eric Biggers <ebiggers@google.com>,
	Mike Snitzer <snitzer@redhat.com>
Subject: [PATCH 3.18 08/67] dm bufio: fix integer overflow when limiting maximum cache size
Date: Tue, 28 Nov 2017 11:19:07 +0100	[thread overview]
Message-ID: <20171128100421.374129485@linuxfoundation.org> (raw)
In-Reply-To: <20171128100420.274075224@linuxfoundation.org>

3.18-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Eric Biggers <ebiggers@google.com>

commit 74d4108d9e681dbbe4a2940ed8fdff1f6868184c upstream.

The default max_cache_size_bytes for dm-bufio is meant to be the lesser
of 25% of the size of the vmalloc area and 2% of the size of lowmem.
However, on 32-bit systems the intermediate result in the expression

    (VMALLOC_END - VMALLOC_START) * DM_BUFIO_VMALLOC_PERCENT / 100

overflows, causing the wrong result to be computed.  For example, on a
32-bit system where the vmalloc area is 520093696 bytes, the result is
1174405 rather than the expected 130023424, which makes the maximum
cache size much too small (far less than 2% of lowmem).  This causes
severe performance problems for dm-verity users on affected systems.

Fix this by using mult_frac() to correctly multiply by a percentage.  Do
this for all places in dm-bufio that multiply by a percentage.  Also
replace (VMALLOC_END - VMALLOC_START) with VMALLOC_TOTAL, which contrary
to the comment is now defined in include/linux/vmalloc.h.

Depends-on: 9993bc635 ("sched/x86: Fix overflow in cyc2ns_offset")
Fixes: 95d402f057f2 ("dm: add bufio")
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 drivers/md/dm-bufio.c |   15 ++++++---------
 1 file changed, 6 insertions(+), 9 deletions(-)

--- a/drivers/md/dm-bufio.c
+++ b/drivers/md/dm-bufio.c
@@ -876,7 +876,8 @@ static void __get_memory_limit(struct dm
 		buffers = c->minimum_buffers;
 
 	*limit_buffers = buffers;
-	*threshold_buffers = buffers * DM_BUFIO_WRITEBACK_PERCENT / 100;
+	*threshold_buffers = mult_frac(buffers,
+				       DM_BUFIO_WRITEBACK_PERCENT, 100);
 }
 
 /*
@@ -1764,19 +1765,15 @@ static int __init dm_bufio_init(void)
 	memset(&dm_bufio_caches, 0, sizeof dm_bufio_caches);
 	memset(&dm_bufio_cache_names, 0, sizeof dm_bufio_cache_names);
 
-	mem = (__u64)((totalram_pages - totalhigh_pages) *
-		      DM_BUFIO_MEMORY_PERCENT / 100) << PAGE_SHIFT;
+	mem = (__u64)mult_frac(totalram_pages - totalhigh_pages,
+			       DM_BUFIO_MEMORY_PERCENT, 100) << PAGE_SHIFT;
 
 	if (mem > ULONG_MAX)
 		mem = ULONG_MAX;
 
 #ifdef CONFIG_MMU
-	/*
-	 * Get the size of vmalloc space the same way as VMALLOC_TOTAL
-	 * in fs/proc/internal.h
-	 */
-	if (mem > (VMALLOC_END - VMALLOC_START) * DM_BUFIO_VMALLOC_PERCENT / 100)
-		mem = (VMALLOC_END - VMALLOC_START) * DM_BUFIO_VMALLOC_PERCENT / 100;
+	if (mem > mult_frac(VMALLOC_TOTAL, DM_BUFIO_VMALLOC_PERCENT, 100))
+		mem = mult_frac(VMALLOC_TOTAL, DM_BUFIO_VMALLOC_PERCENT, 100);
 #endif
 
 	dm_bufio_default_cache_size = mem;

  parent reply	other threads:[~2017-11-28 10:20 UTC|newest]

Thread overview: 69+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-28 10:18 [PATCH 3.18 00/67] 3.18.85-stable review Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 01/67] s390/disassembler: increase show_code buffer size Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 02/67] ipv6: only call ip6_route_dev_notify() once for NETDEV_UNREGISTER Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 03/67] sched: Make resched_cpu() unconditional Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 04/67] lib/mpi: call cond_resched() from mpi_powm() loop Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 05/67] x86/decoder: Add new TEST instruction pattern Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 06/67] ARM: 8721/1: mm: dump: check hardware RO bit for LPAE Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 07/67] ALSA: hda: Add Raven PCI ID Greg Kroah-Hartman
2017-11-28 10:19 ` Greg Kroah-Hartman [this message]
2017-11-28 10:19 ` [PATCH 3.18 09/67] dm: fix race between dm_get_from_kobject() and __dm_destroy() Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 10/67] MIPS: Fix an n32 core file generation regset support regression Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 12/67] autofs: dont fail mount for transient error Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 13/67] nilfs2: fix race condition that causes file system corruption Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 14/67] eCryptfs: use after free in ecryptfs_release_messaging() Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 15/67] bcache: check ca->alloc_thread initialized before wake up it Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 16/67] isofs: fix timestamps beyond 2027 Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 17/67] NFS: Fix typo in nomigration mount option Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 18/67] nfs: Fix ugly referral attributes Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 19/67] nfsd: deal with revoked delegations appropriately Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 20/67] ext4: fix interaction between i_size, fallocate, and delalloc after a crash Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 21/67] ALSA: usb-audio: Add sanity checks to FE parser Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 22/67] ALSA: usb-audio: Fix potential out-of-bound access at parsing SU Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 23/67] ALSA: usb-audio: Add sanity checks in v2 clock parsers Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 24/67] ALSA: timer: Remove kernel warning at compat ioctl error paths Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 25/67] fs/9p: Compare qid.path in v9fs_test_inode Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 26/67] iscsi-target: Fix non-immediate TMR reference leak Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 27/67] KVM: nVMX: set IDTR and GDTR limits when loading L1 host state Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 29/67] clk: ti: dra7-atl-clock: Fix of_node reference counting Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 30/67] clk: ti: dra7-atl-clock: fix child-node lookups Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 31/67] IB/srpt: Do not accept invalid initiator port names Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 32/67] NFC: fix device-allocation error return Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 33/67] time: Always make sure wall_to_monotonic isnt positive Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 34/67] i40e: Use smp_rmb rather than read_barrier_depends Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 35/67] igb: " Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 36/67] igbvf: " Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 37/67] ixgbevf: " Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 38/67] i40evf: " Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 39/67] fm10k: " Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 40/67] ixgbe: Fix skb list corruption on Power systems Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 41/67] parisc: Fix validity check of pointer size argument in new CAS implementation Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 42/67] powerpc/signal: Properly handle return value from uprobe_deny_signal() Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 43/67] media: Dont do DMA on stack for firmware upload in the AS102 driver Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 44/67] media: rc: check for integer overflow Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 45/67] media: v4l2-ctrl: Fix flags field on Control events Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 46/67] net/9p: Switch to wait_event_killable() Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 47/67] mtd: nand: Fix writing mtdoops to nand flash Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 48/67] USB: fix buffer overflows with parsing CDC headers Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 49/67] iio: iio-trig-periodic-rtc: Free trigger resource correctly Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 50/67] e1000e: Fix error path in link detection Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 51/67] e1000e: Fix return value test Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 52/67] e1000e: Separate signaling for link check/link up Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 53/67] RDS: RDMA: return appropriate error on rdma map failures Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 54/67] PCI: Apply _HPX settings only to relevant devices Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 55/67] net: 3com: typhoon: typhoon_init_one: make return values more specific Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 56/67] net: 3com: typhoon: typhoon_init_one: fix incorrect return values Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 57/67] drm/armada: Fix compile fail Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 58/67] ALSA: hda - Apply ALC269_FIXUP_NO_SHUTUP on HDA_FIXUP_ACT_PROBE Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 59/67] mac80211: Remove invalid flag operations in mesh TSF synchronization Greg Kroah-Hartman
2017-11-28 10:19 ` [PATCH 3.18 60/67] mac80211: Suppress NEW_PEER_CANDIDATE event if no room Greg Kroah-Hartman
2017-11-28 10:20 ` [PATCH 3.18 61/67] staging: iio: cdc: fix improper return value Greg Kroah-Hartman
2017-11-28 10:20 ` [PATCH 3.18 62/67] netfilter: nft_queue: use raw_smp_processor_id() Greg Kroah-Hartman
2017-11-28 10:20 ` [PATCH 3.18 63/67] netfilter: nf_tables: fix oob access Greg Kroah-Hartman
2017-11-28 10:20 ` [PATCH 3.18 64/67] btrfs: return the actual error value from from btrfs_uuid_tree_iterate Greg Kroah-Hartman
2017-11-28 10:20 ` [PATCH 3.18 65/67] ASoC: wm_adsp: Dont overrun firmware file buffer when reading region data Greg Kroah-Hartman
2017-11-28 10:20 ` [PATCH 3.18 66/67] s390/kbuild: enable modversions for symbols exported from asm Greg Kroah-Hartman
2017-11-28 10:20 ` [PATCH 3.18 67/67] xen: xenbus driver must not accept invalid transaction ids Greg Kroah-Hartman
2017-11-28 19:37 ` [PATCH 3.18 00/67] 3.18.85-stable review Shuah Khan
2017-11-29  6:33   ` Greg Kroah-Hartman
2017-11-28 21:50 ` Guenter Roeck

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=20171128100421.374129485@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=ebiggers@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=snitzer@redhat.com \
    --cc=stable@vger.kernel.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: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).